Environmental and human determinates of vegetation distribution  in the Hadhramaut region,  Republic of Yemen

Abdul Wali  AlKhulaidi

The principal objectives of the research are to analyse the distribution and dynamics of plants of the Hadhramaut region and to evaluate the role of the physical parameters and human action on their distribution, survival and conservation. The study area is located in Hadhramaut Governorate and lies in the eastern part of the Republic of Yemen. This is a remote and inaccessible region; however, there has been rapid development in recent years with the discovery of oil, which has had a significant effect on the vegetation and landscape. The Hadhramaut region represents an important area of eastern Yemen, linking eastern and western phyto-geographical units, representing a key transition zone between northeast Africa and Southeast Asia. Previous studies in the study area have only dealt with individual species and there has been no complete botanical survey. Recent floristic studies are turning up new species with many endemic and near endemic plant species. The Hadhramaut region is a desert region, dissected by deep valleys where agriculture is possible and the main towns are surrounded by rocky, dry limestone plateaus. The northern section passes into the deserts of the Rub ‘al Khali or Empty Quarter. Hadhramaut has a long history of human occupation with ancient civilisations well reflected in the archaeological records. Archaeological sites suggest that agriculture, with a related development of irrigation technology, was more widespread during a period when rainfall was more abundant. Initially, a reconnaissance survey of the whole Hadhramaut Governorate was undertaken, leading to the selection the Wadi Hadhramaut for detailed study. Within this study area, three sites were selected for intensive survey. These sites were considered representative of the major landforms and vegetation of the area and reflect the principal patterns of land use. The three sites represent tracts of land that were either unaffected, undergoing change or already altered as a result of oil-related development. Transects were designed to cross each site, from the valley bottoms to the plateau surfaces, passing across the representative landforms and vegetation. Surveys were made of the vegetation associations, their structure and biodiversity, as well as their relationship with environment and human impact. Two preliminary transects were made across the entire region, from the southern coast to the plateau in the northwest and from east to west, in order to place the study area in a regional context. The research is the first detailed vegetation survey in the Hadhramaut region and has revealed relevant data that can be used for further studies in similar habitats or for further management and conservation activities. In the study area, major vegetation associations, their composition and biodiversity were identified and in addition, vegetation and land use maps were generated including local endemic, near-endemic and rare plant species. About 469 plant species have been identified from the Hadhramaut region. There are 107 taxa which are endemic and near-endemic; 68 of these are endemic to Yemen, of these 41 are confined to Hadhramaut region. A total of 134 species belonging to 42 families (about 30% of flora of Hadhramaut region) were recorded in the study area and, of these, seven species are endemic to Yemen (four of them endemic to Hadhramaut region). The study revealed 15 vegetation associations and thirty sociological species groups. The main wadis are covered by desert alluvial shrubland comprising Fagonia indica, Tephrosia apollinea, Cymbopogon schoenanthus, Boerhavia elegans and Dichanthium insculptum with scattered trees of Acacia campoptila. In contrast, much of the fertile lands of the main wadis, such as the bottom of the rocky slopes, are intensively cultivated with palm trees and other annual crops, notably sorghum and wheat. The rocky slopes facing the main wadis and the plateau surface are covered by stony and gravelly desert vegetation dominated by herbaceous plants, namely Stipagrostis hirtigluma, Farsetia linearis, Aristida triticoides, Fagonia paulayana, Boerhavia elegans and Dichanthium insculptum. Within the plateau there are some sloping sites and secondary wadis which support dense vegetation. The vegetation here comprises shrubland or grassland dominated by Jatropha spinosa with Zygophyllum decumbens, Commiphora foliacea, Commiphora kua, Maerua crassifolia. Dichanthium insculptum, Stipagrostis hirtigluma and Farsetia linearis. The research in the Hadhramaut region has revealed the importance of this region in terms of plant biodiversity, and particularly of endemic, rare and near-endemic species, which urgently require further management and conservation activities.

Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi Environmental and human determinates of vegetation distribution in the Hadhramaut region, Republic of Yemen by Abdul Wali Ahmed Al Khulaidi A thesis submitted for the degree of Doctor of Philosophy School of GeoSciences University of Edinburgh September 2006 Abstract The principal objectives of the research are to analyse the distribution and dynamics of plants of the Hadhramaut region and to evaluate the role of the physical parameters and human action on their distribution, survival and conservation. The study area is located in Hadhramaut Governorate and lies in the eastern part of the Republic of Yemen. This is a remote and inaccessible region; however, there has been rapid development in recent years with the discovery of oil, which has had a significant effect on the vegetation and landscape. The Hadhramaut region represents an important area of eastern Yemen, linking eastern and western phyto-geographical units, representing a key transition zone between northeast Africa and Southeast Asia. Previous studies in the study area have only dealt with individual species and there has been no complete botanical survey. Recent floristic studies are turning up new species with many endemic and near endemic plant species. The Hadhramaut region is a desert region, dissected by deep valleys where agriculture is possible and the main towns are surrounded by rocky, dry limestone plateaus. The northern section passes into the deserts of the Rub ‘al Khali or Empty Quarter. Hadhramaut has a long history of human occupation with ancient civilisations well reflected in the archaeological records. Archaeological sites suggest that agriculture, with a related development of irrigation technology, was more widespread during a period when rainfall was more abundant. I Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi Initially, a reconnaissance survey of the whole Hadhramaut Governorate was undertaken, leading to the selection the Wadi Hadhramaut for detailed study. Within this study area, three sites were selected for intensive survey. These sites were considered representative of the major landforms and vegetation of the area and reflect the principal patterns of land use. The three sites represent tracts of land that were either unaffected, undergoing change or already altered as a result of oil-related development. Transects were designed to cross each site, from the valley bottoms to the plateau surfaces, passing across the representative landforms and vegetation. Surveys were made of the vegetation associations, their structure and biodiversity, as well as their relationship with environment and human impact. Two preliminary transects were made across the entire region, from the southern coast to the plateau in the northwest and from east to west, in order to place the study area in a regional context. The research is the first detailed vegetation survey in the Hadhramaut region and has revealed relevant data that can be used for further studies in similar habitats or for further management and conservation activities. In the study area, major vegetation associations, their composition and biodiversity were identified and in addition, vegetation and land use maps were generated including local endemic, near-endemic and rare plant species. About 469 plant species have been identified from the Hadhramaut region. There are 107 taxa which are endemic and near-endemic; 68 of these are endemic to Yemen, of these 41 are confined to Hadhramaut region. A total of 134 species belonging to 42 families (about 30% of flora of Hadhramaut region) were recorded in the study area and, of these, seven species are endemic to Yemen (four of them endemic to Hadhramaut region). The study revealed 15 vegetation associations and thirty sociological species groups. The main wadis are covered by desert alluvial shrubland comprising Fagonia indica, Tephrosia apollinea, Cymbopogon schoenanthus, Boerhavia elegans and Dichanthium insculptum with scattered trees of Acacia campoptila. In contrast, much of the fertile lands of the main wadis, such as the bottom of the rocky slopes, are intensively cultivated with palm trees and other annual crops, notably sorghum and wheat. The rocky slopes facing the main wadis and the plateau surface are covered by stony and gravelly desert vegetation dominated by herbaceous plants, namely Stipagrostis hirtigluma, Farsetia linearis, Aristida triticoides, Fagonia paulayana, Boerhavia elegans and Dichanthium insculptum. Within the plateau there are some sloping sites and secondary wadis which support dense vegetation. The vegetation here comprises shrubland or grassland dominated by Jatropha spinosa with Zygophyllum decumbens, Commiphora foliacea, Commiphora kua, Maerua crassifolia. Dichanthium insculptum, Stipagrostis hirtigluma and Farsetia linearis. The research in the Hadhramaut region has revealed the importance of this region in terms of plant biodiversity, and particularly of endemic, rare and near-endemic species, which urgently require further management and conservation activities. II Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi Acknowledgements .First, I would like to express my gratitude to my suppervisors Professor Peter Furley and Tony Miller for their continuous support and great help in finalising the study. Especial thanks to Sabina Knees (Flora of Arabia and SW Asia expert), Christopher Ellis (Lichen Ecology expert) and Alan Sneath (Computer expert) from the Royal Botanic Garden Edinburgh (RBGE) and Dr. Colin Legg and Chris Place from the University of Edinburgh for their great help. Especial thanks also to Eng. Mohamed Al-Suneidar the head of Sustainable Environmental Management Program (YEM/97/100), Sub-Program II, Sanaa and to the staff of British Council, Sanaa, especially Mrs Elizabeth White and Ms Raya Almu Ayyad. During my work I have encountered many obstacles and have been provided with support and assistance from different institutions and companies. Personal, special thanks to DOVE Energy Oil Company, in particular the manager, Mr Stewart Ahmed and Ahmed Al Shami, who supported me during my field work with transport and accommodation. Especial thanks also to Mr Muhammed Atran, Eng. Fuad Abbad and Eng. Ahmed Bataher for their support during the field work. I would like to express special thanks to the staff of the Agriculture Research and Extension Authority (AREA) for their great support, in particular the chairman of AREA Dr. Esmaeil Muharram, Dr. Fadhl Mutlak who died recently in a tragedy accident, Mr Ahmed Rezq, Mohammed Hizam Al Mashriqi, and Dr. Muhammed Qaed al Sabiri the head of AREA Regional Station, Taiz. Last, but not least, I thank my family, in particular, my wife who has stood beside me throughout my research and who has spent long periods looking after our children alone. Abdul Wali Ahmed Al Khulaidi III Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi Glossary Hima: An Arabic term for protected areas, it is exclusively for the conservation of natural resources, wildlife and forests. Mahajer: A local Arabic name for protected areas that are located on mountain slopes or on land adjacent or around cultivated fields. These protected areas are applied for a certain period, mostly in rainy season and mainly used for grazing. Sabakha or Sabaka: an Arabic term for a marsh or bare mud covered by salt crust sue to evaporation (Murray, 1951; Scholte et al., 1991) Abiotic: non-living components that affect the living organisms such as soil, pH, water, light, temperature and topography (Rojas et al., 2001; Deyn et al., 2004). Biodiversity: The variety of all living things; the different plants, animals and micro organisms, the genetic information they contain and the ecosystems they form. Biotic: living components that affect the living organisms such as plants, animals and human activities (Deyn et al., 2004). CITES: Convention on International Trade in Endangered Species of Wild Flora and Fauna, it is the main tool for controlling international trade in wildlife. CITES species is the species that grouped according to how threatened they are by international trade. Cliff: nearly vertical rocky slopes facing the main wadis. Diversity: Diversity is a measure of the number of species within a unit area and has been used by different plant ecologists to evaluate plant species in different ecosystems of the world Dwarf shrubland: A group of small shrubs usually less 0.5m tall, forming more than 10% cover. Trees and shrubs generally less than 10% cover). Edaphic factor: Any physical or chemical property of the soil that influences plants growing in it (Rajakaruna, 2004) by controlling the floristic and the structure of the vegetation (Beadle, 1953). Endangered: A species present in such small numbers that it is at high risk of extinction in the near future (Hunter, 2002). Endemic species: Species exclusively native to a place or a species that is unique to that place or region, found naturally nowhere else. Escarpment: steep to moderately steep slope mountain or cliff, facing mainly south. IV Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi Ex-situ conservation: maintaining organisms outside of their natural habitats (Hunter, 2002). Evenness: How equally abundant are each of the species in a community. Extinct plant: The last remaining member of the plant species had died, or is presumed to have died beyond reasonable doubt (Hunter, 2002). Frequency: Frequency is defined as the number of times a plant species is present within a given number of sample quadrats of uniform size placed repeatedly across a stand of vegetation (Mueller-Dombois et al. 1974; Daubenmire, 1968). Graasland: A group of herbaceous (herb, grass), forming more than 10% cover. Trees, shrubs, and dwarf-shrubs less than 10% cover. Hamada or hammada: is Arabic for ‘stone desert’ and is applied to surfaces covered by stone fragments and pebbles. Hamada consists mostly of fine particles and soil produced by weathering (Zohary, 1962; Batanouny, 1981) Importance Value: Importance values refer to how important a species is in terms of the structure of a community or species composition (Nautiyal et al., 2003 Kareef: Depression for rainwater catchments systems Number of individuals: Number of individuals refers to the density of each species that been recorded in the sample sites during the vegetation survey. Plant association: A group of plant species or a plant community that occupies the landscape Plateau: A relatively flat upland surface, often rocky in Yemen. Rock outcrop: Big bare rock found on the surface of rocky slope mountains. Secondary wadi: It is the inter-plateau wadi or a “Canyon”, with very steep to moderately steep rocky slopes. Shrubland: A group or individual of Shrubs generally greater then 0.5m tall, forming more than 10% cover, trees generally less than 10% cover. Sociological plant species: Group of plant species that show a similar distribution across the sample plots or group of plant species (taxa) that are more or less similar in behaviour (Zonneveld, 1986) Species accumulation curves: This is a method for determining whether the sampling is sufficient to have collected all the plant species from a region or for comparing species richness between different communities (Moreno and Halffer, 2000; Willott, 2001). V Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi Species richness: Refers to the number of species present in a community (Krebs, 1999). Total Vegetation %: It is a totals of the trees, shrubs, dwarf shrubs and herbaceous. Vulnerable plant: Plant faces a high risk of extinction in the medium-term Wadi: An Arabic term for a valley or streambed in that remains dry except during the rainy season. Woodland: A group or individual trees usually less than 7 m tall, forming 15-40% cover. The shrub cover is less than 10%. List of Acronyms AREA CBD DNO DOVE EPA FAO GEF GSPC ITCZ IUCN OECD Agricultural Research and Extension Authority Convention on Biological Diversity Det Norske Oljeselskap AS Dove Energy Limited Environmental Protection Authority Food and Agriculture Organization of the United Nations Global Environment Facility Global Strategy for Plant Conservation Inter-Tropical Convergence Zone Interntional Union for Conservation of Nature and natural resources Organization for Economic Co-operation and Development RBGE Royal Botanic Garden of Edinburgh RSCZ UNDP UNEP UNESCO USDA WWF Red Sea Convergence zone effect United Nations Development Program United Nations Environment Program United Nations Educational, Scientific and Cultural Organization United States Department of Agriculture World Wildlife Fund VI Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi Abstract ........................................................................................................................... I Acknowledgements ..................................................................................................... III Glossary ....................................................................................................................... IV List of Acronyms ......................................................................................................... VI List of appendices ........................................................................................................ XI List of tables ................................................................................................................ XI List of figures..............................................................................................................XII List of plates .............................................................................................................. XV Chapter 1: Introduction ............................................................. 1 1.1. Introductory comments 1.2. Research aims 1.3. Context 1.1.3. Hypotheses 1.1.4. Research questions 1.1.5. Structure of the thesis 1.4. Characterization of the Wadi Hadhramaut study area 1.4.1. Introduction 1.4.2. Landscapes and topography The Plateaus The wadis 1.5. The study sites 1.5. 1. Main wadi 1.5.2. Slopes facing the main wadi and its tributaries: 1.5.3. Cliffs: 1.5.4. Plateaus: 1.6. Climate 1.6.1. Introduction 1.6.2. Hadhramaut Rainfall Evaporation Temperature Wind speed 1.7. Geology 1.7.1. Introduction 1.7.2. Quaternary units (wadi deposits) 1.7.3. Tertiary units (Hadhramaut group) 1.7.4. Cretaceous units (Tawilah group) 1.7.5. Sand dunes 1.8. Land use 1.8.1. Agriculture 1.8.2. Animal husbandry system in the study areas Nomadic animal husbandry Sedentary animal husbandry 1.9. Environmental, land unit characteristic and the problems for plant growth 1.9.1. Introduction 1 2 2 3 3 4 5 5 7 7 7 9 9 9 9 9 12 12 14 14 18 21 23 24 24 24 24 25 25 26 26 32 34 35 36 36 VII Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi 1.9.2. The Physical impact of Infrastructural activities 1.9.3. Floods 1.9.4. Grazing 1.9.5. Plant collection 1.10. Soil 1.11. Floristics 1.11.1. The geographical distribution of the vegetation in the study area 1.11.2. The vegetation of the plateau and the Wadi zones 1.11.3. The vegetation of the desert Zone 1.12. Summary 37 37 39 40 41 43 44 45 47 47 Chapter 2. Assessment of previous work on vegetation in the Yemen. ....................................................................................... 50 2.1. Introduction 2.2. Botanical Exploration 2.3. Vegetation studies in Hadhramaut 2.4. Botanical Literature 2.4.1. The Vegetation of the Republic of Yemen (Western Part) 2.4.2. Flora of Republic of Yemen 2.4.3. Flora of the Arabian Peninsula and Socotra Volume 1 2.4.4. Vegetation of the Arabian Peninsula 2.4.5. A Handbook of the Yemen Flora 2.5. Paleoenvironmental studies 2.6. Vegetation maps 2.6.1. Drawing up a preliminary vegetation map of Arabia 2.6.2. Middle East Vegetation Map, scale: 1: 800 0000. 2.6.3. Geobotanical Outline map of the Middle East 2.7. Summary 50 50 55 56 56 58 59 59 60 62 66 67 68 68 70 Chapter 3: The vegetation of the Hadhramaut ........................ 71 3.1 Introduction 3.2. Methodology 3.3. Climate 3.4. Topography 3.5. Results 3.5.1. Vegetation composition of the 2 transects 3.5.2. Environmental data analysis 3.5.3. Threatened plant species in the region. 3.5.4. Comparison with previous vegetation studies 3.5.4. a. Comparison with previous vegetation studies in Yemen 3.5.4.b. Comparison with the vegetation of the region and East Africa 3.6. Conservation and management of plant species and biodiversity 3.6.1. Introduction 3.6.2. Important sites of Hadhramaut region 3.7. Currently existing protected areas in Yemen 3.8.1. National policy on protected areas 3.8.2. Future Activities 3.9. Conclusions 71 71 73 74 75 75 83 87 92 92 94 94 94 95 98 100 100 101 VIII Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi Chapter 4. Floristic analysis and plant biodiversity ............ 106 4. 1. Introduction 106 4.2- Methodology 107 4.3. Results 111 4.3.1. Flora 111 4.3.2. Number of individuals per hectare 119 4.3.3. Frequency and Importance values: 122 4.3.5. Floristic diversity 131 4.3.6. Species richness and density over the landform gradient 138 4.3.7. Species accumulation curves 141 4.3.8. Similarity between the three study sites and between the different landforms 144 4.4. Summary 147 Chapter 5. Vegetation associations: structure and environment analysis.................................................................................... 151 5.1. Introduction 151 5.2. Methodology 153 5.2.1. Transect approach 153 5.2.2. Braun-Blanquet approach 156 5.2.3. Data collection 156 5.2.4. Methods of soil Analysis 157 5.2.5. Vegetation analysis 158 5.2.6. Vegetation structure 163 5.3. Results 165 5.3.1. Vegetation associations and Socio-Ecological species groups 165 Salsola imbricata 172 5.4. The vegetation of the different land forms 182 5.4.1. The vegetation of the main wadis (w) 182 5.4.2. The vegetation of the mountain slopes facing the main wadis (ms) 183 5.4.3. The vegetation of the plateaus (pla) 184 5.4.4. The vegetation in narrow secondary wadis of the plateaus (wp) 186 5.4.5. The vegetation of slopes in plateaus (sp). 186 5.5. Main vegetation associations 188 5.5.1. Alhagi graecorum - Phoenix dactylifera association 188 5.5.2- Citrullus colocynthis - Zygophyllum album association 190 5.5.3. Convolvulus glomertaus - Merremia hadramautica - Crotalaria persica association 191 5.5.4 -Tephrosia dura - Tephrosia apollinea- Fagonia indica association 192 5.5.5- Acacia ehrenbergiana - Rhazya stricta - Acacia campoptila association 193 5.5.6- Salsola imbricata - Blepharis edulis association 194 5.5.7- Tephrosia nubica - Acacia campoptila - Merremia hadramautica association 196 5.5.8. Forsskaolea tenacissima - Boerhavia elegans association 197 5.5.9. Fagonia paulayana - Boerhavia elegans - Cleome scaposa association 197 5.5.10. Indigofera spinosa - Stipagrostis hirtigluma association 199 5.5.11. Seetzenia lanata - Dichanthium insculptum association 200 5.5.12. Chrysopogon aucheri- Stipagrostis hirtigluma association 202 IX Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi 5.5.14- Cornulaca amblyacantha - Stipagrostis hirtigluma association 5.5.15. Zygophyllum decumbens - Dichanthium insculptum association 5.6. Summary of the vegetation associations. 5.7. Similarity between the main vegetation associations 5.8. The distribution of vegetation associations throughout the study sites 5.9. Environmental data analysis 5.9.1. The sampling plots 5.9.2. Plant species 5.9.3. Main vegetation associations 5.9.4. Vegetation structure and cover percentage 5.10. Summary of main findings 206 207 208 213 215 218 218 220 225 226 233 Chapter 6. Distribution and mapping of the vegetation and land cover. ....................................................................................... 239 6.1. Introduction 6.2. Methods of study: 6.3. Result 6.3.1- Relief and drainage 6.3.2. The vegetation associations of the study sites 6.3.3. Agricultural fields 6.3.4- Plant species distribution within the study sites 6.3.5. The distribution of endemic, near-endemic and rare plant species in the Hadhramaut region 6.8. Summary 239 240 242 242 246 254 256 277 287 Chapter 7. Discussion and conclusion ................................. 290 7.1. Introduction 7.2. Floristic and phytogeography 7.2.1. Floristic 7.2.2. Phytogeography 7.3. The effect of biotic and abiotic factors on plant species 7.3. 1. Climate 7.3.2. Topography 7.3.4. Human activities 7.3.5. Soil 7.4. Conservation and management of biodiversity 7.4.1- Conservation: traditional knowledge and land ownership issues 7.4.2- Types of Threats 7.4.3. National and International Framework for Protection of Yemen’s Flora and vegetation 7.5. Conclusions 290 291 291 292 307 307 314 326 342 345 346 347 350 356 X Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi List of appendices page Appendix 1: rainfall data of Wadi Hadhramaut (Sayun) Appendix 2: rainfall data from 28 stations provided by the Wadi Hadhramaut Agriculture Project Appendix 3. Daily rainfall (mm) between 1981 and 2002 from Sayun Appendix 4: List of plant species from Hadhramaut region Appendix 5: Name of local people contributed to the study Appendix 6. Progressive doubling of quadrat size for minimal area of species-area curves for the different landforms 381 List of tables Table 1.1. Table 1.2. Table 1.3. Table 1.4. Table 1.5. Table 1.6. Table 1.7. Table 1.8 Table 2.1. Table 3.1. Table 3.2. Table 4.1. Table 4.2. Table 4.3. Table 4.4. Table 4.5a. Table 4.5b. Table 4.6. Table 4.7. Table 4.8. Table 4.9. Table 5.1. Table 5.2. Table 5.3. Table 5.4a. Table 5.4b. Table 5.4c. Table 5.5. Table 5.6. Table 5.7. 382 383 386 396 397 page 14 Main Agro-climatic Zones of Hadhramaut (AREA, 1999). UNESCO (1979) climatic classification based on ratio between average annual precipitation (P) and annual reference evaporation (Eo). 16 Monthly temperature data between 1981 and 2002 at Sayun weather station. 21 Agriculture Area (Hectare) and number of holdings 28 The total cultivated area and percentage cover in W. Hadhramaut in hectares for the period 1977 –1978. 28 Cropping colander for rain fed fields in sites 2 and 3. 29 Main soil classification types. 43 The vegetation types across a hypothetical transect across Wadi Adim and adjacent plateau. 45 The climate history of Africa, Asia and Arabica during the last 150,000 64 The sample plots with their environmental characteristic and vegetation structure. 85 Provisional list of Hadhramaut Endemic and near Endemic plant species with regional and global assessment 88 List of species in alphabetical order of families for the 3 study sites 113 Number of individuals per hectare for the most dominant species over the different landforms 119 Number of individuals per hectare (density) of each species in the three study sites. 120 The frequency per hectare for each plant species 124 The most high frequency species over the 3 study sites 125 Most important plant species in the study areas with their total importance values. 125 Importance Value Index for each plant species in different landforms 127 Simpson's and Shannon's indices of the 3 study sites. 132 Jaccard distance Index between the 3 study areas. 144 Most important plant species with their important value 148 ITC modified Braun - Blanquet Scale cover/density. 157 An initial matrix of plant species and sample plots. 159 A final matrix of floristic vegetation associations. 160 Number of sample plots for each landform in the study sites. 165 Number of bare sample plots for each landform 165 The GPS coordinates in degree, minute, second format of the sample plots. 165 The classification matrix according to TWINSPAN. 169 Description of the main Socio-Ecological groups of plant species. 171 The vegetation matrix. 175 XI Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi Table 5.8. Laboratory analysis results of the soil collected. Table 5.9. Average vegetation cover for each sample plots in the main wadis. Table 5.10. Average vegetation cover for each sample plots in the mountain slopes facing the main wadis of the three study sites. Table 5.11. Average vegetation cover for each sample plots in the plateaus of the three study sites. Table 5.12. Average vegetation cover for each sampling plots of the secondary wadis in the plateaus of the three study sites. Table 5.13. Average vegetation cover for each sample site of the slopes in the plateaus of the three study sites. Table 5.14. Similarity matrix between the main floristic associations. Table 5.15.The distribution of the floristic associations throughout the study sites. Table 5.16.The distribution of the vegetation structure classes over the study sites. Table 5.17.Species covering more than 1% per 100 m² for each landform unit. Table 5.18.The relationship between vegetation association and structure. Table 5.19.Main floristic vegetation associations with characteristic soil, strata and landform data. Table 6.1. The distribution patterns of the vegetation associations across the study sites. Table 6.2. The area in hectare of each association and other land form. Table 6.3. The area in hectare of each association and other land form of site 2. Table 6.4. The area in hectare of each association and other land form of site 3. Table 6.5. The distribution of the plant species along the elevation gradient. Table 7.1. Phytogeographical classification for the Middle East. 178 178 List of figures page 6 8 8 11 15 17 17 18 19 19 Figure 1.1 The location of the study area and of the 3 study sites. Figure 1.2 The altitude of Wadi Hadhramaut and surrounding plateaus, Figure 1.3 Wadi Hadhramaut with its tributaries wadis. Figure 1.4 Maps illustrating the main topographical units of the study sites Figure 1.5 The location of the weather stations of Wadi Hadhramaut and its tributaries. Figure 1.6 Annual rainfall of Sayun weather station. Figure 1.7 A climate diagram of mean monthly rainfall Figure 1.8 Average monthly evaporation for Sayun. Figure 1.9 Average annual evaporation and precipitation between 1981 and 2000 (Sayun). Figure 1.10 monthly evapotranspiration (PET) in relation to monthly rainfall for 2 stations in Hadhramaut region Figure 1.11 Average annual rainfall of different locations along Wadi Hadhramaut Figure 1.12 Average monthly temperatures data between 1981 and 2002 at Sayun weather station. Figure 1.13 Average monthly relative humidity at Sayun weather station. Figure 1.14 Wind speed of 6 dry ecological zones in Yemen Figure 1.15 Geology cross-section for W. Hadhramaut and surrounding plateaus Figure 1.16 The irrigated fields along the Wadi Hadhramaut from site 1 (west Shibam) to the point where Wadi Adim falls into the Wadi Hadhramaut Figure 1.17 Comparison of the changes in land use between 1973 and 2004. Figure 1.18 Agro-Pastoral systems on Irrigated fields. Figure 1.19 Agro-Pastoral system on spate irrigation (floods) fields sites 2 & 3 Figure 1.20 The movement of the nomads from surrounding areas and coastal plain XII 184 185 186 187 214 215 227 230 237 238 246 250 250 250 257 292 20 21 22 23 26 30 31 33 33 35 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi Figure 3.1 An image showing the transects from the coastal region to the plateau in the northern part of the Hadhramaut region and from site 1 towards west. Figure 3.2 A climate diagram of mean monthly rainfall and mean annual temperature for Sayun (Wadi Hadhramaut, north) and Al Rayan (coastal area, south). Figure 3.3 A map showing the locations of the sites along the first transect. Figure 3.4 A map showing the locations of the sites along the second transect. Figure 3.5 Ordination of the location sites along the 2 transects and environment yielded by CCA Figure 3.6 The vegetation cover percentage along the altitude gradient of both transects Figure 3.7. The change of number of species (richness) along the altitude gradient of both transects Figure 3.8. The vegetation association and climate of 6 different ecological areas of Yemen Figure 3.9. The important areas for conservation in Hadhramaut region Figure 3.10 Examples of distributions of characteristic regional plant from reconnaissance vegetation survey Figure 4.1 Diagram showing the structure of chapter 4 Figure 4.2 Number of individuals per hectare for the most dominant species across the study sites. Figure 4.3 The mean frequency in each landform for the three most abundant species. Figure 4.4 Importance value curves for the three study sites Figure 4.5 The distribution of total importance values in each landform in the study sites Figure 4.6 The distribution patterns of number of individuals between the plant species in the three study sites. Figure 4.7 Cumulative abundance curves, known also as k-dominance curves for the study sites Figure 4.8 Diversity of the different landforms using Shannon index Figure 4.9 Cumulative abundance curves, known also as k-dominance curves for the different landforms in the study sites Figure 4.10 Number of species per hectare over the landform gradient Figure 4.11 Species density per hectare along the landform gradient Figure 4.12 Curve of species versus number of sample sites for site 1 Figure 4.13 Curve of species versus number of sample sites for site 2 Figure 4.14 Curve of species versus number of sample sites for site 3 Figure 4.15 Curve of species versus number of sample sites for the three study sites Figure 4.16 Cluster analysis based on the Jaccard distance analysis of the 3 study sites Figure 4.17 Cluster analysis based on the Jaccard Analysis for the different landforms. Figure 4.18 Diversity, evenness and number of species of the different land forms Figure 5.1 Diagram showing the structure of chapter 5 Figure 5.2a General 3D view, showing the different landscape over the plateau and the main wadi Figure 5.2b Idealized representations transect illustrating the main landform units in the study sites Figure 5.2c The length of the 3 transects across the study sites. Figure 5.3 Example of CCA ordination diagram Figure 5.4 Three dimensional diagram representing of structure vegetation types Figure 5.5 TWINSPAN classification based on presence and absence data Figure 5.6. The result of DCA ordination analysis of numbers that represent the main groups according to a TWINSPAN classification Figure 5.7. The result of DECORANA ordination analysis of the numbers that represent the main groups according to a TWINSPAN classification Figure 5.8a Soil cross section of the study area. Figure 5.8b. The variation in soil properties over the different land forms XIII 72 74 76 79 83 86 86 93 97 103 106 121 123 126 130 132 133 133 134 139 140 142 142 143 143 144 145 146 151 154 154 155 163 164 168 170 170 180 181 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi Figure 5.9 Figure 5.10 Figure 5.11 Figure 5.12 Figure 5.13 Diagram of floristic vegetation associations across the main topographical units. Main vegetation associations of site 1 Main vegetation associations of site 2 Main vegetation associations of site 3. Similarity dendrogram produced by the Bray-Curtis similarity method using GroupAverage clustering, showing the similarity between the main floristic associations Figure 4.14. A bar diagram of floristic vegetation associations Figure 5.15 Canonical Correspondence Analysis (CCA) ordination biplot of sampling points on main landforms and environmental variables Figure 5.16 Canonical Correspondence Analysis (CCA) ordination biplot of plant species and landform environmental variables. Figure 5.17 Interpretation of the effect of erosion on the species using 2 variables (erosion and altitude). Figure 5.18a.Canonical Correspondence Analysis (CCA) ordination biplot of plant species and 4 environmental variables. Figure 5.18b Canonical Correspondence Analysis (CCA) ordination biplot of main landforms and 4 environmental variables. Figure 5.19. Ordination of main vegetation associations and environment yielded by CCA. Figure 5.20. Number of sample plots and vegetation structure over the different landforms. Figure 5.21a Canonical Correspondence Analysis (CCA) of the percentages cover of vegetation structure and different landforms. Figure 5.21b The percentage of each vegetation structure type over the different landforms. Figure 5.22. Average vegetation cover percentage per 100 m² against log. species rank. Figure 5.23. Average vegetation cover percentage per 100 m² for each landform. Figure 5.24. Vegetation cover percentage per 100 m². Figure 5.25. Numbers of species, individuals, and the total vegetation cover percentages of the main floristic associations. Figure 6.1. Vegetation and land cover map using ERDAS IMAGINE Figure 6.2. Topography features of the study site 1 Figure 6.3. Topography features of the study site 2 Figure 6.4. Topography features of the study site 3. Figure 6.5. Vegetation associations of site 1 Figure 6.6. Vegetation associations of site 2 Figure 6.7. Vegetation associations of site 3 Figure 6.8. Percentage covers of associations 11 and 12 in each land form. Figure 6.9. Percentage cover of associations 8 and 10 in the mountain slope facing the main wadi Figure 6.10. Percentage cover of associations 11, 12, 13 and 15 in the plateau area of site 2. Figure 6.11. Percentage covers of associations 11, 12, and 14 in the plateau area of site 3. Figure 6.12. Percentage covers of associations 11 and 14 in the slopes of the plateau of site 3. Figure 6.13. Maps of the cultivated fields of the 3 study sites Figure 6.14a. The distribution patterns of endemic, near-endemic and rare species in Hadhramaut region, including the 3 study sites Figure 6.14b. The distribution of 4 species of Ochradenus in Arabian Peninsula Figure 6.15. Remarkable disjunction between 2 species of Fagonia Figure 7.1a. General limits of the main phytogeographical regions of Africa and south west Asian Figure 7.1b.The Phytogeography of the Hadhramaut region XIV 208 210 211 212 214 217 219 221 222 223 224 225 227 228 228 231 232 232 236 241 243 244 245 247 248 249 252 252 253 253 254 255 277 279 287 294 296 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi Figure 7.2. Average monthly rainfall of 6 areas in the coastal and northern and eastern dry zones Figure 7.3. Satellite images showing the off-road effect. Figure 7.4. Vegetation degradation processes in the main wadis. Figure 7.5. Vegetation degradation processes on the mountain slopes facing the main wadis Figure 7.6. Vegetation degradation processes in the mountain slopes facing the main wadis. Figure 7.7. Vegetation degradation processes on the slope of the plateau. Figure 7.8. Vegetation degradation processes on the slopes of the plateau. Figure 7.9. The change in Prosopis area in the Wadi Hadhramaut from 1998 to 2001. 312 321 334 335 336 337 338 341 List of plates page 35 135 136 189 Plate 1.1. Pods of Prosopis juliflora in big sacks, Plate 4.1. Dense vegetation cover on the slope of plateau, site 2 irrigated by contaminated water Plate 4.2. bare dolomite and chert fragments surface with sand loam textured on the plateau Plate 5.1. Association 1 A date gardens. Widespread along the Wadi Hadhramaut. Plate 5.2. Association 1. In the foreground fallow land with Alhagi graecorum and Senna incana with cultivated fields with scattered date palms in the background 189 Plate 5.3. Association 2. in the foreground scattered Zygophyllum album and Prosopis farcta; next to the sample plot, in the background palm trees with Acacia ehrenbergiana 190 Plate 5.4. Association 3. Fallow land on relatively stony wadi bed, dominated by dwarf shrubland of Merremia somalensis and Crotalaria persica. 191 Plate 5.5. Association 4. Scattered species of Tephrosia apollinea. with Acacia campoptila 192 Plate 5.6. Association 5. Foreground, almost bare compacted hard soil with very poor cover of Fagonia indica and Cassia italica. In the background open woodland of Acacia campoptila with very few Ochradenus baccatus. 193 Plate 5.7. Association 6 Foreground woodland of Acacia campoptila with dwarf shrub Salsola imbricata, Indigofera spinosa and Fagonia indica. 194 Plate 5.8. Association .6. Foreground Acacia campoptila. Background herbaceous cover of Indigofera spinosa, Blepharis edulis and Stipagrostis hirtigluma. 195 Plate 5.9. Association .6. Foreground almost flat, wide, very shallow rocky, open wadi bed (Hamada) dominated by Blepharis edulis. Background single tree of Acacia campoptila. 195 Plate 5.10.Association 7. Foreground Tephrosia nubica and Merremia hadramautica. Background tr of Acacia campoptila 196 Plate 5.11.Association 8. Very eroded slope with vegetation concentrated on the drainage lines and depressions. Foreground plants are Reseda sphenocleoides; and background plants area Stipagrostis hirtigluma, Aerva javanica and Dichanthium insculptum. 197 Plate 5.12. Association 9. (sample plot 310). Very eroded rocky slope with high soil loss. The vegetation cover is less than 1%. 198 Plate 5.13. Association 9. Eroded rocky steep slope with vegetation cover less than 2% in stone pavement. 199 Plate 5.14. Association 10. Very eroded and unstable steep slopes with vegetation cover less than 2%. The vegetation is mainly found on runnels and rills. 200 Plate 5.15. Association 11. Foreground, eroded surface with Stipagrostis hirtigluma and Farsetia linearis (sample plot 124), background undisturbed surface with very poor or no vegetation cover. 201 Plate 5.16. Association 11 (sample site 237) dominated by Dichanthium insculptum and Farsetia linearis 201 XV Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi Plate 5.17. Association 12 (sample site 130). Alluvial fan with good cover on the plateau of site 1. Foreground Cymbopogon schoenanthus and Acacia oerfota. Background Commiphora foliacea. Plate 5.18 Association 12. Narrow drainage line on the plateau, with predominantly pavement rock covering 80 to 99 % of the site. Foreground Dichanthium insculptum, Stipagrostis hirtigluma and Barleria aff bispinosa. Background Acacia hamulosa. Plate 5.19. Association 12. Very wet site with dense vegetation cover reaching 86%. Foreground, Chrysopogon aucheri, Dichanthium insculptum, Farsetia linearis, Aerva javanica and others. Background trees of Acacia oerfota , Acacia mellifera and Commiphora foliacea. Plate 5.20. Association 13 (sample site 245). Dry, shallow, eroded slope with vegetation less than 5%. Background sparse cover of Jatropha spinosa with Dichanthium insculptum and Farsetia linearis Plate 5.21. Association 13 (sample site 238). Runnels on slope, reach on vegetation cover. Foreground herbs dominated by Indigofera spinosa. Background trees mainly Acacia mellifera, Maerua crassifolia, and Jatropha spinosa Plate 5.22. Association 14 (sample site 326). Very eroded rocky slopes with vegetation <4%. Background Jatropha spinosa with Stipagrostis hirtigluma. Plate 5.23. Association 15 (sample site 224), very rocky slope with sparse vegetation cover dominated by Dichanthium insculptum and Zygophyllum decumbens. Plate 7.1. Dense fog on the southern mountains of Hadhramaut region Plate 7.2. Methods of tackling water shortage. Roots of Dipterygium glaucum Plate 7.3. Vegetation growing along the rain water reservoir on the plateau. Plate 7.4. An example of microclimate change is stone ring on the plateau, Plate 7.5. The vegetation is concentrated only on gullies and runnels, where moisture is available. Plate 7.6. The surface of the secondary wadi of the plateau. The surface is covered by bare pavement rocks with soil less than 3%. Plate 7.7. Stones and rock outcrops (Shieb Sultan, site 1), under the arid condition of the study area. Plate 7.8. A desert pavement (a surface gravel deposit of tightly packed pebbles, devoid of vegetation) is found on wadi bed surfaces site 2. 7.6. Plate 7.9a. Roads created by oil working activities Plate 7.9b. Roads created by oil working activities have resulted in compaction of the soil surface Plate 7.10.A wadi bed of site 3 (plot 305). The active surface on the right and another less active, higher on the left only receive water during the most intense flooding events. Plate 7.11.Vegetation dominated by Zygophyllum album on extreme flooded wadi beds (site 1). Plate 7.12. The impact of waste water on vegetation. Plate 7.13.The impact of concentration of waste water on the vegetation cover. Plate 7.14. Prosopis juliflora has recently invaded abandoned agricultural land and flooded wadi beds in many parts of Wadi Hadhramaut Plate 7.15. A close-up view of the plateau pavement (exposed flat surface) shows that the gaps between rock fragments are small or rarely visible References 363 XVI 201 203 203 204 205 206 207 300 305 310 316 317 317 318 318 320 320 323 325 328 328 339 344 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi Chapter 1: Introduction 1.1. Introductory comments The Hadhramaut region is one of the least known regions of the Middle East yet lies on a highly significant path way of plant dispersal from the Arabian Peninsula to the Horn of Africa and from the Indo-Gangetic marchlands to the Mediterranean. It is an area that has been occupied by human settlement for millennia, but where climate has changed radically over historical time and where the current vegetation is a reflection of harsh physical environmental conditions and increasing human pressure. The impact of recent human disturbance has increased stresses on the fragile biota and makes taxonomic recognition of plant resources an urgent priority. The climate of the study areas is arid and semi-arid characterised by high temperatures and low average annual rainfall below 100 mm per year. As a result of the climate and the harsh landscape, which consists of desolate limestone plateaus, relatively fertile wadis and areas of desert with sand-dunes and gravel plains, plants have developed ingenious mechanisms to survive. Like dry savanna woodland, the vegetation in the main wadis consists of open woodland with low woody-herb cover and scattered thorny trees (e.g. Acacia campoptila, and Ziziphus leucodermis). The vegetation of the plateau consists mainly of communities dominated by grassland species with trees and shrubs on the plateaus confined to secondary wadis and steep slopes. The study area shows distinctive patterns of plant distribution relating which relate to the different vegetation associations. There are changes in vegetation associations across the area; a short drive or walk across the plateau and main wadi reveals gradational changes in the vegetation associations from one end to the another, and even from one side of a road or wadi to the other. Generally the vegetation of the plateau is sparse and dense vegetation is confined mainly to areas where moisture is available, such as depressions and seasonal drainage lines. The vegetation of the study areas has never been systematically studied or surveyed by botanists. Botanically it is the least-known area of the Arabian Peninsula, 1 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi representing perhaps the least explored region in South West Asia (WWF, 1994). Almost 40 endemic plant species are known from the Hadhramaut region (Miller, 1991), however, this figure was certainly an underestimate and this figure can be at least doubled. Earlier botanical studies concentrated on checklists or general descriptions of the vegetation of certain habitats, without following any rigorous scientific methodology. 1.2. Research aims The principal objectives of the research are to analyse the distribution and dynamics of the plants of the Hadhramaut region, and to evaluate the role of physical parameters and human actions on the vegetation. The area selected for detailed study is representative of the Hadhramaut as a whole and is further subdivided into 3 sites for intensive survey. 1.3. Context Hadhramaut has a long history of human occupation. The development of ancient civilisations is well reflected in the archaeological records. Archaeological sites suggest that agriculture, with a related development of irrigation technology, was more widespread during a period when rainfall was more abundant; this goes back perhaps as early as 7000 – 8000 B.P (McCorriston, 2000). The decline of vegetation, population, and collapse of agriculture was possibly due to reduced precipitation and evaporation from the Indian Ocean associated with the weakening of the IndianAfrican monsoon system at the end of the last ice age (Moehl, 1996). Today, oil exploration and population pressure are putting enormous pressures on the natural resources of the region. Oil exploration activities have caused major impacts to the soil surface, ground water and local ecosystems in the region. The impacts on the landforms, soil, vegetation, habitat, and cultural resources are already obvious in many parts of the study sites. Thus the status of the vegetation cover is at present precarious and evidence of changes in species distribution and abundance is highly relevant to the discussion of vegetation conservation and management. The study area is important for a number of reasons: (a) It is a key transition zone between northeast Africa and southwest Asia. 2 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi (b) It is an area that has never been systematically studied or surveyed by botanists. (c) Previous studies have concentrated only on preparing reconnaissance checklists of the flora and general descriptions of the vegetation. (d) Botanically the region is the least-known of the entire Arabian Peninsula and possibly the least explored in South West Asia (WWF, 1994). (e) A considerable number of endemic plant species are known from the area (Miller, 1991). (f) Human population and settlement in the region is low but have a distinct impact on plant life and distribution. Oil exploration has had a recent but significant impact on the vegetation. 1.1.3. Hypotheses It is intended to test the following hypotheses: 1. The number and abundance of individual plant species has varied over time, especially over the past two decades. 2. Plant communities and vegetation structure have remained relatively constant over the same period, although the populations of different plant species and their local distributions has changed. 3. Human activity has exploited the plant resources of the entire area but has preferentially targeted particular types of vegetation. 1.1.4. Research questions In order to test these hypotheses and to achieve the principal aims, several specific questions will be addressed: What are the types of vegetation, the structure of plant communities  and their distribution?  Which areas have the greatest diversity?  contemporary plant distributions?  What is the composition and diversity of plant species? What are the principal physical and anthropogenic factors affecting To what extent have the plants and plant communities changed over the past few decades and historically? 3 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi   What are the processes affecting vegetation change?  plant composition and distribution? What has been the nature of land use change and how has it affected To what extent can this information helpfully inform conservative and management users? 1.1.5. Structure of the thesis In the first chapter the characteristics of the study area and the main problems are explained. The achievements and the limitations of previous studies on the vegetation in the Hadhramaut region and Yemen are evaluated in Chapter 2. The general characteristics of the vegetation, their significance, and the endemic and near- endemic plants of Hadhramaut region are discussed with recommendations for conservation and management in Chapter 3. Chapter 4 discusses the floristic analysis and the biodiversity of the study sites. The floristic associations, vegetation structure and environmental factors of the study sites are explained in Chapter 5. In Chapter 6, the vegetation and the distribution of plant species of Hadhramaut region are presented in maps. The thesis ends with Chapter 7 which comprises discussion and conclusion. 4 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi 1.4. Characterization of the Wadi Hadhramaut study area 1.4.1. Introduction The Republic of Yemen is located at the South-Western edge of the Arabian Peninsula between 12°40 and 19° N latitude and between 42°30 and 55° E longitude . Besides the mainland, it includes many islands, the larger of these being Soqotra in the Arabian Sea and Kamaran in the Red Sea. The country is divided into 21 Governorates, with a total area of 547,000 km² (approximately 55 million hectares). Hadhramaut Governorate, where the study area is located, lies in the southeast of Yemen, between the latitudes 14-19 degrees north and the longitudes 48-51 degrees east (Figure 1.1). It is bordered by Saudi Arabia to the north, the Al Mahra Governorate to the east and the Shabwa Governorate to the west. It extends northwards deep into the Empty Quarter (Rub’ Al Khali) with its southern coasts overlooking the Arabian Sea, however the dominant feature is the Jol, a desolate limestone plateau, which dominates the centre of the governorate. Hadhramaut is the largest Governorate in the Republic of Yemen, at about 161,749 km² (about 30% of the whole area of Yemen), it reaches over 2500m above sea level and includes some areas of the Empty Quarter in the north and coastal areas in the south. The directorates are divided into two main groups: 14 directorates in Wadi Hadhramaut and surrounding desert areas, 13 directorates in the coastal plain and facing mountains and 2 directorates on the Soqotra archipelago (Central Statistical Organisation, 2001). 5 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi Figure 1.1. The location of the study area and of the 3 study sites. (I) regional context, (II) Yemen and neighbouring states (GraphicMaps.com), (III) the study area, (IV to VI) the 3 study sites in details (landsat TM of 1998). The red (false colour) and the green (true colour) on the satellite images represent the cultivated fields. 6 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi 1.4.2. Landscapes and topography The Plateaus The Oligocene-Miocene plateau (SELKHOZPROMEXPORT, 1984) is divided into the northern and southern plateaus of the Jol by one wadi (MAW, 1996), named Wadi Hadhramaut in the upper section and Wadi al Masila in the lower section. The altitude of the southern Jol ranges between 860 and 1580 m and is higher than the northern jol, which rarely exceeds 1200m (Figure 1.2). The altitude in the northern Jol, where study areas 1 and 3 are located, decreases gradually towards the Rub al Khali desert in the north-east and increases in the west towards the Ramalat Assaba’tain desert. The southern plateau increases in height towards the south. The plateau was compressed during the Tertiary into broad anticlines and synclines (DOVE, 2001) and comprises an undulating, almost flat surface, dissected by numerous deep narrow secondary wadis (drainage lines) that flow to the main Wadi Hadhramaut. All drainage lines flow north into the Wadi Hadhramaut before flowing again east to Wadi Masila and then south east to the coast (Figure 1.3). The wadis There are also many large wide wadis, the biggest of which is Wadi Hadhramaut and its many tributaries. Wadi Hadhramaut has an almost flat bottom with steep sidewalls, it varies in width, from 1.5 km in the east to 6 km in the west and lies about 300 to 400 m below the level of the plateau. The wadis are cut by unstable rocky drainage lines; the natural vegetation is mainly concentrated along the sides of these drainage lines but dense vegetation may develop in certain locations, where favourable moist conditions occur. The altitude of the bottom of Wadi Hadhramaut ranges from between 600 to 800m (Figure 1.2). The headwaters of Wadi Hadhramaut open westward into a wide desert plain called the Ramlat Assaba’tain. The Wadi Hadhramaut is considered to extend between Wadi Saar in the west to just west of Tarim in the east (SOGREAH, 1979). 7 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi Figure 1.2. The altitude of W. Hadhramaut and surrounding plateaus, including the desert of Ramlat Assaba’tain (west). The altitude ranges between 500 to 1300m asl. This altitude image was generated using the available DEM image and the softwares of ERDAS IMAGINE 8.4 and ArcView 3.2a. Figure 1.3. Wadi Hadhramaut with its tributary wadis. Wadi Hadhramaut divides the plateau region into two. The altitude of the southern plateau increases towards the south and the northern plateau decreases towards the north. Periodic surface flow. 8 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi 1.5. The study sites Generally the topography of the 3 detailed study sites is very similar and comprises the following (Figure 1.4). 1.5. 1. Main wadi This is nearly flat and is cut by deep drainage lines (wadi beds) where most of the natural vegetation is concentrated. The width of the wadi varies from three to five km in site 1, from 0.64 km to 1 km in site 2 and from 1.5 to 3 km in site 3. The terrain is generally flat but cut by drainage lines that are covered by bare stones and gravels. The slope percentage of the wadi ranges between 1 and 3% facing east, north and north-east. The altitude ranges between 673 and 733 m asl. The wadi bed cuts into the valley of the main wadi, dividing it into flood plains and terraces on either side. In some locations, this wadi bed runs close to the mountain slopes. The main wadi contains terraces, sandy plains, depressions, and some bare land. Almost all of the agricultural fields are located in the main wadi. 1.5.2. Slopes facing the main wadi and its tributaries: These are steep to moderate steep rocky slopes, with slope percentages ranging between 20% and 60%. The surface cover is very shallow and is mostly rocky (87 – 93%), sometimes covered with large bare rock outcrops (5 – 9%). 1.5.3. Cliffs: These are very steep to vertical bare rocky areas facing the main wadi. 1.5.4. Plateaus: These are undulating to almost flat rocky areas with gradients range between 1% and 3%, dissected by numerous drainage lines (secondary wadis). The Wadi Hadhramaut divides the plateaus into southern and northern sections, the distance between both plateaus is wider to the west towards the Ramlat Assaba’tain desert plain. The northern plateau slopes to the south east, towards the Wadi Hadhramaut, with altitudes ranging between 910 and 995 m asl at the west corner and between 903 9 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi and 955 m asl at the east corner. The southern plateau slopes to the north east, towards the Wadi Hadhramaut with altitudes ranging between 858 and 977 m asl. 1.5.5. Slopes on plateaus: There are steep to moderately steep rocky slopes (gradients between 20% and 50%) on the plateaus. Their surfaces are shallow and very rocky (between 95% – 97%) with large rock outcrops (between 1% and 4%). The lengths of the slopes range from 50 m to almost 400 m. 1.5.6. Secondary wadis on the plateaus: These are canyon-like valleys with undulating to almost flat rocky surfaces with gradients of 1% to 5%. Most of the surface is covered by bare rocks. These secondary wadis flow into the main Wadi Hadhramaut and its tributaries. 10 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi Figure 1.4. Maps illustrating the main topographical units of the study sites. 11 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi 1.6. Climate 1.6.1. Introduction Climate has a major influence over plant distributions throughout the area. It is known that climate has changed considerably, particularly as regards air temperature and precipitation during the Holocene (Anderson, 1997). Scientists have hypothesized that the climate of the region would have been wet during a period beginning 13 000 BP to about 9000 BP (Kenneth et al., 2001). They suggested that the present desert climate of Arabia probably began about 6000 BP and has continued to gradually dry out to the present time. Today Yemen has a predominantly semi-arid to arid climate, with rain during spring and summer. The annual rainfall rises from less than 50 mm along the Red Sea and Gulf of Aden coasts to a maximum of 500-1000 mm in the western highlands and decreases steadily to below 50 mm towards the eastern desert. Yemen's climate is affected by three large bodies of water (Ayele and Al Shadily, 2000):    the Indian Ocean (including the Gulf of Aden and the Arabian sea), the Red Sea which contributes to the spring rainy season (March-May) and the Mediterranean Sea, The presence of the Red Sea produces what it is called Red Sea Convergence zone effect (RSCZ) and contributes to the first spring rainy period between March and May. Monsoonal winds reinforce the Inter-Tropical Convergence Zone (ITCZ) effect during summer, causing a second rainy period between July and September. (Ayele and Al Shadily, 2000). Dry climates predominate in Yemen. According to Koppen's climate classification system (1936), almost the entire area of Yemen belongs to the climate class Bwh of tropical / subtropical desert (Ayele and Al Shadily, 2000). Some zones, such as the Ibb area (about 193 km south of Sanaa, the capital) have a relatively high rainfall and a steppe climate (Bsh). The Empty Quarter (Ar Rub’al Khali) represents a 12 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi low-latitude desert (Walker, 1998) or hyper-arid (AREA, 1999; Mandaville, 1986) and the coastal area represents the tropical semi-desert type. There are differences between various authors over the definitions of aridity and the boundaries of arid regions; and so these terms and boundaries have to be treated with caution (Furley, 1983). The definition of arid lands used by researchers has also varied according to the aims of the enquiry (Heathcote, 1983). Aridity is measured in terms of the ratio of rainfall to potential evapotranspiration. Aridity is usually taken as a situation in which rainfall is less than half the value of potential evapotranspiration (Findlay, 1996). According to Isani (1999) and Cook and Warren (1973), arid regions are defined as barren lands lacking enough water for vegetation to grow. Arid regions within deserts are usually identified using a combination of average precipitation, evaporation, and temperature features (Havstad, 1999), and characterised by low precipitation, which is usually associated with considerable insolation (Isani, 1999). More generally though, arid regions are defined as areas receiving less than 250 mm of annual precipitation over a long-term average (Havstad, 1999) or between 125 mm and 380 mm (Furley, 1983). Arid regions can be divided to five climatic classes (Cloudsley, 1976): 1. Tropical or subtropical deserts 2. Cool coastal deserts 3. Rain shadow deserts 4. Interior continental deserts 5. Polar deserts Some classifications divide deserts into three classes: semi-arid with annual rainfall less than 600mm, arid with annual raingall less than 200mm and hyper-arid with annual rainfall less than 25mm (and where rainfall occurs only sporadically over for long periods). The Agricultural Research and Extension Authority (AREA) divided the Hadhramaut region into three Agro-climatic Zones (AREA, 1999) as follows: Zone 1 represents Wadi Hadhramaut and the surrounding plateaus (700-1100 m asl) Zone 2 represents the coastal areas (0-200 m asl) 13 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi Zone 3 represents the Ramlat as Saba’tayn (in the western part of Wadi Hadhramaut) and the Empty Quarter (the northern part of Wadi Hadhramaut) (800-1100 m asl). Zone Annual Potential Mean Sunshine Average rainfall evaopo- Relative hours/day temperature (mm) transpiration humidity per year mm/year 1 50-125 1650-1800 20-40% 8.5-10.5 31˚C 2 10-200 1400-1600 60-80 8.5-10 31˚C 3 < 100 2000 20-3- 9-10.5 30 Table 1.1. Main Agro-climatic Zones of Hadhramaut le 1.1 (AREA, 1999) 1.6.2. Hadhramaut The Hadhramaut is an arid region characterised by low rainfall, high temperatures and high evaporation. One Arabic Omani proverb (Hawley, 1978) which reflects the cruel environment of the Hadhramaut is: “Escaping from death, he got lost in the Hadhramaut.” Rainfall Rainfall in the area occurs primarily because of convective storms and is generally infrequent, of short duration and of high spatial variability modified by the monsoon system (Komex International Ltd, 1999). Weather records in the Hadhramaut cover only the last 64 years. The main weather stations are located in Sayun (Wadi Hadhramaut) and in Al Rayan (coastal area); 27 new rainfall stations were established recently by Wadi Hadhramaut Agricultural Development Project in different locations of Wadi Hadhramaut and its tributaries (Figure 1.5), 14 stations were established in 1985 and 13 in 1989. 14 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi Figure 1.5. The location of the weather stations in Wadi Hadhramaut and its tributaries. All stations except Sayun have only recently been established by Wadi Hadhramaut Agricultural Development Project. The average annual rainfall in Wadi Hadhramaut is 74.2 mm with average annual temperature of about 27 Cº. 83% of the annual rainfall falls in the warmest 7 months (spring and summer). The average temperature of the coldest month is 22.3 Cº. Therefore, based on the climate classification evolved by Koppen’s (1936), Wadi Hadhramaut belongs to the arid Bwh climate type. Despite the presence of the oil industry on the plateau since 1995, there is no climatic data for the plateaus; nevertheless, the plateau is regarded as a trade wind desert (Kanfoush, 1998). According to a map produced by Cook & Warren (1973), the whole of the Hadhramaut is located within arid lands. The western border of the study area 1 (Ramlat Assaba’tain desert), is covered by sand dunes and is characterised by an annual rainfall of less than 30 mm representing a hyper-arid area. 15 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi The climate classification of UNESCO 1979 (Table 1.2), which is based on the ratio between average annual precipitation (P) and annual potential evapotranspiration (Eo) (Table 1.1), defines Wadi Hadhramaut as arid and the desert area west of the wadi as hyper-arid. More accurate definitions cannot be given because of the following reasons:  Only one weather recording station in the study area has complete rainfall data stretching over a long period; other locations have only recent rainfall data (for a short period of not more than 4 years and often with a lot of gaps in the records).   No accurate or complete data is available for some years A variation in annual rainfall data from year to year is very obvious Name index Hyper – arid 0.03 Arid 0.25 Semi - arid 0.5 Sub humid 0.75 Humid 0.75 Table 1.2. UNESCO (1979) climate classification based on the ratio between average annual precipitation (P) and annual potential evaporation (Eo). According to rainfall data from 28 stations (Appendix 2) provided by the Wadi Hadhramaut Agriculture Project, some areas receive on average less than 50 mm a year, which classifies them as deserts (Hamis, 1979); these areas are mainly located in the northern and eastern part of the study area. The variation in annual precipitation is striking. It is noted from information provided by the meteorological station at the Agricultural Research and Extension Authority in Sayun, Wadi Hadhramaut, that in the last 22 years, the highest annual rainfall recorded was 175.6 mm, diminishing gradually in volume to 1989 with a low of 8.4 mm in 1988. Spring (March to April) and Summer (July to August) are the peak periods with average monthly rainfall between 7.8 and 17 mm (Appendix 1 and Figures 1.6 & 1.7). 16 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi As in all arid and desert regions (Harris 1979, Western, 1988), the rainfall distribution in Sayun, varies from one year to the next and from one month to another (Figures 1.6 and 1. 7). Monthly precipitation is also variable in both amount and time of occurrence from year-to-year, the average ranging from 0.1 mm in December to 17 mm in March. The maximum monthly rainfall occurred in March 1989 when 112 mm was recorded. Annual rainfall 200 180 160 rainfall mm 140 120 100 80 60 40 20 19 81 19 82 19 83 19 84 19 85 19 86 19 87 19 88 19 89 19 90 19 91 19 92 19 93 19 94 19 95 19 96 19 97 19 98 19 99 20 00 20 01 20 02 0 years Figure 1.6. Annual rainfall of Sayun weather station. The rainfall distribution varies from one year to another, the highest annual rainfall recorded was in 1989 and the lowest was in 1988. Figure 1.7. A climate diagram of mean monthly rainfall (the lower line) and mean annual temperature (the upper line) for Sayun. Names from left to right are: the station with elevation above sea level, mean annual rainfall and mean annual temperature. The dotted area represents the dry period while the vertical lines indicate periods of moisture surplus; it can be seen that for most of the year the area is in moisture deficit). Most of the rain in the study areas falls during the summer when the temperatures are high there is also a period of rain in March when the temperature is lower. 17 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi Evaporation The average monthly evaporation ranges between 4.5 to 6 mm/day in winter, and between 7.5 to 9.7 mm/day during spring and summer (Figure 1.8). Generally average annual evaporation exceeds the average annual rainfall except in 1989 (Figure 1.9). The annual evaporation from an open stretch of water in Sayun was 2550 mm/year (SOGREAH, 1979). The study area has a potential evapotranspiration (PET) ratio in the range of 3 to 3.5 mm/day during the cool period and 6 to 6.5 mm/day during the months June-July, while the coastal area has PET ratio of 3 to 4 mm/day during the cool period and 4.5 to 5 mm/day during the months June-July. Figure (1.10) shows the variation in potential evapotranspiration (PET) and precipitation for the study area (Sayun) and for coastal area (al Rayan). 12 mm/day 10 8 6 4 2 0 J F M A M J J A S O N D Figure 1.8. Average monthly evaporation for Sayun. It ranges between 4.5 to 6 mm/day in winter, and between 7.5 to 9.7 mm/day during the spring and summer. 18 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi Average annual evaporation and precipitation between 1981 and 2000 (Sayun) e v a p o r a t i o n 10.0 16.0 14.0 8.0 12.0 8.0 4.0 6.0 rainfall 10.0 6.0 4.0 2.0 2.0 year 2000 1999 1998 1997 1995 1996 1994 1992 1993 1991 1990 1989 1988 1987 1986 1985 1984 1982 1983 0.0 1981 0.0 average EVA average PRE PET/ P mm Figure 1.9. Average annual evaporation (EVA) and precipitation (PRE) between 1981 and 2000 (Sayun). Generally average annual evaporation exceeds the average annual rainfall (except in 1989). The rates of evaporation range from between 0.2 and 2 times the mean annual rainfall. In summer, when evaporation is highest, there is a moisture deficit in the soil, and plants must be able to stand severe drought for several weeks. Sayun 200 180 160 140 120 100 80 60 40 20 0 PET rainf all J F M A M J J A S O N D Al Rayan 160 140 PET/ P mm 120 100 PET 80 rainfall 60 40 20 0 J F M A M J J A S O N D Figure (1.10): monthly evapotranspiration (PET) in relation to monthly rainfall for 2 stations in Hadhramaut region. PET exceeds the monthly rainfall in both Sayun (the study area) and Al Rayan (the coastal area). Averaged is over 20 years for Sayun and over 13 years for Al Rayan. 19 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi With regard to the PET data, Figure 1.10 shows that rainfall in both Sayun and Al Rayan areas is insignificant. The calculation of evaporation and infiltration at the end of 60 days in wadi basins shows that evaporation losses (60 mm) are almost negligible compared to infiltration (1.5 m) (SOGREAH, 1979). Figure 1.11. Average annual rainfall of different locations along Wadi Hadhramaut, based on the data collected between 1985 and 2000 by the Wadi Hadhramaut Agriculture Project (see Appendix 1). The 30 mm isohyet in Wadi Hadhramaut runs in the west (the beginning of Ramlat Assaba’tain desert). The 50 mm isohyet passes almost through the main Wadi Hadhramaut, then the isohyet lines increase from 60 mm to almost 300mm towards the southern plateau. The average annual rainfall at Wadi Amd (west) is about 80 to 100 mm, and at WadiDawa’n and Wadi Ala’in around 200 to 300mm. The 60 mm isoheyt runs through Wadi Bin A’li, and at Hikma and ArRadood villages (Wadi Adim). Compared to the southern plateau, the northern plateau has low rainfall, the average annual recorded in the wadis of the northern plateau (from Wadi Saar towards the east) is about 60 mm, and the average from Wadi Saar towards west is 40 mm, assuming that the annual rainfall on the northern plateau surface at the top is more than the wadi bottom down, the annual rainfall on the plateau may estimated to be 70 mm. SOGREAH (1979) also suggested that the annual rainfall on the northern plateau is less than 70 mm. During the floods of 7th April 1977, a total depth of 35.5 mm of rainfall was collected at Sayun and the volume was 40 mm3 in Adim and 4 mm3 in Shibam (SOGREAH 1979). It is not only the lack of rainwater in the study areas which is a problem, but also the fact that rainfall does not occur regularly. The rainfall of the study area is characterised by poor distribution over a month with a long dry periods (see Appendix 3 20 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi 3). The devastating floods which occasionally occur in the main wadis may be explained by high intensity of rainfall on the catchments of the wadis on the plateau and slopes. As a result of this high intensity of rainfall infiltration decreases and runoff increases. However, records showing the rate of the rainfall per hour are not available so it is not possible to get an accurate measurement of the rainfall intensity. Generally, as in other desert areas in the world, the important limiting factor in the study area is the quantity of moisture available to the plant. In order to survive in this climate, a plant must, on one hand, utilize the little rainwater available to it, and on the other hand, survive the very long dry periods. Temperature The mean monthly maximum temperature ranges between 29˚C to 35˚C in winter and 37 to 43 ˚C in summer. The mean monthly minimum temperature ranges between 9.8 ˚C and 16 ˚C in winter and between 18 ˚C and 26 ˚C in summer. The hottest months in the region are June and July. The average mean monthly temperature over 21 years at Sayun ranges from 19 ºC in January to 34 ºC in July and August. The average maximum monthly temperature exceeds 42 ºC in July and August and drops below 10 ºC in January (Table 1.3 and Figure 1.12). The average monthly relative humidity ranges between 29 in July and 47 in January (Figure 1.13). Years 1981-2002 J F M A M J J A S O N D mean 19 22 25 28 29 32 34 34 30 26 22 20 maximum 29 32 35 37 40 42 43 43 40 35 32 29 minimum 9.8 13 16 18 20 23 26 25 21 16 12 10 Table 1.3. Monthly temperature data between 1981 and 2002 at Sayun weather station. 45.0 40.0 temperature Cº 35.0 30.0 mean 25.0 maximum 20.0 minimum 15.0 10.0 5.0 0.0 J F M A M J J A S O N D months Figure 1.12. Average monthly temperatures data between 1981 and 2002 at Sayun weather station. 21 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi % 50.0 45.0 40.0 35.0 30.0 25.0 20.0 15.0 10.0 5.0 0.0 J F M A M J J A S O N D months Figure 1.13. Average monthly relative humidity at Sayun weather station. The rainfall in the study areas mainly falls in summer when the temperature is very high, the rate of evaporation exceeds rainfall throughout the year, In such environmental conditions, much of the rainfall is returned to the atmosphere by evaporation and only vegetation which is adapted to a dry environment can survive. Many plants found in the arid climate that characterizes Wadi Hadhramaut have ecological adaptations that help them to survive (see chapter 7). In arid regions, the quantity of rain is less important than the amount of water remaining in the soil after evaporation, and thus available to plants. The amount of this remainder is a function of the soil texture; coarser soils hold more water below the surface, so unlike in a humid environment, clay soils are actually the driest environments in arid regions (FAO, 1989.). The depth of wetting the arid and semiarid zones will depend on soil type and rainfall distribution (Fischer and Turner, 1978). Sandy textured soil and loamy fine sand have high permeability and coarse texture. Sandy loam and loamy textures are dominant in the study sites. Unlike conditions in temperate regions, the rainfall distribution in arid zones varies from one year to another and from summer and winter (FAO, 1989). For example, the study area receives rain during the summer and spring, while the cold months are almost devoid of rainfall. Winter rains can penetrate the soil to underground storage, while the summer rains in fall on a hot soil surface and are lost by evaporation. Therefore, the effective rainfall available to plants is low. 22 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi Wind speed Compared to similar habitats in Yemen, wind speed in the study area (as measured at Sayun) is low and ranges between 0.8 to1.3 m/sec. (Figure 1.14. The wind speed in the study area is generally high in June and July and low in November and December. w indspeed in 6 ecological zones 3 2.5 m/sec 2 Sayun Marib 1.5 alJaw f alKod alRayan 1 w est. Coastal area 0.5 0 J F M A M J J A S O N D months Figure 1.14 Wind speed of 6 dry ecological zones in Yemen. Wind speed at the study areas (Sayun) was the lowest. These ecological sites are located in the arid and semi-arid zones of Yemen. Marib and Al Jawf are located at the desert zone of northern part of Yemen and characterised by having vegetation similar to the study area, while the western coastal area has vegetation similar to the coastal area of Hadhramaut region (AlRayan). Because of its damaging effects on vegetation and loss of fertile top soil, wind erosion is an important factor in arid and semi-arid areas (Balba, 1995; Bell, 2003). Other climatic variables such as temperature and precipitation have also effects on wind erosion. For example, the presence of soil moisture can reduce the soil’s erodibility from wind (Toy et al., 2002). Precipitation can thus be sufficient to keep the soil moist and reduce the wind erosion. Nevertheless, with the high temperatures that characteriz the study area and the easily erodible sandy soil, wind can be a problem 23 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi (Toy et al., 2002). Despite this fact, information on the amount of wind on the plateau and the amount of soil removed by wind erosion under the current climate conditions in the study area is not available. 1.7. Geology 1.7.1. Introduction The present relief of the southern part of Arabia formed as a result of active neotectonic movements of the earth crust that dissected the pre-existing platform into separate structural landforms (SELKHOZPROMEXPORT,1984) and shows that there has been little movement of earth since their deposition some 30 million years ago (Ghazanfar 2004). The present plateau surface is either of Oligocene-Miocene age, overlying LowerEocene deposits (SELKHOZPROMEXPORT, 1984) or an Eocene limestone shelf (Vogt et al 1997). The plateau occupies a vast area to the north and south of Wadi Hadhramaut. The geology of Wadi Hadhramaut and surrounding areas consists of the following formations (Comrie 1992, Komex 1999, MAW, 1996): (see Figure 1.15). 1.7.2. Quaternary units (wadi deposits) Wadi deposits contain gravels, cobbles, sands, silts and clay-grade materials. Deposition of materials eroded from the plateaus is the main process in the current development of the wadi (Al Mashreki 1999). Coarse materials are laid down in the wadi bed, while finer materials are deposited on top of the wadi terraces when the wadi overflows its banks. What is visible in site 2 is the lower Jeza formation, the Umm Er Radhuma limestone cliffs, and wadi alluviums with some gravel in the wadi beds (DOVE 2001). There is an outcrop of Mukalla sandstone where Wadi Adim (site 2) opens out into Wadi Hadhramaut (Al-Ghuraf area). 1.7.3. Tertiary units (Hadhramaut group) These form an extensive and almost continuous cover in the eastern half of Yemen (WRAY 1995, Komex 1999) and consist of the following formations: 24 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi Umm er Rahduma formation: Palaeocene carbonate rocks covering the plateau and comprising limestones and dolomites; the thickness ranges between 300-400m. The Wadi systems have been eroded in this formation to a depth of 300m. Jeza formations: Inter-bedded, limestones and shales with poor aquifer properties. The thickness varies from 40 and 110m. The Jeza formation is underlain by the Umm er Radhma aquifer and divided into lower and upper formations:Upper Jeza (lower Eocene). Approximately 35 m thick with a limestone or dolomite cap with chert nodules, sometimes underlayen by marl, gypsum, clay and sands. Lower Jeza (upper Paleocene to lower Eocene). Approximately 40 m thick with a nodular, chalky, well-bedded limestone cap and thin beds of brown and grey shale. The upper and lower members of the Jeza formations are found at lower elevations, along the Wadi escarpment and cover the upper level of the plateau. 1.7.4. Cretaceous units (Tawilah group) Mukalla formation: Consists of sandstones and continental conglomerates; its thickness varies from 165 m to 520 m with fine to medium grained, white cross bedded sandstones. Sharwayn formation: A uniform sandstone formation topped by 50 m of thick shale; it forms the uppermost stratigraphic unit of the Tawilah group, and has a thickness of between 11 to 28 m. 1.7.5. Sand dunes This formation occurs in limited areas to the west of Shibam (site 1) and southwest of site 2. For the location of the study sites see Figure 1.1 page 6. 25 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi Figure 1.15 Geologic cross-section through W. Hadhramaut and the surrounding plateaus (after SOGREAH 1979). The Quaternary alluvial layer has a favourable for plant growth, it has a good depth, is fine textured, with good nutrient supplies as well as greater soil water availabilities; conversely the limestone plateau is characterised by very thin soils with less water availability for plants. 1.8. Land use 1.8.1. Agriculture The current land use was determined by interviewing local people and officials from the Agricultural Research and Extension Authority (AREA) at Sayun (see Appendix 5). The eastern part of Yemen at altitudes below 1600 m is characterised by unproductive plateaus with wadis which flood after the heavy but infrequent rainfall in the region. Small oases of date palms are found along the entire Wadi Hadhramaut. Farming, supported by wells and spate irrigation (from the wadis), is all that remains of the large areas formerly irrigated by the large-scale flood control techniques of pre-Islamic times. These date back to the time of the Sabaean kingdom in the first millennium BC (Wissmann, 1968) A recent archaeological survey (Kharaiti, 2000) in Hadhramaut has documented a variety of types of agriculture practised in the distant and recent past. The topographic layout of the valley allowed several, small protected areas to survive on low lying, fluvial terraces along the wadi banks (Komex 1993, Vogt et al., 1997). 26 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi Wadi Hadhramaut is one of the most important agricultural regions in the eastern part of Yemen. Most of the wadi bottom is heavily cultivated. Trees such as Acacia campoptila, Acacia ehrenbergiana, Conocarpus lancifolius, Prosopis juliflora, Ziziphus spina-christi and others are grown around and in cultivated fields. The irrigated fields are mainly cultivated for palm trees, wheat, sorghum, alfalfa (Medicago sativa) and vegetables. Other trees less frequently encountered near or inside the cultivated fields include: Maerua crassifolia, Ziziphus leucodermis, Tamarix aphylla, Moringa peregrina, Pithecellobium dulce, and Anogeissus bentii. The indigenous trees are used to provide forage, fuel wood and as wind breaks and border markers. According to (Bataher 1998) Acacia tortilis (A. campoptila) is the most frequent tree in the wadis followed by Ziziphus spina-christi. Another important plant of cultivation is the date. These are grown in dategardens, which consist of pure stands of Phoenix dactylifera or sometimes with annual crops such as alfalfa, and sesame and and various vegetables (e.g. Onion, Potato, Tomato, Garlic, Okra) planted beneath. Alfalfa is considered to be a major fodder crop in Wadi Hadhramaut (Awadelkarem, 1997). The main crops in the region are: 1. Cereals (wheat, sorghum and sesame), 2. Vegetables (Onion, potato, tomato, garlic), 3. Fodder (alfalfa) 4. Date palms (Phoenix dactylifera). Anthony et al., (1963) reports the following main crops from Hadhramaut: wheat as a winter crop, sorghum as summer and winter crops, date palms, and lucerne. Sesame and melons are cultivated under flood-irrigation with millet, lucerne and onions as secondary crops under well-irrigation. Most of these crops are currently planted in the study areas. Some fields are left fallow during the dry seasons and irrigated by spate-irrigation during the seasonal floods which affect the region. In fields, which are irrigated by flood-irrigation, only sorghum, sesame, melons and Vigna unguiculata can be cultivated. According to farmers these crops provide harvests even if the fields are 27 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi flooded only once. Tomatoes used to be widely cultivated, but are now only rarely grown. Irrigation by floods was mainly practiced in sites 2 and 3, no purely rainfed fields were observed in site 1. The cultivated areas of Hadhramaut region are estimated to be 23050 ha in 2003, about 2% of total cultivated area of Yemen (Table 1.4). SOGREAH estimated the cultivated fields in Wadi Hadhramaut and its tributaries as 23679 ha (Table 1.5). Years Floods Well Rainfed Cultivated Area 2003 2996 8529 10373 23050 Table 1.4. Agriculture Area (Hectare) and no of holdings (statistical year-book, from 2001 to 2003 respectively). Wheat Lucerne Market gardeni Summer sorgh Winter sorghu Sesame Palm gr Cultivated area only 11226 705 440 6312 3261 243 1492 23679 47 % 3% 2% 27 % 14 % 1% 6% (21.6 % ) Table 1.5. The total cultivated area and percentage cover in W. Hadhramaut in hectares for the period 1977 –1978 (SOGREAH 1979). Irrigated fields are mainly cultivated for palms, sorghum, wheat, alfalfa (Medicago sativa) and other vegetables. Generally sorghum is cultivated in summer and wheat in winter. Alfalfa is grown for about 3 to 5 years and harvested several times every year. Compared to sites 2 and 3, more cultivated fields are found in site 1. The cultivated areas of the study sites were measured using ArcGIS and found to be as following (see chapter 6): Site 1: 1877 ha Site 2: 467 ha Site 3: 237 ha 28 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi Month J F M A Sesame winter Sesame summer J J p P P P Sorghum summer A S O N h h h h h D h Sorghum winter Citrullus M p h Dolichos lablab P h p P h p P h Table 1.6. Cropping calendar for rain fed fields in sites 2 and 3. P= planting, h= harvesting. During the rainy season water is diverted to the fields and left for about week to dry, the following crops are then sown: sesame, sorghum, lablab and citrullus. Cultivated agriculture is practiced mainly in the wadis. Generally the main crops in the region are: Cereals (wheat sorghum and sesame), Vegetables (Onion, potato, tomato, garlic), Fodder (alfalfa) and Date palms (Phoenix dactylifera). The characteristic annual crops are Alfalfa (Medicago sativa) and dense gardens of palm trees (Phoenix dactylifera) irrigated mainly by pumping, In spateirrigation areas, only Sorghum, Sesame, Citrullus and Vigna unguiculata are cultivated. In the past the best available estimates of areas under cultivation in Wadi Hadhramaut were from the work of SOGREAH (1979), which relied on the interpretation of arial photographs. However, now, with the availabilty of satellite images and GIS software programs such as ArcView and ERDAS, the cultivated areas can be more accurately calculated and changes in areas under cultivation can be monitored from year to year. The area of land under irrigation between the west of site 1 (west Shibam) and the point where Wadi Adim falls into the Wadi Hadhramaut was calculated using a satellite image from April 2001 and ArcGIS software. The calculation shows about 73.5 km2 (7350ha) of irrigated fields (Figure1.16). 29 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi Figure 1.16. The irrigated fields along the Wadi Hadhramaut from site 1 (west Shibam) to the point where Wadi Adim falls into the Wadi Hadhramaut. The cultivated fields are estimated to cover about 73.5 km2 (7350ha). The uncultivated areas are about 68%. There is no obvious spread of agricultural land onto the rangeland or onto the natural habitats in the Wadi Hadhramaut; in actual fact comparison between the different satellite images shows a decrease in the areas under cultivation in some areas; in particular in areas that have been and, still are affected by floods. For example, the Landsat imageries of 1973 and 1998 of site 1 were classified using ERDAS Imagine (Figure 1.17); the result shows that the cultivated areas decreased dramatically from 1973 to 2001, and, as a result, these areas have reverted to natural vegetation dominated by Zygophyllum album (association 2). These areas now provide important grazing for local animals. However, in some areas fields have been created in formerly neglected or marginal areas. 30 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi Figure 1.17. Comparison of the changes in land use between 1973 and 1998 in site 1. Due to the extreme floods, some agriculture field areas in 1973 (right) have been washed away and replaced by almost bare sandy areas covered by sparse vegetation (left). Compared to the cultivated lands of Wadi Hadhramaut (Figure 1.16), uncultivated areas represent about 22961 ha (68%). SOGREAH (1979) estimated the uncultivated areas in Wadi Hadhramaut and its tributaries at more than 78%. According to Alwan (1990) in Wadi Hadhramaut (from Wadi al Khon east to the desert of Ramlat Assaba’tain), only 5.5 % of the land is utilized as agriculture fields and 25.5 % as grazing lands. 31 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi From these figures it can be seen that uncultivated areas in the Wadi Hadhramaut and its tributaries range between about 70 and 80%, the figure is higher (about 94%) towards the desert of Ramlat Assaba’tain. These uncultivated areas are distributed as follows:  Gravely wadi beds and adjacent banks; these are the main grazing lands, as most natural vegetation is concentrated here.  Bad lands, which are abandoned fields with deep gully erosion; the vegetation here is sparse mainly herbaceous.  Sand dune areas, covered by sparse vegetation and isolated trees. 1.8.2. Animal husbandry system in the study areas As the study area is situated in arid and semi-arid areas, the ability to sustain a production system in the study area depends upon the integrated management of cultivated fields and rangelands (Powell et al., 2004; Dixon et al., 2001). Dry lands (Dixon et al., 2001) in the study area comprise 70-80% of the total land area. Local people have developed their own ways of reducing the effect of the harsh environment by using crop and forage plants in agro-pastoral (Figures 1.18 and 1.19) or agroforestry systems. Livestock are raised mostly on the rangelands near the settlements and on agro-pastoral systems and comprise a major source of income. The agropastoral systems in the study area are practised widely by almost all farmers. The system involves combining forage crops, such as alfalfa, and annual crops, such as sorghum and wheat, in rotation. Animals usually graze on the sorghum stubble and weeds after harvesting is finished. Agro-forestry, is less widely used. This involves the planting of trees, such as Ziziphus spina-christa, Ziziphus leucodermis and date palms, inside or around cultivated fields. Supplementary feeding, using alfalfa leaves of Prosopis fracta, seeds of Acacia spp. and Prosopis juliflora is fairly common in the study area. . 32 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi Figure 1.18. Agro-Pastoral system on irrigated fields. Alfalfa is on of the main forage cash crop for livestock, sorghum stubble and weeds are also important sources. Figure 1.19. Agro-Pastoral system on spate irrigation (floods) fields sites 2 & 3. The floods mainly occur in March and July (summer and spring). Animals graze on the sorghum stubble and weeds. The photo shows a mixture seeds of sorghum, sesame, melons and lablab. A typical mixture includes sesame, sorghum and lablab. Sometimes, in particular in site 2, Citrullus spp. are added to this combination. 33 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi There are two main forms of animal husbandry in the study areas: 1- Nomadic animal husbandry, in which the herdsmen with their herds of goats, sheep and camels spend the entire year in the Hadhramaut region. The nomads wander constantly from place to place, seeking food and watering places for their herds. 2- Sedentary animal husbandry, in which the goats and sheep graze in the areas surrounding villages throughout the year. Nomadic animal husbandry This is practised and still important in the Hadhramaut region. 70 years ago (Stark, 1936), nomads used to move around but generally settled for 4 months in an area between Wadi Amed and near al Ahqaf (in the western part of the study area) searching for fodder, mainly with their camels. Nomadic herders were also noted in the northern and eastern part near Ghayl Bin Yamain (Komex International Ltd, 1999). This type of grazing is no longer practiced; most of the nomads in the study area settled about 5 to 10 years ago. Nevertheless, up to the present time, animal husbandry has been the most productive use of the study area. Due to the low rainfall, the vegetation density of Hadhramaut is generally low, as is the supply of forage and drinking water for animals. These factors force the herdsmen to wander continuously with their herds of camels, goats, and sheep. At the beginning of the rainy season, the herdsmen migrate with their herds towards areas with higher rainfall or with better forage and fewer herders (see Figure 1.20). In the past, particularly during the rainy seasons, nomads used to come to the study area from various parts of the Hadhramaut region (Figure 1.20). These areas are:   al Mukalla (southern coastal area)  Ghayl Bin Yamin (southern coastal area)  Qaf Al Amiri (north of Tarim) and Qaf Al Kuthairi (north of Shibam)  Ashihr, (southern coastal area)  Ghayl ba Wazir (southern plateau)  Thamood (northern part of the study areas)  Ramlat Assaba’tain desert areas (western part of the study areas) Al Masila areas (eastern part of the study areas) 34 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi The length of stay in the study areas ranged between one and five years, but some pastoralists have been in the study areas (especially site 3) for more than 10 years. Figure 1.20. The movement of the nomads from surrounding areas and coastal plain to the Wadi Hadhramaut. Recently, nomads and their herds have started coming to Wadi Hadhramaut, from areas such as Al Masila, in the eastern part of the study area, or from the southern coastal areas, to graze on the pods of Prosopis juliflora or even to buy these pods from farmers whose land has been invaded by this tree (plate 1.1). Plate 1.1. Pods of Prosopis juliflora in large sacks, ready to transfer to Al Mukalla and Al Shihr (coastal area). Recently herders have started to come to the study area with their animals to graze on Prosopis juliflora or even buy few sacks of pods. For more on Prosopis juliflora and its effect on plant ecology see chapter 7. Sedentary animal husbandry 35 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi The development of arable farming encourages sedentary animal husbandry. Village animals graze around the fields in the wadi beds and surrounding slopes. The grazing starts generally in the morning and ends in the evening, some farmers let their livestock graze from the morning to the afternoon and, after a short break, continue to graze from the afternoon until the evening. In the dry season, herders mainlufeed their animals on alfalfa (Medicago sativa) and sorghum stalks. Animals also graze on the sorghum stubble remaining in the fields after harvest as well as on weeds. This practice is mainly found in cultivated fields belonging to the government. In some places, especially in Wadi Athahab (site 3), goats and sheep are kept indoors and supplied with forage that has been harvested or bought. This is due to a number of reasons: 1. where the herders were previously young girls who now go to school, 2. where a farmer has few animals 3. where there is an agreement between farmers. This agreement prevents grazing outside, especially during farming period. This law has recently been broken by Bedouins from who have come in from other parts of the Hadhramaut, as a result more pressure has been placed on the rangelands, which may negatively effect the plant ecology in future time. 1.9. Environmental, land unit characteristic and the problems for plant growth 1.9.1. Introduction According to UNEP & WMO (1996), the effects of climate change may be unimportant compared with the past and present impacts of human activities, including livestock grazing. Human impacts such as roads and other form of infrastructure have a strong impact on plant species distribution and wildlife (Nellemann et al., 2001). According to the World Resources Institute (WRI, 1992) overgrazing is the most important cause of soil degradation in arid and semi-arid regions. In in semi-arid and arid regions of Africa and Australia, overgrazing causes 49% and 80% respectively of soil degradation (FAD, 2000). In Jabal Bura’ (in the western part of Yemen), 42 km2 of woodlands have severely degraded between 1973 and 1987, due to the conversion of woodland to croplands (Herzog, 1993). 36 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi 1.9.2. The Physical impact of Infrastructural activities The exploration and production of petroleum has caused major negative impacts to the soil, surface and ground waters of the local ecosystems in the United States of America (Yousif and James, 2003; James and Zielinski, 2003). Current oil exploration and related activities are probably leading to the same problem in the Hadhramaut. Oil companies have now been present in the region for over 10 years. The impact on the landforms, soil, vegetation, water bodies, habitat, and cultural resources are already obvious in many parts of the study area. The activities of oil companies in the study area, particularly in Wadi Adim (site 2), have made a considerable impact on the landscape through clearing land for the construction of service tracks, pipeline inspections and access roads. These activities have already heavily eroded the landscape, particularly on the plateau. Many roads on the plateau are new and cover more ground than is necessary. As a result, bulldozers have scraped away the top surface of soil and stones, destroying the vegetation as well as archaeological sites on the plateau. The materials tipped down slopes and over cliffs during the construction of road and pipelines have covered and damaged the vegetation in the wadi beds and adjacent lower slopes. During the construction of oil wells, the primary emissions are waste water, solid waste, waste oil, chemicals and dust (DOVE, 2001). Most of these waste products have ended up in the adjacent wadis. 1.9.3. Floods Flooding is a highly dynamic action influencing the plant species of the Wadi Hadhramaut. Information on the effect of flooding on the distribution and richness of plant species is not available. Floods in Wadi Hadhramaut originate exclusively from the northern and southern limestone plateaus (SOGREAH, 1979), although the northern plateau generates fewer floods than southern plateau. The floods of 7th April and 23rd October 1977 devastated areas in both Wadi Adim (site 2) and Qassam (east 37 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi of Tarim in the eastern part of the study area). Measurements show that Wadi Adim suffer much more frequent flooding than the other wadis (SOGREAH, 1979). Historically, Wadi Hadhramaut has often suffered from several major floods which caused severe damaged to cities and, in particular, to the historic town of Shibam. Many people and livestock were lost and palm trees destroyed. The strongest and most severe floods were recorded in 1249, 1356, 1493, 1495, 1582, 1590 and 1687 (Lewcock, 1987; INTERCONSULT and MacDonald, 1993). Major floods in recent years occured in 1974, 1977,1987,1989,1992 and 1999. These periods correspond with the patterns of annual rainfall data recorded from Sayun (Appendix 1), which range between 110 mm and 175.6 mm. The floods of March 1989, which lasted for about 9 hours, are considered by far the most severe in Wadi Hadhramaut (INTERCONSULT and MacDonald, 1993). The annual rainfall recorded in 1989 from different stations within W. Hadhramaut (Appendix 2) also shows amounts higher than 100 mm. A strong flood was recorded by SOGREAH (1979) on 2nd March 1978 in Wadi Thibi (site 3) and on 11th July 1978 in Wadi Adim (site 2). Floods occur over the greater part of the wadi systems once or twice a year, with the strongest in the larger wadis usually once a year. The floods are insignificant in small or tributary wadis (SELKHOZPROMEXPORT, 1984). The vegetation of some parts of the Wadi Hadhramaut is composed of different successional stages reflecting their subjection to annual floods. For example, species richness and distribution along the Wadi Hadhramaut (in particular site 1) has been severely affected by annual flooding and sedimentation. Here the structure of the vegetation has changed from woodland dominated by date palm to grassland dominated by Zygophyllum album. Strong floods in Wadi Adim destroyed some important trees, for example Anogeissus bentii reported by local people to be abundant in Wadi Adim and Wadi Hikma (site 2). This tree species has been destroyed by floods, in particular the flood of 7 April 1977 (SOGREAH, 1979), and not a single tree of Anogeissus bentii was observed during the field work. 38 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi 1.9.4. Grazing Grazing can alter the spatial heterogeneity of vegetation, influencing ecosystem processes and biodiversity (Adler et al., 2001). Overgrazing has been indicated as one of the main factors causing range degradation by accelerating erosion, reducing biodiversity, and altering soil properties (Ozgul and Oztas, 1990). In the study areas, and particularly in Study Site 2, traditional grazing on the plateau has now largely been forbidden because of oil-related activities. Grazing is now mainly confined to the wadis and the adjacent slopes. As a result grazing intensity has increased in the wadis of all Study Sites.. Generally, grazing is continuous; herders going out early every morning and returning in the late afternoon. Very few people now graze animals on the plateau, and then usually only after good rainfall. According to local people, overgrazing has reduced the abundance of some species notably Merremia hadramautica, Boerhavia elegans and Tribulus arabicus. The first species is abundant on the wadi beds of site 3 and is found near cultivated and fallow lands at sites 1 and 2. The second species is widespread over in the entire study sites but with low cover. Goats and sheep dominate the area, with camels and cows limited in number. In addition to grazing, animals are fed with alfalfa, sorghum and crop residues. Grazing time varies but is practiced daily; Herders (mainly women) take their animals (goats and sheep) to the neighbouring area (wadi beds and adjacent slopes) and usually return in the evening. In the dry season, most of the herders feed their animals on alfalfa (Medicago sativa) and sorghum. Recently some farmers have started feeding their animals inside their houses most of the year. This is due to a number of reasons, such as a lack of herders and because of the high cost of herders. Also children, who used to herd the animals, have stopped herding any more mainly because farmers prefer to send their children to school. There are also fewer animals and many farmers now have smaller herds and prefer to buy forage rather than to herd them outside. A further factor is that grazing on the plateau is not allowed because of intensive oil activities and the complex security measures that surrounding them. 39 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi According to DOVE (2001) herders are allowed to graze on the plateau only if they first obtain permission from the Military commander in Sayun. Other Bedouin say that there is no formal restriction on access to the plateau. In general, military rule since 1994 has restricted the travel patterns of nomadic tribes throughout the plateaus (Komex International Ltd. 1999). Access to grazing in oil company sites on the plateau is not allowed; in addition the pipelines form a physical barrier to their camels. Generally, for most of the Bedouin, the restriction on grazing on the plateau is a small impact on their livelihoods, nevertheless, for some Bedouin, these restrictions create a problem since their traditional routes between grazing areas often cross the plateaus. 1.9.5. Plant collection Collecting plants for medicinal or other purposes has an impact on certain species. For example the roots of Rhazya stricta are crushed, mixed with water and drunk to heal stomach pains. Boerhavia elegans, which is reported by local people to have been abundant in the past, has disappeared in some locations or is now rare. The seeds of this plant have been and are still collected extensively by local people to be mixed with traditional food (called As’id). In the past, people used to protect the areas and slopes bordering their fields from livestock during the period of cultivation. Livestock, mainly belonging to the owner of the fields, was then allowed to graze after harvesting was completed. In addition, people used to harvest the seeds of Boerhavia elegans from these protected slopes in large quantities. Now however, this practice has changed and the slopes are no longer protected. Local people believe that the low availability of this plant in recent times is due to this change in the traditional patterns of grazing. 40 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi 1.10. Soil A predominance of physical weathering processes in the desert and semidesert areas keeps soil development at an early stage (Zohary 1962). Hamadas are the dominant rock type in the study area, in particular over the plateau. This type of rock occupies large areas of North Africa, Central Asia (Zohary, 1962) and in the Arabian Peninsula (Batanouny, 1981). Hamada or hammada is Arabic for ‘stone desert’ and is applied to surfaces covered by stone fragments and pebbles. Hamada consists mostly of fine particles and soil produced by weathering. The vegetation of the hamada is characterised by very poor cover, and is concentrated mainly in runnels, depressions and other drainage lines (Zohary, 1962 and Batanouny, 1981). Many of the alluvial and colluvial soils formed by erosion and deposition of materials came from the plateau surfaces and small tributaries. (Al Mashreki, 1999; Andrew et al., 2005). According to Coode Blizard Ltd (1997), SOGREAH (1981) and Al Mashreki (1999), Wadi Hadhramaut and its tributaries lie on alluvium deposits and consist of four alluvial units and one aeolian unit; 1- Basal gravels and soils of fluvial origin, which form the beds of the main wadis and the lower terraces laying over two metres of hard limestone gravel. 2- Lower silt deposits; with up to 14 m thick of fine alluvial deposition of early Holocene age, the surface of this layer is light reddish brown make up of heavy silt loams to very fine sandy loams. 3- Middle silt (irrigation silts) and laid down during millennia by the activities of early hydraulic civilization in the Hadhramaut giving coarser textured, sandy loam, silt loam soils which dominate the upper horizons of this deposit. 4- Recent flood silts; recent sediments deposited by temporary floods in the main wadis and their tributaries; they form the upper metre of the soil and comprise of loamy sand, silty loam and sandy loam textures. 5- Recent Aeolian deposits. They occur in a number of locations such as Wadi Adim and (site 2) and Wadi Khashamer, Shibam (site 1). The soil of narrow tributary wadis and the alluvial fans formed at the foot of the limestone cliffs, as defined by SOGREAH (1981) and Al Mashreki (1999) are of alluvial-colluvial origin, formed by recent erosion and deposition of materials from 41 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi small tributaries flowing from the plateau down into the wadi. An investigation of soil in Wadi Adim, site 2 (Coode Blizard Ltd, 1997) shows that the soils have varied textures but are predominantly sandy loam with sand-filled cracks that cover the entire area, and along which gullying spreads. Soils in the study area generally are formed from Tertiary and Quaternary material. Parent material consists of sedimentary material brought from elsewhere, mainly gypsum-rich limestone and marl and some sandstone (Al Mashreki, 2005). According to FAO World Soil Resources (FAO, 2003), the soil classification of the study area is Gypsisols, Calcisols (GY). The plateau has soil less than 10 cm and mountain slopes facing the main wadis have less than 20 cm of soil with the loss of rock material, both are over Umm er Radhuma and Jeza limestone rocks. According to the FAO classification the plateau type of soil may classified as Lithosols and the mountain slope type of soil as Calcaric (Zouzou and Furley, 1975). Main soil Classification for each land form unit in Wadi Hadhramaut and the surrounding mountain slopes and plateaus is given in Table 1.7 (Al Mashreki, 2005). 42 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi Table 1.7: Main soil types. The soil of the study area were classified according to Soil Taxonomy Classification systems (USDA, 1994) and correlated to World Reference Base (WRB, 1998). Physiographic Unit Dominant ST classification FAO classification Soil Order Wadi Hadhramaut Entisols Aridisols Surrounding mountain slop Entisols Aridisols Plateaus Aridisols Entisols Typic Torrifluvents Calcaric Fluvisol Typic Torriorthent Calcaric Regosol Typic Calciorthids Haplic Calcisol Typic Campiorthids Calcaric Campisols Typic Torriopsmments Calcaric Arenosols Typic Torriorthent Calcaric Regosol Typic Torrifluvents Calcaric Fluvisol Typic Haplogypsid Haplic Gypsisol Fluventic Campiorthids Calcaric Campisols Lithic Calciorthids Eutric Calcisol Typic Petrogypsid Petric Gypsisol Typic Haplogypsid Haplic Gypsisol Haplic Campiorthids Haplic Campisols Lithic Calciorthids Eutric Calcisol Lithic Torriorthent Eutric Regosol 1.11. Floristics Largely because of its great variability in elevation, topography, soil and climate, Yemen is one of the most ecologically diverse countries in the Arabian Peninsula. Major vegetation communities include drought-deciduous and evergreen woodland, tropical valley forest, monsoon woodland, coastal and montane grasslands, and some of the most arid deserts on earth. This plant diversity in turn supports a rich biodiversity, including about 3000 plant species (Scholte et al., 1991). Yemen also supports high rates of endemism, large numbers of rare species and many species with restricted distributions. There have been estimated to be about 429 endemic plant species in Yemen (Miller and Nyberg, 1991) this is approximately 14% of the total; of these 307 occur in the Soqotra Archipelago (Miller and Morris, 2004). 43 In Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi Hadhramaut, the Jol plateau is considered to be a local centre of endemism (Miller & Nyberg, 1991) where endemics such as the following are found: Pulicaria nivea, P. rauhii, P. lancifolia, Anogeissus bentii, Kissenia arabia and Ochradenus (see chapter 3). There is no flora covering the whole of Yemen and no list of threatened plants (IUCN, 1986). However, a recent study by Al Khulaidi (2000) estimated the total number of plant species at 2810, belonging to 1006 genera and 173 families, of which 2559 are naturalized, 121 cultivated and 111 introduced. 1.11.1. The geographical distribution of the vegetation in the study area According to White (1983), the flora of Yemen is, a mixture of paleotropical and holarctic origin with three main three main elements represented: the SomaliaMasai regional centre of endemism and the Afromontane archipelago-like regional centre of endemism are of paleotropical origin and the Saharo-Sindian regional zone is a mixture of both paleotropical and holarctic elements. The Somalia-Masai and the Afromontane dominate in the western mountains, parts of the highland plains and the coastal areas; Saharo-Arabian plants dominate in the eastern mountain and the eastern and northern desert plains. The map in White & Léonard (1991) shows two of these regional zones occur in Hadhramaut – the Saharo-Sindian and the Somalia-Masai The following are species occurring in Hadhramaut considered to be of Saharo-Sindian affinity (White, 1983, Zohary, 1973): Ochradenus baccatus, Rhazya stricta, Panicum turgidum, Aerva javanica. Zygophyllum simplex, Fagonia indica, Salsola spp., Phoenix dactylifera, Hyphaene thebaica, Dipterygium glaucum, Heliotropium rariflorum, Seddera latifolia, Pulicaria undulate, Capparis decidua, Tamarix aphylla, and Tribulus arabicus. A few species of Sahara-Sahel linking element (White, 1983) are found in the region, these are: Blepharis edulis, Forskohlea tenacissima, Maerua crassifolia, Leptadenia phyrotechnica, Chascanum marrubifolium, Cleome scaposa, Cornulaca monacantha. 44 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi Somalia-Masai species in Hadhramaut include the following (White, 1983): Acacia spp., Caralluma spp., Aloe spp., Maerua crassifolia, Balanites aegyptiaca, Moringa peregrina, Commiphora gileadensis, Commiphora habessinica, Commiphora kua, Cadaba farinosa, Ficus cordata L. subsp. salicifolia and Ziziphus spina –christi. 1.11.2. The vegetation of the plateau and the Wadi zones The only recent work in the study area was a brief vegetation survey carried out in Wadi Adim and the surrounding plateau, in the southern part of Wadi Hadhramaut, by the oil company DOVE (2001). The total number of plant species recorded in this study was about 101 species, belonging to 77 genera. The results of this study are summarized in Table 1.8. No 1 2 3 4 5 Vegetation description and type Sparse grassland Iphiona scabra & Zygophyllum simplex Woodland, shrubland and open woodland Rhazya stricta & Acacia tortilis Dense grassland Schouwia purpurea & Pulicaria undulata Dense woodland to sparse shrubland Tephrosia sp. & Acacia tortilis Open woodland and grassland Seetzenia lanata & Bothriochloa insculpta Altitude (m) 907 980 Floor of main wadis 677 718 Floor of main wadis: fallow fields 643 700 Floor of spur wadis and drainage lines 643 940 Flat to steep slopes, floor of drainage lin 850 and small wadis on plateau 949 Sites Flat areas on plateau 6 Shrubland Dry slopes on plateau and facing the main 940 Grewia villosa & Acacia mellifera wadis 950 7 Shrubland to sparse shrubland Shoulders of main wadis and dry slopes o936 Jatropha spinosa & Zygophyllum coccineum plateau 962 8 Dense woodland Wadi floor, Prosopis invaded fields or 680 Prosopis juliflora & Zygophyllum album riverbed 9 Shrubland Oasis stream / river 731 Kanahia laniflora & Pluchea dioscoroides 10 Dense woodland Wadi floor, cultivated / date palm 680 Salvadora persica & Phoenix dactylifera 734 Table 1.8. The vegetation types across a hypothetical transect across Wadi Adim and adjacent plateau (DOVE,. 2001). Not: Acacia campoptila is incorrectly named as A. tortilis. . Three major plant communities from the plateau zone south eastern Hadhramaut have also recorded by Komex International Ltd. (1999). These communities are: 45 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi 1- Euphorbia balsamifera - Caesalpinia erianthera community, with Euphorbia hadramautica, E. rubriseminalis, Farsetia longisiliqua, Tephrosia heterophylla, Lycium shawii and others 2- Grewia erythraea - Jatropha spinosa community, with Adenium obesum, Commiphora kua, Acacia mellifera and A. hamulosa. 3- Depression community with Stipagrostis paradisea, Fagonia paulayana, Cleome brachycarpa, Reseda sphenocleoides and Farsetia linearis. This community is similar to the sparse grassland Iphiona scabra & Zygophyllum simplex vegetation type described by DOVE (2001). In the Wadi zone the following communities were also recorded by Komex International Ltd (1999): 1-Typha domingensis community with Tamarix nilotica, Juncus acutus, Scirpus corymbosus, Cyperus laevigatus, Eleocharis geniculata and others; this community is very similar to the Kanahia laniflora & Pluchea dioscoroides type that was described on wadi beds by DOVE (2001) and 2- Acacia ehrenbergiana community with Acacia tortilis, Ziziphus leucodermis, Turraea parvifolia and Lycium shawii. Several other communities of the main wadis were described by Gabali and Gifri (1991), these are: 1. Panicum turgidum community with Rhazya stricta and Tamarix spp., 2. Cassia italica-Tephrosia spp. community 3. Anisotes trisulcus -Solanum dubium community. 46 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi 1.11.3. The vegetation of the desert Zone The plant cover is very poor and sparse and is concentrated in depressions and wadis where more moisture is available; on the open plains the vegetation is generally a sparse desert grassland. According to Gabali and Al Gifri, (1991), the main communities in this region are: 1. Acacia spp.-Tamarix spp. with an association of Salvadora persica, Calotropis procera, Balanites aegyptiaca, Leptadenia pyrotechnica and others in depressions 2. Panicum turgidum-Lasiurus hirsutus in relatively deep soil. 3. Salvadora persica-Tamarix spp. on the sand dunes. 1.12. Summary Generally the topography of the 3 detailed study sites is very similar and comprises the following: Steep-sided valleys with flat bottoms cut by deep drainage lines (wadi beds) where most of the vegetation is concentrated,  steep to moderately steep, rocky mountain slopes facing the main wadis and their tributaries,  very steep to vertical bare rocky cliffs,  undulating to almost flat rocky plateaus,  steep to moderately steep rocky slopes on the plateaus and,  canyon-like secondary wadis on the plateaus. As in other arid zones and desert regions (Harris, 1979; Western, 1988), the rainfall distribution in the study area varies from one year to another and from month to another. For example, the Sayun area receives most of the rain during the warm spring and summer months when the temperature is very high. The cold winter periods 47 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi are almost devoid of rainfall. The rate of evaporation exceeds rainfall throughout the year, The main crops in the region are: 1. Cereals (wheat, sorghum and sesame), 2. Vegetables (Onion, potato, tomato, garlic), 3. Fodder (alfalfa) and 4. Date palms (Phoenix dactylifera). The area under cultivation in the floor of the wadi Hadhramaut and its tributaries ranges between 20 to 30%; the figure is much lower (about 6%) towards the desert of Ramlat Assaba’tain. As elsewhere in arid and semi-arid areas, the ability to sustain a production system in the study area depends upon the integrated management of cultivated fields and rangeland. The local people have developed their own systems to reduce the effect of harsh environment by using crop and forage in agro-pastoral combinations. There are two main forms of animal husbandry in the study areas: 1- Nomadic animal husbandry, in which the herdsmen with their herds of goats, sheep and camels spend the entire year in the Hadhramaut region but wander constantly from place to place, seeking food and watering places for their herds. 2- Daily grazing, in which the goats and sheep graze by local people in the areas surrounding the villages throughout the year. The intensity of grazing in the main wadi is high, while on the plateau it is low. Sedentary animal husbandry is restricted to the wadis, whilst most Bedouin (nomads) graze on the plateaus. Generally the oil activities have restricted the movement of herders on the plateaus. Exploration for oil and production of petroleum has caused major negative impacts to the soil, surface and ground waters and the local ecosystems in the study area (in particularly site 2 and currently site 1). The impact of waste water on vegetation seems generally to be beneficial: the release of water in certain wadis has produced vigorous 48 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi growth of plant species. However, over the long term the vegetation may be intolerant of salts and chemicals in the waste water. This is likely to affect growth and reduce seeds germination. Flooding is by far the most extreme and active environmental factor influencing the plant species of Wadi Hadhramaut. The structure of vegetation in some areas has been changed from woodland to grassland as a result of floods and this has resulted in the reduction and disappearance of some common species from the study area. Thus the vegetation of the Hadhramaut appears to the result of the interaction of past and present human activities on a remarkable set of landforms and a set of severe climatic conditions. 49 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi Chapter 2. Assessment of previous work on vegetation in the Yemen. 2.1. Introduction There is little published information on the plants of the Hadhramaut Governorate or indeed of Yemen as a whole. Such paucity is well illustrated by the account in the IUCN publication Plants in Danger (IUCN, 1986). This book provides information about data sources on plants for each country and island group in the world. In the introduction it states that they are “unable to present figures for only a handful of countries” – one of these is Yemen. In short, the only Floras to cover the area are Flora Arabica (Blatter, 1936), Flora of Aden (Blatter, 1914-1916), Flora Tropischen Arabien (Schwartz, 1939) and the Flora of the Arabian Peninsula and Soqotra (Miller & Cope, 1996). However, the first two, although called “Floras”, are little more than annotated checklists and the last is incomplete with only the first of five volumes so far published. An appraisal of the available literature forms the objective of this chapter and it will be seen that this largely follows the history of botanical collectors. 2.2. Botanical Exploration The botanical exploration of Yemen goes back over two hundred years to the pioneering research of Pehr Forskahl, the botanist on the ill-fated Danish expedition to SW Arabia in 1761-1767, and the posthumous publication in 1775 of his Flora Aegyptiaco-Arabica (Miller & Cope 1996, Wood, 1980). Yemen was then known as Arabia Felix - “Happy Arabia” - the fabled land of frankincense and myrrh (Wood, 1997). There is little written about the plants of Yemen in the Early Arab literature. The most important source is the Kitab al Nabat (the “book of plants”) by the Persian botanist Abu Hanifa al-Dinawari published around 895 AD. This paid particular attention to plant species of medicinal valu but also covered diverse topics such as soils, crops (including cereals and the date palm), astronomy and meteorology. Yusuf Bin Al Rasul al Mudhaffar, the king of Al Rasulya kingdom, who ruled Yemen in the seventeen century, and Al Zabidi in the Taj al Arus (around 1780 A.D.) gave valuable information on plant uses, in particular as medicine. 50 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi These early sources dealt principally with the western part of Yemen making little if any reference to the Hadhramaut region (which lies in the eastern part of the country). This situation, with a relative wealth of literature referring to the western part of the country, but more or less none to the eastern part, was little changed until the later part of the 19th century when the first plant collectors ventured into the remote and often inhospitable Wadi Hadhramaut (Gabali and Al Gifri, 1991). In fact, until this present study, the vegetation of Hadhramaut had never been systematically studied or surveyed by botanists. Previous studies concentrated mainly on compiling floristic checklists or making general descriptions of some of the principal vegetation communities. As explained above, the account of the botanical exploration of Hadhramaut is essentially a history of botanical collectors and their publications. These, unfortunately, are often of limited scientific value, typically being little more than travelogues. An excellent summary of these early collectors, together with notes on their publications, is given in Flora Arabica (Blatter, 1936) and a more up to date list, together with notes on where specimens are deposited is given by Wickens in his Guide to Collectors in Arabia (Wickens, 1982). There follows a brief history of the principal collectors and the significance of their collections and publications. The first recorded collections were made in 1881, by the German Explorer Georg Schweinfurth, famous for his explorations in north east and central Africa, who visited Yemen as a member of the Austrian Riebeck Expedition. He collected plant specimens from the coastal area of Hadhramaut (Wickens, 1982) but did not venture into the interior. His results were published as the Sammlung Arabish-Aethiopischer Pflanzen Ergebnisse von Reisen 1881, 1888, 1889, 1891,1892 in a series of supplements to the Bulletin de l' Herbier Boissier (Bulletin of the Boissier Herbarium) These publications contain the names of plants collected by Schweinfurth but no description of the vegetation was given. 51 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi In 1894, the German explorer Leo Hirsch collected over 100 specimens from Hadhramaut. He was the first collector to visit the interior, penetrating northwards to Shibam in Wadi Hadhramaut, to some of the main wadis further north and also venturing onto the Jol plateau. His specimens were kept in the Berlin Herbarium but his travels resulted in no scientific description of the area. In 1893-94 a British expedition headed by the archaeologist and traveller James Theodore Bent, accompanied by the British gardener William Lunt, collected some 150 species of flowering plant including 25 new species and 2 new genera. These were collected on the coastal plain, the southern plateau as well as in the main wadis of Hadhramaut. Their specimens were deposited in the Kew herbarium and the British Museum and a brief notice of their expedition to the Hadhramaut, including a list of plants collected appeared in Kew Bulletin in 1894. In 1917 Philby John was a plant collector visited the region, his plant collection considered some of the earliest data for this region. The next significant contributions to the botany of Hadhramaut were made by two Austrian explorers, Rathjens and Wissman (1934). They made a series of visits to Hadhramaut in the 1930s. They collected several species from Hadhramaut and their collections written up in the Flora tropischen arabien – which as explained above is little more than a annotated checklist with no descriptions of the vegetation. In 1946 the first collections from the desert interior were made by Wilfred Thesiger and other members of the Desert Locust Survey including Guichard and Tilliland (Thesiger, 1959). Their collections were deposited in the British Museum. Their explorations lead to no scientific descriptions of the vegetation but nevertheless the travelogues of Thesiger make interesting reading and some information on the vegetation can be gleaned from them. From the 1960s to the present day a series of expeditions by succulent plant enthusiasts have made important contributions to the botanical exploration of Hadhramaut. Principal amongst these collectors was the South African amateur botanist John Lavranos. He first visited the region in 1962, when he travelled from Al-Mukallah 52 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi to Mawla Mattar (on the escarpment above Al-Mukallah). During this trip he collected herbarium material and live specimens of various succulent plants, some of which proved to be new to science. In April 1964 he visited Al-Mukallah with the late Prof. Werner Rauh, of the University of Heidelberg, Germany. They spent a week travelling up the Wadi Himem to Bayn al-Jibal and then on to the Wadi Hadhramaut via the Wadi Dawa’n. Much material was collected during this trip and several new species described as a result. Between 1996 and 2001 he made several journeys to the Hadhramaut, covering the Jol plateau along both routes from Mukalla via Bayn al-Jibal and also along the main road via Ra's Huweira, including also part of the lower valley of the Wadi Maseilah and the foothills of the Jol between Ash-Shehr and Sayhut. These journeys were in part in the company of a Yemeni botanist Dr. Abdul Nasser Al-Gifri and also with Dr. Bruno Mies of Essen University in Germany, Mr Thomas McCoy and Mr Giuseppe Orlando of UNICEF. These expeditions resulted in the discovery of several new species and greatly increased our understanding of the distributions of many taxa. The results of these trips are written up in various journals aimed at succulent plant enthusiasts but nevertheless they contain important descriptions of the vegetation (see Lavranos and Al Gifri, 1999; Lavranos, 1966, 1971, 1974 and 1993). During the last ten years, a series of collectors have visited the region. Important collections have been made but no significant publications have resulted: these are listed below:- In 1975 an Egyptian UN expert Fathallah collected about 350 specimens from Hadhramaut. His specimens are kept in both Cairo University and al Kod Agricultural Research Station. He made important contributions to the botanical exploration of Hadhramaut. In 1993 as well as in 2003 and 2004 Tony Miller a botanist working on the Arabian Peninsula vegetation visited the coastal area, Jol plateau and many other places of Hadhramaut. He travelled from the coastal plain to the northern plateau and collected a considerable number of plant species and also live specimens, some of which proved to be new to science. He made important contributions to the botanical exploration of 53 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi Hadhramaut. His collections are kept in the Herbarium of Royal Botanic Garden of Edinburgh (RBGE). In 1992 and 1998 Prof Mats Thulin first visited the region in October 1992 (with Eriksson, Al Gifri and Långström) and the second visit was in November 1998 (with Beier and Mohammed Hussein). In his first visit he collected 412 plant specimens from different parts of Hadhramaut region. In his second visit to the region, he collected over 100 specimens. In his 2 visits, he covered several places from the coastal plains to the northern plateau along such as Al Mukalla, near Riyan and 25 km W of Al Ridah,Tarim, Shibam and finally along Wadi Hajr. His expeditions resulted in the discovery of several new species and greatly increased our understanding of the distributions of many taxa. The results of these trips are written up in various journals and contain important descriptions of the vegetation. (see Thulin, 2001 and 2002; Thulin and Al Gifri, 1993 and 1995). In 1985, 1992, 1993, 1995 and 2001 a local Taxonomist, Prof. Abdul Nasser Al Gifri, visited Hadhramaut several times, in which he studied the vegetation and flora of Hadhramaut region. On some of his trips he was accompanied by Dr. Lutfi Boulos of Kuwait University (1985) and Prof Mats Tthulin of Uppsala (1992), with Harald Kurschner of Frie Berlin Uni. (1993) and with Lavranous the Arabian succulent expert (1995). The results of these trips are written up in various journals (see Thulin and Al Gifri, 1993 and 1995; AI-Gifri and Kurschner, 1996; Gabali. and AI Gifri, 1990; and Lavranous and Al Gifri, 1999). From 1996 to 2002 the German botanists Norbert Kilian and Peter Hein (Kilian et al., 2002) have made a series of expeditions to Hadhramaut region, during which they collected a considerable number of plant species and added new records of plant species to the region. Their trips included the southern summit plateau above al Mukala and al Rayan. The specimen collections were deposited in the herbarium of the Botanic Gardens and Botanical Musium Berlin-Dahlem, Germany. Their work has greatly contributed in understanding the distribution of many endemic, near-endemic and rare species and made a valuable contribution to appreciating the importance of the southern part of the region. 54 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi Other local botanists included Al Gabali, Bataher, Bilaidi, Obadi, Bazara’a, and Al Dubaei, who are conducting very limited surveys on vegetation of the Hadhramaut; they mainly joined in the research of other international botanist and explorers. 2.3. Vegetation studies in Hadhramaut There have been two published descriptions of the vegetation of Hadhramaut:1. A study made under the auspices of ACSAD (The Arab Center for the Studies of Arid Zones and Dry Lands). In 1986 a study team from ACSAD, Syria and the Ministry of Agriculture from the former South Yemen made a comprehensive collection of plants. In total 72 sites in 4 governorates (Lahj, Shabwa, Hadhramaut and Al Mahara) were visited and 500 specimens of 280 plant species belong to 50 families and 157 genera were collected. Brief information on location, topography, soil and altitude at each site was collected. The resulting report (Barkuda and Sanadiqi, 1986) contains a map showing the location of the sites visited. However, the location of the sites is not precisely given and in consequence the collections are of limited value for vegetation description or mapping. Despite the incorrect and incomplete identification of some species, the study has proved valuable, especially in increasing our understanding of the distribution of species in southern and eastern Yemen. Interestingly the invasive plant Prosopis juliflora was recorded only from the coastal plain and not from the Wadi Hadhramaut, which show that this species probably was not existence in the study area in 1985. Specimens were deposited in the following places: Agriculture Research & Extension Authority, Al Kod branch, Yemen International Central for Genetic resources, Kew, London. ACSAD Damascus, Syria 2. A survey of the vegetation of Hadhramaut, Yemen (Gabali and Al Gifri, 1991). This study lists dominant and common species and describes some common plant associations. It divides the region into five main ecological zones based on their topography, soil, and climate. It includes a comprehensive review of botanical collecting in Hadhramaut. However, there are significant drawbacks. There is no description of the methodology used, the sites visited are not locally identified and 55 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi many species are only identified to genera (e.g. Ficus spp, Cymbopogon spp., Cassia spp., Barleria spp. etc). 2.4. Botanical Literature The above outline of the botanical exploration of Hadhramaut shows how an understanding of the floristic composition of the region was gradually built up by a series of rather unfocussed expeditions spanning some 120 years. However, they did little to improve our understanding of the composition of the vegetation and there is still no comprehensive description of the vegetation. To build up a picture of the vegetation and to place it in a regional context, I have had to rely on publications from surrounding regions. The following proved most useful. 2.4.1. The Vegetation of the Republic of Yemen (Western Part) (Scholte, Al Khulaidi and Kessler, 1991). This publication is the result of a joint study undertaken by DHV, International consultancy and engineering from the Netherlands and the Yemen Government (Environmental Protection Council and Agricultural Research Authority). It comprises a vegetation map at a scale of 1:500,000 and a photographic description of the main vegetation types. The map covers most of the western part of Yemen of what was formally called the Yemen Arab Republic. The study involved both extensive field surveys over most parts of the western part of the country and an analysis of available satellite imagery, in particular SPOT images, The booklet begins with an identification of the eight main vegetation landscapes in Yemen. These are defined as following: 1. Tihama Coastal Plain (< 400 m) 2. Tihama Foothills and Low Altitude Western Mountains (< 1000 m) 3. Medium Altitude Western Mountains (1000-1800 m) 4. High Altitude Western Mountains (1800 m) 5. Highland Plains (2000 m) 6. High Altitude Eastern Mountains and Highlands (> 1800 m) 7. Medium Altitude Eastern Mountains (1200-1800 m) 8. Eastern Desert Plain (< 1400 m). 56 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi Each major zone is further subdivided into separate ecological zones. For example, the Tihama coastal plain consists of: T1 as mangrove Avicenna woodland; T2 as sabakhas Sueda sparse dwarf-shrub and bare land; T3 as palm groves, Phoenix-Salvadora as woodland; T4 as salt-bush lands, Salsola-Odyssea dwarf-shrub land; T5 as Panicum sparse grassland and bare land; T6 as Dactyloctenium cultivated land (irrigated); T7 as Ziziphus-Dobera cultivated land (mainly rainfed); T8 as Acacia-Commiphora, open woodland and bare land. Each of these ecological zones is described in general terms, together with information on the importance of the vegetation, for instance, as a resource for forage, timber and firewood together with comments on sustainable management. The vegetation structure of each unit is illustrated by a photograph, from which the main species can be identified by an accompanying black and white sketch of the layout of the photograph. The geography of each ecological zone is described in terms of physiography, geology, climate and the main vegetation types. The problems facing each zone for instance, over-grazing, cutting of trees for firewood and timber and increasing cultivation are identified. These problems are graphically illustrated by a series of pictures showing examples of healthy and problematic sites across the different ecological zones. Recommendations are also given for future management strategies. The work provides a basic source of material for vegetation studies and a comparative base line for similar studies in the rest of Yemen. Of significant importance is the vegetation map at a scale of 1:500,000. The work has been and is still used as one of basic and important references for biodiversity and forest activities in Yemen. The work was prepared before the unification of Yemen (May 1990) and the fieldwork has only carried out in the former Yemen Arab Republic. No surveys were undertaken in Hadhramaut, and the eastern part of Yemen is not covered. However, this 57 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi publication has proved invaluable in helping place the vegetation of Hadhramaut in a regional context. 2.4.2. Flora of Republic of Yemen (Al Khulaidi, 2000). A comprehensive checklist of the Flora of Republic of Yemen was compiled by EPA (Yemen Environmental Protection Authority) and UNDP (United Nations Development Programme). This work was made possible through financial support of the Sustainable Environmental Management Program, Yemen/97/100. The work is based on nine main studies on the flora of Yemen carried out between 1975 and 1997 (Boulos, 1988; Diccon, & Miller, 1998; Gabali and Miller, 1992; Gabali, 1998; Hepper, 1975; Mies, 1994; Miller and Cope, 1996; Wood, 1997 and Cope, 1985) and on extensive fieldwork throughout Yemen carried out by the author. In total 2810 species, belonging to 1006 genera and 173 families are listed for Yemen, of which some 415 are endemic, including 236 confined to the Soqotra archipelago. Under each species entry there are also notes on the distribution of species within Yemen, local names of many species and other references. This is the first checklist to cover the whole of Yemen and is regarded as an important resource for any environmental studies in Yemen. A similar study was carried out in the former South Yemen, (Boulos, 1988). This publication was the earlier checklist of flora for the former south Yemen followed by a study written by Gabali and Miller (1992). This was a provisional checklist of former South Yemen prepared as a draft. The checklist was based on 10 studies carried out between 1939 and 1990. The former checklists of the Flora of Yemen have contributed greatly to our knowledge of the flora of the Hadhramaut region and have helped to build up understanding of the flora of Hadhramaut. 58 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi 2.4.3. Flora of the Arabian Peninsula and Socotra Volume 1 (Miller, A.G. & Cope, T.A. 1996). The Flora of the Arabian Peninsula and Socotra is the first comprehensive study covering the flora of the entire region. It will appear in five volumes; so far only volume 1 has been published (Miller and Cope, 1996). The Flora covers the entire native flowering plants and ferns, as well as major cultivated and economic plants. It includes Yemen, Saudi Arabia, Oman, the United Arab Emirates, Qatar, Bahrain and Kuwait. Volume 1 contains introductory chapters covering topography, geology, climate, vegetation, floristic, phytogeography, a history of botanical exploration and conservation. The second part of the volume consists of accounts of 63 families. A brief description of families and genera are included, as well as distribution dot maps for all species, the book is comprehensively illustrated and contains keys to genera and species. It is an organized, excellent and useful study, which is set to become the essential reference source on the Arabian flora. On the down side, local names of the plant species are not included, there are no colour plates and, so far only volume 1, covering only 63 families, has been published. Unfortunately, the Flora is very expensive and because of this it is not readily available to Yemeni botanists. 2.4.4. Vegetation of the Arabian Peninsula (Ghazanfar and Fisher, 1998) The Vegetation of the Arabian Peninsula contains the most recent vegetation study, which covers part of Hadhramaut region. It is the first comprehensive overview of the, phytogeography and vegetation of the Arabian Peninsula. It is written and edited by 15 specialists on the botany and environment of the Peninsula. It covers Yemen, Saudi Arabia, Oman, the United Arab Emirates, Qatar, Bahrain and Kuwait and contains 11 chapters covering climate, geology, phytogeography, vegetation, and ecology of the mountains, wadis, sand 59 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi deserts, gravel plains, coasts and sabakhas (an Arabic term for a marsh or bare mud). It includes chapters covering the flora and vegetation of all the major habitats and landforms. The chapters on climate and geology provide important background information for understanding the dynamics and history of the Arabian vegetation. The book is an important reference source for students, academics and scientists interested in Arabian botany. On the downside it does not cover the vegetation of Soqotra archipelago. The description of the vegetation of Yemen, in particular the western mountains areas, is based mainly on the studies of Deil and Muller-Hohenstein (Muller-Hohenstein and Rappenhoner, 1991; Muller-Hohenstein, 1984, 1986, 1988, 1989; Deil 1986 and 1991, Deil and Muller-Hohenstein, 1988, 1991; Deil & Rappenhone, 1988, Deil and Al Gifri, 1998) and does not take into account the results of ecological studies carried out by competent bodies in Yemen such as AREA (Agricultural Research and Extension Authority), Land and Water Conservation Project, EPA (Environmental Protection Authority), which contain a wealth of detailed rangeland vegetation studies, based on systematic methodologies such as Braun-Blanquet and supported by land ecological big scale vegetation maps, such as Wistinga and Thalen, 1980; Kessler, 1987 and Al Khulaidi, 1989.. 2.4.5. A Handbook of the Yemen Flora (J.R.I Wood, 1997) A handbook of Yemen Flora was published in 1997. The book contains an account of all native and naturalized species of fern and flowering plants, found in the former Yemen Arab Republic, In addition, descriptions of all commonly cultivated species are provided. It also contains 40 colour illustrations painted by Hugo HaigThomas. The book starts with a chapter on the discovery of the Yemen flora, giving information on how Arab and European scientists and writers became interested in discovering the Yemen flora. 60 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi Wood based this work on his extensive collections made during visits to of all parts of the country during his stay in Yemen during the years 1974-1981. The text was finalised in 1982, but unfortunately not published until 1989. The book is of importance not only to botanists, but also to all those with an interest in the plants of Yemen, including rangeland ecologists, foresters, pharmacists and agriculturalists. Before the publication of this book, there was no complete Flora covering the Yemen. Even though it is intended to cover only the plants of the former north Yemen, it can also be used to identify a significant proportion of the plants of the former south Yemen. The identification keys in the book are based mainly on technical characteristics, and so can only be used by those with some basic botanical knowledge. Few illustrations (apart from the colour plates of some 40 species) are included but where they are, for instance line drawings of the leaves of the critical genera such as Grewia and Indigofera, they are particularly useful. This Flora is most successful used in Yemen today. The Flora also contains an index to genera as well as of Arabic local names. Perhaps its strongest and most useful feature is the very detailed notes given on the distributions of individual species which often proved very helpful in identifying unknown plants. Unfortunately former South Yemen is not included, making the title of the book, published eight years after the unification, misleading. Many plants are unknown outside southwest Arabia and most of them have never been described before. Generally the accounts are very accurate, however difficulties have been experienced in critical genera such as Maytenus, Grewia and Commiphora. Unfortunately, the book is very expensive and so not widely available in Yemen. A flora covering the whole Yemen is urgently recommended. This can be done with the cooperation of Environmental Protection Authority (EPA), Agricultural Research and Extension Authority (AREA) and University of Aden. 61 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi 2.5. Paleoenvironmental studies In the past, distributions of plants have shifted dramatically in response to changes in climate (USGS, 2002). Understanding the way plants have responded to past climate change can help predict the way vegetation will respond to future changes. Paleoenvironmental studies help us understand the causes of climatic change as well as providing information on the present day nature of the plant cover. It does this by looking a range of lines of evidence including offshore deposits, sea level change, cave deposits, palynology, isotopes, geochemistry, sedimentology, paleontology, etc. (Last and Smol, 2005). It helps to reconstruct past environments using fossil materials of plants and animals, or other indicators of past environments (USGS, 2002). Unfortunately, due to the unsuitable conditions for the preservation of fossils and pollen over most of Arabian Peninsula, there is little direct evidence for reconstructing the past vegetation and climate of the region. However, evidence from a variety of studies from surrounding areas, in particular north and east Africa, helps us to paint a picture of the past climate and vegetation of the Arabian Peninsula. It is believed that Arabia experienced its main floristic invasions during the Tertiary (Ghazanfar & Fisher, 1998). The paleo-African vegetation presumably extended eastwards into the Arabian Peninsula throughout the middle-late Eocene and Oligocene (Mandaville, 1990). Fossil pollen and spore samples from the eastern Rub al Khali, were found to be dominated by the unclassified fern genus, Psilatricolporites with pollen showing species of Myrtaceae and Palmae and spores of the water fern Ceratopteris also present indicating a much wetter climate in central Arabia than that of today (Mandaville, 1990). The palaeoecological atlas of northern and western Africa in the Middle Holocene (8000 BP) (Agwu and Beug, 1982) shows a mosaic of vegetation with the Mediterranean region enlarged to the south and characterized by deciduous and coniferous forests. The favourable climate in the northern part of Africa and Arabia lasted for about 8000 years (between 14000 and 6000 years ago). According to the maps presented by Adams (2002), the climate in these regions began to become dry again some 5000 years 62 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi ago. During this period a tropical extremely arid desert expanded and covered almost the whole of the Arabian Peninsula. According to Schroder (2002) and McCorriston, 2000), 10000 to 5000 years ago the Southern Arabia was moister than today. The maps presented by Jallow (2002) showed that about 8000 to 5000 years ago, the Sahara desert receded to small isolated areas in northern Africa and Lake Chad and the forest areas were much larger. About 20000 years ago the Sahara had expanded southward into the current Sahel zone and rivers in this region dried up. The climate began to improve (wetter) about 15000 years ago; this improvement lasted for about 5000 years. Generally, for most of geological time, the Arabian Peninsula and Africa had a common tropical flora and vegetation, allowing the migration and exchange of the palaeo-African, Indo-Malayan and Mesogean floras across Arabia (BIOLOG, 2004). Then, after the Arabian Peninsula separated from the African continent during the Oligocene (Meister et al., 2002) Arabia was isolated from the Africa. The reconstruction of biomass from pollen and plant macrofossils in Africa and the Arabian Peninsula showed only minor changes, compared to the present day for eastern and central Africa and major changes for the region north of 15º N, with steppe in many low elevation sites that are now desert in the Saharan mountains (Jolly et al, 1998). The investigation of 320 charcoal samples from prehistoric sites in the Eastern Sahara furnishes evidence for a fundamental change of vegetation during the early and middle Holocene (Neuman, 1989). Pollen records from a crater lake in the Sudanian zone of northeast Nigeria provide evidence for the persistence of woodland savanna throughout the Holocene. Wetter conditions from c. 10000 B.P. to c. 6800 B.P. enabled the establishment of a dense Guinean savanna, though the occurrence and rapid spread of the montane element Olea hochstetteri indicates cool climatic conditions prior to c. 8800 B.P. (Salzmann, 2000). The changes in the region have been summarised by Adams (2002). According to Adams (2002), the climate history of Africa, Asia and Arabia during the last 150,000 years was as follows (see Table 2.1): 63 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi The period 110000-11000 years ago 43000-40000 years ago 28000-25000 years ago 23000-14500 years ago 13800-13500 years ago 12000-11000 years ago 10000 years ago 9000-8000 years ago 8000-7000 years ago 7500-6000 years ago Since 5000 years ago The climate Drier and cooler than present extreme arid with extensive dune & desert moister than at present aridity and cooling and partly forest increase in temperature and moisture forest had returned moister, more rain than present with pluvial episode moister with forest moister than at present the Sahara was vegetated , rainforest was greater similar to the present Table 2.1. The climate history of Africa, Asia and Arabia during the last 150,000 years Source: Adams (2002). Paleoenvironmental studies in Hadhramaut Earlier in this section it was stated that, due to the unsuitable conditions for the preservation of fossils and pollen over most of Arabian Peninsula there is little direct evidence for reconstructing the past vegetation and climate of the region. However, an important piece of research by Dr Kenneth Cole has now revolutionised palaeoenvironmental studies in Hadhramaut (Kenneth et al., 2001). Cole discovered that middens made by packrats in Colorado, USA could be dated back to 50,000 using radiocarbon dating. The fossils contained with the middens preserve a record of the plants and animals that lived within a packrat’s range (usually up to 100m) and provided a powerful tool for reproducing past biotic communities. He was able to use a similar technique to age hyrax middens in Hadhramaut. The background to his research in Arabia is a result of the archaeological importance of Hadhramaut and an active programme to understand the recent vegetation history of the region. This research is coordinated through the Roots of Agriculture in Southern Arabia (RASA) program, headed by Dr. Joy McCorriston, the Yemeni Ministry of Antiquities, and the American Institute for Yemeni Studies (AIYS). One of the most interesting aspects of this research is the way that fossil hyrax middens have been used to reconstruct paleo-environments in Hadhramaut (Kenneth et 64 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi al., 2001). The task of this project is to reconstruct the vegetation history through plant macrofossils and pollen contained within hyrax middens in order to understand and to reconstruct the past vegetation and climate and the vegetation and the climatic changes taking place in the area and the possible impacts of domesticated animals (primarily goats and camels) on the vegetation. The preliminary results of vegetation at their survey site in Wadi Sana (east of the present study area) indicates that the area is currently dominated by Ziziphus leucodermis, Acacia hamulosa, and A. ehrenbergiana (Kenneth. et al 2001). Their studies have shown that Acacia ehrenbergiana, which today is found throughout Yemen, did not appear in Wadi Sana until 2159 yr B.P. They interpret this as indicating a return to moister climates following an extremely arid period in the late-middle Holocene. The study also suggests that the period of maximum Holocene aridity in the southern Arabian highlands may have been between 5000 and 2500 yr B.P. Archaeological sites from the study area (Kenneth et al., 2001) suggest that in the early mid Holocene (7500 to 6000 years ago) the climate was wetter than at present and that the climate of the entire Middle East would have been more favorable to plant growth from a period beginning 13,000 BP to about 9000 BP. Fieldwork for this remarkable new study began in February-March, 1998, and is still ongoing. Their final results should give valuable information on the past vegetation and climate in the region and act as a baseline for the vegetation and the climatic changes taking place in the study area at present. Climate change in Arabia The Intergovernmental Panel on Climate Change (IPCC) project that annual temperatures in the Arabian Peninsula will increase by 1-2°C and precipitation will increase slightly (<0.5 mm/day) throughout the region by 2030-2050 (1PCC, 1996). Furthermore, it predicts that soil moisture will decrease because the insignificant increase in precipitation will be counteracted by the increase in evaporation which will come with rising temperatures - for western Asia, reductions between 2% and 10% are expected (Dai et al 2001). More recent projections (IPCC, 2001) give increases in global mean temperatures in the range of 2 to 4°C by the end of this Century. 65 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi Thus, despite uncertainties and a lack of direct evidence, there seems to be agreement on considerable change over the Quaternary in the Hadhramaut region, particularly in the early and middle Holocene. Moister conditions would have encouraged greatly plant growth, species diversity and relict elements of which can be seen today (e.g. Acacia spp.) in much drier environment. 2.6. Vegetation maps Several maps which cover Hadhramaut have been produced. However, all are to a lesser or greater degree inaccurate and none have been based on fieldwork in the region. Most have been apparently based on information collected from Blatter and Schwartz. (described in section 1 above). - A map scale 1:100 000 showing the woodland types of some areas in Hadhramaut has been produced (Hunting Technical Services Limited, 1992). These types are presented in format of the dominant species, the association species and a general description. A drawing of a vegetation profile and associated species are also described as a characteristic of the class. The report shows also a photograph of this characteristic example. The following sites have been described:    Western branch of Wadi Bin Ali near Shibam as Acacia open woodland type, Wadi Hadhramaut and tributary wadis (e. g. eastern side of Sayun, wadi Al A’yn, near Tarim as date palm agriculture type, South of Tarim, on terrace of wadi Masilah main channel as Tamarix woodland type. The dominant species, density, canopy cover, height range and thickets of multistemmed of each type were mentioned. The description and illustrations of the woodland types as they appear on the ground were presented, mainly in a format of the dominant trees, the associated species and a general description followed by descriptive statistics of density, stem density, canopy cover, height range, stem diameter and bole length. A drawing of a vegetation profile for the characteristic example of the type was given, with photographs showing the characteristic example and other sites displaying some of variation in the type. 66 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi 2.6.1. Drawing up a preliminary vegetation map of Arabia (Novikova. N.M. 1970). The map was based on analysing and studying the publications of botanists and travellers in the Arabian Peninsula. Based on the association of the main plant communities and the habitats, the author divided the Arabian Peninsula into 10 physiogeographic areas. According to this map the following vegetation communities were described from Hadhramaut: Ficus sycomorus woodland (15), part of the footslopes near the coastal areas. Dipterygium glaucum stony-rock Hamada (30) most of the plateau Leptadenia pyrotechnica stony Hamada (31) Tribulus longipetalus high ridge sands(32a), the northern desert area Tamarix spp. Wadi type (37) Wadis cutting the plateaus Acacia spp. in Wadis (38) Boswellia carterii woodland (20) Commiphora opobalsamum woodland (19). Some of the plant species mentioned in his communities are not common in the Hadhramaut. For example, Ficus sycomorus is rare in Hadhramaut but abundant in wadis in Taiz, al Dhalaa, Ibb, Lahg, Hajjah, Abyen, Huf, (on the SW escarpment mountains of Yemen) and in the monsoon woodlands in al Mahara (on the Oman border) where the rainfall is high. Dipterygium glaucum is not common on the plateau. This species forms a community on the desert of western part of Hadhramaut and is thinly distributed on most of Wadis in Hadhramaut region (Gabali and Gifri, 1991). The coverage of Hadhramaut is in general very patchy and cannot be relied upon. 67 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi 2.6.2. Middle East Vegetation Map, scale: 1: 800 0000. (Compiled Tavo, University of Tubingen, 1989). The map divided the study area into :1- mixed formation of herbaceous and semi-woody salt-swamps and coastal dunes (the coastal area) 2- xeromorphic very open scrub (semi-desert shrublands) and open grassland. (the footslope) 3- thorn woodlands (the southern plateau) 4- mixed formation of sclerophyllous-rich extremely xeromorphic woodlands, succulent scrub and xeromorphic open grassland (the middle of the plateau) 5- dry deciduous extremely xeromorphic woodlands (the middle and the northern part of the plateau) 6- mixed formation of xeromorphic very open scrub (semi-desert shrublands), (north Eadi Hadhramaut ) 7- mixed formation of xeromorphic very open dwarf shrubland (the desert) 8- Scarce vegetation of sand dunes and desert (al Ruba al Khali). Scientific name of the plant species were not mentioned. In general, it is good source of information on the basic vegetation structure of the communities in Hadramaut. The map was not based on field work. 2.6.3. Geobotanical Outline map of the Middle East (Zohary, 1973). The map divided the Middle East to 8 vegetation types; and classified the Hadhramaut region as Sudanian and sub-Sudanian vegetation (Tropical deserts, savana and forests) in which it is divided to the following subunits:  subdivision Acacietea tortilis ssp. sub-sudanica, this covers the study area subdivision Savana of Acacietea sudano-arabica (mainly Acacia – Commiphora scrub) and covers south-facing slopes of Hadhramaut  region and subdivision mosaics of Haloxylotea salicornici and Suaedetea desert covering the coastal area of Hadhramaut region. 68 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi The map is based on literature review of old travellers to Arabia, such as Deflers (1896), Schwartz (1939) and Wissmann (1968). According to this map, the region is classified as Acacietea tortilis ssp. subsudanica. This class comprises mainly a Saharo-Arabian with few Sudanian species (Zohary, 1973). This vegetation type is confined to wadis or depressions in several regions such as Egypt, S. Sinai, S. Palestine, the coastal of Arabian Peninsula and S. Eilat, Israel. In the study area there are few of the leading plant species of this class, namely Acacia tortilis (in our case the very similar species A. campoptila), Ziziphus spinachristi, Balanites aegyptiaca, Ochradenus baccatus, Citrullus colocynthis, Cleome droserifolia, Panicum turgidum, Iphiona scabra, Zygophyllum coccineum, Salvadora persica, Moringa peregrina, Capparis cartilaginea, Calotropis procera, Tephrosia apollinea and T. nubica. This type of vegetation was called by Zohary (1973) as pseudo-savana. Compared to the previous maps, this map gives a more accurate reflection of the composition of the vegetation and the phytogeographical classification of the study area. 69 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi 2.7. Summary Most of the mentioned studies are little more than annotated checklists and the botanical exploration of Hadhramaut is essentially a history of botanical collectors and their publications, which has limited scientific value for the study area. Nevertheless, during the last ten years a series of collectors and botanists have visited the region. Important collections have been made and these trips have resulted in the discovery of several new species and greatly increased our understanding of the distributions of many taxa. The results of these trips are written up in various journals (see the references). However these studies, including the previous preliminary flora checklists of the southern Governorate (formally south Yemen), have given incomplete inventories of the plant species of the Hadhramaut region. The botanical exploration of Hadhramaut shows how an understanding of the floristic composition of the region was gradually built up by a series of rather unfocussed expeditions spanning some 120 years. However, they did little to improve our understanding of the composition of the vegetation and there is still no comprehensive description of the vegetation. Few vegetation maps which cover the Arabian Peninsula and include the study area have been produced. These maps are not accurate and were based on the information of previous collectors and were not based on field trips to the region. To build up a precise and accurate picture of the vegetation of the Hadhramaut region, a reconnaissance survey of almost the entire region was conducted fallowed by detailed vegetation surveys of 3 study sites in the next chapters. 70 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi Chapter 3: The vegetation of the Hadhramaut 3.1 Introduction As indicated earlier the Hadhramaut region is too extensive a region (161,749 km²) to allow for a detailed floristic survey within the scope of the present research. Consequently, and in view of the near-absence of previous quantitative work on the topic, it was decided to undertake a preliminary reconnaissance survey, which would provide on overview of the phytogeography and permit the selection of representative study sites for more detailed research. This chapter presents the results of this reconnaissance survey of the vegetation of the Hadhramaut region. Two transects were chosen, starting from the coastal area in the south and ending towards the west. Sample points were established across transects, one from the south-east (coastal area) to the north-west (up to Wadi Dawa’n) and the other from the east (near to Shibam) towards the west (Ramlat Assaba’tain) (Figure 3.1). A comparison was also made of the vegetation habitats which were similar to those found in the study area. 3.2. Methodology The two transects were selected to represent the main landscape features and associated vegetation of the Hadhramaut region. The two transects passed along the only accessible roads in the Hadhramaut connecting south with north (transect 2) and west with east (transect 1). Practical considerations, such as access issues, species richness, and travel time were taken into account when the two transects and the sampling points were selected. Vegetation composition and structure were studied across the two transects along an altitudinal gradient, from the coastal area to the Wadi Dawa’n at altitude of 1000 m asl, then from west of Shibam towards Ramlat Assaba’tain at altitude of about 770 m asl (Figure 3.1). The areas to be sampled were selected on the basis of altitude and vegetation changes; one 10 X 10 m sample site was selected at approximatley every 100 m asl. Natural 71 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi vegetation and abundance data were collected using Braun-Banquet’s cover scale (See Chapter 4); slope percentage, exposure and stoniness were also recorded. The sites along the two transects and their relationships to environmental factors were analysed by Canonical Correspondence Analysis (CCA), using the statistical software of MVSP version 3.1 (Figure 3.5). Data were available on: altitude, slope, aspect, stoniness, total vegetation cover % and the total number of species. The location of the sites and the endemic, near-endemic and rare plant species were plotted using ArcView and Diva-GIS softwares. Figure 3.1. An image showing the transects from the coastal region to the plateau in the northern part of the Hadhramaut region and from (Shibam) towards the west (the desert of Ramlat Assaba’tain). 72 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi 3.3. Climate The difference in altitude across the different topographical units of the two transects strongly influence climatic conditions. Average temperatures decrease more or less linearly with altitude; temperature was taken to be 1 degree C colder for every 100 m of altitude. The orographic rise of air masses from the onshore wind provides an effective cooling mechanism, which causes rainfall. For example, areas exposed to the sea such as the southern slopes, receive more rainfall than the zones facing the interior plateau. Local topographic features cause similar effects, at correspondingly smaller scales (Ayele and Al Shadily, 2000). A considerable amount of moisture can be added to the ground flora from fog, which affects the escarpment and highest points on the plateau. Some of the richest areas for endemics are found on these escarpments. There is also an important difference in timing of rainfall from the north (Wadi Hadhramaut) to the south (coastal area); the rainfall in the north falls in spring and summer, while in the coastal area it falls mainly in spring and winter. Wadi Hadhramaut is affected by both the Red Sea Convergence zone (RSCZ), which contributes to the first spring rainy period between March and May, and by the Inter-Tropical Convergence Zone (ITCZ) which contributes to the second summer rainy period between July and September, while the coastal area (Al Rayan) is affected mainly by RSCZ (Figure 3.2). The climate along this transect varies from a mean annual rainfall of about 69 mm at Al Rayan on the coastal area to a mean annual rainfall figure of less than 75 mm at Sayun in the interior. There are no meteorological data over the plateaus; nevertheless the abundance of floods in the secondary wadis flowing across the plateaus suggests that the annual rainfall over these landforms is more than 200 mm (SOGREAH 1979) but highly sporadic. The climate along the second transect varies from a mean annual rainfall of about 74 mm to extreme (hyper-) aridity with a mean annual rainfall of less than 50 mm. 73 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi F Figure 3.2. A climate diagram of mean monthly rainfall (the lower line) and mean annual temperature (the upper line) for Sayun (Wadi Hadhramaut, north) and Al Rayan (coastal area, south). The rainfall in the north falls in spring and summer, while in the coastal area falls mainly in spring and winter. 3.4. Topography The transect from the coastal plain (south) to Wadi Hadhramaut (north), then towards Ramlat Assaba’tain (west Wadi. Hadhramaut) traverses four main topographical units. These units can be broadly described as the coastal plain, the southfacing escarpment, the Jol plateau and the alluvial wadis. 3.4.1. The coastal plain: This is an undulating to almost flat plain, cut by shallow drainage lines. The plain consists of gravels and alluvial deposits with dolomite rocks (Vogt and Sedov 1997). 3.4.2. South-facing slope escarpments: These comprise s-facing mountainous areas which drop from 1500 m to the coastal plain. Cliffs and steep slopes are the dominant features on these mountains. 3.4.3. The plateau (Jol): This plateau consists of rocks of Oligocene-Miocene age (USSR and PDRY, 1984) It is an undulating to almost flat limestone plateau, dissected by numerous deep secondary wadis (gullies and canyon-like secondary wadis). The altitude gradually reduces towards the north and ranges between 900 and 1500 m above sea level. The plateau is dissected by Wadi Hadhramaut and Wadi Masila (MAW, 1996). The altitude of the southern Jol ranges between 860 and 1580 m and is higher than the northern one which rarely exceeds 1200 m. 3.4.4. The wadis: The main wadis are Hadhramaut and Al Masila. Wadi Hadhramaut divides the plateau area into two, northern and southern. Wadi Hadhramaut runs nearly 74 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi 165 km into Saihout in the al Mahara Governorate, and is located at latitude 16 degrees north and longitude 48 degrees east (Verba et al., 1995). Upstream Wadi Hadhramaut opens westward into a wide desert plain called the Ramlat Assaba’tain. Wadi Hadhramaut is considered to extend between Wadi Saar (west) to west of Tarim (east) (SOGREAH, 1979). The wadi bottom is mostly flat with steep to vertical sidewalls. The valley varies in width, from 1.5 km in the east to 6 km in the west and lies about 300 to 400 m below the level of the plateau. 3.4.5. Sand dune area: An extensive expanse of sandy desert, known as the Ramlat Assaba’tain, is found west of Wadi Hadhramaut. The altitude of the Ramlat Assaba’tain gradually increases from almost 700 m in the east to more than 800 m in the west. 3.5. Results 3.5.1. Vegetation composition of the 2 transects The following section describes the main vegetation composition of the two transects made across the Hadhramaut region. A comparison of the vegetation types found in Hadhramaut with those of the surrounding regions will be also described. The first transect extends from the east towards the west. Generally the vegetation cover towards the west is poor and sparse and gradually changes from woodland, characterised by Acacia campoptila and Fagonia indica, to sparse dwarf shrubland or grassland dominated by Zygophyllum album and Dipterygium glaucum. The second transect extends from the coastal region to the plateau in the northern part of Hadhramaut region. The altitude here varies from sea level in the south to about 1600 m above sea level then to about 600 m above sea level at Wadi Hadhramaut in the northwest of the region. Species richness and species diversity dramatically decrease from east towards the west (transect 1).. The species richness and species diversity of transect 2 increases towards the south, in particular south-facing slopes and drainage lines. The vegetation cover here is very rich especially on slopes or wadis facing south, which are sustained by the regular fogs. 75 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi 3.5.1a. East west transect (see Figure 3.3) Figure 3.3. A map showing the locations of the sites along the first transect from east to west. Views from 6 locations are also shown. 76 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi Key 1- Undulating concave flooded wadi, covered by sand dunes (670 m). Cover: Trees 1% and dwarf shrubs 30% (40%) Dwarf shrubland dominated by Zygophyllum album and Dipterygium glaucum, associated with Alhagi graecorum, Tribulus arabicus, Cleome scaposa, Panicum turgidum, Calotropis procera, Tephrosia apollinea, Senna italica, Prosopis juliflora and Tamarix arabica. 2- Fallow lands (703 m) Cover: Trees 55% and dwarf shrubs 2% (57%) Open woodland dominated by Acacia campoptila and Acacia ehrenbergiana, associated with Prosopis juliflora, Zygophyllum album, Capparis spinosa and Prosopis farcta. 3- Undulating to almost flat sand dune area (730 m) Cover: Trees 3% and dwarf shrubs 1% (4%) Open woodland dominated by Acacia campoptila and Rhazya stricta associated with Acacia ehrenbergiana. 4- Almost flat sandy plain near cultivated field (740 m) Cover: Trees 1%, shrubs 3% and herbs 1% (5%). Sparse shrubland dominated by Acacia ehrenbergiana and Zygophyllum album with Acacia campoptila, Rhazya stricta and Prosopis juliflora. 5- Almost flat sand dunes area (750 m) Cover: Trees <1%, shrubs 2%, dwarf shrubs 20% and herbs 3% (26%) Dwarf shrubland dominated by Rhazya stricta in association with Acacia ehrenbergiana and Acacia campoptila. 6- Wide sandy wadi bed, almost flat, probably old fallow land (769 m) Cover: Tree 1%, dwarf shrubs 7% and herbs 15% (23%) Dwarf shrubland dominated by Tephrosia apollinea and Rhazya stricta in association with Acacia campoptila, Panicum turgidum Dipterygium glaucum, Aerva javanica, Tribulus arabicus, Senna italica, and Indigofera sp. 77 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi 7- Almost flat sandy plain (735 m) Cover: Shrubs 15% and herbs 5% (15%) Shrubland dominated by Rhazya stricta with Aerva javanica, Calotropis procera, Prosopis juliflora, Acacia campoptila, Boerhavia sp., Tephrosia sp., Panicum turgidum, Corchorus depressus and Fagonia sp. 8- Almost flat sandy plain (640 m) Cover: Herbs 3%. Sparse shrubland dominated by Rhazya stricta with Acacia campoptila, Aerva javanica and Dipterygium glaucum 78 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi 3.5.1.b. South-east to north-west transect (Figure 3.4) Figure 3.4. A map showing the locations of the sites along the aecond transect from south-east to northwest. Views from 8 locations are also shown 79 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi Key: 9. Wadi Daua’n 1000 m. Gully and dissected limestone plateau with trees restricted to gullies. Cover: Trees 20%, shrubs 5%, herbs 5%, (30%) Woodland dominated by Anogeissus bentii and Ziziphus leucodermis, with Cymbopogon schoenanthus, Cassia sp., Indigofera spinosa, Rhazya stricta, and Tephrosia sp. 10. Dissected plateau 1160 m. Cover: Trees 5%, herbs 15% and dwarf shrubs 7% (27%) Open woodland dominated by Acacia hamulosa and Pulicaria somalensis with Acacia mellifera, Commiphora sp., Cymbopogon schoenanthus, Ochradenus arabicus, Barleria bispinosa, Heliotropium sp., Pavonia sp., Tarenna graveolena, Farsetia sp., Aerva javanica, Periploca visciformis, Corallocarpus glomeruliferus, Sarcostemma viminale, Farsetia dhofarica and Fagonia sp. 11 - Sandy wadi bed in plain (1210 m). Cover: Trees 20%, shrubs 5%, dwarf shrubs 5%, herbs 2% (32%) Open woodland dominated by Acacia campoptila and Grewia erythraea with Acacia hamulosa Solanum schimperianum, Senna italica, S. holosericea, Barleria sp., Indigofera spinosa, Pulicaria somalensis, Ceropegia botrys, Zehneria anomala, Pulicaria cylindrica, Balanites aegyptiaca, Convolvulus glomeratus, Gnidia somalensis var. sphaerocephala and Hibiscus sp. 12 - Rocky flat plain at summit limestone plateau (1500 m). Cover: Dwarf shrubs 10% and herbs <1 (11%). Sparse shrubland dominated by Jatropha spinosa, with Barleria bispinosa Barleria proxima, Barleria sp., Hibiscus sp., Grewia erythraea, Cymbopogon schoenanthus, Gnidia somalensis var. sphaerocephala, Heliotropium strigosum, Ochradenus arabicus, Pulicaria somalensis, Sarcostemma viminale, Tetrapogon villosus, Zygophyllum decumbens and Caralluma sp. 80 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi 13- Rocky almost flat plateau (1496 m) Cover: Trees 2%, shrubs 25% and herbs 5% (32 %). Shrubland dominated by Dodonaea viscosa with Grewia erythraea, Limonium cylindrifolium, Gnidia somalensis var. sphaerocephala, Barleria proxima, Fagonia schweinfurthii, Euphorbia schimperi, Balanites aegyptiaca, Pulicaria somalensis, Euphorbia balsamifera, Aloe sp., and Periploca visciformis. 14 – Rocky secondary wadi on plateau (1344 m) Cover: Trees 5%, shrubs 20%, dwarf shrubs 5% and herbs 10% (40%) Open woodland dominated by Dodonaea viscosa with Dracaena serrulata and Acacia etbaica, Aloe sp., Asparagus sp., Balanites aegyptiaca, Barleria proxima, Barleria sp., Capparis cartilaginea, Ecbolium sp., Euphorbia sp., Euphorbia hadramautica, Fagonia schweinfurthii, Gnidia somalensis var. sphaerocephala, Grewia erythraea, Indigofera sp. Kleinia odora, Launaea sp., Launaea spinosa, Limonium cylindrifolium, Lycium shawii, Maerua crassifolia, Ochradenus arabicus, Periploca visciformis, Pulicaria cylindrica, Pulicaria somalensis, Seddera arabica, Tephrosia heterophylla, Ephedra foliata and Vernonia arabica. 15- Plateau near to the previous valley (1300m) Cover: Trees 1% and shrubs 15%, (16%) Sparse Shrubland dominated by Zygophyllum decumbens, with scattered Euphorbia balsamifera, Pulicaria somalensis, Acacia etbaica, Heliotropium sp. and Acacia oerfota. 17 – Lip of escarpment overlooking coastal plain (856m) Cover: Trees 3%, shrubs 5%, dwarf shrubs 5% and herbs 20% (33%) Open Woodland with Fagonia sp., Aerva javanica, Crotalaria sp., Caesalpinia erianthera, Delonix elata, Reseda sphenocleoides, Adenium obesum, Fagonia sp., Acacia hamulosa, Heliotropium sp, H. longiflorum, Kleinia odora, Commiphora kua, Cometes abyssinica, Convolvulus glomeratus, Aristida sp., Schweinfurthia spinosa, Euphorbia meuliana, Aerva artemisioides, Atractylis kentrophylloides, Blepharis edulis, Crotalaria persica, Convolvulus glomeratus, and Kickxia ramosissima. 18 - Stony slope and drainage line with gully erosion (614m). 81 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi Cover: Trees 2%, shrubs 5%, dwarf shrubs 2% and herbs 10% (17%) Sparse shrubland dominated by Jatropha spinosa, with Fagonia schweinfurthii, Acacia oerfota, Pulicaria somalensis, Sterculia africana, Zehneria anomala, Iphiona anthemidifolia, Solanum schimperianum, Tephrosia heterophylla, Abutilon sp., Aerva javanica, Indigofera spinosa, Acacia mellifera, Commiphora kua, Senna holosericea, Cucumis canoxyi, Commelina albescens, Cadaba longifolia, Commicarpus sp, Vernonia sp., Merremia hadramautica., Reseda sp. and Cleome brachycarpa 19 - Rocky slope and depression (235m) Cover: Trees 1%, shrubs 5% and dwarf shrubs 5% (11%) Shrubland dominated by Pulicaria somalensis and Caesalpinia erianthera with Cadaba longifolia, Saltia papposa, Aerva javanica, Euphorbia schimperiana, Euphorbia schimperi, Commiphora gileadensis, Cleome droserifolia, Fagonia sp., Acacia tortilis, Senna holosericea, Jatropha spinosa, Acacia oerfota, Rhazya stricta, Reseda sp., and Zygophyllum sp. 20- Stony plateau (1600m) Cover: Shrubs 2% and dwarf shrubs 5% (7%) Sparse shrubland dominated by Euphorbia balsamifera with Jatropha spinosa, Ochradenus arabicus, Psiadia arabica, Pulicaria cylindrica, Ruta amoena, Pulicaria somalensis and Kalanchoe caespitosa. 21 - Undulating sand hummocks and runnels on the coastal area (15m) Cover: Trees <1%, dwarf shrubs 20% and herbs 1% (22%). Sparse shrubland dominated by the endemic Limoniastrum arabicum with Cleome macradenia , Odyssea mucronata, Limonium cylindrifolium, Prosopis cineraria, Alhagi graecorum, Aerva javanica, Pulicaria somalensis , Zygophyllum simplex, Jatropha spinosa, Aristolochia rigida, Tribulus sp., Tamarix aphylla, Salvadora persica, Pluchea dioscoroides, Aeluropus lagopoides and Halopyrum mucronatum. 82 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi 22- Drainage line on the coastal plain (15m) Cover: Trees 3%, and shrubs 2% (5%) Open woodland dominated by Tamarix aphylla associated with Aerva javanica, Halopyrum mucronatum, Prosopis cineraria, Pluchea dioscoroides, Salvadora persica, and Aeluropus lagopoides. 3.5.2. Environmental data analysis In Figure 3.5 the main floristic associations and environment variables are presented. The points represent the sites and arrows represent the environmental variables. Figure 3.5.Ordination of the location sites along the 2 transects and environment yielded by CCA. The most important variables on axis 1 are altitude and slope percentage; on the second axis the exposure (aspect) and stoniness are important. Group A is found to the right top of the ordination. The species of this group are shown to grow on almost flat surfaces with low altitude and non or very low stoniness, while group (B), top left, is strongly correlated with stoniness and high altitude with relatively flat to slightly steep slopes. Group (C) is found mainly on south-facing, gently sloping plateaus and secondary wadis. Group (D) is found on moderately steep slopes of secondary wadis and adjacent slopes. 83 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi The most important variables determining changes in species composition in Figure 3.5 are altitude and stoniness. Group A of the second transect (from west Shibam towards the Ramlat Assaba’tain desert) is found towards the right bottom of the ordination. Species within this group, such as Rhazya stricta, Acacia campoptila Dipterygium glaucum, Tribulus arabicus and others, are shown to grow on almost flat sandy surfaces with low altitude and none or very low stoniness, while group (B) top left is strongly correlated with stoniness and altitude with relatively flat to steep slopes; some characteristic species of this group are Acacia etbaica, A. oerfota, Dodonaea viscosa, Dracaena serrulata and Kleinia odora. Group (C) is mainly found on the south-facing, gently sloping plateaus, coastal plains and secondary wadis from the coast to the high altitude mountains. Some species within this group are widespread along a variety of landforms and altitudes but most of them grow well on relatively flat, stony areas. Some characteristic species of this group are Limoniastrum arabicum, L. cylindrifolium, Euphorbia balsamifera, Pulicaria somalensis, Ochradenus arabicus, Grewia erythraea, Jatropha spinosa and Periploca visciformis Group (D) is found on moderately steep to almost flat slopes of the secondary wadis and adjacent slopes at moderate altitudes; characteristic species include Tamarix aphylla, Zygophyllum simplex, Cymbopogon schoenanthus, Ziziphus leucodermis, Senna holosericea, Xerotia arabica, Adenium obesum, Atractylis kentrophylloides and Caesalpinia erianthera. Table 3.1 shows the sample sites with their environmental characteristics and vegetation structure. 84 total vegetation cover % total species Altitude m Slope % exposure Stoniness % Trees % Shrub % dwarf shrub % Herb % Sampling sites total species total vegetation cover % Herb % dwarf shrub % Shrub % Trees % Stoniness % exposure Slope % Altitude m Sampling sites Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi 1 675 1 e 0 1 0 30 0 31 11 12 1530 1 w 50 0 0 10 1 11 15 2 703 1 e 0 5 0 2 0 7 6 13 1500 3 ne 80 2 25 0 5 32 12 3 730 1 e 0 3 0 1 0 4 3 14 1344 2 sw 80 5 20 5 10 40 30 4 740 2 e 0 1 3 0 1 5 5 15 1300 2 sw 95 1 15 0 0 16 6 5 750 2 e 0 1 2 20 3 6 769 1 e 0 1 0 7 7 735 15 e 1 0 0 15 5 20 10 8 766 5 1 1 0 1 1 3 9 1000 20 ne 85 25 5 0 5 35 7 0 7 15 27 19 21 15 2 sw 1 1 0 20 1 22 17 20 5 5 2 22 15 1 sw 1 3 2 0 0 5 e 10 1160 2 nw 60 5 11 1210 2 nw 1 26 3 16 614 3 sw 85 2 5 2 10 19 22 15 23 9 17 856 30 s 18 614 25 se 80 2 5 2 10 19 22 19 235 30 sw 70 1 5 5 0 11 18 4 32 15 80 3 5 5 20 33 23 20 1600 2 se 90 0 2 5 0 7 6 7 Table 3.1. The sample plots with their environmental characteristics and vegetation structure. Figure 3.6 shows the increase of vegetation cover along the altitude gradient. The flat limestone plateau of high altitude areas has a low vegetation cover compared to that of other landforms (e.g. drainage lines and mountain slope) at the same altitude. Figure 3.7 shows the change of number of species (richness) along the altitudinal gradient. The species richness of the plateau at high altitude areas is low compared to other landforms at the same altitude. The low number of species in the drainage sites 9, 10 and 11 results from slope direction (NW and NE) and low moisture content, as these sites are located further north on altitudes less than 1210 m asl, where the rainfall is lower. Other environmental conditions such as landform type, soil and moisture should be taken to consideration when there is an attempt to analyse the change in vegetation cover or richness along altitude gradients of Hadhramaut region. These factors are taken into more detailed study is the analysis of the selected study sites (Chapters 4 and 5) 85 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi Figure 3.6. The figure shows the increase of vegetation cover percentage along the altitudinal gradient of both transects. Compared to other landforms at the same altitude, the vegetation cover of the plateau at high altitude areas was low. Numbers refer to sampling sites. Figure 3.7. The figure shows the change of number of species (richness) along the altitudinal gradient of both transects. Compared to other landforms at the same altitude, the species richness of the plateau at high altitude areas was low. Numbers refer to sampling sites. 86 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi 3.5.3. Threatened plant species in the region. Recently the IUCN Red List Categories and Criteria have been promoted as a widely understood system for classifying species at high risk of global extinction. The general aim of the system is to provide a clear framework for the classification of the plant and animal species according to their extinction risk (IUCN, 1994). However, there is inadequate information to make an assessment of the risk of extinction for the plant species in the Hadhramaut region. Extensive information on the distribution, abundance and population of the plant species, in particular endemic and near-endemic plants, needs to be collected to get a clear idea of just how vulnerable to extinction these species are and then to evaluate them according to the IUCN Red List Categories. Nevertheless, a provisional list of Arabian endemic plants and threatened nonendemic plants was published recently by Miller et al (2006). A provisional evaluation of endemic and near-endemic plant species list of Hadhramaut region (Table 3.2) was obtained based on that work and on the available information from field work and other sources (e.g. Lavranos, 2001, Lavranos and Mies, 2001, Lavranos et al., 2004; Thulin, 2001 and 2002, Thulin and Al Gifri, 1993 and 1995; AI-Gifri and Kurschner , 1996; Gabali. and AI Gifri, 1990; and Lavranous and Al Gifri, 1999 and Kilian et al., 2002). 87 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi Table 3.2. Provisional list of Hadhramaut Endemic and near Endemic plant species with regional and global assessment (after Miller et al., 2006). Endemic and near plant endemic plant species Family Distribution Hadhramaut Acacia campoptila Schweinf. * Aloe abyssicola Lavr. & Bilaidi * Aloe doei Lavr. * Aloe eremophila Lavr. * Aloe fleurentinorum Lavr. & Newton. ** Aloe inermis Forssk. * Aloe luntii Baker * Aloe mahraensis Lavr. & McCoy ** Aloe mccoyi Lavr. & Mies * Aloe serriyensis Lavr. * Anogeissus bentii E.G.Baker * Anticharis linearis (Benth.) Hochst. ex Asch. ** Aristolochia rigida Duch. Asystasia petalidiodes Defl. * Atractylis kentrophylloides (Bak.)F.G.Davis ** Barleria farinosa Defl. * Mimosaceae Amaranthaceae Aloeaceae Aloeaceae Aloeaceae Aloeaceae Aloeaceae Aloeaceae Aloeaceae Aloeaceae Aloeaceae Combretaceae Sapotaceae Aristolochiaceae Acanthaceae Asteraceae (Compositae) Acanthaceae Bauhinia ellenbeckii Harms. Boerhavia elegans Choisy ssp. elegans ** Boscia arabica Pestalozz. ** Cadaba baccarinii Chiov. Calligonum crinitum Boiss. ** subsp. arabicum Campylanthus pungens Schwartz ** Caralluma adenensis (Defl.)Burg. ** Caralluma arabica N.E.Br. ** Caralluma dolichocarpa Schwartz. * Caralluma flava N.E.Br. ** Caralluma foulcheri-delboscii var. greenbergiana Lavr.* Caesalpiniaceae Nyctaginaceae Capparaceae (Capparidaceae) Capparaceae (Capparidaceae) Polygonaceae Scrophulariaceae Apocynaceae (Asclepiadaceae) Apocynaceae (Asclepiadaceae) Apocynaceae (Asclepiadaceae) Apocynaceae (Asclepiadaceae) Apocynaceae (Asclepiadaceae) Aerva artemisioides Vierh. & Schwartz subsp. artemisioides * Yemen Y Arabia H H Assessment Africa Regionally NT Provisional NT Provisional NT Provisional NT Provisional NT Provisional NT Provisional NT Provisional NT Provisional NT Provisional NT Y H Y Y S Y O H H H H Y H O Y H O Y H Y Y Y Y Y Y Y Y Y H globally Provisional NT S, O O O O, U O O O O EN NE Somalia NE NE NE NE Ethiopia Somallia NE EN VU Somalia NE NE CR VU NE VU NE 88 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi Table 3.2. continue Endemic and near plant endemic plant species Family Distribution Hadhramaut Caralluma lavrani Rauh & Wertel * Caralluma penicellata (Defl.)N.E.Br. ** Caralluma quadrrangula (Forssk.)N.E.Br. ** Caralluma shadhabana Lav. & Newton. ** Caralluma subulata (Forssk.)Decne. ** Celtis africana Burm.f. Ceratonia oreothauma Hillc., Lewis & Verdc. ** Ceropegia botrys K. Schuhmann Ceropegia subaphylla K.Schuhmann Cleome hadramautica Thulin * Cleome macradenia Schweinf. * Cleome pruinosa T.Anders. ** Conocarpus lancifolius Engl. Convolvulus sericophyllos T. Anderson. * Corallocarpus glomeruliforus Schweinf. Cryptolepis yemenensis Venter & R.L.Verh. * Cucumis canoxyi Thulin & Gifri. * Cystostemon kissenioide (Delf.)A.Miller & H.Riedl. * Dracaena serrulata Baker. Ecbolium strictum Schwartz * Echidnopsis bentii N.E.Br. * Echidnopsis globosa Thulin & Hjertson * Echidnopsis seibanica Lavr. * Echiochilon arabicum (Schwar.)I.M.Johns. * Ephedra milleri Freitag & Mairer-Stolte. ** Euphorbia applanata Thulin & Gifrii. * Euphorbia fodhliana Defl. * Euphorbia meuleniana O. Schwartz * Euphorbia quaitensis S.Carter * Apocynaceae (Asclepiadaceae) Apocynaceae (Asclepiadaceae) Apocynaceae (Asclepiadaceae) Apocynaceae (Asclepiadaceae) Apocynaceae (Asclepiadaceae) Ulmaceae Caesalpiniaceae Apocynaceae (Asclepiadaceae) Apocynaceae (Asclepiadaceae) Cleomaceae Cleomaceae Cleomaceae Combretaceae Convolvulaceae Cucurbitaceae Apocynaceae (Asclepiadaceae) Cucurbitaceae Boraginaceae Agavaceae Acanthaceae Apocynaceae (Asclepiadaceae) Apocynaceae (Asclepiadaceae) Apocynaceae (Asclepiadaceae) Boraginaceae Ephedraceae (Gnetaceae) Euphorbiaceae Euphorbiaceae Euphorbiaceae Euphorbiaceae Assessment Yemen Y Y Y Y Y Y Y Y Y Y Arabia Africa S S, O S S Africa O S, O Somalia N Africa H Y Y Somalia H Y Y Y Y Y Y O N Africa S, O N Africa H Y H H Y H H H H O O Regionally NE VU LC NE NE NE VU NE DD NE NE NE NE NE VU NE VU NE EN NE VU NE VU LC NE NE NE NE NE globally VU LC EN VU Provisional NT VU Provisional LC Provisional NT Provisional NT Provisional NT 89 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen by Dr. Abdul Wali Ahmed Al Khulaidi Table 3.2. continue Endemic and near plant endemic plant species Family Distribution Hadhramaut Euphorbia riebeckii Pax. ** Euphorbia rubriseminalis S.Carter * Euphorbia seibanica Lavr. & Gifri. * Fagonia hadramautica Beier & Thulin * Farsetia linearis Dene ex Boiss ** Farsetia dhofarica Jonsell & Miller. ** Gaillonia jolana Thulin * Gymnocarpos rotindifolius Petruss & Thulin. ** Halothamnus bottae Jaub. & Spack ** Heliotropium bottae Deflers. * Heliotropium fartakense Schwartz ** Heliotropium makallense Schart. * Heliotropium paradoxum Vatke. * Heliotropium wagneri Vierh. * Heliotropium wissmannii Schwartz * Huernia hadhramautica Lavr. * Indigofera nephrocarpoides J.B.Gillett * Indigofera rubromarginata Thulin ** Iphiona anthemidifolia (Bak.)A.Anderb. * Iphiona senecionoides (Bak.)A. Anderb. ** Iphiona teretefolia A.Anderb. * Justicia areysiana Defl. ** Kleinia deflersii Defl. * Kleinia odora (Forssk.)A.Berger ** Launaea castanosperma F.G.Davies. ** Lavandula subnuda Benth. ** Limoniastrum aramicum J. Edmondson * Merremia hadramautica (Baker) R.R. Mill * Ochradenus arabicus Chaudhary, Hillc. & A.G. Mill.** Ochradenus gifrii Thulin. ** Ochradenus spartioides (Schwartz)Abdulla. * Orbea luntii (N.E. Brown) Bruyns = (Caralluma luntii N.E.Br.) ** Pavonia subaphylla Schwartz * Euphorbiaceae Euphorbiaceae Euphorbiaceae Zygophyllaceae Brassicaceae (Cruciferae) Brassicaceae (Cruciferae) Rubiaceae Caryophyllaceae (Illecebraceae) Chinopodiaceae Boraginaceae Boraginaceae Boraginaceae Boraginaceae Boraginaceae Boraginaceae Apocynaceae (Asclepiadaceae) papilionoideae (Fabaceae) Asteraceae (Compositae) Asteraceae (Compositae) Asteraceae (Compositae) Asteraceae (Compositae) Acacnthaceae Asteraceae (Compositae) Asteraceae (Compositae) Asteraceae (Compositae) Lamiaceae (Labiatae) Plumbaginaceae Convolvulaceae Resedaceae Resedaceae Resedaceae Apocynaceae (Asclepiadaceae) Malvaceae Assessment Yemen Y Y Arabia O Y Y S, O, U O H H H H Y Y Y O S, O, U O H H H H H Regionally NE NE NT NT LC NE VU NE NE LC NE NE VU NE globally LC Provisional NT Provisional NT VU VU Provisional NT Y Y NE O Y O Y Y Y O Y Y S.O, UAE S. O H H H H H H H H H Africa S, O O O O NE LC NE LC Provisional DD Provisional CR VU NE NE CR NE NE NE NE NT NE 90 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Table 3.2. continue Endemic and near plant endemic plant species Pentzia arabica Thulin * Pluchea arabica (Boiss.)Qaiser & Lack. ** Pulicaria cylindrica (Bak.)O. Schwartz ** Pulicaria lancifolia Schwartz. * Pulicaria nivea Schwartz * Pulicaria rauhii Gamal-Eldin. * Reseda sphenocleoides Defl. ** Rhus flexicaulis Baker * Rhus glutinosa A. Rich subsp neoglutinosa Gilbert Rhytidocaulon mccoyi Lavr. & Mies * Saltia papposa (Forssk.) Moq. * Schweinfurthia latifolia (Baker. Ex) Oliver * Schweinfurthia spinosa Miller, Sutton & Short ** Seddera hadramautica R.R.Mill * Sideroxylon mascatense (A. DC.) T.D. Penn. * Stachys yemenensis Hedge. * Stultitia araysiana Lavr. & Bilaidi * Taverniera glauca Edgeworth * Taverniera multinoda Thulin * Taverniera schimperi Jaub. & Spach. * Tephrosia dura Baker Tephrosia hadramautica M.Thulin * Teucrium eximium Schwartz. * Teucrium rhodocalyx Schwartz. * Verbascum luntii E.G.Baker. * Vernonia areysiana Defl. * Xerotia arabica Oliver ** Ziziphus leucodermis (E.G.Baker) Schwartz ** Family Asteraceae (Compositae) Asteraceae (Compositae) Asteraceae (Compositae) Asteraceae (Compositae) Asteraceae (Compositae) Asteraceae (Compositae) Resedaceae Anacardiaceae Anacardiaceae Apocynaceae (Asclepiadaceae) Amaranthaceae Scrophulariaceae Scrophulariaceae Convolvulaceae Sapotaceae Lamiaceae (Labiatae) Apocynaceae (Asclepiadaceae) Papilionoideae (Fabaceae) Papilionoideae (Fabaceae) Papilionoideae (Fabaceae)) Papilionoideae (Fabaceae)) Papilionoideae (Fabaceae) Lamiaceae (Labiatae) Lamiaceae (Labiatae) Scrophulariaceae Asteraceae (Compositae) Caryophyllaceae (Illecebraceae) Rhamnaceae Distribution Hadhramaut H H H H H H Yemen Arabia O O Y Y O H H H H H Africa Ethiopia Y Y O Y Y Y Y Y H H H H H Somalia Y H H S O Asessment Region globally NE NE NE NE NE NE NE VU NE NE NE NE NE NE NE NE NE NE NE NE NE NE NE NE NE NE NE NE Key: IUCN categories NOT EVALUATED (NE) DATA DEFICIENT (DD) VULNERABLE (VU) LEAST CONCERN (LC) NEAR THREATENED (NT) ENDANGERED (EN) CRITICALLY ENDANGERED (CR) Country Y= Yemen, H= Hadhramaut region S= Saudi Arabia O= Oman U= Unite Arab Emirate UAE= united Arab Emirate Note: Some species in the list are found outside Arabia only in Somalia, Ethiopia and North Africa. 91 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 3.5.4. Comparison with previous vegetation studies As discussed in Chapter 2, many vegetation studies have been conducted in different parts of Yemen. Some of these studies were carried out in habitats that are similar to those found in the study area, such as at Marib and the southern and western coastal areas. A resemblance in vegetation composition was found between these areas and the Hadhramaut study area, and the following account indicates some of the similarities. 3.5.4. a. Comparison with previous vegetation studies in Yemen Indigofera spinosa, Jatropha spinosa and Maerua crassifolia form an open woodland and grassland on some isolated foothills of southern and western mountains and on the coastal plains of Yemen (e.g. on wadis, drainage lines, along the edges of the wadi and sand dunes) between 500 and 1900 m (Al Khulaidi, 2000); these areas have a relatively higher annual rainfall than Wadi Hadhramaut (Figure 3.8). Rhazya stricta is restricted to wadi beds and sands of northern east and eastern desert areas of Yemen (e.g. Marib and Rada’) and the desert of Oman (Ghazanfar, 2004). In lower altitudes of the Hormozgan province in Iran there is a savanna-like community of Acacia tortilis and A. ehrenbergiana, and Rhazya stricta is frequent along seasonal rivers and shallow valleys (Zohary, 1973). The Acacia campoptila community association with Panicum turgidum, Aerva javanica, Indigofera spinosa, Fagonia indica, Ochradenus baccatus, Dipterygium glaucum Calotropis procera and Rhazya stricta is widespread in the wadi beds at Marib, Harib, Ataq and Rada’ in Yemen (Al Khulaidi, 1989; Wistinga and Thalen 1980; Al Hubaishi and Muller-Hohenstein, 1979; Barkuda and Sanadiqi, 1986). These areas are characterised by having more rainfall than the Hadhramaut region. Some characteristic species of wadi beds in these areas, such as Leptadenia pyrotechnica, Lycium shawii, Jatropha spinosa and Acacia oerfota, were absent at similar habitats in Hadhramaut. The last two species were confined to slopes of the plateaus and were absent on the wadi beds in Hadhramaut. Some characteristic species of the Hadhramaut region, such as Tephrosia apollinea subsp. longistipulata, Ziziphus leucodermis, and Merremia hadramautica were not recorded at either Marib or Rada’. 92 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 30 mean annual tepmerature C 29.5 29 28.5 28 27.5 27 26.5 26 25.5 1 2 3 4 5 6 25 45 69 70 74 83 100 156 223 350 359 mean annual rainfall mm Figure 3.8. The main vegetation associations of 6 ecological areas in Yemen: 1- Al Mokha (western coastal area), 2- Al Kod (southern coastal area), 3- Marib (northern arid area), 4- Al Hudaydah, 5- Zabid and 6- Al Jaruba (western coastal area). The ap (left) shows the location of these areas. The main vegetation associations: 1- Salsola spinescens - Suaeda fruticosa: is dwarf shrubland; sassociated species include Salvadora persica. 2- Dipterygium glaucum - Panicum turgidum forms a grassland on sand dunes plain; associated species include: Fagonia indica, Aerva javanica and Indigofera spinosa. 3- Acacia campoptila – Aerva javanica forms grassland and woodland on drainage lines; associated species include: Fagonia indica, Panicum turgidum, Indigofera spinosa, Dipterygium glaucum, Rhazya stricta, Ochradenus baccatus and Acacia oerfota. 4- Odyssea mucronata - Panicum turgidum forms a shrubland; associated species include: Dipterygium glaucum and Cadaba rotundifolia. 5- Salvadora persica – Tamarix aphylla forms a woodland along Wadis; associated species include: Calotropis procera, Cadaba rotundifolia and Cissus quadrangularis. Leptadenia pyrotechnica – Panicum turgidum. forms a grassland or shrubland found on sand dunes; associated species include: Aerva javanica and Odyssea mucronata. 6- Acacia tortilis – Lasiurus scindicus forms a grassland or open woodland; associated species include: Jatropha glauca, Acacia ehrenbergiana, Indigofera spinosa, Leptadenia pyrotechnica, Cassia senna, and Indigofera oblongifolia. 93 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 3.5.4.b. Comparison with the vegetation of the region and East Africa The vegetation of the Hadhramaut region is very similar to the vegetation of wadis, in Sudan, Palestine and Arabia (White, 1983; Zohary 1973). Species of Acacia (e.g. Acacia tortilis and A. hamulosa) are accompanied by other typical species such as Panicum turgidum, Leptadenia pyrotechnica. On a sand wadi between Thaif and Bisha (Saudi Arabia) Rhazya stricta, Aerva javanica, Pennisetum divisum and others form a savanna-like vegetation (Zohary, 1973) and similar vegetation is also found in Wadi Bike, Egypt (Abdul El-Ghani, 2000). The Acacia tortilis-Panicum turgidum community is widespread throughout the tropical Sahara and the Sahel regions (White, 1983). The vegetation of the Tihama Plain (along the Red Sea coast of Yemen) is very similar to the vegetation of coastal area of Hadhramaut and the semi-arid grasslands of East Africa found in Kenya, Somalia and to a small extent in NE Uganda (Lind and Morrison, 1974). Acacia tortilis and A. hamulosa form savanna-like communities in association with Panicum turgidum, Leptadenia pyrotechnica, Aerva javanica, Calotropis procera and Ziziphus spina-christi which are widespread throughout tropical regions along the south of the Sahara desert and in wadis in Sudan, Egypt, Palestine and Arabia (Zohary 1973; White, 1983). A similar habitat to the open woodland dominated by Dracaena serrulata is found on Jebel Samhan, a south-facing escarpment in southern Oman, at altitudes between 600-1100 m asl (Al-Zdjali, 1995; Michael, et al., 2001). 3.6. Conservation and management of plant species and biodiversity 3.6.1. Introduction Plant biodiversity is an important natural resource for humans and wildlife. Human activities have dramatically affected natural resources caused great damage to the environment resulting in a great loss of biodiversity in many parts of the world. Because of the industrial revolution, the impact has become global rather than regional. This global impact is taking place through 4 primary processes (Otten, 2001):- overharvesting, unknown species introduction, pollution, and habitat destruction. Plant species that have taken millions years to develop, are now rapidly destroyed by human 94 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen activities and climate change (Bryant, 2004). Plant species in Hadhramaut are being threatened by a rapidly growing human population, increasing climatic hazards, economic development and, in by particular, oil-related activities and devastating seasonal floods in the main wadis. Currently, many plant species are rare, threatened, or endangered in the region; however, further studies are needed to ascertain the number of species under threat: this will form the basics for a conservation and management strategy. 3.6.2. Important sites of Hadhramaut region A working list of known plant species is considered to be a fundamental requirement for plant conservation. At present we do not have a complete inventory of the plants of the Yemen. With the continuing discovery of new endemic plant species every year from all parts of Yemen: for example, over the last 11 years new endemic plant species for Yemen include: Acacia harala Thulin & Gifri. (2000), Acacia mahrana Thulin & Gifri (2000), Aloe irafensis Lavr. McCoy & Gifri (2004), Aloe luntii Baker (2002), Campylanthus antonii Thulin (1995), Cleome hadramautica Thulin (2002), Cleome socotrana Balf. (2002), Cucumis canoxyi Thulin & Gifri (1994), Echidnopsis globosa Thulin & Hjertson (1995), Euphorbia applanata Thulin & Gifrii. (1995), Euphorbia quaitensis S.Carter (1995), Fagonia hadramautica Beier & Thulin (2005), Gaillonia jolana Thulin (1998), Gaillonia yemenensis Thulin (1998), Indigofera rubromarginata Thulin (1995), Nesocrambe socotrana AG Mill. (2002), Ochradenus gifrii Thulin (1994), Pentzia arabica Thulin (2001), Rhytidocaulon mccoyi Lavr. & Mies (2001), Tephrosia hadramautica Thulin (1994) and many others in particular, many new species continue to be found in previously unexplored areas in the Soqotra Archipelago, Hadhramaut and Al Mahara Governorates. The total number of known species in Yemen is currently estimated to be around 2900. The number of known plant species recorded by myself (Al Khulaidi, 2000) was 2810 species. Further fieldwork remains essential in some presently unexplored areas of Yemen to enable more comprehensive assessments to be undertaken. Of particular concern is that much work has still to be done to achieve the year 2010 Global strategy of plant conservation (CBD and UNEP, 2002) which aims to achieve the following goals: 95 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen (a) Understanding and documenting plant diversity (b) Conserving plant diversity. (c) Using plant diversity sustainably (d) Promoting education and awareness about plant diversity (e) Building capacity for the conservation of plant diversity. Important areas to be protected are identified using criteria which include endemism, rarity, near-endemism, species richness, and threat to and richness of habitats. 469 plant species have been identified from the Hadhramaut region (Appendix 4). There are 107 taxa which are endemic and near-endemic (Table 3.2); 67 of these are endemic to Yemen, of these about 40 are confined to Hadhramaut region. Apart from work recently carried out by Miller and Al Khulaidi (2004), there have not been attempts to apply the IUCN threatened categories to plant species in Yemen. However, the categories for Soqotra Island have been completed (Miller and Morris, 2004). The vegetation of the Hadhramaut plateau (Jol), on the south-facing mountains or monsoon-affected coastal mountains of the southern part of Hadhramaut region is sustained by the regular fogs that envelop the region. The summits of these mountains, which just below 2000 m asl (Figure 3.9), are considered to have the the richest flora in the region; particularly as regards endemic and near-endemic and newly recorded plant species. The summits have a unique vegetation notably in around Kor Seiban (14°82’N 48°81’E, 1850 m), Sharj al Alif (14° 48' N 48° 46' E, 1840 m), Al Qumra (14° 48' 29N 8° 44' 58E, 1968 m) and Maula Matar (14° 46' 6N 48° 47' 52E, 1651 m), and in the coastal area east of Al Mukalla (14° 42 ‘83’’N, 49° 29’ 91”E, 15 m). Generally, the escarpments south of Maula Mattar to around Bain al-Jibal are recorded as the richest, probably because they are relatively easy to reach (Lavranos ,1966; Lavranos and Mies, 2001; Thulin, and Al Gifri, 1993). However, there are other areas along the escarpment which are hard to get to and which are much less exposed to human pressures. It is argued that these sites should be included in the list of the proposed protected areas in Yemen in future strategies. Other Interesting sites are found in the coastal area between Al Mukalla and Al Hami, where there are further endemic and near-endemic plant species. Examples of remarkable species in these sites include: Euphorbia seibanica, Rhytidocaulon mccoyi, Aloe eremophila , Huernia hadhramautica, Echidnopsis seibanica, Anogeissus bentii, Justicia areysiana, Schweinfurthia latifolia, Cucumis canoxyi, Caralluma lavrani, Euphorbia applanata, Cryptolepis yemenensis, Launaea castanosperma, Ochradenus arabicus, Echidnopsis 96 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen globosa, Limoniastrum arabicum and Enicostema axillare (Al Gifri and Gabali 1999; Lavranos and Mies, 2001; Meister et al., 2005, and Kilian et al., 2002): (For more see the list of plant species, Table 3.2). Figure 3.9. The important areas for conservation in Hadhramaut region, Kor Seiban, Sharj al Alif, Al Qumra), Maula Matar and coastal area. The green dots show the distribution of rare, endemic and nearendemic plant species in the Hadhramaut based on field work records as well as on the records of Kilian et al (2002), Thulin et al (2001), Thulin and Al Gifri (1995), Miller and Al Khulaidi, 2004 and Lavaranos (2001). 97 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 3.7. Currently existing protected areas in Yemen At present there are no government protected areas in Yemen. Actively protected areas in Yemen are normally managed by the local communities and fall into the following categories: (Al Khulaidi and El-Ghouri, 1996):- At present there are no fully protected areas by legislation although three are at present being proposed (see below and Chapter 7)  woodlands and/or shrublands that are formed and protected naturally and mostly situated in remote areas,  woodlands that are situated around agricultural fields and surrounding areas,  National parks artificially afforested small sites and areas run by the public sector, these are planted mainly by introduced trees and shrubs. In many parts of the country, traditional small protected areas can be found. These areas are mainly situated around or near villages and are privately or communally owned. These kinds of protected areas are known in Islamic history as (hima) and almost every village in the Arabian Peninsula was associated with one or more himas. Management of the himas was highly adaptive to the particular requirements of a given village and the potential of a given area. For example, himas in the Arabian Peninsula were managed to protect and conserve plants by limiting animal grazing but allowing the cutting of fodder; permitting seasonal grazing and fodder cutting after plants had flowered, limiting grazing to certain types of animals, or protecting trees (Draz, 1969; Sulayem and Joubert; 1994, Llewellyn, 2000) Traditional rules are used to protect these areas, which are mainly used for grazing. The implementation of these rules differs from one place to another; the common rules are formed by agreement with a fine between 1000 and 10 000 Yemeni Reyal for cutting trees from the protected areas. These protected areas represent the better managed sites of the Yemen, and are normally managed by local communities, village leaders or Sheikhs. Different types of protected areas exist in Yemen called locally Mahajer in Arabic (hima). This type of protected area can be divided into the following categories (Kessler, 1988a):  Temporary protected areas: located adjacent or around cultivated fields, the main reasons to set up these areas are 98 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen a) to protect the crops from grazing b) to reserve forage for dry season This type used to be common in the past in different parts of the Hadhramaut region.  Permanent protected areas: located on rocky slopes, and set up for the following reasons: a) to harvest rain from the slopes c) to be used as personal grazing and firewood collecting d) as a result of a dispute about the ownership of areas.  Temporary village protected areas: mountain slopes declared as protected areas for a certain period (mostly rainy season and mainly used for grazing). Currently the Yemeni government is carrying out biodiversity assessment activities on proposed protected areas. Two of these areas will be declared as protected areas shortly (UNEP, 2004); these area are Jabal Bura’ (western mountain area) and Huf, Al Mahara (eastern part of the country). The protected areas will be under the IUCN Category definition II, which aims for the area to be managed mainly for ecosystem protection and recreation (IUCN, 2004, Anon, 1994). 3.8. Conservation of plants Most of the conservation carried out in Yemen is based on traditional protected areas that are mainly made for traditional grazing activities. In these protected areas, conservation of a particular habitat or plant species was not the aim of the preservers. Over 600 endemic, near-endemic and non-endemic plant species from Yemen have been evaluated recently for conservation status to CITES 1 and 2. (Miller and Al Khulaidi, 2004), but still a lot of work remains to be done in the future to cover the unexplored areas. However, assessments of the endemic plants of the Soqotra Archipelago have been completed in line with the IUCN categories (Miller and Morris, 2004). The Hadhramaut is an important area of biodiversity, especially the southern plateau which has a considerable number of endemic and near-endemic plant species. 99 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen All endemics from the Hadhramaut region must ultimately be included in the IUCN Red Data world list, but to achieve this further studies are required. Examples of endemics and near-endemics species are: Cucumis canoxyi, Cleome macradenia, Cleome pruinosa, Commicarpus stenocarpus, Commiphora foliacea, Cystostemon kissenioide, Ecbolium strictum, Echidnopsis bentii, Echidnopsis globosa, Echidnopsis seibanica, Echiochilon arabicum, Echiochilon colona, Echiochilon strigosum,, Ephedra milleri, Eragrostis maharana, Eriochloa fatmensis, Eruca sativa, Erucastrum arabicum, Euphorbia applanata, Euphorbia arabica, Euphorbia balsamifera, Euphorbia fodhliana, Euphorbia meuleniana, Euphorbia quaitensis, Euphorbia rubriseminalis, Euphorbia seibanica, Fagonia hadramautica and others (see Table 3.2). Only a few species from the Hadhramaut are already included in the IUCN Red Data list as Vulnerable (VU); (Miller and Al Khulaidi, 2004; Beier, 2005; Ghazanfar and Al Kiyumi, 1999), these species are Caralluma adenensis, Echidnopsis bentii, E. seibanica Heliotropium wagneri and Fagonia hadramautica. 3.8.1. National policy on protected areas Natural protected areas are the in-situ method for the conservation of biological diversity and special entities with unique or endemic or rare flora and fauna. Protected areas offer protection for all the plant species from over-use and degradation and they provide suitable sites for field studies and research on biodiversity and natural resources, and provide suitable sites for training and monitoring environmental change. It is therefore, an ancillary objective of this research to propose clearly defined areas which deserve protected status. 3.8.2. Future Activities Research activities would ideally be carried out by the Environmental Protection Authority (EPA) and Agricultural research and extension Authority (AREA). Activities are needed to provide data for planning and formulating programs of applied projects, selecting suitable sites for protection as nature reserves and assessing the current status of plants and habitats. This will provide a baseline for monitoring change in the natural environment and should provide information for the evaluation and protection of natural resources. 100 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen The research should include the following:  Inventories and collections of plant species from the southern coastal area to the northern plateau bordering the Empty Quarter.  Surveys of the biodiversity and natural resources in selected areas  Mapping of land vegetation, land use and topography of selected areas  Identification of the vegetation associations and the vegetation structures  Compilation of a check list of plant species  Plant taxonomic studies  Identification and assessment of rare, endemic, near-endemic and endangered species  Social economic studies related to biodiversity  Consideration of possible legislative measures to prevent pollution and landscape damage 3.9. Conclusions Due to the sharp climate gradient from east to west, the changes in the structure and composition of the vegetation E-W (transect 1) are more marked than those from south (coastal plain) to the north (transect 2). Transect 2 extends from Al Rayan on the coastal region east of al Mukalla in the south across the central plateau to Sayun in Wadi Hadhramaut north. The altitude varies from see level south then rises to about 1600 m on the escarpment above al Mukalla and then falls to about 1000 m above sea level in Wadi Hadhramaut in the north. The rainfall along this transect varies from a annual mean of about 69 mm at Al Rayan on the coastal plain in the south to a mean annual less than 75 mm in Sayun in Wadi Hadhramaut in the north. There is no rainfall data from the limestone plateau across the middle part of the transect, nevertheless, the abundance of floods on the wadis running off from the plateaus and the landforms suggests an annual rainfall of more than 200 mm (SOGREAH, 1979). The climate along the first transects decreases from a mean annual rainfall of about 74 mm in the east to an extreme arid climate with a mean annual less than 50 mm in the west. The altitude varies from 670 m above sea level in the east to about 770 m above sea level in the west. Generally species richness and diversity decrease from east to west. In the west the vegetation cover is sparse but moving eastwards it gradually increases until in the far east it is transformed into woodland characterised by Acacia campoptila and 101 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Fagonia indica, sparse dwarf shrubland or grassland dominated by Zygophyllum album and Dipterygium glaucum in the west Despite similar parent rock and landscape, the plant composition along the second transect differed in both vegetation structure and plant associations. This may be ascribed to difference in moisture availability along the altitudinal gradient from south to north. The floristically richest part of the transect was on the south-facing escarpments and the high ridges above al Mukala. These areas are regularly blanketed in cloud which significantly supplements the moisture available for plant growth. This is a widespread phenomenon across southern Arabia where fogs play an important role in sustaining vegetation. For instance, in the monsoon woodland of Hauf and Dhofar on the Yemen-Oman border and on the low limestone plateau of the Jiddat al Harasis in the deserts of central Oman and Dhufar (Oman) (Ghazanfer, 2004; Kurschner et al.,1998). The vegetation of the drier parts of the transect comprises open shrubland dominated by Acacia tortilis, A. ehrenbergiana and Ziziphus leucodermis associated with Prosopis cineraria, Zygophyllum spp., Rhazya stricta, Iphiona scabra, Tephrosia apollinea, Panicum turgidum and Pulicaria glutinosa. Similar vegetation has been reported in similar habitats with an annual rainfall of less than 50 mm in the eastern parts of the Oman desert (Ghazanfer, 2004). Rhazya stricta is a Saharo-Sindian characteristic plant species of the desert dry wadis that have spring and summer rainfall. This species was not seen in the coastal areas of al Mukalla, which is much more humid that the interior and has rainfall mainly in spring rainfall with little in winter. The limit of distribution of this species towards the south seems to be at about 1000 m asl (Figure 3.10). This Sahro-Sindian species is widely distributed from north Jeddah (Saudi Arabia) to the northern part of the United Arab Emirates (Ghazanfar and Fisher, 1998; Western, 1985). Pulicaria somalensis, Ochradenus arabicus and Ochradenus baccatus are other interesting species that show well-defined distributions (Figure 3.10). The first species is widespread between the southern coastal areas to a limit distribution at about 1200 m towards the north; the second is widespread from the foothills of the coastal escarpment to almost the same one point in the north as P. somalensis; the third species has a distribution from the eastern part of W. Hadhramaut towards the north. Pulicaria 102 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen somalensis is widespread on the stony gravel coastal plains of some Gulf countries (Ghazanfar and Fisher, 1998), while Ochradenus arabicus is only known from northern Oman, Saudi Arabia united Arab Emirate and the Hadhramaut region. Figure 3.10. Examples of distributions of characteristic plant of the region from the reconnaissance vegetation survey:. Pulicaria somaliensis, Ochradenus arabicus and Ochradenus baccatus (right) and Rhazya stricta (left) Several changes in the appearance of the vegetative with increase in altitude were noticed. In the southern transect, there was a clear succession of plant species from the low coastal area to high elevation sites; only 4 plant species (e.g. Pulicaria somaliensis, Jatropha spinosa, Fagonia schweinfurthii and Indigofera spinosa) were found at almost all elevations and the plant species of the highest elevation are almost completely distinct from those at the lowest elevation. Examples of high altitude species (over 1150 m) include Euphorbia balsamifera, Gnidia somalensis var. sphaerocephala, Ochradenus arabicus, Barleria bispinosa, Barleria proxima, Aloe sp., Acacia etbaica, Dodonaea viscose, Dracaena serrulata and Ephedra foliata. Some of these species are restricted to specific landform types, for example Euphorbia balsamifera grows only on almost flat limestone plateau summits while others, like Dracaena serrulata and Ephedra foliata, prefer the rocky slopes above secondary wadis. On the other hand 103 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen there was no clear succession of plant species from the east towards Ramlat Assaba’tain (transect 1). The following are widespread on this habitat: Acacia campoptila, Acacia ehrenbergiana, Dipterygium glaucum, Panicum turgidum Tephrosia apollinea and Prosopis juliflora. Other species are confined to specific landform and altitude. For example, Aeluropus lagopoides, Halopyrum mucronatum, Odyssea mucronata, Limoniastrum arabicum, Aristolochia rigida are only seen on the sandy areas along the coast. Whilst Salvadora persica, Tamarix aphylla, Prosopis cineraria and Pluchea dioscoroides grow are typically found on almost flat wadi beds at moderate altitudes in the study area further north. The greatest vegetation cover percentage and species richness were observed in the south and south-east facing mountain rocky slopes and drainage lines and also in the north-east and north-west facing transition sites, where species from both high and low altitudes were recorded. Shrubland is the dominant structural form in transect 1; trees percentages were high at the middle altitude (1000-1200 m) of the north-west and north-east facing secondary wadi beds. Generally as in other areas of the world (Hopper and Maslin, 1978), the geographical isolation of this region has played and still is playing an important role in isolating the area and therefore protecting the plant species from human activities. About 469 plants are found in Hadhramaut region, of which about 23% (107 taxa) are endemic and near-endemic, and many have their origin in the African zone (see Appendix 4). Adaptive radiation and speciation by geographical separation has occurred in many genera and each mountain range in the western escarpment of Arabian Peninsula has its own endemic species (Ghazanfar and Fisher, 1998), the most endemicrich family in the Hadhramaut region is Apocynaceae (19) , and the genera with most endemics are Caralluma (11 taxa), and Aloe (9 taxa). In this Chapter, a general description of the vegetation composition across the Hadhramaut region has been presented. Several changes in vegetative characteristics due to altitude gradients and climate change were noticed. The Hadhramaut region which is not explored very well is very rich in endemics and near-endemics, in 104 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen particular the southern summit. Further surveys covering the descriptions of soil, land use, moisture content and landform should be carried out in this extraordinary region so that a more accurate assessment can be made of the distribution, abundance and composition of its plants and, in particular, of its rare, endemic and near-endemic species. The distribution maps of some important plant species in Hadhramaut region can be seen in chapter 6. The data that are presented in this chapter have been valuable in providing a general assessment of the main vegetation types in the Hadhramaut and have supported the selection of sites for further detailed studies. They have also been useful for framing recommendations for future management and conservation activities. The next two chapters will illustrate in detail the biodiversity and the main vegetation associations of the three selected study sites within Wadi Hadhramaut. 105 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Chapter 4. Floristic analysis and plant biodiversity 4. 1. Introduction Studies on the vegetation of Hadhramaut go back to the last two decades of the nineteenth century (Gabali and Al Gifri, 1991) but the present work is the first systematic survey of the vegetation of the region. As indicated earlier, previous studies concentrated on compiling flora checklists or on general descriptions of the vegetation communities of certain habitats without following any consistent scientific methodology. The study area is still the least known part of the whole of Yemen. It is intended that the present research should form the basis for any future management plan for the region. The aims of this chapter are to answer the following research questions:    What are the principal plant associations? Which sites contain the greatest species diversity? What are the principal physical and anthropogenic factors affecting contemporary plant patterns?. The structure of the presentation is illustrated in Figure 4.1. Vegetation analysis and plant diversity Shannon & Simpson indices (MVSP) software Species accumulation Curves (BioDiversity software) Species diversity Abundance K-dominance curve (BioDiversity software) Abundance, species richness & density Importance Value Index (density, frequency and dominance) Jaccard similarity index (BioDiversity software) Similarity between the study sites & landforms Phyto geographic patterns and physical anthropogenic determination Figure 4.1. Diagram showing the structure of this chapter. 106 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 4.2- Methodology A large number of techniques have been employed to describe and assess the plant associations of an area and to understand plant diversity. The study area (Wadi Hadhramaut) was subdivided into 3 representative study sites for detailed survey and vegetation analysis. This chapter deals with a relation of the most appropriate methods that serve to explain the research questions outlined in the introduction. Number of individuals per hectare Number of individuals refers to the density of each species that has been recorded in the sample sites during the vegetation survey. The numbers of individuals in each site were calculated per hectare. This is achieved by dividing the total number of individuals by the total areas surveyed in the different land forms. Frequency Frequency is defined as the number of times a plant species is present within a given number of sample quadrats of uniform size placed repeatedly across a stand of vegetation (Mueller-Dombois and Ellenberg 1974; and Daubenmire, 1968). Plant frequency, by itself, is useful for monitoring vegetation changes over time at the same locations or for comparisons of different locations. Frequency was calculated by dividing the number of plots in which a species occurs into the total number of plots sampled. Plant species diversity Diversity is a measure of the number of species within a unit area and has been used by different plant ecologists to evaluate plant species in different ecosystems of the world. Diversity has two components: species richness (S), and species evenness (equitability), or how well distributed is abundance among species within a community (Wilsey et al., 2000). 107 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Diversity can be considered as a factor alongside species composition, disturbance, soil type and climate that influences ecosystem functioning (Tilman, 2000). Human activity is obviously reducing the diversity of species within many habitats worldwide, and is speeding up extinction (Tilman, 2000; Vitousek et al., 1997). Diversity increases when 1- The number of species increases (species richness) or 2- The species abundance become more equal (evenness) Many species diversity indices have been developed. The most popular diversity index is the Shannon index (Shannon-Wiener index). It is the most widely used diversity index in plant ecology because it includes both species richness and abundance (Krebs, 1999). Richness refers to the number of species present; evenness refers to the distribution of species among different habitats. The Simpson diversity index is another diversity measure used by plant ecologists. Like Shannon’s index, Simpson's index accounts for both abundance and evenness of species. These indices are closely related (Hill, 1973). The values of the Simpson index are between 0 in a community of one species, to 7 in a more diverse community (Kutiel et al., 2000). The more the area is dominated by one of the components, the closer the value is to 1. The Simpson’s index is more sensitive to more abundant species and less sensitive to species richness and therefore weighs abundant species more than rare species (Gage, 2002). The Shannon index takes into consideration the relative contribution of all species. The more the abundance is equitably distributed among species, the higher the value of diversity (Kutiel et al., 2000). The total number of individuals and both the Shannon and Simpson indices were used to illustrate the plant diversity of the study sites. The plant diversity of the study sites over different landforms was calculated using the Shannon index (Table 4.11 and Figure 4.9a). This plot provides a measure of species diversity, allowing comparison between the different habitats. Since each surveyed habitat has a different area, individuals were standardised per hectare (Koellner et al., 2004). 108 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Percentage cumulative abundance was plotted against log species rank, using BioDiversity Professional Beta (McAleece, 1997), It was also used for comparing diversity between the 3 study sites (Figures 4.7) and between the different landforms (Figure 4.9). Abundance and density Numbers of individuals and frequency were recorded. Further, importance value indices (IVI) of each plant species were calculated in all study sites by summing the relative density value, the relative dominance value, and the relative frequency value. Importance Value and Species Richness Importance values refer to how important a species is in terms of the structure of a community or species composition. To determine the importance of each species sampled, the importance value index (IVI) of each plant species was calculated in all study sites by summing the relative density value, the relative dominance value, and the relative frequency value (Mahdi & Al Khulaidi, 1999; Nautiyal et al., 2003) (see Table 4.6). Importance value curves were obtained using BioDiversity software by plotting abundance against species rank for the three study sites (Figure 4.6) and for the different landforms (Figure 4.5). These values are expressed in an absolute form and as relative density, relative dominance, and relative frequency, which shows the percentage of an individual species with respect to the total species. These values are calculated by the following equations: Absolute density = Number of individuals of a species per hectare  Absolute dominance = Total basal cover of a species per hectare  Relative density = Density (number of individuals) of a species/ total density for all species x 100  Relative dominance = Dominance for a species/ total dominance for all species x 100 109 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen  Absolute Frequency = Number of sample sites in which a species occurs/ total number of sample sites  Relative Frequency = Frequency value for a species/ total of frequency values for all species x 100  Importance Value Index (IVI) = relative density + relative dominance + relative frequency Species richness was determined as the total number of species distinguished at each habitat. The number of individuals was determined at each sample site, and then the numbers were averaged and weighed per hectare. Species accumulation curves This is a method for determining whether the sampling is sufficient to have collected all the plant species from a region or for comparing species richness between different communities (Moreno and Halffer, 2000; Willott, 2001). The expected species richness in each study area was estimated by species accumulation curves. The analysis used software developed by the BioDiversity Professional Beta (McAleece, 1997). Species richness, diversity and the proportion of rare species in a community influenced the shape of species accumulation curves (Thompson, et al., 2003). Species accumulation curves can provide three basic pieces of information: 1) How many species there are, 2) How species are distributed, 3) How adequately the habitat was sampled. The level at which a curve levels off indicates the number of species in the habitat. The rate at which the curve begins to level off indicates how species are distributed. For example, if a curve rises quickly and levels off it means that most of the species are found in just the first few sampling plots. The curve for species versus the number of sample sites was plotted for each site and for each land form in the 3 study sites in order to calculate the minimum number of samples that should be taken to be truly representative of the habitats in the study area. 110 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Similarity between study sites Data of presence and absence using the BioDiversity Program software was analysed by the Jaccard distance similarity index in order to assess the similarity between the 3 study sites and between the different habitats within these areas. According to Chao et al., (2005), the Jaccard distance measure gives reasonable similarity clusters result if presence/absence data are used. 4.3. Results 4.3.1. Flora A total of 134 species belonging to 42 families (about 30% of flora of Hadhramaut region) were recorded in the study area and, of these, seven species are endemic to Yemen (five of them endemic to Hadhramaut region). A total of 117 plant species were collected in the whole sample plots of which 4 have not yet been identified. There are 74 species in site 1, 83 species in site 2 and 61 species in site 3, with 39 plant species recorded in all 3 study sites. The areas surveyed was estimated as 0.46 ha, 0.53 ha and 0.34 ha respectively, which correspond to 46, 53 and 34 X 100 m² sample plots. Two voucher specimens were collected from each plant species, one set is deposited in the herbarium of Royal Botanic Garden of Edinburgh, UK (E) and the other is deposited in the herbarium of the Agricultural Research and Extension Authority (AREA), regional station, Taiz, Yemen. The species records from the study area were classified using both BraunBlanquet and Twinspan into: (1) groups that show a similar distribution across the sample plots, these were termed sociological species groups and (2) into groups of sample plots with a large similarity termed vegetation associations (see chapter 5). The following plant species were considered rare, because they were only recorded once: Cadaba heterotricha, Cadaba farinosa, Aerva artemisioides, Euphorbia rubriseminalis, Glossonema varians, Trichodesma calathiforme, Salvadora persica, 111 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Leptadenia arborea, Withania somnifera, and Portulaca oleracea. All species were found in site 2 except the endemic Aerva artemisioides, which was only found in site 3. However this species was recorded by DOVE (2001) on the plateau at site 2. The last three species are weedy species and are found in large quantities in wetter areas, especially in or near cultivated and fallow lands.  The largest families in terms of species in the study sites are:  Poaceae (Gramineae) with 13 species  Zygophyllaceae with 11 species  Mimosaceae with 8 species  Fabaceae with 12 species  Capparaceae with 9 species Asteraceae with 7 species Poaceae (Gramineae) is the family with highest number of individuals; 17 families were recorded with one single species (Table 4.1). 112 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Table 4.1. List of families and number of species in alphabetical order of families for the 3 study sites Family Acanthaceae (3) Amaranthaceae (2) Apocynaceae (5) Arecaceae (Palmae) (1) Asteraceae (Compositae) (7) Balanitaceae (1) Boraginaceae (5) Brassicaceae (Cruciferae) (3) Burseraceae (2) Caesalpiniaceae (3) Capparaceae (6) Caryophyllaceae (1) Chenopodiaceae (3) Cleomaceae (3) Convolvulaceae (5) Area Scientific name 1,2,3 1. Barleria aff. bispinosa (Forssk.)Vahl. 2. Blepharis edulis Forssk. 3. Peristrophe paniculata (Forssk.)Brummitt 4. Aerva artemisioides Vierh. & Schwartz subsp. artemisioides. * 5. Aerva javanica (Borm.f.)Juss. ex Schult. 6. Calotropis procera (Ait.)Ait.f. 7. Glossonema varians (Stocks)Benth. ex Hook. 8. Leptadenia arborea (Forssk.)Schweinf. 9. Rhazya stricta Decne 10. Periploca visciformis (Vatke)K.Schum. 11. Phoenix dactylifera L. 12. Helichrysum pumilum (Klatt.)Moes. 13. Hochstetteri schimperi DC. 14. Iphiona anthemidifolia (Bak.)A.Anderb. * 15. Iphiona scabra DC. 16. Launaea sp. 17. Pluchea dioscorides (L.)DC. 18. Pulicaria undulata (L.)C.A.Mey. 19. Balanites aegyptiaca Del. 20. Arnebia hispidissima (Lehm.)DC. 21. Heliotropium ramosissimum (Lehm.)Sieb. ex DC. 22. Heliotropium longiflorum Steud. & Hochst. ex Bunge 23. Heliotropium rariflorum Stocks 24. Moltkiopsis ciliata (Forssk.)I.M.Johnston. 25. Trichodesma calathiforme Hochst. 26. Erucastrum arabicum Fisch. & Mey. 27. Farsetia linearis Dene 28. Farsetia longisiliqua Decne. 29. Commiphora foliacea Sprague. *** 30. Commiphora kua (J.F.Royale) Vollesen . 31. Cassia senna L. 32. Senna holosericea (Fres.)Greuter 33. Senna italica Miller 34. Cadaba heterotricha Stocks ex Hook. 35. Cadaba farinosa Forssk. 36. Capparis cartilaginea Decne. 37. Capparis spinosa L. 38. Dipterygium glaucum Decne. 39. Maerua crassifolia Forssk. 40. Cometes abyssinica (R.Br.) Wallich. 41. Cornulaca amblyacantha Bunge. 42. Halothamnus bottae Jaub. & Spach 43. Salsola imbricata Forssk. 44. Cleome brachycarpa Vahl. ex DC. 45. Cleome droserifolia Del. 46. Cleome scaposa DC. 47. Convolvulus arvensis L. 48. Convolvulus glomeratus Choisy 1,2,3 3” 3 1,2,3 1,2 3 2 1,2,3 2” 1,2,3 2 2 1 1 2 2 2 2” 2 1,2,3 3” 1,3 1 2 1” 1,2,3 3” 1,2,3 2,3 3 1 1,2,3 2 2 1,2 1,.2 1,2,3 1.2.3 1,2 2,3 1,2,3 2 1,2,3 1,2,3 1,2,3 1 2,3 113 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Table 4.1. Continue Cucurbitaceae (3) Cuscutaceae (Convolvulaceae) (1) Cyperaceae (1) Euphorbiaceae (4) Fabaceae (Papilionoideae) (12) Loranthaceae (1) Malvaceae (4) Mimosacea (8) Moraceae (1) Moringaceae (1) Nyctaginaceae (1) Pedaliaceae (1) 49. Cressa cretica L. 1 50. Merremia hadramautica Hall.f. * 1,2,3 51. Seddera latifolia Hochst. & Steud. 2 52. Corallocarpus glomeruliflorus Schweinf. 2,3 53. Citrullus colocynthis (L.)Schrad. 1,2 54. Cucumis canoxyi Thulin & Gifri. * 55. Cuscuta campestris Yunker. 3” 56. Cyperus conglomeratus Rottb. 2,3 57. Chrozophora tinctoria A.Jjuss. 1,2,3 58. Euphorbia granulata Forssk. 1 59. Euphorbia rubriseminalis S.Carter. ** 2 60. Jatropha spinosa (Forssk.)Vahl 2,3, 61. Alhagi graecorum Boiss. 1 62. Crotalaria persica (Burm.f.)Merr 1,2 63. Crotalaria saltiana Andr. 1,3 64. Indigofera oblongifolia Forssk. 3 65. Indigofera spinosa Forssk. 1,2,3 66. Indigofera spiniflora Hochst. & Steud. ex Boiss. 2 67. Medicago sativa L. 1,2,3” 68. Rhynchosia memnonia (Del.)DC. 2,3 69. Tephrosia apollinea (Del.)DC. subsp. longistipulata 1,2,3 70. Tephrosia dura Baker 1,3 71. 72. Tephrosia nubica (Boiss.)Baker subsp. arabica (Boiss.)Gillett 1,3 3 Tephrosia subtriflora Hochst. ex Baker. 73. Plicosepalus curviflorus (Benth. ex Oliv.)Tiegh. 1,2 74. Abutilon bidentatum Hochst. ex A.Rich 2 75. Abutilon fruticosum Guill. & Perr. 76. Abutilon pannosum (Forrsk.) Schlechl. 2” 77. Senra incana (Cav.)DC. 1,2 78. Acacia campoptila Schweinf. ** 1,2,3 79. Acacia ehrenbergiana Hayne. 1,2,3 80. Acacia hamulosa Benth. 1,2,3 81. Acacia mellifera (Vahl)Benth. 1,2,3 82. Acacia oerfota (Forssk.)Schweinf 1,2 83. Prosopis cineraria (L.)Druce 1,2” 84. Prosopis farcta (Banks. & Sol.)Mc Bride. 1 85. Prosopis juliflora (Sw..)DC. 1 86. Ficus cordata L subsp. salicifolia 87. Moringa peregrina (Forssk.)Fiori 1” 88. Boerhavia elegans Choisy subsp. elegans. *** 1,2,3 89. Sesamum indicum L. 2” 2” 2” 1,2,3” 114 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Table 4.1. Continue Poaceae (Gramineae) (13) 28 31 Portulacaceae (1) Ranunculaceae (1) Resedaceae (2) Rhamnaceae (3) Rubiaceae (2) Salvadoraceae (1) Scrophulariaceae (3) Solanaceae (2) Sterculiaceae (1) Tamaricaceae (2) Tiliaceae (2) Urticaceae (1) Verbenaceae (1) Zygophyllaceae (11) 90. Aristida triticoides Henrard. 1,2,3 91. Chloris barbata Sw. 1 92. Chrysopogon aucheri (Boiss.)Stapf var. quinqueplumis 2,3 93. Cymbopogon schoenanthus (L.) Spreng subsp. schoenanthus 1,2,3 94. Cynodon dactylon (L.)Pers. 1,2 95. Dactyloctenium aegyptium (L.)P.Willd. 3” 96. Dichanthium insculptum (A.Rich.)Clayton 1,2,3 97. Enneapogon desvauxii J.E.Smith 1,2,3 98. Eragrostis sp. 2,3 99. Lasiurus scindicus Henrard 1,3 100. Panicum turgidum Forssk. 1,3 101. Setaria verticillata (L.)P. Beauv. 1 102. Stipagrostis hirtigluma (Steud. ex Trin. & Rupr.)De Winter 1,2,3 103. Portulaca oleracea L. 2 104. Nigella sativa L. 3” 105. Ochradenus baccatus Del. 2,3 106. Reseda sphenocleoides Defl. 107. Ziziphus leucodermis (E.G.Baker) Schwartz 108. Ziziphus spina-christi (L.)Willd. 1,2,3 109. Kohautia retrorsa (Boiss.)Bremek. 1,2,3 110. Tarenna graveolens (S.Moore)Bremek. subsp. arabica 2 111. Salvadora persica L. 2 112. Anticharis glandulosa (Ehrenb. & Hempr.)Aschers. 2,3 113. Lindenbergia indica (L.)Kuntze 1 114. Striga asiatica (L.)Kuntze 3” 115. Datura innoxia Mill. 1 116. Withania somnifera (L.)Dun 1 117. Hermannia paniculata Franch. 3 118. Tamarix aphylla (L.)Karst. 1,2 119. Tamarix arabica Bunge 2 120. Corchorus depressus (L.)Christ 1,2,3 121. Grewia erythraea Schweinf. 2 122. Forskohlea tenacissima L. 1 123. Chascanum marrubifolium Fenzl ex Walp. 2 124. Fagonia bruguieri DC. 125. 126. Fagonia hadramautica Beier & Thulin * Fagonia indica Burm.f. 127. Fagonia paulayana Wagner & Vierh. 1,2,3 1,2,3 128. Seetzenia lanata (Willd.) Bullock 1,2 129. Tribulus arabicus H.Hosnlined. 1,2,3 130. Tribulus terrestris L. 2” 131. Zygophyllum album L. var. amblyocarpum 1 132. Zygophyllum coccineum L. 1,3 133. Zygophyllum decumbens Del . 2 134. Zygophyllum simplex L. 1,2 1 *** *** 1,2,3 1” 2” key: bold * ** *** Species found outside of the sample sites Endemic to Hadhramaut Endemic to Yemen Endemic to Arabian Peninsula 115 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen A total of 134 species belonging to 42 families (about 30% of the known flora of Hadhramaut region) were recorded in the study area. A total of 117 plant species were recorded in the 3 sites of the study area, of which 5 are endemic to Hadhramaut. The density and the distribution of these species were different from one study site to another, as well as from one landform to another. The following endemic taxa for the Hadhramaut were found in the study sites: Aerva artemisioides subsp artemisioides, Iphiona anthemidifolia, Cucumis canoxyi, Merremia hadramautica and Fagonia hadramautica. Observations outside the cross sections but within the study sites revealed the following 22 plant species: 1. Abutilon fruticosum 12. Heliotropium longiflorum 2. Abutilon pannosum 13. Medicago sativa 3. Balanites aegyptiaca 14. Moringa peregrina 4. Cucumis canoxyi * 15. Nigella sativa 5. Cuscuta campestris 16. Periploca visciformis 6. Dactyloctenium aegyptium 17. Peristrophe paniculata 7. Erucastrum arabicum 18. Prosopis cineraria 8. Fagonia bruguieri 19. Sesamum indicum 9. Fagonia hadramautica * 20. Striga asiatica 10. Farsetia longisiliqua 21. Tephrosia subtriflora 11. Ficus cordata L subsp. salicifolia 22. Tribulus terrestris *= taxa endemic to the Hadhramaut. Most of these species are weeds found near or in the cultivation. Two of the species (Medicago sativa and Nigella sativa) are cultivated by local people. The species Fagonia indica, Tephrosia apollinea, Cymbopogon schoenanthus, Dichanthium insculptum and Boerhavia elegans subsp. elegans, occur throughout the entire area, in particular on the mountain slopes facing the main wadis, the plateaus and their slopes and the secondary wadis, while Acacia campoptila, Panicum turgidum, 116 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Aerva javanica, Indigofera spinosa, Fagonia indica, Ochradenus baccatus, Dipterygium glaucum and Rhazya stricta are widespread species of wadi beds. Phytogeography attempts to divide the globe into natural floristic units (Miller and Cope, 1996). According to Zohary (1973), the Hadhramaut region falls within SaharoArabian (covering the northern part) and Nubo-Sindian province of Sudanian-region (covering the southern part). Most of the Sahro-Arabian elements are found in the main wadis, while most of the Sudanian-region elements are found on the plateau. The northern limit of the Sudanian-region in the Hadhramaut region can be drawn with latitude ranging from 15° 22’ in Wadi Daua’n west to 15°35’ in Wadi Sah east but considerable numbers of African are found further north between 15° 49’ to 15° 50’ on the plateau of site 2 (see Figure 7.1). About 28% (42) of species recorded from the SW of the Egyptian desert (Shede, 2002) are found in the study area and most of them are Saharo-Sindian + SaharoZambezian (White, 1983). Among the Saharo-Sindian elements in the study area that extends to Egypt, Palestine, Arabia, southern Iraq and south western Syria and Iran (Zohary, 1973) are:Panicum turgidum, Aerva javanica. Phoenix dactylifera, Capparis cartilaginea, Cassia italica, Forskohlea tenacissima, Dipterygium glaucum, Leptadenia arborea, Periploca visciformis, Rhazya stricta, Fagonia indica, F. paulayana, F. bruguieri, Tribulus arabicus, Withania somnifera, Zygophyllum simplex, Z. album, Z decumbens, Z. coccineum, Lasiurus scindicus, Cyperus conglomeratus, Cymbopogon schoenanthus, Blepharis edulis, Pulicaria undulata, Heliotropium rariflorum, Chrozophora oblongifolia. These species are well represented in the main wadis. The Sudanian element is concentrated on the plateau. According to Zohary (1973), this element is confined partly to families and genera that are absent elsewhere in the Asian part of the Middle East. The Sudanian elements in the study area can be as a result of immigration; this may have happened in the period about 4,000 to 8,000 years ago when Sudanian plants started penetrating to the deserts of Palestine and Israel (Shmida and Aronson, 1986). Examples are Acacia mellifera, A. hamulosa, A. ehrenbergiana Calotropis procera, Salvadora persica, Tamarix aphylla, Tephrosia 117 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen apollinea, Calotropis procera, Capparis spinosa, Ziziphus leucodermis, Ziziphus spinachristi, Moringa peregrina, and Ochradenus baccatus. The following are endemic or near-endemic species recorded in the study sites: Acacia campoptila, Boerhavia elegans subsp. elegans, Aerva artemisioides subsp artemisioides, Euphorbia rubriseminalis, Farsetia linearis, Iphiona anthemidifolia, Reseda sphenocleoides, Ziziphus leucodermis., Merremia hadramautica, Cucumis canoxyi and Fagonia hadramautica; the last two species are found only on the plateau of site 2. Other elements are the Central American and Caribbean native species, such as the exotic Prosopis juliflora, or the Hadhramaut-Somalia connection such as Tephrosia dura and a Hadhramaut-Dhofar (Oman) connection such as Ziziphus leucodermis and Reseda sphenocleoides. 118 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 4.3.2. Number of individuals per hectare Site 1 2 3 Landform w Acacia campoptila 186 0 23 0 20 0 400 20 0 33 0 217 0 0 0 0 Aristida triticoides 7 0 115 33 0 100 14 0 0 1033 238 50 50 0 0 0 Blepharis edulis 7 0 8 0 0 0 1307 0 0 0 48 633 0 0 0 0 Boerhavia elegans 36 0 154 333 80 100 57 20 0 33 62 75 150 0 250 100 Dichanthium insculptum 79 0 138 417 360 50 0 200 0 433 743 50 80 0 800 957 171 67 200 200 0 0 250 0 0 83 19 342 0 0 0 0 7 0 115 350 80 0 0 60 25 267 214 0 30 0 0 14 14 0 0 0 0 0 393 60 0 0 271 608 0 0 0 0 Fagonia indica Farsetia linearis Indigofera spinosa Stipagrostis hirtigluma f ms pla wp sp w ms pla wp sp w ms pla wp sp 0 0 215 1433 120 250 129 140 250 350 376 0 640 100 1500 571 Tephrosia apollinea 836 33 192 0 60 71 300 0 33 14 71 92 200 0 0 0 No of sample plots 14 6 13 6 5 2 14 5 4 6 21 12 10 3 2 Table 4.2. Number of individuals per hectare for the most dominant species over the different landforms. w= main wadi, f= field or fallow lands, ms= slope facing main wadis, pla= plateau, wp= secondary wadi, sp= slope on plateau. The species Stipagrostis hirtigluma and Dichanthium insculptum are well represented in the 3 study sites, particularly on plateaus and mountain slopes facing the main wadis (Table 4.2). Aristida triticoides, Blepharis edulis, Dichanthium insculptum, Farsetia linearis and Indigofera spinosa are well represented in site 2, with Stipagrostis hirtigluma in site 3 and Boerhavia elegans and Tephrosia apollinea in site 1. Acacia campoptila, Tephrosia apollinea, Fagonia indica, Indigofera spinosa and Blepharis edulis are highly represented on wadi beds while Dichanthium insculptum and Stipagrostis hirtigluma are well represented on the slopes of plateaus. Table 4.3 shows number of individuals per hectare of each plant species in the 3 study sites:- 119 7 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Table 4.3. Number of individuals per hectare (density) of each species in the three study sites. Plant species Abutilon bidentatum Acacia campoptila Acacia ehrenbergiana Acacia hamulosa Acacia mellifera Acacia oerfota Aerva artemisioides Aerva javanica Compositae aff Flaveria Heliotropium rariflorum Alhagi graecorum Anticharis glandulosa Aristida triticoides Arnebia hispidissima Barleria aff bispinosa Blepharis edulis Boerhavia elegans Cadaba farinosa Cadaba heterotricha Calotropis procera Capparis cartilaginea Capparis spinosa Cassia senna Chascanum marrubifolium Chloris barbata Chrozophora tinctoria Chrysopogon aucheri Citrullus colocynthis Cleome brachycarpa Cleome droserifolia Cleome scaposa Cometes abyssinica Commiphora foliacea Commiphora kua Convolvulus arvensis Convolvulus glomeratus Corallocarpus glomeruliflorus Corchorus depressus Cornulaca amblyacantha Cressa cretica Crotalaria persica Crotalaria saltiana Cymbopogon schoenanthus Cynodon dactylon Cyperus conglomeratus Datura innoxia Dichanthium insculptum Dipterygium glaucum Enneapogon desvauxii Eragrostis sp. Erucastrum arabicum Euphorbia granulata Euphorbia rubriseminalis Fagonia indica Fagonia paulayana Farsetia linearis Forskohlea tenacissima Glossonema varians Site 1 2 3 0 65 33 26 11 11 0 63 0 13 85 0 43 0 11 4 111 0 0 13 7 9 4 0 30 39 0 7 48 11 4 7 9 0 4 0 0 13 0 28 7 2 87 65 0 2 159 20 24 0 65 4 0 143 28 89 7 0 4 115 17 6 8 9 0 19 8 0 0 4 215 6 36 364 45 2 2 8 2 2 0 6 0 23 43 4 17 8 32 4 30 21 0 11 8 9 15 0 32 0 17 2 9 0 362 8 4 8 0 0 2 83 9 123 0 0 0 70 3 3 3 0 3 41 0 3 0 5 43 0 5 205 97 0 0 0 0 0 0 0 0 3 8 0 46 14 165 0 3 5 0 3 14 5 16 0 0 27 27 0 5 0 262 8 165 8 0 0 0 138 124 11 0 3 Plant species Grewia erythraea Halothamnus bottae Helichrysum pumilum Heliotropium ramosissimum Hermannia paniculata Hochstetteri schimperi Indigofera oblongifolia Indigofera spinifolia Indigofera spinosa Iphiona anthemidifolia Iphiona scabra Jatropha spinosa Kohautia retrorsa Launaea sp. Leptadenia arborea Lindenbergia indica Maerua crassifolia Merremia hadramautica Moltkiopsis ciliata Ochradenus baccatus Panicum turgidum Peristrophe paniculata Phoenix dactylifera ‎ Plicosepalus curviflorus Pluchea dioscorides Portulaca oleracea Prosopis farcta Prosopis juliflora Pulicaria undulata. Reseda sphenocleoides Rhazya stricta Rhynchosia memnonia Salsola imbricata Salvadora persica Seddera latifolia Seetzenia lanata Senna holosericea Senna italica Senra incana Setaria verticillata Stipagrostis hirtigluma Striga asiatica Tamarix aphylla Tamarix arabica. Tarenna graveolens Tephrosia apollinea Tephrosia dura Tephrosia nubica Tephrosia subtriflora Tribulus arabicus Trichodesma calathiforme Withania somnifera Ziziphus leucodermis Ziziphus spina-christi Zygophyllum album Zygophyllum coccineum Zygophyllum decumbens Zygophyllum simplex Site 1 2 3 0 2 0 11 0 0 0 0 4 2 33 0 22 0 0 4 2 2 4 0 7 0 41 4 0 0 72 15 0 9 9 0 0 0 0 15 0 20 57 39 272 0 4 0 0 320 107 9 0 2 0 2 2 13 170 65 0 22 15 9 25 49 0 25 0 38 217 0 9 55 21 4 2 0 21 30 0 36 0 0 0 9 4 2 0 0 23 0 30 2 17 2 6 2 28 21 2 0 255 0 6 8 9 45 0 0 0 26 2 0 6 2 0 0 115 0 0 32 0 22 11 0 5 0 197 0 16 30 43 0 0 0 5 105 0 5 89 3 0 0 0 0 0 0 0 0 59 8 0 0 0 0 16 51 0 0 370 3 0 0 0 114 3 332 8 30 0 0 14 5 0 5 0 0 120 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen The number of individuals per hectare (Table 4.3) is calculated by dividing the total number of individuals into the total areas for each site as surveyed over the different landforms. 38 plant species are common to all 3 study sites. Variations in number can also be observed. Some species occur at high numbers in one site and at low numbers or are absent in other sites. Stipagrostis hirtigluma and Dichanthium insculptum are common in the 3 sites while Blepharis edulis and Indigofera spinosa are common in site 2 and 3 especially on wadi beds (Figure 4.2). Compared to sites 1 and 2, Cleome scaposa, Enneapogon desvauxii, Fagonia paulayana, Tephrosia nubica and Merremia hadramautica are common in site 3. Site 2 Figure 4.2. Number of individuals per hectare for the most dominant species across the 3 study sites. 121 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Variations in number of individuals are observed in the three study sites. Also some species occur at high numbers in one landform and at low numbers or are absent in other landforms. For example, the number of individuals of Acacia campoptila, Acacia ehrenbergiana and Tephrosia apollinea were high in relatively flat wadi beds and favourable moisture sites, while Dichanthium insculptum was high on the slopes of the rocky slopes of the plateau. Also, the low density of Enneapogon desvauxii, Merremia hadramautica, Stipagrostis hirtigluma and Rhazya stricta in sites 1 and 2 may be due to high human activities in these sites. Plant density is affected not only by the availability of the main limiting factors, namely water and minerals, but is also determined by the species composition of the annual vegetation (Kigel, et al., 2004). Variation in the density of plant species is observed during the field work; this can be due to timing of rainfall distribution or over-grazing. For example the density of Blepharis edulis, Boerhavia elegans, Tephrosia apollinea and Tribulus arabicus was high after the rain or flooding and the high density of Boerhavia elegans in site 1 can be due to low grazing pressure. Generally, many factors such as climatic conditions, grazing and topographic conditions have determined the density of the plant species and total seed bank under the soil surface. The long-term study of the effect of climate change and human activities such as grazing on the plant density and on the seed bank changes is needed. This can lead to better understanding of the dynamics of the seed bank and of the density of plant species. (For more see chapter 7) 4.3.3. Frequency and Importance values: Table 4.4 shows the frequency of each plant species in the different ecological zones across the 3 study sites. The number of individuals per hectare (density) explains how many plants there are in a unit of area but cannot tell us about how these plants are distributed; to know that, the frequency of each plant species per hectare was calculated (Table 4.4). Variations in abundance can also be observed. Some species occur at high frequencies in one landform and at low frequencies or absent in other landforms. For many of them, differences in mean abundance were obvious. 122 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen . Figure 4.3. The mean frequency in each landform for the three most abundant species in the study sites. Variations in abundance can be observed. The frequency was high for Stipagrostis hirtigluma (79%) on the slope of the plateaus (sp) and for Dichanthium insculptum (76%) and Boerhavia elegans (58%) in the secondary wadis of the plateaus (wp). Acacia campoptila was one of the most abundant species recorded in the wadis, where it had a mean frequency of 81%, but it occurred at less than 20% on the mountain slopes facing the main wadis and on the secondary wadis of the plateaus. The frequency of this species was the highest in site 3. Studies conducted in two tributary wadis in Hadhramaut by Bataher (2004), found that the frequency of Acacia campoptila was the highest. Other important species of the wadis are Fagonia indica with mean frequency of 67%, and Tephrosia apollinea with mean frequency of 46%. Stipagrostis hirtigluma, Dichanthium insculptum and Boerhavia elegans were widespread species, with a considerable variation in abundance from one landform to another. The mean frequency of these three species significantly increased from the main wadis to the adjacent slopes, then decreased on the plateaus and increased again to reach a peak over the slope of the plateaus in the case of the first species, and at the secondary wadis of the plateaus in the case of the last two species (see Figure 4.3). The low frequency of these species on the main wadis and adjacent slopes can be due to overgrazing by herders from local areas, who mainly herd their animals not far from their villages. Generally, the frequency of these species was low in the main wadis and over the plateaus. Dichanthium insculptum and Boerhavia elegans were missing on the 123 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen plateaus of sites 2 and 3, while Stipagrostis hirtigluma was missing on the main wadis of sites 1 and 3. Table 4.4. The frequency per hectare for each plant species. Variations in frequency can be observed. Some species occur at high frequencies in one site and at low frequencies or absent in other site. plant species / study site Abutilon bidentatum Acacia campoptila Acacia ehrenbergiana Acacia hamulosa Acacia mellifera Acacia oerfota Aerva artemisioides Aerva javanica Alhagi graecorum Anticharis glandulosa Aristida triticoides Arnebia hispidissima Barleria aff bispinosa Blepharis edulis Boerhavia elegans Cadaba farinosa Cadaba heterotricha Calotropis procera Capparis cartilaginea Capparis spinosa Cassia senna Chascanum marrubifolium Chloris barbata Chrozophora tinctoria Chrysopogon aucheri Citrullus colocynthis Cleome brachycarpa Cleome droserifolia Cleome scaposa Cometes abyssinica Commiphora foliacea Commiphora kua Convolvulus arvensis Convolvulus glomeratus Corallocarpus glomeruliflorus Corchorus depressus Cornulaca amblyacantha Cressa cretica Crotalaria persica Crotalaria saltiana Cymbopogon schoenanthus Cynodon dactylon Cyperus conglomeratus Datura innoxia Dichanthium insculptum Dipterygium glaucum Enneapogon desvauxii Eragrostis sp. Erucastrum arabicum Euphorbia granulata Euphorbia rubriseminalis Fagonia indica Fagonia paulayana Farsetia linearis Forskohlea tenacissima Glossonema varians 1 0 26 9 11 7 4 0 20 11 0 15 0 4 4 30 0 0 13 7 4 2 0 4 11 0 7 17 9 4 7 7 0 4 0 0 2 0 7 2 2 26 4 0 2 24 11 4 0 2 2 0 41 9 30 4 0 2 2 30 9 2 6 6 0 11 0 2 15 4 6 13 19 2 2 6 2 2 0 2 0 6 9 2 4 6 6 2 17 8 0 4 4 2 2 0 4 0 4 2 6 0 28 8 2 2 0 0 2 26 6 25 0 0 3 0 35 3 3 3 0 3 32 0 3 9 0 3 15 38 0 0 0 0 0 0 0 0 0 3 0 32 15 24 0 3 3 0 0 9 3 6 0 0 3 18 0 3 0 26 3 24 0 0 0 0 21 24 9 0 3 plant species / study site Grewia erythraea Halothamnus bottae Helichrysum pumilum Heliotropium ramosissimum Heliotropium rariflorum Hermannia paniculata Hochstetteri schimperi Indigofera oblongifolia Indigofera spinifolia Indigofera spinosa Iphiona anthemidifolia Iphiona scabra Jatropha spinosa Kohautia retrorsa Launaea sp. Leptadenia arborea Lindenbergia indica Maerua crassifolia Merremia somalensis Moltkiopsis ciliata Ochradenus baccatus Panicum turgidum Phoenix dactylifera ‎ Plicosepalus curviflorus Pluchea dioscoroides Portulaca oleracea Prosopis farcta Prosopis juliflora Pulicaria undulata. Reseda sphenocleoides Rhazya stricta Rhynchosia memnonia Salsola imbricata Salvadora persica Seddera latifolia Seetzenia lanata Senna holosericea Senna italica Senra incana Setaria verticillata Stipagrostis hirtigluma Tamarix aphylla Tamarix arabica. Tarenna graveolens Tephrosia apollinea Tephrosia dura Tephrosia nubica Tribulus arabicus Trichodesma calathiforme Withania somnifera Ziziphus leucodermis Ziziphus spina-christi Zygophyllum album Zygophyllum coccineum Zygophyllum decumbens Zygophyllum simplex 1 0 2 0 11 7 0 0 0 0 2 2 15 0 9 0 0 4 2 2 4 0 4 9 2 0 0 15 9 0 2 4 0 0 0 0 4 0 9 7 7 30 4 0 0 37 15 4 2 0 2 2 11 11 11 0 11 2 6 6 4 17 0 0 9 0 9 28 0 6 17 11 2 2 0 11 4 0 15 0 0 6 2 2 0 0 6 0 13 2 11 2 4 2 8 6 2 0 32 2 4 2 15 0 0 8 2 0 6 2 0 0 25 0 3 0 15 0 12 3 5 0 3 0 26 0 3 12 15 0 0 0 3 32 0 3 18 0 0 0 0 0 0 0 0 12 3 0 0 0 0 12 15 0 0 41 0 0 0 12 3 32 14 0 0 12 0 0 3 0 0 124 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Variations in frequency are characteristic. Some species occur at high frequencies in one site and at low frequencies or absent in other sites. For example the frequency of Stipagrostis hirtigluma, Dichanthium insculptum and Boerhavia elegans was high across the 3 study sites. Boerhavia elegans, Seetzenia lanata and Aristida triticoides was high in site 1, Farsetia linearis, Ochradenus baccatus and Indigofera spinosa were high in site 2 and Cleome scaposa, Enneapogon desvauxii, Fagonia paulayana and Merremia hadramautica were high in site 3 (Table 4.5a). plant species site 1 plant species site 2 plant species site 3 Fagonia indica 41 Stipagrostis hirtigluma 32 Stipagrostis hirtigluma 41 Tephrosia apollinea 37 Acacia campoptila 30 Boerhavia elegans 38 Boerhavia elegans 30 Dichanthium insculptum 28 Acacia campoptila 35 Farsetia linearis 30 Indigofera spinosa 28 Aerva javanica 32 Stipagrostis hirtigluma 30 Fagonia indica 26 Cleome brachycarpa 32 Acacia campoptila 26 Farsetia linearis 25 Merremia somalensis 32 Cymbopogon schoenanthus 26 Zygophyllum decumbens 25 Tephrosia nubica 32 Dichanthium insculptum 24 Boerhavia elegans 19 Dichanthium insculptum 26 Aerva javanica 20 Commiphora foliacea 17 Indigofera spinosa 26 Cleome brachycarpa 17 Heliotropium ramosissimum 17 Cleome scaposa 24 Aristida triticoides 15 Jatropha spinosa 17 Enneapogon desvauxii 24 Iphiona scabra 15 Aristida triticoides 15 Fagonia paulayana 24 Prosopis farcta 15 Ochradenus baccatus 15 Fagonia indica 21 Tephrosia dura 15 Tephrosia apollinea 15 Cymbopogon schoenanthus 18 Calotropis procera 13 Blepharis edulis 13 Panicum turgidum 18 Acacia hamulosa 11 Rhazya stricta 13 Blepharis edulis 15 Alhagi graecorum 11 Aerva javanica 11 Cleome droserifolia 15 Chrozophora tinctoria 11 Kohautia retrorsa 11 Halothamnus bottae 15 Dipterygium glaucum 11 Maerua crassifolia 11 Kohautia retrorsa 15 Heliotropium ramosissimum 11 Salsola imbricata 11 Senna italica 15 Ziziphus spina-christi 11 Acacia ehrenbergiana 9 Tribulus arabicus 14 Zygophyllum album 11 Chrysopogon aucheri 9 Heliotropium ramosissimum 12 Zygophyllum coccineum 11 Hochstetteri schimperi 9 Jatropha spinosa 12 Zygophyllum simplex 11 Indigofera spinifolia 9 Rhazya stricta 12 Acacia ehrenbergiana 9 Commiphora kua 8 Senna holosericea 12 Cleome droserifolia 9 Dipterygium glaucum 8 Tephrosia apollinea 12 Fagonia paulayana 9 Senna holosericea 8 Ziziphus leucodermis 12 Table 4.5a. The most high frequency species over the 3 study sites. Stipagrostis hirtigluma and Dichanthium insculptum are the most important plant species in the study sites (Table 4.5b). Table 4.5b. Most important plant species in the study areas with their total importance values. plant species site 1 plant species Site 2 plant species Site 3 Stipagrostis hirtigluma 193 Stipagrostis hirtigluma 306 Stipagrostis hirtigluma 346 Dichanthium insculptum 112 Farsetia linearis 132 Dichanthium insculptum 209 Cleome brachycarpa 111 Dichanthium insculptum 129 125 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Figure 4.4. Importance Value Curves for the three study sites. The curves have steep slopes because most of the percentage importance values in the study sites are distributed among just a few species (see table 4.10). For example, site 3 has a steep curve because most of the percentage importance values are distributed among just Stipagrostis hirtigluma and Dichanthium insculptum. Steeper curves in Figure 4.4 show that the site is only dominated by a few species, while straight curves show that the site is dominated by more species. The Importance value curves of the three study sites have steep slopes because most of the percentage importance values are distributed among just a few species. Site 1 is dominated by only three species, notably Stipagrostis hirtigluma, Dichanthium insculptum and Cleome brachycarpa; site 2 is dominated by three species notably, Stipagrostis hirtigluma, Farsetia linearis, Dichanthium insculptum, and site 3 is dominated by two species, notably Stipagrostis hirtigluma and Dichanthium insculptum. Stipagrostis hirtigluma and Dichanthium insculptum are widely distributed over the entire study sites in particular on the plateau zone. Table 4.6 shows the importance value of each species in different landforms of the three study sites. 126 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Table 4.6. Importance Value Index for each plant species over different landforms of the three study sites. w= main wadi, fal= field or fallow land, ms= mountain slope facing main wadi, pla= plateau, wp= secondary wadi, sp= slope on plateau. The study site 1 Plant species / landform w fal ms pla wp sp 2 w ms pla wp sp 3 w ms pla wp sp Abutilon bidentatum 0 0 0 0 0 0 0 0 0 3 0 0 0 0 0 0 Acacia campoptila 19 0 8 0 5 0 52 20 0 5 0 27 0 0 0 0 Acacia ehrenbergiana 9 0 0 0 0 0 7 0 0 0 1 0 0 0 6 0 Acacia hamulosa 7 0 0 0 9 0 0 0 0 4 0 0 0 0 6 0 Acacia mellifera 0 0 0 0 19 0 0 0 0 4 3 0 0 0 6 0 Acacia oerfota 0 0 0 0 15 0 0 0 0 4 3 0 0 0 0 0 Aerva artemisioides 0 0 0 0 0 0 0 0 0 0 0 0 0 53 0 0 Aerva javanica 1 0 31 10 0 0 4 0 0 5 2 11 12 0 13 0 Alhagi graecorum 1 34 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Anticharis glandulosa 0 0 0 0 0 0 0 0 0 0 2 0 0 0 0 4 Aristida triticoides 1 0 19 7 0 27 2 0 0 39 16 4 7 0 0 0 Arnebia hispidissima 0 0 0 0 0 0 0 0 0 0 2 0 0 0 0 0 Barleria aff bispinosa 4 0 0 0 0 0 1 0 0 14 0 2 0 0 0 0 Blepharis edulis 1 0 3 0 0 0 55 0 0 0 4 23 0 0 0 0 Boerhavia elegans 5 0 25 29 13 37 7 14 0 5 7 6 30 0 19 14 Cadaba farinosa 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 Cadaba heterotricha 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 Calotropis procera 7 0 0 5 0 0 3 0 0 0 1 0 0 0 0 0 Capparis cartilaginea 1 0 0 0 9 0 0 0 0 2 0 0 0 0 0 0 Capparis spinosa 0 9 0 0 0 0 1 0 0 0 0 0 0 0 0 0 Cassia senna 2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Chascanum marrubifolium 0 0 0 0 0 0 0 0 0 0 2 0 0 0 0 0 Chloris barbata 0 14 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Chrozophora tinctoria 0 0 1 13 23 0 0 0 0 1 0 0 0 0 0 0 Chrysopogon aucheri 0 0 0 0 0 0 0 0 0 30 0 0 0 0 10 0 Citrullus colocynthis 4 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Cleome brachycarpa 0 0 16 5 0 90 1 42 0 0 0 9 23 0 0 0 Cleome droserifolia 0 0 8 7 0 0 1 0 0 6 0 3 7 0 6 0 Cleome scaposa 1 0 3 0 0 0 8 0 0 0 0 5 45 0 0 0 Cometes abyssinica 0 0 5 5 0 0 0 0 0 0 2 0 0 0 0 0 Commiphora foliacea 0 0 0 0 12 18 0 0 0 3 13 0 0 0 0 4 Commiphora kua 0 0 0 0 0 0 0 0 0 2 5 0 0 0 0 7 Convolvulus arvensis 0 7 1 0 0 0 0 0 0 0 0 0 0 0 0 0 Convolvulus glomeratus 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 127 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Table 4.6 Continued The study site Plant species / landform 1 2 w fal ms pla wp sp 3 w ms pla wp sp w ms pla wp sp Corallocarpus glomeruliflorus 0 0 0 0 0 0 0 0 0 3 2 0 0 0 15 6 Corchorus depressus 3 0 0 0 0 0 0 0 0 0 0 2 0 0 0 0 Cornulaca amblyacantha 0 0 0 0 0 0 0 0 0 5 0 0 0 0 0 13 Cressa cretica 0 16 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Crotalaria persica 2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Crotalaria saltiana 0 0 0 0 5 0 0 0 0 0 0 0 0 0 20 0 14 0 0 0 47 0 2 0 0 6 0 8 0 0 0 4 Cynodon dactylon 0 28 0 0 0 0 0 14 0 0 0 0 0 0 0 0 Cyperus conglomeratus 0 0 0 0 0 0 0 23 0 7 0 0 0 73 0 0 Datura innoxia 0 3 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Dichanthium insculptum 7 0 17 24 36 28 0 76 0 30 50 4 14 0 Dipterygium glaucum 8 0 0 0 0 0 3 0 0 0 0 2 0 0 0 0 Enneapogon desvauxii 0 0 5 6 0 0 2 0 0 0 0 1 29 0 6 15 Eragrostis sp. 0 0 0 0 0 0 0 0 0 0 2 0 0 0 0 0 Erucastrum arabicum 0 18 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Euphorbia granulata 0 0 4 0 0 0 0 0 0 0 0 0 0 0 0 0 Euphorbia rubriseminalis 0 0 0 0 0 0 0 0 0 2 0 0 0 0 0 0 Cymbopogon schoenanthus Fagonia indica 34 101 19 8 29 20 0 0 24 0 0 3 3 19 0 0 0 0 Fagonia paulayana 0 0 3 0 27 0 0 0 0 1 4 0 40 0 10 9 Farsetia linearis 1 0 24 38 18 0 0 25 79 13 23 0 8 0 0 4 Forskohlea tenacissima 0 0 7 0 0 0 0 0 0 0 0 0 0 0 0 0 Glossonema varians 0 0 0 0 0 0 0 0 0 0 0 0 0 0 6 0 Grewia erythraea 0 0 0 0 0 0 0 0 0 0 6 0 0 0 0 0 Halothamnus bottae 0 0 3 0 0 0 1 0 0 7 0 1 0 0 10 20 Helichrysum pumilum 0 0 0 0 0 0 0 0 0 0 5 0 0 0 0 0 Heliotropium ramosissimum 2 0 7 0 0 0 3 0 0 2 5 3 3 0 0 6 Heliotropium rariflorum 5 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 Hermannia paniculata 0 0 0 0 0 0 0 0 0 0 0 0 0 0 18 0 Hochstetteri schimperi 0 0 0 0 1 0 0 0 0 3 7 0 0 0 0 0 Indigofera oblongifolia 0 0 0 0 0 0 0 0 0 0 0 2 0 0 0 0 Indigofera spinifolia 0 0 0 0 0 0 0 0 0 5 6 0 0 0 0 0 Indigofera spinosa 2 0 0 0 0 0 24 33 0 0 21 30 0 0 0 0 Iphiona anthemidifolia 1 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 Iphiona scabra 0 0 5 20 16 0 0 0 0 2 3 0 0 0 14 0 Jatropha spinosa 0 0 0 0 0 0 0 0 0 0 18 0 0 0 0 26 Kohautia retrorsa 5 0 0 0 9 0 0 0 0 8 4 6 0 0 23 0 Launaea sp. 0 0 0 0 0 0 0 0 0 0 2 0 0 0 0 0 Leptadenia arborea 0 0 0 0 0 0 0 0 0 2 0 0 0 0 0 0 Lindenbergia indica 0 0 3 5 0 0 0 0 0 0 0 0 0 0 0 0 128 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Table 4.6 Continued The study site Plant species / landform Maerua crassifolia Merremia hadramautica Moltkiopsis ciliata Ochradenus baccatus Panicum turgidum Peristrophe paniculata Phoenix dactylifera ‎ Plicosepalus curviflorus Pluchea dioscorides Portulaca oleracea Prosopis farcta Prosopis juliflora Pulicaria undulata. Reseda sphenocleoides Rhazya stricta Rhynchosia memnonia Salsola imbricata Salvadora persica Seddera latifolia Seetzenia lanata Senna holosericea Senna italica Senra incana Setaria verticillata Stipagrostis hirtigluma Striga asiatica Tamarix aphylla Tamarix arabica. Tarenna graveolens Tephrosia apollinea Tephrosia dura Tephrosia nubica Tephrosia subtriflora Tribulus arabicus Trichodesma calathiforme Withania somnifera Ziziphus leucodermis Ziziphus spina-christi Zygophyllum album Zygophyllum coccineum Zygophyllum decumbens Zygophyllum simplex 1 w 2 fal 1 1 0 0 3 0 0 2 0 0 6 6 0 0 4 0 0 0 0 0 0 4 1 0 0 0 3 0 0 48 25 3 0 1 0 0 0 4 38 13 0 3 0 0 0 0 0 0 69 0 0 0 35 0 0 0 0 0 0 0 0 0 0 3 22 19 0 0 0 0 0 4 0 0 0 0 0 3 0 7 0 0 0 0 ms 0 0 0 0 0 0 0 0 0 0 0 0 0 5 0 0 0 0 0 0 0 3 0 0 32 0 0 0 0 22 0 0 0 0 0 0 0 0 0 4 0 8 pla 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 8 0 0 0 0 89 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 8 wp 0 0 9 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 15 0 0 0 0 7 0 0 0 0 0 0 3 0 0 0 0 0 sp 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 39 0 0 0 0 57 0 0 0 0 0 0 0 0 0 0 0 5 0 0 0 0 0 w 3 ms 0 0 0 16 0 0 0 4 0 0 0 0 1 0 14 0 10 1 1 0 6 8 0 0 8 0 3 0 0 10 0 0 0 9 0 0 2 0 0 0 0 0 pla wp sp w ms pla wp sp 0 0 0 10 0 0 0 0 4 0 0 0 0 21 0 0 6 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 0 0 0 0 0 0 0 0 12 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 0 0 0 0 0 0 0 0 3 0 0 0 0 0 0 0 0 0 2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 17 2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 11 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 5 3 0 0 0 0 0 0 0 9 0 0 10 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 50 221 23 27 0 47 174 59 62 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 4 2 0 0 0 0 0 0 0 0 3 0 0 0 0 0 0 0 3 3 7 14 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 53 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 5 10 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 0 4 0 0 0 0 3 0 0 0 0 2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 5 0 0 0 0 0 16 23 0 0 0 0 0 0 0 0 0 0 0 0 0 0 129 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Figure 4.5 The distribution of total Importance Values over each landform in the study sites. w= main wadi, fal= field or fallow land, ms= mountain slope facing main wadi, pla= plateau, wp= secondary wadi, sp= slope on plateau. Numbers refer to the study site. pla2 has steep curves because it is dominated by only two species Stipagrostis hirtigluma and Farsetia linearis. Table 4.6 and Figure 4.5 show the importance values for each plant species over different landforms for the three study sites. The Importance Values can be seen to vary distinctly between the different landforms. As can be observed, there are a few species in these habitats that have more important value than any other species. Steeper curves show that the landform is dominated by only a few species or that most of the percentages Importance Values are distributed among just a few species; while straight curves show that the landform is dominated by more species or that most plants species within this landform have similar percentage importance values. For example, the (pla2), (pla3), (ms2) and (sp3) curves (Figure 4.5) have a steep slope because most of the percentages Importance Values in these habitats are distributed among just a few species. The first habitat was dominated by Stipagrostis hirtigluma and Farsetia linearis, the remaining three habitats were dominated by Stipagrostis hirtigluma and Dichanthium insculptum (Table 4.6). 130 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 4.3.5. Floristic diversity In total 117 plant species were found in the study area (about 10 taxa per hectare). The mean number of species per sample plots was 7.6. The maximum number of species at any one sample plot was 20 and the minimum was two. Many species were present only in one or very few sample plots. Out of 117 plant species, only 7 (6%) were present in more than 30% of the sample plots. These species are: Acacia campoptila, Fagonia indica, Tephrosia apollinea, Boerhavia elegans, Dichanthium insculptum, Farsetia linearis and Stipagrostis hirtigluma. The first 3 species are mainly found on wadi beds, the rest of the species are widely distributed throughout the landforms, with a high frequency on the plateau and on the rocky slopes facing the main wadis. The calculation of plant diversity using numbers of individuals for each plant species was achieved by using Shannon and Simpson indices. The use of presence and absence data failed to give logical results, for example the evenness values for all areas were 1. The Simpson index was calculated as 0.96 for site 1, and 0.94 for sites 2 and 3, while the Shannon index is calculated as 3.55 for site 1, 3.41 for site 2 and 4.24 for site 3. The three study sites (Table 4.7) have different numbers of plant richness (74 in 1, 83 in 2 and 61 in 3) with different total numbers of individuals (1275 in 1, 1555 in 2 and 1156 in 3). As can be seen, site 2 has more species and individuals. However, both Shannon’s and Simpson’s diversity indices show that the study site 1 was more diverse than sites 2 and 3. This is because the total number of individuals in site 1 is almost equally distributed between the most plant species (Figure 4.6), in other word has more evenness (Wilsey et al., 2000) than sites 2 and 3 (Table 4.7). For example In site 2, 50 % of the individuals belong to only 5 species (Table 4.3 and Figure 4.6), namely Aristida triticoides, Blepharis edulis, Indigofera spinosa, Dichanthium insculptum and Stipagrostis hirtigluma, while in site 3, 50% of the individuals belong to only 4 species, namely Blepharis edulis, Dichanthium insculptum, Stipagrostis hirtigluma Tephrosia nubica, Indigofera spinosa, Cleome scaposa and Enneapogon desvauxii, while the distribution was among 8 plant species in site 1. 131 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Simpson's index Shannon's index No of No of Area Index Evenness Index Evenness Species individuals 1 0.96 0.968 3.55 0.824 74 1275 2 0.94 0.953 3.41 0.773 83 1555 3 0.94 0.959 4.24 0.789 61 1156 Table 4.7. Simpson's and Shannon's indices of the 3 study sites. Both Simpson and Shannon indices show that site 1 is more diverse than site 2 and 3. The number of species of site 2, is high, but the evenness is low, as a result the diversity was low. Figure 4.6. The distribution patterns of number of individuals between the plant species in the three study sites. The total number of individuals in site 1 is more evenly distributed. 50% of individuals were distributed among 5 plant species in site 2 and among almost 6 plant species in site 3, while the distribution was among 8 plant species in site 1. The comparison in diversity between the 3 study sites can be further illustrated by plotting percentage cumulative abundance against log species rank over the different landforms (Figure 4.7). In this case, the lower line has the higher diversity (Shaw et al., 1983). The output (Figure 4.7) shows that site 1 has higher diversity than the other two sites, and this result supports the Shannon and Simpson indices result shown earlier in Table 4.7. 132 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Figure 4.7. Cumulative abundance curves, known also as k-dominance curves (John and Gallegos, 2001) for the three study sites showing cumulative ranked abundances plotted against log species rank. The lower line, which in this case represents site 1, has the higher diversity. The biodiversity of different landforms was also calculated using the Shannon index (Figure 4.8). Figure 4.8. Diversity of the different landforms using Shannon index. w= main wadi, fal= field or fallow lands, ms= slope facing main wad, pla= plateau, wp= secondary wadi, sp= slope on plateau. Numbers refer to the study site. w1, wp2 and sp2 have the highest diversity values, while pla2 and pla3 have the lowest. A habitat with high evenness has a larger diversity index than a habitat of higher or the same number of species with lower evenness (e.g. fal 1 & pla 1 and w2 & w3). 133 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Figure 4.9. Cumulative abundance curves, known also as k-dominance curves (John and Gallegos, 2001) for the different landforms in the study sites showing cumulative ranked abundances plotted against log species rank. Almost all landforms of site 1 have higher diversity and all landforms of site 2 have low diversity, except the secondary wadi of site 2 (wp2). This due to high number of species found over this landform. The Shannon diversity index of main landforms varied from 0.3 to 2.8, the mean was 2.1. Figure 4.8 and 4.9 show that main wadis, wp2, sp2 and ms1 are the most diverse habitats. These index values correspond to the high number of plant species at each habitat, while the landforms pla2, pla3 and sp1, which are characterized by a low number of species (from 2 to 7), have lower diversity. In addition, a habitat with high evenness has a larger diversity index than a habitat of higher or the same richness with lower evenness. For example fal1 has the same species of pla1, but fal1 is more diverse than pla1, and w3 has less species than w2, but w3 is more diverse than w2 because both fal1 and w3 have more evenness than pla1and w2 (Figure. 4.8). The diversity indices calculated for the main wadis are generally high and range between 2.4 and 2.8. There are no significant differences between the diversity of these wadis. The relatively high diversity of the wadi in site 1 is due to the fact that 12 species in this landform are confined to cultivated and palm field areas, with species composed of weeds such as Alhagi graecorum, Phoenix dactylifera, Cressa cretica, Capparis spinosa, Cynodon dactylon, Chloris barbata, Convolvulus arvensis, Datura innoxia and Withania somnifera. (see vegetation association 1, chapter 5). 134 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Site 1 was the most diverse, while site 2 was the richest in plant species. Plant species of some locations on slopes (sp) and secondary wadis of plateaus (wp) have been found to be more diverse with a very dense cover when surface soil is irrigated by contaminated water as in case of site 2 (plate 4.1). In these sites, species that require high moisture such as Abutilon bidentatum, Pluchea dioscoroides, Leptadenia arborea, Pulicaria undulata and Tamarix arabica are abundant. On the other hand, common species of slopes and wadis of plateaus such as Stipagrostis hirtigluma, Indigofera spinosa and Maerua crassifolia are absent here. Plate 4.1. Dense vegetation cover dominated by Dichanthium insculptum and Farsetia linearis on the slope of plateau, site 2. Bare ground surface (foreground) is due to intolerance of the vegetation to salts and chemicals in the waste water. All plant species of the mountain slopes facing the main wadis (ms) are herbaceous, except for trees of Acacia campoptila, which are found as individuals in favoured locations (in particular in gullies at the foot of these slopes). Due to the steep slopes, the soil is probably affected by runoff and exposed to direct evaporation as a result the lower moisture (Danin et al., 1997), and the shallow soil cover is not sufficient to support woody or shrubby vegetation. The dominant species here are Dichanthium insculptum, Boerhavia elegans, Cleome brachycarpa Stipagrostis hirtigluma and Farsetia linearis and they are very well developed on such dry habitats (see Associations 8 and 9, chapter 5).The frequency of these species range between 15% and 60%. 135 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Almost half of plant species on mountain slopes facing the main wadis at site 1 are annual herbs and only 15 plant species are found here. Compared to the rest of the slopes, the slopes of site 1 have high diversity (Figures 4.8 and 4.9). The mountain slope is subjected to overgrazing by local herders and Bedouins. The structure of vegetation of this habitat cannot resist overgrazing, and consequently the vegetation cover has been reduced dramatically. The diversity of the slopes on the plateaus (sp) is medium to high and ranges between 1.6 and 1.8 (Figure 4.8). Here the diversity on slopes of site 1 is the lowest. The diversity of slopes of site 2 was high due to the increase of moisture content in one sample plot. This plot is regularly irrigated by contaminated water from the oil workings (plate 1); 23% of plant species that occur on this habitat are confined to this location. The plateaus generally have the lowest diversity. This could be due to physical factors, which are characterised by high proportion of exposed hard jagged limestone, dolomite and chert fragments (Komex International Ltd 1999), and a high proportion of sand and stones (Plate 4.2). Similar landscapes with almost bare vegetation are found in the desert areas throughout the world (Danin, 1983). The plateau of site 2 contained the lowest diversity; this could be (in addition to the physical factors), due to the intensive activities of oil workings. Plate 4.2. Bare dolomite and chert fragments surface with sandy loam textured on the plateau of area 1. (A) General view, (B) a close view. 136 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Generally most of surfaces on the plateau of sites 2 and 3 are almost devoid of vegetation. The diversity calculated for the plateaus (pla) of site 1 is higher than sites 2 and 3. The higher diversity of plateau in site 1 may be due to one or more of the following reasons: 1- The moisture content and clay content of the soil samples collected is higher (see soil data table chapter 5). 2- The landform of the plateau in site 1 consists of small hills with alluvial fans and drainage lines formed at the foot of these hills, which support a habitats characterise by relatively deep, moist soil. The E.C of the alluvial deposits in these drainage lines is relatively high (3 %). This salt probably built up as water is evaporated very rapidly from the surface. Halophyte plants that have adapted to tolerate a high salt regime, such as Zygophyllum simplex, Fagonia indica Dichanthium insculptum aand Tamarix arabica re confined here. 3- Plant species of some locations have been found to be more diverse due to the removal of the surface layer by local people to make protection barriers to guard the road. The exposed soil below the surface layer is able to support greater plant life (i.e. the subsurface horizons are the most favourable). 4- Disturbed areas can be invaded easily by new species, which increase species richness. (Focht and Pillar, 2003). Species of disturbed areas such as Calotropis procera, Farsetia linearis and Chrozophora oblongifolia, were only found here. In addition to high organic matter, soil fertility and high moisture content (Furley and Newey, 1983; Krishnaswamy and Richterb, 2002; Norton et al., 2003; Walker and William, 2002), the increase of plant species on the alluvial fans that are located at the bottom of the plateaus, can be also due to immigration of some species from the plateaus down to the wadis. For example, species such as Maerua crassifolia, which is usually confined to the plateau, is found at the foot of the wadi bed in site 1. 137 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen It is understood from the previous section that study site 1 was more diverse than sites 2 and 3, while site 2 was the richest in plant species. This is because the total number of individuals in site 1 is almost equally distributed between the most plant species. The diversity of the slopes on the plateaus (sp) is medium to high. The plateaus generally have the lowest diversity. The main wadis, secondary wadis of site 2, the slope of plateaus of site 2 and the mountain slopes of site 1 are the most diverse habitats. The index values of these landforms correspond to the high number of plant species at each habitat, while the plateaus of site 2 and 3 and the slopes of plateau of site 1, have lower diversity, these landforms are characterized by a low number of species. Plant species of some locations have been found to be more diverse with a very dense cover when surface soil is irrigated or when rocky surface layer is removed. Generally, abiotic and biotic factors account for variation in species composition and diversity. 4.3.6. Species richness and density over the landform gradient Figure 4.10 shows that number of species per hectare varies across the different landforms. The number of species over the plateaus increases steadily from exposed flat surfaces, to the slopes, then to the bottom of the secondary wadis. In the main wadis, the number of species for the three study sites increases from zero on the cliffs, to the slopes that face the main wadis, then to the main wadi beds. The peak in the number of species (1000 species/ha) is found at the bottom of the secondary wadis that cut the plateaus of site 3 (wp3). The lowest is found at the plateaus of sites 2 and 3. Species richness was high on the valley floor on both the main wadis and the secondary wadis and was low on the mountain slopes, plateaus and slopes of the plateaus. This is possibly due to the movement of large volumes of water and sediments down the slopes during the rainy season (Robert & Thomas 1997); as a result, the amount of moisture is increased at the subsurface store. 138 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Figure 4.10. Number of species per hectare over the landform gradient. w = wadi, ms = mountain slopes facing main wadis, cliff = very steep cliffs facing the main wadis, wp = secondary wadi, sp = slope on plateaus, pla = plateau. The number of species decreases from main wadi to the cliff and on the plateau it increases from exposed flat surfaces to the bottom of the secondary wadis. The continuous deposition over the foot slopes also increases the depth of the soil over the wadi bottom as well as the amount of organic matter and moisture. Conversely, increases in erosion over the slopes decreases the depth of the soil of these lower slopes (Furley & Newey, 1983). The low number of species at wp1 is due to the landform characteristics of relatively narrow shallow wadis with steep short sides; in addition, the surface of this wadi has solid pavement with a soil cover of less than 2%. Figure 4.11 shows the fluctuation in number of individuals over the altitude gradient. Elevation has no obvious effect on the distribution of numbers of individual species. The high number of individuals at 690 and 930 m and low number of individuals at 965 m are due to landforms characteristics; in the first two, the landform is a deep valley bottom and in the second, the landform is an exposed surface of the plateaus. A sharp decline in numbers of individuals from 930m to 965m of the plateau areas in sites 2 and 3 is due to landform changes from relatively wet wadi bottom to the slope and then to the exposed dry surface of the plateaus. The high number of individuals at the plateau of site 1 is due to human disturbance (see the previous 139 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen section). Elevation has an effect on the distribution of some individual species. For example some species such as Acacia mellifera, Corallocarpus glomeruliflorus, Jatropha spinosa and Chrysopogon aucheri are confined to high altitude areas (over 900m), while others like Acacia ehrenbergiana, Zygophyllum album, Rhazya stricta and Alhagi graecorum are confined to low altitude areas (less than 700m). Figure 4.11. Species density per hectare along the landform gradient. w = wadi, ms = mountain slopes facing main wadis, cliff = very steep cliffs facing the main wadis, wp = secondary wadi, sp = slope on plateaus, pla = plateau. The high number of individuals at 690 and 930 m and low number of individuals at 965 are due to landforms characteristics; in the first two, the landform is a deep valley bottom and in the second, the landform is an exposed surface of the plateaus. A sharp decline in numbers of individuals from 930m to 965m of the plateau areas in site 2 and 3 is due to landform changes from relatively wet wadi bottom to the slope and then to the exposed dry surface of the plateaus. The high number of individuals at the plateau of site 1 is due to human disturbance. The previous section showed that the number of species per hectare varies across the different landforms. Species richness was high on the valley floor on both the main wadis and the secondary wadis and was low on the mountain slopes, plateaus and slopes of the plateaus. This is due to the movement of large volumes of water and sediments down the slopes during the rainy season, as well as the depth of the soil over the wadi bottom and the amount of organic matter and moisture were high. The low number of species is due to the landform characteristic of relatively narrow shallow wadis with steep short sides. Generally the density and the distribution of plant species were different from one study site to another, as well as from one landform to another. The landforms, with their component, such as moisture, were the main factors affecting the 140 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen distribution and the density of plant species. Human activities play an important role in some sites. 4.3.7. Species accumulation curves Curves for species versus the number of sample plots were plotted for each area in order to ascertain the minimum number of samples that should be taken to be representative of the habitats in the study areas (Willott, 2001). The total number of species increases with area, leading to an increased probability of finding more species with larger areas. At some spatial scale, all or most of species will be counted before the entire area is sampled. The curves of species versus number of sample plots for the three study sites tend to flatten, indicating that the number sample sites was sufficient to record their floristic composition (Mueller-Dombois and Ellenberg, 1974; Thompson et al., 2003). The curves rise quickly at the early stage, meaning that most of the species are recorded in just the first few sampling plots, and subsequently they almost level out. The total numbers of samples that have been taken in the 3 study sites were as following: site 1 (49), site 2 (54) and site 3 (37). Figures 4.12 to 4.14 show that 20 sample plots (totalling 0.2 hectares) would be satisfactory to cover almost 74%, 69% and 84% of plant species recorded for site 1, site 2 and site 3 respectively. The curves increase very slowly with continued sampling; for example conducting a further 20 sample sites would add 19 (26%), 26 (31%) and 10 (16%) new species respectively to the three study sites. 141 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Figure 4.12. Curve of species versus number of sample sites for site 1. At the beginning, 41 plant species (55.4%) were counted in the first 10 sample plots (totalling 0.10 ha), at a rate of almost 4 new species per sample site. 20 sample plots was satisfactory to account for 80% (almost 60 plant species). By the time 60 species are counted, new species are being recorded at a rate of one new species per sample plot. After 65 species had been recorded, the rate of new species recording drops sharply. Figure 4.13. Curve of species versus number of sample plots for site 2. At the beginning, 41 plant species (49.4%) were counted in the first 10 sample sites (0.10 ha), at a rate of almost 5 new species per sample sites. 20 sample plots were satisfactory to account 71% (almost 59 plant species). After 68 species had been recorded, the rate of new species recording drops sharply, almost 1 new species every sample plots, then the rate slightly drop again to about 1 new species every 2 to 3 sample plots. 142 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Figure 4.14. Curve of species versus number of sample plots for site 3. At the beginning, 33 plant species (56%) were counted in the first 10 sample plots (0.10 ha), at a rate of almost 4 new species per sample sites. 20 sample plots were satisfactory to account 83% (almost 49 plant species). After 57 species were recorded, the rate of new species recording drops sharply, to almost 1 new species every 2 sample plots. Figure 4.15 Curve of species versus number of sample plots for the three study sites. Site 2 was the most species rich and site 3 the least. From above figure it can be seen that site 2 is the richest in terms of plant species and site 3 the least. 143 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Most of the species of the study sites have been recorded in just the first few sampling plots, and the survey has captured the vast majority of plant species presented in the study sites. The number of species encountered did not increase with area, signifying that most species were counted before the total area was sampled. This suggests that most of plant species in the study areas are uniformly dispersed throughout the habitats (Scheiner et al., 2000). The great numbers of species were encountered at 30 sample plots. Thus, optimal number of 10 X 10m samples, which should be taken to be truly representative of each study site, is concluded to be 30. 4.3.8. Similarity between the three study sites and between the different landforms From Figure 4.16, it can be observed that sites 2 and 3 are very similar to each other, with similarity of 50%. Site 1 showed a similarity of less than 46% with sites 2 and 3. The result is also shown in Table 4.8. Figure 4.16. Cluster analysis based on the Jaccard distance analysis of the 3 study sites. Step Clusters 1 2 2 1 Distance 50 54.8 Similarity 50 45.2 Joined 1 2 1 Similarity Matrix site 1 site 2 site 3 site 1 * 44.0 45.2 site 2 * * 50 site 3 * * * Table 4.8. Jaccard distance Index between the 3 study areas. A decline of similarity with increasing distance between areas is obvious. High similarity (50%) between area 2 and 3 and low similarity (44%) between area 1 and 2 can be seen. 144 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen From Figure 4.17 we can see that there are 4 groups of similarities between the different landforms. These groups can be summarised as follow: Group 1 comprises the main wadis of the 3 study sites (w1, w2, &w3) and cultivated fields of site 1(fal1) with a similarity between 45% and 51%, Group 2 comprises mountain slopes of the 3 study sites (ms1, ms2 & ms3), plateau of site 1 (pla1) and slopes on the plateaus of site 1 and 3 (sp1 & sp3) with similarity between 33% and 52%, Group 3 comprises secondary wadis of the three study sites (wp1, wp2 & wp3) and slopes on the plateau of site 2 (sp2) with similarity between 30% and 36% and group 4 comprises plateaus of site 2 and 3 (pla2 & pla3) with similarity of 25%. Figure 4.17. Cluster analysis based on the Jaccard Analysis for the different landforms in the study sites. w = main wadi, fal 1 = palm groves and fields of area 1, ms = slope facing main wadi, pla = plateau, sp = slope on plateau, wp = secondary wadi. The number in front of each landform symbol refers to the study site. A decline of similarity with increasing distance between areas is obvious. The cluster above shows 4 landform groups of similarities, ms1 and pla1 and w2 and w3 are the most similar habitats, with similarity ranges between 51% to 52% and distance ranges between 48.3 and 48.9 145 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen The cluster above shows 4 landform groups of similarities in which ms1 and pla1 and w2 and w3 are considered as the most similar habitats over the study sites. The formers (ms1 & pla1) are characterised by low number of species ranging from 17 to 27 and the latter (w2 & w3) by high number of species ranging between 33 and 35 (Figure 4.9a). The least rich habitats are pla2 and pla3 with a number of species less than 3. Groups 1 and 3 are characterised by high number of individuals and richness. As mentioned in previous floristic diversity section, site 1 was more diverse than sites 2 and 3; this is because the total number of individuals in site 1 is almost equally distributed between the most plant species (see section 4.3.5). The similarity between the different landforms has also close link with the number of plant species, for example the main wadis which are characterised by high diversity, high richness and relatively high evenness were grouped together (group 1), and the secondary wadis of the plateau, which characterised by high evenness, relatively high diversity and richness were grouped together (group 3), because of their high diversity index, the slopes of the plateau of site 2 were put with this group. The plateaus of sites 2 and 3 (group 4), characterised by low diversity and richness (see Figures 4.8 and 4.18). Figure 4.18. Diversity, evenness and number of species (richness) of the different landforms using Shannon index. w= main wadi, ms= slope facing main wad, pla= plateau, wp= secondary wadi, sp= slope on plateau. Main wadis and secondary wadis characterized by high diversity and richness, while the plateaus by low diversity and richness. 146 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 4.4. Summary A total of 117 plant species were recorded in the 3 sites of the study area, of which 5 are endemic to Hadhramaut. The density and the distribution of these species were different from one study site to another, as well as from landform to another. The landforms with their component such as moisture were the main factors affecting the distribution and the density of plant species. Human activities play an important role in some sites. Most of the plant species of the study sites are desert-adapted species which botanists have classified as Saharo-Sindian, or Saharo-Arabian (Zohary, 1973). Some of these species are of African origin. These species have adapted to the current harsh climate. As a result, many of these plants are now so adapted that they cannot live outside of the arid and semi-arid region that characterises the study area. On the other hand, some of these species do not exist in other parts of the world and these are considered as endemic. The following are endemic and near-endemic species of the study area: Acacia campoptila, Boerhavia elegans subsp elegans, Anogeissus bentii, Aerva artemisioides subsp artemisioides, Cucumis canoxyi Euphorbia rubriseminalis, Farsetia linearis, Iphiona anthemidifolia, Reseda sphenocleoides, Ziziphus leucodermis. Merremia hadramautica and Fagonia hadramautica. The plant species composition varies according to topography and drainage, soil nutrient status and environmental hazards (Furley, 1999). Variation in plant species from one landform to another and within one landform is obvious in the study sites. For example, the following 2 species are well represented in the 3 study sites: Stipagrostis hirtigluma and Dichanthium insculptum; they are particularly on plateaus and mountain slopes facing the main wadis, with a considerable variation in abundance from landform to another (see table 4.9). Both species are considered to be the most important plants in the study sites. Acacia campoptila is one of the most abundant species recorded in the wadis, where it has a mean frequency of 81%. The Importance Value Index of each plant species was calculated to determine the importance of each plant species in the study sites; this indicated that Stipagrostis hirtigluma, Dichanthium insculptum, Cleome 147 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen brachycarpa and Farsetia linearis are the most important plant species in the study sites. Species richness is higher on the valley floor in both the main wadis and the secondary wadis and was low on mountain slopes, plateaus and slopes of the plateaus. w1 ms1 pla1 Tephrosia apollinea 48 Stipagrostis hirtigluma 32 Stipagrostis hirtigluma 89 Zygophyllum album 38 Aerva javanica 31 Farsetia linearis 38 Tephrosia dura 25 Fagonia indica 29 Boerhavia elegans 29 Acacia campoptila 19 Boerhavia elegans 25 Dichanthium insculptum 24 Fagonia indica 19 Farsetia linearis 24 Fagonia indica 20 wp1 sp1 field Cymbopogon schoenanthus 47 Cleome brachycarpa 90 Phoenix dactylifera 69 Dichanthium insculptum 36 Stipagrostis hirtigluma 57 Prosopis farcta 35 Fagonia paulayana 27 Seetzenia lanata 39 Alhagi graecorum 34 Chrozophora tinctoria 23 Boerhavia elegans 37 Cynodon dactylon 28 Acacia mellifera 19 Dichanthium insculptum 28 Senra incana 22 w2 ms2 pla2 Blepharis edulis 55 Dichanthium insculptum 76 Stipagrostis hirtigluma 221 Acacia campoptila 52 Stipagrostis hirtigluma 50 Farsetia linearis 79 Fagonia indica 24 Cleome brachycarpa 42 Indigofera spinosa 24 Indigofera spinosa 33 Ochradenus baccatus 16 Farsetia linearis 25 wp2 sp2 Aristida triticoides 39 Dichanthium insculptum 50 Chrysopogon aucheri 30 Stipagrostis hirtigluma 27 Dichanthium insculptum 30 Farsetia linearis 23 Stipagrostis hirtigluma 23 Zygophyllum decumbens 23 Pulicaria undulata. 17 Indigofera spinosa 21 w3 ms3 pla3 Tephrosia nubica 53 Stipagrostis hirtigluma 47 Stipagrostis hirtigluma 174 Indigofera spinosa 30 Cleome scaposa 45 Cyperus conglomeratus 73 Acacia campoptila 27 Fagonia paulayana 40 Aerva artemisioides 53 Blepharis edulis 23 Boerhavia elegans 30 Merremia somalensis 21 Enneapogon desvauxii 29 wp3 sp3 Stipagrostis hirtigluma 59 Dichanthium insculptum 101 Dichanthium insculptum 34 Stipagrostis hirtigluma 62 Kohautia retrorsa 23 Jatropha spinosa 26 Crotalaria saltiana 20 Halothamnus bottae 20 Boerhavia elegans 19 Enneapogon desvauxii 15 Table 4.9. Most important plant species in the 3 study sites with their importance values per hectare. Stipagrostis hirtigluma and Dichanthium insculptum are well represented in the 3 study sites. Acacia campoptila was one of the most abundant species recorded in the wadis. The diversity of main wadis and the secondary wadis is high. In these landform systems, the complex micro-landforms such as minor depressions, terraces, colluvial and alluvial deposits and levees which lead to accumulate moisture and nutrients or salinity, can help in creating significantly higher species richness than the adjacent slopes (Furley, 1974; Goebel et al., 2003). Compared to the rest of the mountain slopes, 148 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen the slopes of site 1 have high diversity. The mountain slope of sites 2 and 3 are more subjected to overgrazing by local herders and nomadic Bedouins. The structure of vegetation of this habitat cannot resist overgrazing, and consequently the vegetation cover has been reduced dramatically. The diversity of the slopes on the plateaus (sp) is medium to high and ranges between 1.6 and 1.8. Here the diversity on slopes of site 1 is the lowest. The diversity of slopes of site 2 is high. The plateaus generally have the lowest diversity. The diversity calculated for the plateaus (pla) of site 1 is higher than for sites 2 and 3. Generally, most of the surfaces on the plateau of sites 2 and 3 are almost devoid of vegetation. The diversity of slopes of site 2 was high due to the increase of moisture content in one sample plot. This plot is regularly irrigated by contaminated water from the oil workings (plate 4.1). The plateaus generally have the lowest diversity. This could be due to physical factors, which are characterised by high proportion of exposed hard jagged limestone, dolomite and chert fragments with thin soil and a high proportion of sand and stones (plate 4.2). The plateau of site 2 contained the lowest diversity; this could be (in addition to the physical factors), due to the intensive activities of oil workings, (see section 4.3.5). Phytogeography attempts to divide the globe into natural floristic units (Miller and Cope, 1996). According to Zohary (1973), the Hadhramaut region falls within SaharoArabian (covering the northern plateau) and Nubo-Sindian province of Sudanian-region (covering the southern plateau). Most of the Sahro-Arabian elements (Holarctic origin) are found in the main wadis, while most of the Sudanian-region (Paleotropical origin) elements are found on the plateau. The northern limit of the Sudanian-region can be drawn with latitude ranging from 15° 49’ to 15° 50’ on the plateau of site 2. The flora of the study area has strong links with adjacent parts of Africa and Arabia but some species and genera have interesting distributions for example, Ochradenus baccatus which is restricted to the main wadis of sites 2 and 3 has a typical Saharo-Sindian distribution being found in the deserts of eastern north Africa, Somalia, Sudan, Egypt, the Gulf countries, Iran and Pakistan (Miller, 1984). The related species is the near-endemic Ochradenus arabicus, and is found in the southern part of the Hadhramaut region and in Oman, Saudi Arabia and UAE, the distribution patterns of this species is associated with fog-affected areas. There are several local endemics of Ochradenus in southern Arabia including O. aucheri subsp. aucheri from UAE and 149 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Oman, O. harsusiticus and O. gifrii from Oman (Ghazanfar, 2003; Miller, 1984) and al Mahara, Yemen (Thulin, et al., 2001). The endemic Aerva artemisioides subsp artemisioides has wide range distribution, but with very low frequency, for example in the study area it was only found in 2 locations. There are 2 sub species of Aerva artemisioides: subsp. artemisioides, which is endemic to Hadhramaut and subsp. batharitica, which is endemic to central Oman. Cucumis canoxyi is an endemic. It has a wide distribution from the southern mountains at 615 m asl to study site 2 at 950 m. It has previously been noticed that some plant species have relatively high values for density and a low frequency. Low frequency of plants means unequal distribution over the study areas. For example Blepharis edulis has high density values in sites 2 and 3 but is not well distributed (low frequency). This plant is only abundant in rocky habitats in the main wadis and less common or absent from other habitats. This also can be seen in other species such as Indigofera spinosa and Tephrosia apollinea, which are found mainly on wadi beds and on relatively moist sites. In this chapter the floristic composition of the study sites, their important values and the abundance have been described and analysed, as well as the plant diversity, and the vegetation fond over each landform in the 3 study sites have been assessed. In the next chapter the plant species that been identified in chapter 4 will be further explored in terms of association and structure, in order to recognize the types of vegetation and the structure of plant communities of the study sites. 150 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Chapter 5. Vegetation associations: structure and environment analysis 5.1. Introduction This chapter presents the results of the study of floristic vegetation associations and structure in the three study sites located at Wadi Hadhramaut, namely Shibam (site 1), Wadi Adim (site 2) and Wadi Athahab (site 3). The aims of this chapter are to answer the following research questions:    What are the types of vegetation and the structure of plant communities? What are the factors and processes affecting vegetation change? What has been the nature of land use change and how has it affected plant composition and distribution? The structure of the presentation is illustrated in Figure 5.1. Fig. 5.1 Diagram showing the structure of this chapter 151 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen The main aims of vegetation sampling were to identify plant species and their density cover, and to explain the locality of these plant species. There are many sampling techniques that study the vegetation and its relationship to environment; each method has its strengths and limitations but the most important are: 1. quadrat methods, 2. transect methods, 3. point-intercept methods, 4. line-intercept methods, 5. plotless methods. Quadrat methods have been the most widely used by plant ecologists (Lisa and Susan, 2001). The size of the quadrat varies from one habitat to another and there is considerable discussion as to the most appropriate quadrat size for local type of vegetation. For example, sparse vegetation required small quadrats (10x10 m) and dense vegetation required larger quadrats (50x50 m) in the work reported by Van Gils et al., (1985) in order to cover the area studied and include significant numbers of plant species. Mueller-Dombois and Spatz, (1975) suggested a size of 400 m2 for shrubland, 200 m2 for grassland and 100 m2 for savanna, While Kent & Coker, (1992) suggested 1mX1m-4mX4m for grassland and 10mX10m for scrub and shrubland, but there can be significant differences in the tree-shrub-herbaceous plant cover in these biomes which would affect the size of sample plot. Species-area curve method (Kent & Coker, 1992, see Appendix 6 was applied to estimate the optimum size of quadrat for the vegetation of the study area, the method showed that 10X10m quadrat was adequate for sampling about 50 to 60% of the plant specie recorded in the study area. Transect methods have been used to measure the vegetation across sites where there are rapid changes in landforms and vegetation and are recommended as being less time-consuming than using quadrats. The point-intercept method is objective and rapid but more time- consuming than quadrat or transect methods. It is not recommended for cover less than 5% or greater than 35% (Diersing et al., 1992). 152 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Line-intercept methods are suitable in sparse vegetation and for species with a relatively large basal area. It is considered suitable for dense herbaceous species. (Canfield, 1941). Plotless sampling is appropriate for the inventory of trees and woody shrubs in sites of forest and savannah woodland ecosystems (Cottam and Curtis, 1956; Kent and Coker, 1992) or for measuring species abundances (Eldridge et al., 2003), and is used if the delineation of a sample plot is not possible (Van der, 2004). A combination of transect and quadrat methods were selected in this study because they are consistent with the Braun-Blanquet approach, which can include information on topography such as slope, aspect, soil and meso-climate. 5.2. Methodology Transect approach Transect surveys has been shown to be extremely successful in studies of vegetation, especially across sloping terrain and other vegetation gradient (Kent & Coker, 1992). In the present study sites, the sharply defined topographic units lend themselves to a cross sectional transect approach. The methodology used for the vegetation transects in the present study has been adapted from that used by the Forest Planning and Management Project in Belize (Furley, 1974; Goldsmith et al., 1986). After preliminary reconnaissance, transects were located over the plateau and the main wadi. Each transect was 10 m in width. Transects were chosen to cross the region and identify major vegetation communities in the different landforms (Figure 5.2b). Because of the sparse cover of vegetation, the transects were 5700 to 9600 m long in almost straight lines over the plateau (pla) and the drainage lines (wp), with an additional 2000 to 3500 m cutting the width of the main wadi and adjacent slopes (ms). In places with insufficient land for a given topographic unit, additional horizontal lines orthogonal to the main lines were made or additional sites were added (Figures 5.2 b and c). 153 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Figure 5.2a. General 3D view, showing the different landscape units over the plateau and the main wadi (Source NASA world wind.com). Figure 5.2b. Idealized representative transect illustrating the main landform units in the study sites. The transect line is shown cutting the same topographic units as depicted in Figure 5.2a. 1= mountain slope, 2= wadi bed, 3= wadi bed, 4= cliff, secondary wadi on plateau and adjacent slope. 154 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen A total of 3 transects, 10 m wide, were selected. At each transect, sampling points were established at 100 m intervals. The locations of the transects were selected according to accessibility to the plateau and to reflect the characteristic vegetation zones. This was considered sufficient to represent the region in an area of relatively uniform physiographic relief and vegetation distribution (Fig. 5.2 a & b). The selected transects represent the vegetation of the entire study area (Wadi Hadhramaut), reflecting the different geological and topographical zones. The sample points were approximately 10x10 m. Figure 5.2c. The length of the 3 transects across the study sites. The transects were 5700 to 9600 m long in almost straight lines over the plateau (pla) and the drainage lines (wp). In places with insufficient land for a given topographic unit, additional horizontal lines cutting the main lines were made or additional plots were added. For example, 4 sample plots were added in sites where oil activities are going on (site 2) and 3 sample plots where there is insufficient natural wadi along selected transect (site 1). The site 1 transect mostly passes through the agricultural fields of the main wadi. 155 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 5.2.2. Braun-Blanquet approach The Braun-Blanquet approach (1932) for classification and data collection of vegetation has been used recently in Yemen by Westinga and Thalen, (1980), Kessler, (1987) and Al Khulaidi, (1989, 1992, 1993, 1996 and 1996). Many vegetation studies in the world have been conducted using this approach and have been accepted as being suitable for illustrating vegetation communities. The Braun-Blanquet approach includes information on topography such as slope, aspect, soil and climate and was and is still very useful for classification of plant communities and can be used to estimate abundance and species composition. 5.2.3. Data collection With the aid of a vegetation data sheet the following variables were measured and recorded in the field:  Terrain (land form by field observation according to the categories, slope percentage using Suunto Clinometer; exposure using compass and altitude; latitude and longitude using Geographical Positioning System (GPS).  Soil (texture, depth, type of erosion, surface stoniness).  Vegetation cover, expressed as a percentage of each plant species estimated by using the ITC modified Braun - Blanquet cover/density scale (Van Gils et al., 1985 and Zonneveld, 1986) (Table 5.1). To facilitate the identification of the maximum number of species, four visits were made to these sample sites, July and December (2003) and January and April (2004). 156 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 1- density if cover more than 5% Cover sample Cover % 1 10 (5 -15) 2 20 (15 - 25) 3 30 (25 - 35) 4 40 (30 - 35) 5 50 (45 -55) 6 60 (55 - 65) 7 70 (65 -75) 8 80 (75 - 85) 9 90 (85 - 95) 10 100 (95 - 105) 2- density if cover less than 5%, m = many, a = abundant, p = poor, r = rare Table 5.1. ITC modified Braun - Blanquet Scale cover/density (Van Gils et al., 1985 and Zonneveld, 1986). Cover density was assessed for each plant species encountered. Soils from 37 sample plots representing the different landforms were collected at depth of 10 to 20 cm and analysed by the soil laboratory of Agricultural and Extension Authority, regional station, Taiz (Yemen). The following analyses were under taken and were considered to give a preliminary over view of the soil conditions. Texture, pH, Ec, CaCO3 %, O.C % and CeC. 5.2.4. Methods of soil Analysis Textural analysis, by hydrometer method. pH- determination: by pH meter using 1: 1soil water ratio suspension. Electric conductivity (Ec in ms /cm): by conductivity meter using 1:1 soil-water ratio suspension. Calcium carbonate: by acid- neutralization using hydrochloric acid. Organic carbon: by the Walkley-Black method. Soil moisture content: oven drying at 105°c for 24 hours. Cation Exchange Capacity by barium acetate saturation and calcium replacement. 157 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 5.2.5. Vegetation analysis The first stage was the grouping together of a set of observational units (in this case, quadrats or sample plots) on the basis of their common floristic composition (Kent & Coker, 1992). There are many and varied methods for carrying out vegetation classification based on floristic characteristics, and the results were initially drawn up as a raw-data matrix. Braun-Blanquet The Braun-Blanquet classification method shows relatively clear vegetation associations and sociological group structures, with dominant species and with species that are confined to specific associations, facilitating the labelling of associations with appropriate names. The following method was used to survey the plant cover based on the ZurichMontpellier (Kent & Coker, 1992) or Braun-Blanquet school (Zonneveld, 1989):- 1. plant species were identified and entered in rows and sample plots were entered in columns of an initial matrix (Table 5.2). 2. density/cover symbols for each plant species and environmental data such as altitude, aspect, exposure, stoniness and rock outcrops for each sample plot were added. 3. sample plots with a high similarity of plant species composition were placed side by side. 4. Plant species with similar patterns in quadrat plots were aggregated. Rearrangement of plots and plants species in the matrix was continued until a diagonal matrix of mutually discriminate clusters of both plots (called vegetation types or associations or plant communities as defined by Kent & Coker, (1992) and a group of plant species (taxa) that are more or less similar in behaviour and called sociological species groups (Zonneveld, 1989) were derived (Table 5.3). Floristic vegetation associations are labelled by the dominant species (the second name) and by the species almost exclusively occurring in the association (the first name). 158 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Even though the tabular rearrangement of association groups was based on the basis of floristic composition data, the associations which resulted reflect certain environmental conditions such as moisture, landform, altitude and soil. 1 2 3 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 Sample plots Prosopis farcta mr c Phoenix dactylifera 2 3 c Convolvulus arvensis r m c 9 r Cressa cretica r c r c Withania somnifera Rhazya stricta r c c r r c r c c c r c Datura innoxia r a c Alhagi graecorum r 1 Citrullus colocynthis r Ziziphus spina-christi r Dipterygium glaucum c r Acacia campoptila r r c c r c Tephrosia apollinea c r mc c a c c 1 a a a r r c r r r a r c c c r c c r c r c r r Plant species Tribulus arabicus r c c Salsola imbricata r c c c r c r r c m Aerva javanica c r c Aristida triticoides c c c r r r c c a c r r c r c r? c r c r Cassia italica r r 1 r r r c r 1 Panicum turgidum Merremia somalensis c r r a 1 c Heliotropium ramosissimum c r r r c c c r c c a c Halothamnus bottae c c a c c c r c c a c r r c r c r c Ochradenus baccatus c c r c c c r c r c c c Zygophyllum coccineum c r r r r r Capparis spinosa c Tamarix aphylla c c c c r r c c Acacia ehrenbergiana a r c c r c c r Plicosepalus curviflorus Indigofera spinosa a c Tephrosia nubica c Cleome droserifolia c c m1 a r r r c r r c r a c r r c r c a r r r a ` r r c r c c c r r c a c m c r c c a 1a r c c a c c c c 2 1 r r c c c r Prosopis juliflora c c c a c c Heliotropium rariflorum Ziziphus leucodermis c c c 3 r c c r r c Blepharis edulis c?c r 1 r r a Senna holosericea a 2 a c r r Acacia hamulosa c c a c r c c a c Calotropis procera r c 1 c c c a m Zygophyllum album Cleome scaposa c c c a c Cynodon dactylon Fagonia paulayana r c c 5 r r r c 1 a a c Barleria aff bispinosa Boerhavia elegans a c c c r Kohautia retrorsa Cleome brachycarpa c c r c c c a c 2 r a c c r a c c r Tephrosia dura Cymbopogon schoenanthus c r r Senna incana Dichanthium insculptum c r m c m c a r r r r c Table 5.2. An initial matrix of plant species and samples in order to demonstrate the methodology employed. key: density if cover less than 5%, m = many, a = abundant, c = occasional (= poor), r = rare; density if cover more than 5%, 1= 10(5-15%), 2= 20(15-25%) etc. (see table 5.1). 159 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 670 690 722 720 707 703 680 700 671 700 670 733 729 723 705 686 683 683 673 673 643 679 686 624 645 650 654 681 632 692 668 668 640 644 644 657 651 579 646 655 620 662 626 620 583 668 95 80 55 15 3 10 0 70 30 95 70 80 90 75 70 0 80 80 70 85 60 95 70 70 95 80 50 90 rock outcrops 0 0 0 0 0 0 0 0 0 0 0 0 8 0 0 0 0 0 0 0 0 2 0 0 0 9 0 0 0 erosion r trees <1 shrub 0 herb r r r r r f f f f f f f f f f f f f f f f f f f f f f f f f f f f f f f f f f f f f f f f 90 30 0 0 0 <1 0 <1 0 0 1 1 5 7 1 2 3 1 10 1 3 7 7 20 30 2 1 1 3 0 0 1 1 <1 2 0 <1 <1 <1 2 1 <1 7 1 <1 0 0 0 0 0 0 0 1 0 0 15 30 20 10 0 7 1 0 0 0 1 0 3 10 5 3 0 0 0 0 0 2 0 0 0 0 0 0 0 1 0 0 2 0 2 0 1 15 30 25 0 0 0 0 0 0 0 0 0 5 2 0 0 2 0 1 0 3 0 1 5 0 2 0 30 0 0 1 1 0 5 2 3 3 15 30 1 10 7 30 2 5 1 1 <1 5 2 5 50 3 2 8 3 10 15 5 2 5 10 5 3 2 1 <1 1 0 1 5 <1 4 5 <1 <1 2 3 3 <1 <1 <1 <1 0 <1 1 <1 1 1 3 dwarf shrub 0 0 0 0 0 0 0 0 0 0 6 7 5 0 0 0 surface stoniness 1 1 0 0 0 0 0 0 3 0 5 85 90 70 60 90 80 slope % e e e e e e n e ne e ne ne ne ne n ne ne n e n ne ne n ne se e nw e sw w s s se se se s sw e sw sw sw w nw sw e se Aspect 0 0 0 0 0 0 2 3 1 1 1 10 3 3 2 2 3 1 2 2 1 3 3 1 3 2 3 2 2 2 1 1 5 3 1 1 2 2 1 2 5 2 5 2 1 2 altitude 30 40 29 31 28 33 27 47 26 32 41 25 34 24 23 42 35 22 20 18 16 19 15 36 14 43 10 17 9 11 37 8 44 7 2 12 21 6 13 3 38 45 5 46 39 1 vegetation association sample plot Prosopis farcta mc 2 r r mc Phoenix dactylifera c 9 3 Alhagi graecorum r a Senna incana a c Cressa cretica r Cynodon dactylon a 2 Capparis spinosa Convolvulus arvensis b c c 1 r a c r c c c c r r Datura innoxia r Withania somnifera r Zygophyllum album c 3 a Prosopis juliflora r Citrulus colocynthis r r aff Heliotropium sp. Tiny r Tamarix aphylla r ziziphus spina-christi 1 c r r r r ` Dipteregium glaucum r c r r c r r r r c c c c c c c r r r r r r Acacia campoptila d Tephrosia apollinea c r r c Calotropes procera Fagonia ovalifolia c a r r c r a a a a a m1 c a c c c a e Tephrosia dura f Cymbopogon schoenanthus r c r c c c a c r c m c c r 2 c g Boerhavia elegana c a c c r c c r r c c r c c r c c c c r r c c? c a r c h Barleria aff bispinosa i Acacia ehrenbergiana c c r r c a r a r r c c Rhazia stricta r c r c c Ochradenus baccatus r c c r r c c c c c c g Cassia adenensis c c Cleome scaposa r a k Tribulus arabicus r c l Plicosepalus curviflorus m Salsola imbricata c Indigofera spinosa c r c r r r c r c c r c Aerva javanica c c c c c ma r r r r 1 c c a c c r 1 1 1 a c c c c c r r r r c c r Blepharis edulis c c r m r c r r r c r c r o Heliotropium ramosissimum r Tephrosia nubica c r c r c r c c r r r c r c c a c c r Halothamnus bottae mc a a r c c c c Cleome droserifolia r c c c a c m2 c 1 a 1 c Merremia somalensis r c c a c c Cleome brachycarpa p Ziziphus leucodermis c r c c Kohautia retrorsa Panicum turgidum r m c r c r r c c r Acacia hamulosa Aristida tritioides r r c c c c c c r r r c r c c r c a a a 1 r c c Dichanthium insculptum n Cassia italica c c a c c c 5 a 1 a r c c c c r c c c r c c a Zygophyllum coccineua r c r c r r r r a c c c c r a c r c r r Table 5.3. The final matrix of vegetation based on plant species and sample plots as derived from Table 5.2 to demonistrate the methodology. vegetation associations (1 - 7) in the columns versus sociological species groups (a to p) in the rows. key: density if cover less than 5%, m = many, a = abundant, c = occasional (= poor), r = rare; density if cover more than 5%, 1= 10(5-15%), 2= 20(15-25%) etc. (see Table 5.1). 160 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Twinspan Traditional approaches to plant classification often use ordination (e.g. Twinspan) analysis to investigate plant communities and the environmental gradients influencing species composition. Twinspan produces good results if rare species are removed (Kent & Coker, 1992). For example, study of the vegetation of Egyptian salt marshes (Abd El-Ghani, 2000) using Twinspan performed well by excluding species that covered less than 5%. The Twinspan method proved to be very sensitive to changes in the parameters defined (Pitkänen, 2000) for the classification. So running Twinspan with different optional settings such as different number of species, different level of divisions or group size produced data with different groups. In Twinspan analysis, the first two or three higher levels tend to be valid with relevant ecological meaning, but the lower levels can sometimes seem quite illogical (Dr. Colin Legg, personal contact). Cluster and Ordination analysis These are exploratory data analysis tools for solving classification problems. Their object is to sort plant species into groups, or clusters, so that the degree of association is strong between members of the same cluster and weak between members of different clusters (Kent & Coker, 1992). Modern computer programs can cope with large databases for classification of vegetation, with methods based on mathematics and statistical analysis (Kent & Coker, 1992), and the results of this classification strongly depend on the structure of the data (Lubomir, 2002). Cluster classification was conducted by means of Cluster Minimum Variance analysis and Square Euclidean Dissimilarity Criterion, using the statistical software of the Multi-Variate Statistical Package (MVSP) version. 4.2, and by Bray-Curtis Similarity using BioDiversity Professional Beta (McAleece, 1997; Kovach, 1985). Kent & Coker, (1992) found that the Minimum Variance analysis method is the optimal for similarity analysis and makes most ecological logic. BioDiversity Professional Beta software using Bray-Curtis similarity GroupAverage clustering was applied to compare the main floristic associations. Bray-Curtis was used because it is offers an acceptable coefficient for biological data on community structure (Cheng, 2004). 161 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Detrended Correspondence Analysis (DCA) is an ordination method and was conducted using statistical software of MVSP version 5.2. The DCA analysis is intended to avoid what is called the arch effect, in which the points are arranged in an arched pattern along the first two axes, rather than a linear pattern (Kent & Coker, 1992). Environmental data analysis Canonical Correspondence Analysis (CCA) is a package of relatively recent ordination techniques developed by ter Braak (1986). It is a multivariate direct gradient analysis method that has become very widely used in ecology (Kent & Coker, 1992; ter Braak and Smilauer, 1998). This method is derived from correspondence analysis (CA), but has been modified to allow environmental data to be incorporated into the analysis. These comparisons are achieved by performing statistical package of MVSP version 5.2. The results of CCA can be presented in a diagram containing the environmental variables plotted as arrows originating from the centre of the graph, along with points for the sample plots and plant species. The relationships between the samples and species are as in CA; each sample point lies at the centroid of the points for species that occur in those samples. The arrows representing the environmental variables indicate the direction of maximum change of that variable across the diagram. For example in Figure 5.3 there is clear arrow for altitude to the left of the diagram; this indicates that altitude is increasing along a gradient from the right to the left. 162 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Figure 5.3. Example of CCA ordination diagram. The arrows represent the environmental variables with the direction of maximum change of that variable across the diagram. In this example the arrow for altitude is increasing along a gradient from the right to the left and a long altitude arrow indicates a large change and indicates also that the change is strongly correlated with axis 1 and therefore with the plots 12 and 13. Slope percentage is strongly correlated with ordination axis 1 and increasing along a gradient from the left to the right. The length of the arrow is proportional to the rate of change, so a long altitude arrow indicates a large change and indicates that change in altitude is strongly correlated with the ordination axes and thus with the association variation shown by the diagram. Points with their vertical projections near to or beyond the tip of the arrow will be strongly associated with the environment that is represented by this arrow. Those at the opposite end will be less strongly associated (Kent & Coker, 1992). The sample sites of the study sites and environmental factors have been analysed by Detrended Canonical Correspondence Analysis (DCCA) using statistical software of MVSP version 5.1. Environmental data were available on altitude, average annual rainfall, surface stoniness, rock outcrops, and slope percentage. The detrended canonical correspondence analysis (DCCA) was applied to avoid the arch effect and to assess the plant speciesenvironment correlation more accurately. Also the main floristic associations of the study sites and environmental factors have been analysed by Canonical Correspondence Analysis (CCA) using statistical software of MVSP version 5.1. Environmental data were available on: Altitude, surface stoniness, CaCO3 %, slope %, pH, CeC, rock outcrops, erosion, moisture %, EC % and O.C. %. 5.2.6. Vegetation structure Vegetation structure was classified according to the diagram given in van Gils et al. (1985) see Figure 5.4. In the study sites woody plants greater than 1 m in height were considered as trees; woody plants under 1 m were considered as shrubs, while woody plants under 0.5 m were considered as sub-shrubs or dwarf shrubs. Non-woody plants (grasses and ground cover herbs) were grouped as the herbaceous stratum. To analyse the relationship between different landforms and vegetation cover, data for percentage cover of each species was entered into a Multi-Variate Statistical Package (MVSP) and analysed using Canonical Correspondence Analysis (CCA). Data of vegetation cover percentage per 100 m² was entered into the BioDiversity Professional Beta software to analyse the abundance of each species in different landforms. The data for cultivated lands was not included into analysis and is treated separately, because it was 163 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen obviously dominated by cultivated palm trees and weeds, namely: Phoenix dactylifera, Prosopis farcta, Cynodon dactylon and Alhagi graecorum. Figure 5.4. Three dimensional diagram representing of structure vegetation types (after van Gils et al., 1985). 1woodland 2- open woodland 3- shrubland 4- sparse shrubland 5- grassland 6- dense woodland 7- dense grassland 8sparse grassland 9--dwarf shrubland. 164 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 5.3. Results 5.3.1. Vegetation associations and Socio-Ecological species groups A total of 116 sample plots in different landforms and ecologies throughout the study sites were combined according to the similarity in their species composition by following the classification methods mentioned above, to form groups of floristic vegetation associations. These floristic vegetation associations groups were made of selected sample plots throughout the study sites. A total of 42 sample plots were analaysed from Shibam (site 1), 43 sample plots from Wadi Adim (site 2), and 31 sample plots from Wadi Athahab (site 3) (Table 5.4). Table 5.4b shows the number of bare sample plots for each landform. Table 5.4c shows the GPS coordinates of the sample plots 16 03 924 48 53 697 315 16 04 525 48 54 084 302 16 04 473 48 54 066 301 16 03 833 48 53 770 313 16 04 920 48 54 450 319 16 03 811 48 53 368 311 Table 5.4c. The GPS coordinates in degree, minute, second format of the sample plots. 165 15 58 295 49 36 659 138 16 03 868 48 53 750 314 16 04 792 48 54 349 310 15 53 740 48 35 670 108 15 48 738 48 59 192 209 16 03 827 48 53 814 312 15 50 617 48 35 640 107 15 48 698 48 59 154 208 16 04 839 48 54 340 309 15 53 617 48 35 607 106 15 48 614 48 59 081 210 16 04 868 48 54 359 317 109 15 48 376 48 59 827 216 16 03 474 48 54 349 334 15 54 - 48 35 - 15 54 004 48 35 794 132 15 58 737 48 36 498 138b 15 48 980 48 58 546 202 15 54 032 48 35 821 133 15 48 324 48 59 615 211 15 48 958 48 58 610 201 15 56 - 48 34 - 15 53 924 48 35 769 131 15 48.364 48 59.646 212 15 49 041 48 58 455 204 15 56 162 48 34 874 119 139 15 48 740 48 59 122 207 15 54 417 48 35 479 104 15 54 716 48 35 443 111 15 49 030 48 58 419 203b 15 55 173 48 35 574 134 15 56 319 48 34 685 123 15 55 196 48 35 626 135 15 49 901 48 58 498 203 15 56 121 48 34 841 118 15 54 768 48 35 444 110 15 48 924 48 58 791 243 16 04 790 48 54 227 307 15 54 891 48 35 217 144 15 48 934 48 59 797 244b 15 54 715 48 35 440 113 16 04 909 48 54 460 320 15 54 608 48 35 414 114 15 48 058 48 58 984 247 16 04 571 48 54 108 303 15 54 543 48 35 414 115 15 54 487 48 35 420 116 15 54 48 35 15 48 058 48 58 984 247 coordinate 16 04 628 48 54 137 304 Sample plots 15 48 935 48 58 836 244 coordinate 16 04 682 48 54 150 305 Sample plots 15 58 322 48 36 736 137 coordinate 16 04 682 48 54 150 306 Sample plots 142 Site / Landform w ms f pla wp sp Total sample plots 14 11 6 4 5 2 Site 1 42 17 3 0 1 6 16 Site 2 43 12 9 0 1 2 7 Site 3 31 Table 5.4a. Number of sample plots for each landform in the study sites. Key: w= wadi, ms = slopes facing the main wadis, pla = plateau, sp = slopes on plateau, wp = secondary wadi on plateau, f = fields or fallow lands. w ms f pla wp sp Site / Landform Total sample plots 0 2 0 2 0 0 Site 1 4 0 2 0 3 0 5 Site 2 10 0 1 0 2 0 0 Site 3 3 Table 5.4b. Number of bare sample plots for each landform in the study sites Key: w= wadi, ms = slopes facing the main wadis, pla = plateau, sp = slopes on plateau, wp = secondary wadi on plateau, f = fields or fallow lands. Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen The aim of vegetation classification is to group the plant species together on the basis of their floristic composition into plant associations (communities) generally known as plant phytosociological units (Kent & Coker, 1992). Following the method of Braun-Branquet, Twinspan and ordination methods, the sample plots were arranged to similarities and then combined in 15 vegetation associations and 29 socio-ecological groups (Table 5.6). The matrix (Table 5.7) shows a diagonal clusters where the boundaries of association and sociological species groups can be detected visually through absence or presence of different plant species. Detailed description of socio-ecological groups is presented in Table 5.6. Sample plot distributions obtained by following Twinspan and ordination methods were not formed to be located exactly in homogeneous patterns as indicated by following the school of Braun-Blanquet. They were therefore re-arranged, combined or merged according to their floristic composition and landform similarities. The 6 group of communities that been found following Twinspan, were re-arranged and combined to five group of communities and some of these groups were further divided to a final grouping of 15 vegetation associations (see the methodology 5.3). Applying lower level of divisions for the vegetation classification using Twinspan produces similar results to those obtained by using Braun-Blanquet and matches the ecological variables of the study sites. However, at the higher level of division the results were not logical but subjective. For example, the high level of division produced a high number of groups (14 groups when 5 level was selected and 9 groups when 4 level was selected), but the classification results are acceptable and similar to those obtained by using Braun-Blanquet. By applying 4 levels of division, this classification revealed 6 groups. Further division to sample plots resulted in several unacceptable and un-meaningfully small groups of associations. For example, a few sample sites were misclassified or in unsuitable locations, these samples ended up in different parts of the tree and gave new groupings that did not fit with the classification of Braun-Blanquet analyses. However applying the lower level of division gave acceptable results with a few exceptions, for example sample plots 208, 209 & 219 of main wadi association 6 as defined by Braun-Blanquet, were classified by Twinspan as group 2 of the plateau and mountain slopes. These sample plots had to be rearranged and put in the most suitable site taking into consideration the species composition of the sample site, the habitat and the appearance of annual species that may be found in the sample plot during the field work. 166 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen The DCA and Decorana (Hill, 1979) ordination analysis methods have revealed some associations and groups in which, although dissimilar to the data obtained by Braun-Blanquet school, failed to generate reasonable and acceptable vegetation associations. Nevertheless they were useful to rearrange some sample plots and plant species of Braun-Blanquet data. Some similarity was found, when comparing the classification results of Twinspan with the distribution of sample sites according to DCA and Decorana ordination (figures 5.6 & 5.7). Some sample plots from the same groups were plotted close to each other; however several sample plots plotted in different locations. As for the Braun-Blanquet method, in DCA ordination analysis and Decorana, groups 5 and 6 were plotted close to each other forming only one group rather than 2 groups as in the case of the Twinspan analysis. There are nearly clear segregations between sample plots of some groups, such as groups 2, 3 and 4, but still with some exceptions. Some sample plots belong to particular group with others belonging obviously to different groups. For example, some sample plots of group 1 were put with group 2 or some sample plots of group 2 were put with group 3 (Figures 5.6 and 5.7). . The classification of Twinspan using level of division 4 and minimum group size 5 generated 6 groups. Similar groups in their plant composition were combined to one group, for example 5 and 6 to group 5 (Table 5.4). The classification of Twinspan was taken to 3 levels (Figure 5.5). At the higher level (level 1), the first division separated cultivated sites (6 sample plots) from all others (109 sample plots). At level 2, the cultivated sites were further divided into two divisions comprising palm grove sites (2 sample plots) and a combination of palm grove and fallow lands (4 sample plots); these 2 groups were not further divided. The remaining plots were divided into two divisions comprising the main wadis (38 sample plots) and the rest of the landforms (71 sample plots). At the lower level (level 3) the main wadis were divided further into two groups; flooded sandy wadi beds (5 sample plots) and rocky wadi beds (33 sample plots). The rest of the landforms were divided into further two groups comprising 4 different landform units- mountain slopes facing main wadis together with plateaus, and some slopes of plateaus and secondary wadis (51 sample plots) and secondly, the remainder of the slopes of plateaus (20 sample plots). 167 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Figure 5.5 Twinspan classification based on presence and absence data for 113 species and 115 sample plots. The classification generated 6 group of communities. At level 1: 2 groups, at level 2: 4 groups, then at level 3: 6 groups. Indicator species are used to name these groups. 168 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Table 5.5. The classification matrix result according to Twinspan. In Braun-Blanquet classification method groups 5 and 6 have been combined because of their species composition similarities (See Table 5.7). 169 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Figure 5.6. The result of DCA ordination analysis of numbers that represent the main groups according to a Twinspan classification. As in the Braun-Blanquet method, DCA analysis groups 5 and 6 close to each other forming only one group rather than 2 groups as in case in Twinspan. However there is still a mix between the sample plots from group1 and 2. Figure 5.7. The result of Decorana ordination analysis of the numbers that represent the main groups according to a Twinspan classification. As the Braun-Blanquet method, Decorana analysis groups 5 and 6 close to each other forming only one group rather than 2 groups as in case in Twinspan. 170 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Table 5.6. Description of the main Socio-Ecological groups of plant species. The plant species were clustered to similar groups implementing the Braun-Blanquet method and the Neighbour-Joining method by using Modelling Patterns in Environmental Data (MOPED) software. Most indicative species Common species Poor species Rare species A. Species of the flat to almost flat fertile wadis and as invaders at moist sites (roadsides, fallow fields and near cultivated fields), between 670 and 750 m above sea level. Datura innoxia Phoenix dactylifera Alhagi graecorum Capparis spinosa Cressa cretica Convolvulus arvensis Chloris barbata Withania somnifera Prosopis farcta Setaria verticillata Senna incana Cynodon dactylon Ziziphus spina-christi B. Species of the temporary flooded streambeds, subjected to severe flooding, consisting predominantly of sandy wadi beds between 670-725 m above sea level, mainly in site 1, some of them also occur on fallow lands and cultivated fields. This group confined to association 2. Zygophyllum album Prosopis juliflora Heliotropium rariflorum Citrullus colocynthis Tamarix aphylla none C. As the group B. but extending to rocky, almost flat wadi beds and disturbed sites, cultivated fields and fallow lands. (between 640 and 750 m above sea level). Dipterygium glaucum Calotropis procera none none D. Species mainly found on the relatively narrow, almost flat, wet drainage lines of site 2, which is cultivated during the rainy season and left fallow during the dry season. Some of these species are found also on moisture sites. Some species have spread to the moist sites of plateaus and their slopes. Crotalaria persica Pulicaria undulata Convolvulus glomeratus none Compositae aff. Flaveria Corchorus depressus Chrozophora tinctoria Indigofera spinifolia E. Species found mainly on rocky wadi beds, almost throughout the study sites; also found on relatively moist sites of rocky mountain slopes facing the main wadis, plateaus, and drainage lines that cut these plateaus, as well as on fallow lands. These species show adaptations to extreme floods. Acacia campoptila Fagonia indica Cymbopogon schoenanthus none Tephrosia apollinea F. Species mainly of the relatively moist wadi bed, and alluvial fans or drainage lines that cut the plateaus, almost throughout the study sites. Kohautia retrorsa Acacia hamulosa Barleria aff bispinosa none 171 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen G. Species of north and northeast facing slopes and undulating to slightly sloping very rocky wadi bed, of site 1 (between 983 and 733 m above sea level). Zygophyllum coccineum penetrates the adjacent north and south west facing mountain slopes. (between 667 and 690 m above sea level). Tephrosia dura Zygophyllum coccineum none none H. Species occurring throughout the entire site, in particular on mountain slopes facing the main wadis, the plateaus and their slopes and secondary wadis, not seen on the sandy strong flooded wadi beds and cultivated sites. Dichanthium insculptum Boerhavia elegans none none I. Species are mainly found on relatively rocky, almost flat, moist wadi beds, along river banks and on depressions near cultivated fields (between 680 and 750 m above sea level). Plicosepalus curviflorus is a parasitic on both Acacia ehrenbergiana and Acacia campoptila. Acacia ehrenbergiana none Plicosepalus curviflorus none J. Species confined to rocky almost flat wadi beds, especially of sites 2 and 3. Senna italica spreads to the cultivated fields, moisture sites of southeast moderately steep adjacent rocky slopes and almost flat rocky secondary wadis of plateaus (between 754 and 926 m above sea level). Rhazya stricta Ochradenus baccatus none none Senna italica Senna holosericea K. As in group J, found on rocky almost flat moist wadi beds of sites 2 and 3, but have spread to adjacent relatively moist very rocky mountain slopes of site 3. (between 590 and 677 m above sea level). Tribulus arabicus is also found near cultivated fields. Cleome scaposa Tribulus arabicus none none L. Species occurring mainly on very rocky wide wadis (between 579 and 690 m above sea level), with low distribution extending to adjacent rocky mountain slopes or on slopes of plateau (over 700m above sea level.). Indigofera spinosa Blepharis edulis none none M. One rare species occurring on almost flat to undulating rocky wadi beds, river banks, and rarely on drainage lines or narrow wadis that cut the plateaus. Ziziphus leucodermis none none none N. One species found only on rocky saline wadi beds of site 2. Salsola imbricata none none none O. Species confined mainly to relatively moist almost flat very rocky wadi beds of site 3 (between 570 and 668 m above sea level), some species of this group are found in low quantities on cultivated, and fallow lands. The species of this group are very palatable and have been negatively effected by over grazing in sites 1 and 2. Merremia hadramautica none Panicum turgidum none Tephrosia nubica 172 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen P. Widespread annual species, occurring in almost all ecological zones, in particular rocky disturbed mountain slopes. Less common or absent on the slopes of the plateaus and on sandy flooded streambeds. Aerva javanica Cleome brachycarpa Heliotropium ramosissimum none Q. Species of stony moderately steep, mountain slopes facing the main wadis, plateaus, and narrow rocky secondary wadis and alluvial fans of the plateaus. Rarely found on the main wadis and on the slopes of the plateaus. Cleome droserifolia Iphiona scabra Zygophyllum simplex none R. Species of stony mountain slopes facing main wadis, plateaus, as well as slopes and narrow rocky secondary wadis of the plateaus. Rarely found on the main wadi beds. Stipagrostis hirtigluma Farsetia linearis Aristida triticoides Fagonia paulayana Enneapogon desvauxii Halothamnus bottae S. Rare species prefer moist sites of very rocky alluvial fans and moister slopes as well as eroded surface sites of the plateaus. Cometes abyssinica Forskohlea tenacissima none Lindenbergia indica Reseda sphenocleoides Euphorbia granulata T. One rare species found only on moist sites such as eroded surface on the plateaus and very rocky slopes of the plateaus. Seetzenia lanata none none none U. One rare species occurring as individuals on very steep rocky slopes of narrow secondary wadis and drainage lines of the plateaus (between 858 and 920 m above sea level.) and spreads to the very steep slopes ridges that facing the main wadis (at 730 m above sea level.). Capparis cartilaginea none none none V. Species occurring on moist almost flat to slightly steep slope narrow rocky secondary wadis of the plateaus (between 855 and 950 m above sea level). Cyperus conglomoratus is found also on almost flat rocky plateaus and north east facing, very rocky mountain slopes facing the main wadis of site 2. Chrysopogon accheri Cyperus conglomeratus none none W. Rare species occurring on wet almost flat and slightly steep rocky narrow secondary wadis of the plateaus (between 914 and 977 m above sea level). Leptadenia arborea is also found as invasive plant near cultivated fields of main wadis of site 1. Hermannia paniculata none none Glossonema varians Abutilon bidentatum Pluchea dioscoroides Leptadenia arborea 173 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen X. Very rare species occurring on the moderately steep slope shallow drainage lines that cut the plateau of site 1 (between 977-980 m above sea level.). Crotalaria saltiana none none Moltkiopsis ciliata Y. Species of moderately steep slope secondary wadis and rocky dry steep slopes of the plateaus (between 865 and 995 m above sea level). Commiphora foliacea Acacia mellifera Corallocarpus glomeruliflorus none Acacia oerfota Zygophyllum decumbens Z. Species mainly of dry rocky slopes of the plateaus, Maerua crassifolia penetrates the adjacent narrow secondary wadis and farther towards the main wadis and adjacent cultivated fields. Jatropha spinosa Maerua crassifolia Grewia erythraea Commiphora kua Hochstet schimperi Helichrysum pumilum Cornulaca amblyacantha AA. Rare species found on moister sites of moderately rocky slopes of the plateaus (between 946 and 953 m above sea level.). Anticharis glandulosa none Tamarix arabica Portulaca oleracea Eragrostes sp. Launea sp. BB. Rare species, only found in site 2 on northeast facing relatively wet rocky slopes of plateaus at 950 m asl and adjacent secondary wadi. Arnebia hispidissima Chascanum marrubifolium none none Tarenna graveolens Cadaba heterotricha CC. Very rare species occurring as individuals on rocky wadi beds Iphiona anthemidifolia, Cassia senna, Salvadora persica, Seddera latifolia, Cadaba farinose, Rhynchosia memnonia, Indigofera oblongifolia and Lasiurus scindicus DD. Very rare species found on slightly steep slopes, rocky narrow secondary wadis of plateau and rocky surface of plateau of site 2 and 3. Aerva artemisioides, Trichodesma calathiforme and Euphorbia rubriseminalis 174 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Table 5.7 2 2 2 2 2 2 2 2 2 2 2 2 337 2 319 3 311 1 310 334 139 140 9 138b 123 104 119 307 118 336 303 2 8 320 306 305 304 302 335 301 209 314 315 216 210 208 204 201 202 203 203b 207 247 244b 243 137 244 247 107 108 138 133 109 106 211 132 131 135 111 212 144 2 312 1 2 1 5 2 1 5 1 7 2 313 6 2 2 2 5 110 1 10 1 134 1 4 25 3 317 2 7 6 2 309 8 115 6 8 3 114 7 1 113 141 142 Scientific name Prosopis farcta Alhagi graecorum A Phoenix dactylifera Cressa cretica Ziziphus spina-christi Senra incana Setaria verticillata Convolvulus arvensis Capparis spinosa Cynodon dactylon Chloris barbata Datura innoxia Withania somnifera Zygophyllum album Prosopis juliflora B Tamarix aphylla Citrulus colocynthis Heliotropium rariflorum Dipteregium glaucum C Calotropes procera Pulicaria undulata Crotalaria persica Compositae aff. Flaveria D Corchorus depressus Convolvulus glomertaus Chrozophora tinctoria Indigofera spinifolia Acacia campoptila E Fagonia indica Tephrosia apollinea Cymbopogon schoenanthus Kohautia retrorsa F Acacia hamulosa Barleria aff bispinosa Tephrosia dura Zygophyllum coccineua Boerhavia elegans H Dichanthium insculptum Plicosepalus curviflorus 1 Acacia ehrenbergiana Rhazya stricta J Senna italica Senna holosericea Ochradenus baccatus Cleome scaposa K Tribulus arabicus Indigofera spinosa L Blepharis edulis M Ziziphus leucodermis N Salsola imbricata Merremia somalensis O Tephrosia nubica Panicum turgidum Aerva javanica P Cleome brachycarpa Heliotropium ramosissimum Cleome droserifolia Q Iphionia scabra Zygophyllum simplex Stipagrostis hirtigluma Farsetia linearis R Aristida triticoides Fagonia paulayana Enneapogon desvauxii Halothamnus bottae Cometes abyssinica Forskohlea tenacissima S Lindenbergia indica Reseda sphenocleoides Euphorbia granulata T Seetzenia lanata U Capparis cartilaginea Chrysopogon aucheri V Cyperus conglomoratus Hermannia paniculata Glossonema varians W Abutilon bidentatum Pluchea dioscoroides Leptadenia arborea X Crotalaria saltiana Moltkiopsis ciliata Acacia mellifera Zygophyllum decumbens Y Commiphora foliacea Acacia oerfota Corallocarpus glomeruliforus Jatropha spinosa Maerua crassifolia Hochstetteri schimperi Z Commiphora kua Grewia erythraea Helichrysum pumilum Cornulaca amblyacantha Anticharis glandulosa Tamarix arabica. AA Launea sp. Portulaca oleracea Eragrostes sp. Arnebia hispidissima BB Chascanum marrubifolium Cadaba heterotricha 116 Table 5.7. A matrix of floristic vegetation associations (1-15) in the columns versus sociological species groups (A-CC) in the rows. numbers refer to the individuals of each species. Twinspan classification 6 6 5 5 5 5 4 4 4 4 4 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 2 3 2 2 3 3 3 3 3 3 3 3 3 3 Association 1 2 3 4 5 6 7 1 1 17 2 1 2 1 10 1 2 2 1 10 20 4 10 1 1 4 40 12 17 1 3 1 1 1 1 2 3 1 3 3 5 1 1 1 1 2 1 2 1 3 1 1 1 1 1 1 2 3 2 1 1 1 1 2 2 1 5 1 5 3 1 1 1 1 3 2 3 6 2 1 1 2 3 3 1 5 2 3 1 1 1 11 6 3 1 4 3 2 5 4 2 4 3 5 2 1 2 2 6 1 25 7 10 10 10 30 12 7 1 1 2 1 1 3 2 1 2 1 1 1 1 2 7 2 10 1 11 1 5 10 4 2 8 1 1 2 2 1 2 3 3 3 1 3 1 1 1 3 1 5 3 3 1 1 6 5 3 2 1 2 1 1 1 3 2 2 1 2 1 25 9 1 3 20 1 1 6 3 2 3 1 10 2 4 20 3 4 2 1 2 2 2 1 4 1 5 1 3 3 6 2 1 7 1 1 3 1 4 4 4 2 1 3 3 3 1 2 2 3 3 1 15 15 1 7 2 1 5 4 5 2 2 1 2 2 5 1 2 2 1 1 1 5 8 3 2 3 1 3 3 2 30 2 1 40 50 60 30 3 1 1 4 1 3 5 5 4 7 10 35 1 1 2 1 2 1 4 5 1 5 10 25 8 4 5 14 3 5 25 15 4 2 1 1 10 1 1 1 3 2 3 1 1 2 15 1 1 3 2 2 1 3 8 1 5 2 8 10 1 1 1 1 1 3 3 1 3 1 1 1 2 4 2 1 1 2 1 2 1 1 1 1 1 1 1 1 3 10 8 2 1 1 1 3 3 2 1 1 1 1 3 5 3 8 2 1 1 55 3 2 3 2 5 8 1 1 1 7 2 25 2 50 1 1 1 2 3 1 2 1 2 1 1 1 1 1 8 2 3 15 1 1 3 55 1 1 2 1 1 3 1 1 1 3 10 27 12 1 1 5 3 1 1 1 1 3 10 1 1 1 7 10 1 1 1 2 5 10 8 2 7 5 5 1 3 1 10 6 3 3 3 1 1 5 8 4 1 6 11 1 3 1 4 2 4 2 3 8 2 15 7 11 8 1 2 2 2 1 10 6 3 3 5 7 3 5 2 5 1 1 1 4 2 1 1 175 2 2 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 1 1 1 1 1 1 1 1 1 1 1 15 35 235 223 227 326 15 331 330 324 233 240 238 14 223b 1 229 245 231 234 218 239 248 130a 130b 237c 220 333 332 236 230 246 246/b 128 129 136b 129b 323 136 328 327 241 237 125 322 124 143 102 128b 122 101 217 215 121 Scientific name Prosopis farcta Alhagi graecorum Phoenix dactylifera Cressa cretica ziziphus spina-christi Senra incana Setaria verticillata Convolvulus arvensis Capparis spinosa Cynodon dactylon Chloris barbata Datura innoxia Withania somnifera Zygophyllum album Prosopis juliflora Tamarix aphylla Citrulus colocynthis Heliotropium rariflorum Dipteregium glaucum Calotropes procera Pulicaria undulata Crotalaria persica aff Flaveria Corchorus depressus Convolvulus glomertaus Chrozophora tinctoria Indigofera spinifolia Acacia campoptila Fagonia indica Tephrosia apollinea Cymbopogon schoenanthus Kohautia retrorsa Acacia hamulosa Barleria aff bispinosa Tephrosia dura Zygophyllum coccineua Boerhavia elegans Dichanthium insculptum Plicosepalus curviflorus Acacia ehrenbergiana Rhazia stricta Senna italica Senna holosericea Ochradenus baccatus Cleome scaposa Tribulus arabicus Indigofera spinosa Blepharis edulis Ziziphus leucodermis Salsola imbricata Merremia somalensis Tephrosia nubica Panicum turgidum Aerva javanica Cleome brachycarpa Heliotropium ramosissimum Cleome droserifolia Iphionia scabra Zygophyllum simplex Stipagrostis hirtigluma Farsetia linearis Aristida triticoides Fagonia paulayana Enneapogon desvauxii Halothamnus bottae Cometes abyssinica Forskohlea tenacissima Lindenbergia indica Reseda sphenocleoides Euphorbia granulata Seetzenia lanata Capparis cartilaginea Chrysopogon aucheri Cyperus conglomoratus Hermannia paniculata Glossonema varians Abutilon bidentatum Pluchea dioscoroides Leptadenia arborea Crotalaria saltiana Moltkiopsis ciliata Acacia mellifera Zygophyllum decumbens Commiphora foliacea Acacia oerfota Corallocarpus glomeruliforus Jatropha spinosa Maerua crassifolia Hochstetteri schimperi Commiphora kua Grewia erythraea Helichrysum pumilum Cornulaca amblyacantha Anticharis glandulosa Tamarix arabica. Launea sp. Portulaca oleracea Eragrostes sp. Arnebia hispidissima Chascanum marrubifolium Cadaba heterotricha 213 Table 5.7. A matrix of floristic vegetation associations (1-15) in the columns versus sociological species groups (A-CC) in the rows. numbers refer to the individuals of each species. Twinspan classification 2 2 1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 1 1 1 1 1 1 1 Association 10 11 12 13 224 Table 5.7. 1 1 1 1 1 3 1 3 4 4 8 4 5 1 1 1 9 1 3 3 3 2 1 4 20 4 3 2 4 1 10 10 8 4 10 15 50 1 33 1 30 1 2 1 8 2 3 5 7 6 1 7 1 8 10 1 7 10 2 1 2 1 3 9 1 5 1 4 1 1 1 2 3 2 10 6 1 1 3 2 4 10 3 5 7 30 1 20 3 6 10 4 1 1 1 1 3 2 1 1 1 2 8 2 23 1 10 1 1 1 1 1 1 1 8 1 2 1 1 12 1 1 1 1 3 2 5 10 40 30 6 1 3 8 11 1 1 1 1 2 3 1 1 3 1 1 2 1 1 1 1 6 3 4 1 3 4 3 1 6 6 1 1 1 3 10 1 2 5 5 4 1 10 3 3 2 8 7 2 2 1 1 2 3 10 8 22 6 2 1 3 3 2 2 1 1 50 3 10 2 2 1 20 30 1 3 2 10 5 2 4 5 5 9 6 2 15 7 1 3 1 1 1 7 2 4 5 5 1 3 7 30 3 2 5 3 2 3 2 1 3 1 4 1 1 1 10 2 2 4 1 2 5 1 2 3 2 2 2 1 2 2 1 10 2 2 1 1 1 1 1 1 2 1 3 2 2 1 8 1 1 2 1 1 2 1 1 15 1 2 2 1 1 3 1 4 1 4 2 1 8 4 4 2 1 1 1 3 1 4 6 1 2 4 2 2 2 2 1 3 2 6 1 4 1 2 1 3 1 1 1 1 3 2 5 3 2 2 2 4 3 1 2 3 10 3 8 1 2 2 1 1 1 5 2 2 2 1 4 2 1 3 1 176 1 5 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 5.3.2. Soil laboratory analysis results Soil samples from the surface layer (0 - 10 cm) and from 10 to 20 cm were taken from 48 plots in the three study sites representing the different landforms (Table 11). The analysis was undertaken by the soil and water department laboratory in the Agriculture Research and Extension Authority, Taiz, Yemen. Generally there is limited or no soil cover on the rocky slopes and plateau areas. Most of the landscapes are covered by accumulations of hard course textured limestone, dolomite and chert fragments left after millions of years of weathering (Komex International Ltd,. 1999). The depth of the soil in the rocky slopes and plateaus is very shallow and does not exceed 50cm, and is usually much less while the depth in the farmable area is more than 50cm. Interpretation The results show that the soils have varied textures being predominantly sandy loam with high proportion of sand and stones. The average pH values range between 7.4 and 8.7 with average of 7.98. There is no significant variation from one horizon to another (Table 5.8). Silty soils are observed in wadis at Wadi Adim (site 2) and Wadi Athahab (site 3). Sand filled cracks with deep gullies are common in site 1 (plate 5.1). The percentage of CaCO3 is generally high and ranges from 36 to 37.4. In cultivated fields and fallow lands the percentage is low and ranges between 29.7 and 37.4 (see the soil cross section Figure 5.8a). 177 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Table 5.8. Laboratory analysis results of the soil collected from 37 sites in the three study areas. w= main wadi, ms= slope facing main wadi, fal= fallow land, f= field, pla= plateau, ps= slope on plateau. SL= sandy loam, L= loam, LS= loamy sand, S= sand, SiL= silty loam. - = no data. Sample plot Sand % 2-0.05 Silt% 0.002 (mm) Clay% (mm) 0-10 52 42 6 10-20c 54 40 41 Land form Depth (cm) ms ms pH (1:1) Ec (1:1) ms/cm CaCO3 % O.C % C.eC cmol/kg Sl 8.2 0.34 37.2 0.46 6 Sl 8.3 0.45 37.2 6 SL 8.25 0.395 37.2 < Textural class Moisture % Stoniness % 4.7 3.4 56 - 4.7 12.9 44 0.46 4.7 8.15 50 Site 1 1 1 average 3 ms 0-10 66 24 10 Sl 8 2.7 34.7 0.35 9.2 7.3 61 3 ms 10-20c 64 26 10 Sl 8 2.5 36 - 6.4 4.4 38 average 65 25 10 SL 8 2.6 35.35 0.35 7.8 5.85 49.5 18 ms 0-10 50 40 10 SL 8.2 1 35.7 - 5.2 1.6 58 21 ms 0-10 50 38 12 L 8.2 0.54 36.5 - 5.7 1.1 50 6 w 0-10 60 34 6 Sl 8.2 0.45 37.2 0.56 4.7 2.8 50 8 w 0-10 48 44 8 Sl 8.2 18.2 37 0.23 5 1.8 44 8 w 10-20c 56 34 10 Sl 8.3 10.4 37 - 5.7 1.1 45 average 52 39 9 SL 8.25 14.3 37 0.23 5.35 1.45 44.5 10 w 0-10 86 4 10 L 8.6 0.36 37 - 2.1 1.2 0.8 10 w 10-20c 88 6 6 LS 8.7 0.23 37..2 0.15 1.4 1.4 0.4 average 87 5 8 L, LS 8.65 0.295 37 0.15 1.75 1.3 0.6 12 f 0-10 44 46 10 L 8.3 3.1 37.2 0.21 5.2 2.1 32 12 f 10-20c 36 42 16 L 8.2 3.4 35.2 - 5.2 1.3 26 average 40 44 13 L 8.25 3.25 36.2 0.21 5.2 1.7 29 14 f 0-10 36 50 14 L 8 5.3 35.7 0.32 7.6 1.3 41 14 f 10-20c 50 40 10 SL 8 7.2 36 - 7.6 2.3 47 average 43 45 12 L, SL 8 6.25 35.85 0.32 7.6 1.8 44 24 pla 0-10 48 32 20 L 7.9 2.9 37 0.34 8 4.1 56 27 pla 0-10 72 24 4 SL 8.1 0.37 34..2 0.52 11 2.3 33 178 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Table5.8 cont. Site 2 1 w 0-10 64 30 6 SL 7.8 0.23 36.7 0.41 3.8 6.5 41 1 w 10-20c 80 16 4 LS 8 0.22 37.4 - 4.7 5.2 93.1 average 72 23 5 SL, LS 7.9 0.225 37.05 0.41 4.25 5.85 67.05 7 w 0-10 60 36 4 SL 7.9 0.39 37 - 5.4 1.2 6 7 w 10-20c 58 36 6 SL 7.9 0.36 36..5 - 5.7 0.8 3 average 59 36 5 SL 7.9 0.375 37 - 5.55 1 4.5 8 w 0-10 30 56 14 SiL 7.9 0.02 37.2 0.52 9.2 0.9 48 9 w 0-10 56 38 6 SL 8.1 0.28 37.2 - 5.6 0.9 53 11 fal 0-10 60 32 8 SL 8.1 0.45 29.7 0.49 6.1 5.5 48 5 fal 0-10 58 36 6 SL 7.6 0.54 37..2 - 5.2 0.9 59 13 ms 0-10 46 48 6 SL 7.9 0.33 37..2 - 6.6 0.5 67 22 pla 0-10 60 32 8 SL 8 0.67 37 0.43 5.6 3.1 43 32 pla 0-10 68 26 6 SL 7.8 0.89 37 - 8.5 0.5 36 18 ps 0-10 52 38 10 SL 8.2 0.2 36..5 - 6.4 2.4 55 24 ps 0-10 56 38 6 SL 8 0.52 37.2 - 5.2 14.2 58 31 ps 0-10 44 48 8 SL 8.1 0.32 37 0.53 3.3 2.4 59 33 ps 0-10 86 8 6 LS 8.1 0.27 37 0.17 3.3 0.7 49 37 ps 0-10 60 36 4 SL 7.4 6.94 37.4 0.2 6.1 3 37.8 37c wp 0-10 59 32 9 SL 8 0.34 37 0.45 4.6 3.7 60 1 w 0-10 50 40 10 SL 7.7 3.1 37 - 10.8 2.1 61 1 w 10-20c 82 14 4 LS 8 0.17 37.4 - 3.3 2.3 57.9 average 66 27 7 SL, LS 7.85 1.635 37.2 - 7.05 2.2 59.45 3 w 0-10 44 44 12 L 7.8 0.63 37.4 - - 4.9 75 3 w 10-20c 88 8 4 S 7.5 0.39 37.4 0.3 3.8 2.2 54.7 average 66 26 8 L, S 7.65 0.51 37.4 0.3 3.8 3.55 64.85 Site 3 5 w 0-10 24 60 10 SiL 7.2 0.65 37.3 0.58 - 3.4 39 5 w 10-20c 36 52 12 SiL 7.5 0.53 37.3 0.64 9.6 14 29 average 30 56 11 SiL 7.35 0.59 37.3 0.61 9.6 8.7 34 16 fal 0-10 60 34 6 SL 7.4 1.3 37.4 0.49 - 6.7 59 10 ms 0-10 46 46 8 SL 7.6 0.52 37.3 - 8.5 7.9 33 11 ms 0-10 48 42 10 L 7.8 0.6 37.4 - 8.2 7.3 18 13 ms 0-10 54 36 10 SL 7.9 0.58 37.4 - 6.1 8.3 57 18 ms 0-10 42 48 10 L 7.7 1.1 37.4 0.46 - 0.9 62 19 ms 0-10 40 40 20 L 8.1 0.59 37.2 - - 6.1 56 24 ps 0-10 56 32 12 SL 8.2 0.41 37.4 - - 1.1 53 28 ps 0-10 48 36 16 L 8.3 0.36 37.4 - - 2.4 43 30 ps 0-10 58 30 12 SL 8.3 0.33 37.4 0.64 - 1.2 57 179 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Figure 5.8a. Soil cross section of the study area. The soil depths of the plateau surface and the rocky slopes are very shallow and range between 10-20cm, while the soil depth of the main wadi beds is more deeper and ranges from 30 to 50cm. The soil of the plateau surface is shallow, mixed with gravel and stones materials. The soil type of the plateau can be classified as: Entisols (Almashreki, 2005). The wadi beds are covered by alluvium deposits of rounded lime-stone gravels, stones and boulders. Most of the land of the main wadis in sites 2 and 3 are covered by bare soil dissected by landslides and deep gullies. Generally the soil of the agricultural fields is very deep alluvium soils, loam and silt loam texture. Most of this soil can be classified as Entisols (see Table 1.7). 180 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen There is little variation in most chemical soil properties over the different land forms, especially in CaCa3%, O.C%, and pH. Despite these similarities, there are important variations in soil depth, soil texture, moisture content and Ec (Figure 5.8b). This is probably complemented by different plant species composition and associations along the cross section. The variation of vegetation associations in the main wadis and adjacent mountain slopes is probably due to significant variation in soil texture in these landforms. High vegetation cover, in particular of palm trees, is seen on the foothills of the mountain slopes. These sites are characterised by deeper soil, high level of water availability and weathering of colluvial deposits from the plateau above which lead to increased plant variety and greater biomass (Furley, 1974). The sparse cover and bare surfaces of the plateau reflect the chert gravels covered by thin soil. This has resulted in one vegetation association with almost uniform vegetation structure. 80 70 60 pla 50 sp 40 wp 30 ms 20 tributeris 10 w fields 0 depth (cm) pH Ec CaCO3 % O.C % C.e.C % moisture stoniness % % Figure 5.8b. The variation in soil properties over the different landforms. There is little variations in most chemical soil properties over the different land forms, especially in CaCa3%, O.C%, and pH. 181 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 5.4. The vegetation of the different land forms 5.4.1. The vegetation of the main wadis (w) The total vegetation cover of the main wadis ranges between 2% and 50%, with an average of 17.8%, while the cover in the palm grove localities reaches almost 95%. The average tree cover is 6.7%, and this occurs mainly as scattered individuals dominated by Acacia campoptila; the shrub cover is almost 3%; the dwarf shrub cover is nearly 4%, and herbaceous cover is 6.4 %). (Table 5.9) The total number of plant species collected is 61 (about 52% of the total number of species recorded in all the study sites), with 44 (63%) in Shibam, (site 1) 35 (50%) in W. Adem (site 2) and 33 (47%) in W. Athahab (site 3). Of these, 17 plant species are distributed throughout the study sites, with Acacia campoptila, Fagonia indica and Tephrosia apollinea occurring at a high frequency. The vegetation is generally concentrated at the edges of wadi beds. About 10 plant species are only recorded near or in cultivated or fallow lands of site 1. The soil of the wadis is characterized by:  High calcium carbonate levels in all horizons, which range between 36.5% and 37.4%, with an average of 37.2% in the upper sections and 37% at lower depths.   High average pH values of 8 with extreme values of 7.8 and 8.7. Low organic matter contents ranging between 0.15% and 0.64%, with an average of 0.42%.   Heterogeneous soil texture, predominantly sandy loam and silty loam. Low to high salinity, generally from 1 to 14.3 ms/cm. 182 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 1 30 0 11 42 5 20 0 16 41 7 10 5 5 27 average 1 15 0 4 20 138 0 0 0 50 50 137 0 0 0 8 8 133 0 0 0 2 2 132 <1 1 0 4 5 131 134 <1 0 0 5 5 109 144 0 0 0 0 0 118 113 0 0 0 30 30 107 115 0 0 0 30 30 106 114 90 0 1 4 95 110 142 <1 0 0 7 7 135 141 trees 30 shrub 15 dwarf shrub0 herb 1 total 46 Plot no 111 116 Site 1 1 0 2 3 6 2 7 0 6 15 3 1 0 11 15 1 0 2 5 8 10 0 0 3 13 8.9 5 0.5 10 23 1 0 2 <1 3 1 2 1 2 6 2 3 0 5 10 1 0 1 2 4 3 1 0 1 5 7 0 3 <1 10 7 3 0 1 11 1 0 0 4 5 3 0 30 8 41 0 0 0 3 8 0 0 2 1 3 <1 0 3 1 4 <1 0 15 1 16 2 1 30 11 44 average <1 0 3 1 4 302 305 2 0 5 1 8 303 <1 0 0 3 3 304 <1 2 2 3 7 314 1 0 30 1 32 306 315 301 7 2 7 3 19 302 <1 0 10 1 11 301 1 0 1 2 4 320 Plot no trees shrub dwarf shru herb total 307 Site 3 average 209 208 216 210 244 247 244b 243 207 202 1 0 1 3 5 0 0 0 <1 8 5.5 1.7 3.1 3.2 14 overall average 30 5 5 2 42 201 203b Plot no trees 20 shrub 10 dwarf shrub 1 herb 4 total 35 204 203 Site 2 2 0 9 2 13 Table 5.9. Average vegetation cover for each sample plots in the main wadis of the three study sites. The average vegetation cover ranges between 13 % in site 3 and 23% in site 1. Trees and herbaceous plants are the dominant strata. The cover in some favoured locations is high, for example the palm grove localities range from 46 to 95%, the abundance of Zygophyllum album on previously cultivated areas has raised the vegetation cover to 50%. 5.4.2. The vegetation of the mountain slopes facing the main wadis (ms) Compared to other ecological zones, the vegetation of the mountain slopes is generally sparse. The total vegetation cover ranges between 1% and 8%, The high percentage (12%) of herbaceous plant in plot 123 is due to the micro climate differences, for example, slight changes in landform characteristics, such as the bottom of the slopes, drainage lines or small gullies, where moisture and soil are accumulated and as a result become favourable for annual grasses such as Dichanthium insculptum and Stipagrostis hirtigluma in the case of site 1 (see association 8 and plate 5.7). The total vegetation cover ranges between 1% and 12%, with an average of 2%. Generally trees, shrubs and dwarf shrubs are not found in this zone. Rarely, individual trees are found in some locations. The average tree cover is less than 1 %, and the herbaceous cover is 3%. (Table 5.10). The vegetation is mainly concentrated on furrows, rills or gullies that cut the slopes. There are 32 plant species collected in total in this zone. (27% of the total plant species collected over the study sites), with 27(84%) in Shibam, (site 1) 9 (28%) in Wadi Adem (site 2) and 16 (50%) in Wadi Athahab (site 3). Of these, 5 plant species occur with high frequency and 183 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen are distributed throughout the study sites, these species are: Dichanthium insculptum, Boerhavia elegans, Cleome brachycarpa, Stipagrostis hirtigluma and Farsetia linearis. The soil of this zone are characterized by:  High calcium carbonate levels in all horizons, which range between 34.7% and 37.4%, with an average of 36.7% in the upper depth and 36% in the lower depth. .     High average pH values of 8 with extreme values of 7.8 and 8.3. Low organic matter, less than 0.35%, with an average of 0.42%. Heterogeneous soil texture, loamy and sandy loam, with stoniness more than 80%. Very low to moderate salinity, generally between 0.34 and 2.7ms/cm. Table 5.10. Average vegetation cover for each sample plots in the mountain slopes facing the main wadis of the three study sites. The average vegetation cover ranges between 2% in sites 1 & 3 and 3 % in site 2. Herbaceous plants are dominant. The high percentage of herbaceous plant in plot 123 is due to the micro climate such as slight change in landform characteristics, such as the bottom of the slopes, drainage lines or small gullies where moisture and soil are accumulated and as a result become favourable for annual grasses such as Dichanthium insculptum and Stipagrostis hirtigluma in the case of site 1 (see association 8 and plate5.7). 5.4.3. The vegetation of the plateaus (pla) The vegetation of the plateaus is generally very low. The total vegetation cover ranges between 2% and 9%, with an average of 4%. Generally trees, shrubs and dwarf shrubs are not found in this zone (Table 5.11); the vegetation, which is mainly herbaceous, is concentrated in furrows, depressions, rills or gullies that cut the plateaus. The total number of plant species found in this zone was 19 (16% of plant species collected over the study sites), with 17(89.5%) in Shibam (site 1), 2 (10.5%) in Wadi Adem (site 2) and 3 (15.8%) in Wadi Athahab (site 3). Stipagrostis hirtigluma and Farsetia linearis are widespread in this zone. The high percentage of total and vegetation cover of site 1 is due to the removal of the surface layer in plot 124 by the local people to make protection barriers to 184 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen prevent the road from water erosion. The exposed soil below the surface layer was able to support greater plant life. The vegetation in this zone is very poor in species, with no significant or specific species to build a clear definition of associations. Almost all species in this zone occur elsewhere in the study sites, except the endemic species Aerva artemisioides, which is confined to the plateaus and adjacent narrow wadis or drainage lines. The soils of this zone are characterized by:  High calcium carbonate levels in all horizons, which range between 34.2% and 37.4%, with an average of 37 % in the upper horizon and 35.6 % in the lower horizon.    High average pH values of 7.9 with extreme values of 7.5 and 8.1. Low organic matter levels of less than 0.3%, with an average of 0.4 %. Heterogeneous soil textures, mainly loamy, sandy and sandy loam, with a high percentage of stoniness. Very low to moderate salinity, generally between 0.17 and 2.9ms/cm. overall average average 329 325 322 3 average 222 232 241/6 average 2 127 125 128b 143/6 124 plot no 1 228 site 126  trees 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 shrub 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 dwarf shrub0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 herb 40 9 2 1 0 0 9 6 0 0 0 2 3 0 0 1 4 total 40 9 2 1 0 0 9 6 0 0 0 2 3 0 0 1 4 Table 5.11. Average vegetation cover for each sample plots in the plateaus of the three study sites. The average vegetation cover, which is only herbaceous, ranges between 1% in site 3 and 9 % in site 1. The high percentage of herbaceous plants in plot 124 of site 1 is due to the removal of the surface layer by the local people to make protection barriers to prevent the road from water erosion. The exposed soil below the surface layer was able to support greater plant life (see Association 11). 185 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 5.4.4. The vegetation in narrow secondary wadis of the plateaus (wp) The total vegetation cover in narrow secondary wadis and drainage lines in the plateaus ranges between 4% and 37%, with an average of about 19%. In moist sites the cover increases to 86%. The average tree cover is 2.5%, the shrub cover is less than 1%; the dwarf shrub cover is nearly 1.1%, and herbaceous cover is 16`% (Table 5.12). The total number of plant species found in this zone was 47 (40% of the total species collected from the study sites), with 19(40%) in Shibam (site 1), 37 (79%) in Wadi Adem (site 2) and 20 (43%) in Wadi Athahab (site 3). Dichanthium insculptum, Boerhavia elegans, and Stipagrostis hirtigluma are widespread plant species in this zone. The following species are only found in this zone: Hermannia paniculata, Crotalaria saltiana, Glossonema varians, Euphorbia rubriseminalis. Abutilon bidentatum, Pluchea dioscoroides and Leptadenia arborea. The last two species are seen in other ecological zones outside the sample sites, such as moisture adjacent slopes or main wadis. Leptadenia 1 1 <1 1.6 <1 <1 1 0 0 0 0 0.2 0 0 0 dwarf shrub 0 0 0 0 0 0 2 10 0 1 1 0 2.3 0 0 0 herb 1 7 11 20 9 20 70 9 35 2 15 25 12 4 8 8 7 13 20 12 23 86 9 37 4 15 29 12 4 8 7 total cover 12 average 333 0 1 239/3 5 0 236 1 0 230 4 0 237c <1 0 246 2 0 129 <1 2 129b 5 0 128 5 shrub 130b trees Plot no 130a 332 3 average 2 246/b 1 average site overall averag arborea is considered by local people as an aggressive weed. Table 5.12. Average vegetation cover for each sampling plots of the secondary wadis in the plateaus of the three study sites. The average vegetation cover ranges between 12% in site 1 and 29 % in site 2. The high percentage of herbaceous plants in site 2 is due to the influence of oil exploration work near to the sample plot, that increases the moisture in the soil at the bottom of the secondary wadi and adjacent slope (see association 12 chapter 5 and plate 5.14). 5.4.5. The vegetation of slopes in plateaus (sp). The total vegetation cover of slopes in plateaus ranges between 1% and 57%, with an average of 14%, the cover on wet slopes that are irrigated by contaminated water from oil exploration work (plate 5.13), reaches almost 86%. The average tree cover is 0.9%, the shrub cover is 0.7%, the dwarf shrub cover is less than 0.2 %, and the herbaceous cover is 12%). (Table 5.13). The total number of plant species of this region is 47 (40% of the total species collected from the three study sites), with 7(15%) in Shibam, (site 1) 42 (89%) in Wadi Adem (site 2) and 16 (34%) in Wadi Athahab (site 3). Of these, the following three species are distributed with relatively high frequency throughout the study sites: Dichanthium insculptum, Boerhavia elegans and Stipagrostis hirtigluma (Table 4.4, chapter 4). 186 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen The soil of this zone is characterized by:  High calcium carbonate in all horizons, which ranges between 36.5% and 37.4%, with average of 37.2% in the upper horizon and 37.4 in the lower horizon. .   High average pH values of 8 with extreme values of 7.5 and 8.3. Low organic matter contents ranging between 0.17% and 0.64%, with an average of 0.41%.  Heterogeneous soil texture, predominantly sandy loam and loamy sand.  Very low salinity, generally less than 1ms/cm. Table 5.13. Average vegetation cover for each sample plots of the slopes in the plateaus of the three study sites. The average vegetation cover is dominated by herbaceous plant and ranges between 3% in site 3 and 17 % in site 2. The high percentage of vegetation in site 2 is due to the influence of oil exploration work near to the sample plot, that increases the moisture in the soil at the bottom of the secondary wadi and adjacent slope (see association 12 and plate 5.14) as well as due to the accumulation of sediments, moisture and soil through the drainage lines and rills that cut the slopes (see plate 5.16). 187 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 5.5. Main vegetation associations 5.5.1. Alhagi graecorum - Phoenix dactylifera cultivated association This association densely planted and forms woodland with sparse grassland and is restricted to cultivated sites and on sites of date gardens (with the palm Phoenix dactylifera) and fallow fields in Shibam (site 1). The vegetation cover ranges between 2% and 46%. In date garden sites, the cover can reach 95%. The average tree cover is 24%, the average dwarf shrub cover is 14%, and the average herbaceous cover is about 3 %. The topography is composed of flat wadis, The soil is loamy and sandy loam with high salinity (between 3.1 and 7.2 ms/cm, the figure is less than 1 in fallow lands), low CaCO3 (between 35 and 37.4%), pH ranges between 7.4 and 8.3, O.C ranges between 0.21 and 0.49 and C.E.C ranges between 5.2 and 7.6 meq/100g . The soil is alluvial-colluvial in origin; the raw mineral soils are Entisols (SOGREAH, 1981) and probably equivalent to Inceptisoil. The altitude ranges from 670 to 722 m above sea level. Weeds are the dominant and most common species of this association. This association is reflected the next association in that it used to be covered by dense palm trees before being degraded by continuous severe floods. The dominant species is the date palm Phoenix dactylifera. The following species are only found in this association: Alhagi graecorum, Cressa cretica, Chloris sp., Capparis spinosa, Datura innoxia, and Withania somnifera. 188 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Plate 5.1. Association 5.5.1 A date garden. Widespread along the Wadi Hadhramaut. Plate 5.2. Association 5.5.1. In the foreground fallow land with Alhagi graecorum and Senna incana with cultivated fields, scattered date palms in the background 189 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 5.5.2- Citrullus colocynthis - Zygophyllum album association This association forms grassland, sparse grassland and sparse shrubland in Wadi Khashamr, Shibam (site 1). The vegetation cover is very poor ranging between 2% and 8%, but in some favourable locations reaches 50%. The average tree cover is less than 1 %, the average shrub cover is less than 1 %, and the average herbaceous cover is 16 %. The topography is undulating to almost flat, subjected to severe flooding, consisting predominantly of sandy wadi beds, with 3-5% cover of small rocks. The soil is loam to loamy sand with a high sand proportion (86-88%), CeC is 1.7, CaCO3 is 37.1, EC is 0.3 and pH is 8.7. The altitude ranges from 670 m to 700 m above sea level. This association is probably a degraded site of wadi beds that used to be covered by dense palm trees with association of Ziziphus spina-christi. The highly invasive tree Prosopis juliflora is found in many locations. The dominant species is Zygophyllum album. The following species are found only in this association: Citrullus colocynthis and Calotropis procera. Plate 5.3. Association 5.5.2. in the foreground scattered Zygophyllum album and Prosopis farcta; in the background palm trees , with Acacia ehrenbergiana. Prosopis juliflora was once planted widely for sand dune stabilization and easily occupies this type of habitat. As for many desert plants (Batanouny, 2000), the exposed roots of these plants adapt to tolerate floods for long time. 190 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 5.5.3. Convolvulus glomertaus - Merremia hadramautica - Crotalaria persica association This association forms a dwarf shrubland, found on fallow lands located at site 2 (between 675 and 676 above sea level). The terrain consists of almost flat, narrow, relatively rocky wadi beds. The soil is sandy loam, with low CaCO3 (30%), high CeC (6.1%) and high moisture content (5.5%) with pH 8.1 and O.C 0.5%. The vegetation cover is high and ranges between 65 to 75%, dominated by a dwarf shrub cover. The average dwarf shrub cover is 60%, and average herbaceous cover is 9 %. The association is dominated by Crotalaria persica and Merremia hadramautica. The following species are abundant in this association and always occur: Pulicaria undulata, Crotalaria persica, Compositae aff. Flaveria, Corchorus depressus, Chrozophora tinctoria, Heliotropium ramosissimum and Convolvulus glomeratus. Plate 5.4. Association 5.5.3. Fallow land on relatively stony wadi bed, dominated by dwarf shrubland of Merremia hadramautica and Crotalaria persica. 191 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 5.5.4 -Tephrosia dura - Tephrosia apollinea- Fagonia indica association This is a mixture of open woodland with sparse shrubland, and grassland, occurring on Wadi Khumur and Shie’b Sultan of site 1 (between 638-733 m above sea level). This association is found on slightly undulating or almost flat rocky wad beds and depressions, often with extensive rocky outcrops. The soil is sandy loam, with stone proportions ranging between 60 to 95% and low moisture content. CeC is 5, O.C. is 0.4%, CaCO3 is 37.1%, EC is 7.4% and pH is 8.3. The vegetation cover ranges between 6 to 42%. The average tree cover is 3%, the average shrub cover is 12%, the average dwarf shrub is 1% and the average herbaceous cover is about 8 %. The leading species are Fagonia indica, and Tephrosia apollinea, in association with Acacia campoptila. The former species, as well as Cymbopogon schoenanthus and Tephrosia dura, are always present. Trees are dominated by Acacia campoptila and are found over large areas with a sparse and scattered distribution but, in some locations, they form open woodland. In areas where water is significant, such as in depressions and near cultivated fields, the vegetation is relatively rich with species such as: Acacia ehrenbergiana, Acacia hamulosa, Barleria aff. bispinosa, Boerhavia elegans, Dichanthium insculptum, Kohautia retrorsa, and Zygophyllum simplex. Zygophyllum album and Senra incana are only seen in high saline sites; on the other hand Kohautia retrorsa and Acacia hamulosa were absent in these sites. Plate 5.5. Association 5.5.4. Scattered species of Tephrosia apollinea. with Acacia campoptila (a tree to the right). 192 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 5.5.5- Acacia ehrenbergiana - Rhazya stricta - Acacia campoptila association This association is a mixture of open woodland, sparse shrubland, and sparse grassland, found mainly on wadi beds and adjacent banks of wadi beds in Wadi Adem and its tributaries Wadi Bayut and Wadi Hikma of site 2 and in a small location at Wadi Nia’m, Shibam north east of site 1. The topography is undulating to almost flat stony over wide wadi beds and banks; subject to strong flooding, the soil is yellowish with pale brown loamy to sandy loam texture, with high CeC (5.5%), low EC (3.8%), low moisture (1%), high clay content (10%) and with pH of 7.9%; no rock outcrops. The altitude ranges between 624 and 686 m above sea level. The vegetation cover varies between 3 and 41%, (the average tree cover is 8 %, the average shrub cover is 2.2, the average dwarf shrub is 1.4% and the average herbaceous cover is about 3 %). Acacia campoptila is the dominant species. Cassia senna, Crotalaria persica and Salvadora persica are rare and only found in this association. Acacia campoptila and A. ehrenbergiana form open woodland in varies locations. In places characterized by non-stony compacted soil with pale brown sandy loam texture, only a few species (about 6) are found, (plate 5.6). Rhazya stricta, Fagonia indica and Ochradenus baccatus occur at a high frequency. Plate 5.6. Association 5.4.5 Foreground, almost bare compacted hard soil with very poor cover of Fagonia indica and Cassia italica. In the background open woodland of Acacia campoptila with very few Ochradenus baccatus. (Wadi Bayut, site 2). 193 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 5.5.6- Salsola imbricata - Blepharis edulis association This association is a mixture of sparse shrubland, dwarf shrubland and sparse grassland, on wadi beds and adjacent banks in site 2. The terrain is almost flat to undulating rocky wadi beds and banks, with sandy loam and silty soil, almost no rock outcrops. The CeC is 6.2, O.C. is 0.5%, CaCO3 is 37%, EC is 3.8%, pH is 7.9 and moisture content is 2.8%. The altitude ranges between 632 and 692m. The vegetation cover is very poor, ranging between 2% and 41%. The average tree cover is 4%, the average shrub cover is 1%, the average dwarf shrub cover is 4% and the average herbaceous cover is 3%. The vegetation cover in some more favourable sites, such as wadi banks (plate 6.7) or at the foot of the limestone slopes where alluvium has been deposited by a small tributaries flowing from the plateau down to the main wadi (alluvial fan), reaches 41%. (plate 5.8). Blepharis edulis is the dominant species especially on the rocky, almost flat open sites (Hamada) with little sediment cover, which are exposed to high winds and sun. This habitat also has very limited plant growth (plate 5.10). Here Acacia campoptila becomes rare or absent. Salsola imbricata is only found in this association. The following occur with high frequency in the association: Acacia campoptila, Salsola imbricata, Boerhavia elegans and Fagonia indica. Plate 5.7. Association 5.5.6 Foreground woodland of Acacia campoptila with dwarf shrub Salsola imbricata, Indigofera spinosa and Fagonia indica. 194 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Plate 5.8. Association 5.5.6. Foreground Acacia campoptila. Background herbaceous cover of Indigofera spinosa, Blepharis edulis and Stipagrostis hirtigluma. (Shie’b Al Salaq, sample plot 216, site 2). Fallen rocks and finer particles accumulate at the base of the mountain slope and make it appropriate for plants. Plate 5.9. Association 5.5.6. Foreground almost flat, wide, very shallow rocky, open wadi bed (Hamada) dominated by Blepharis edulis. Background single tree of Acacia campoptila. 195 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 5.5.7- Tephrosia nubica - Acacia campoptila - Merremia hadramautica association This is a widespread rich association that contains about 37% of total species of the wadi associations and is considered as the second richest association. This association is confined to wadi beds and adjacent banks of Wadi Athahab and its tributaries (site 3). However, the vegetation cover is poor (mainly sparse grassland with dwarf shrubland) and ranges between 3% and 43%. The average tree cover is 2%, the average shrub cover is less than 1 %, the average dwarf shrub cover is 9% and the average herbaceous cover is 3%. The terrain varies from undulating to almost flat rocky wadi beds and banks with soil of sandy loam, silty and loamy texture. The CeC is 10, O.C% is 7, CaCO3 is 37.2%, EC is 1.4%, pH is 7.6% and moisture content is 3.1%. The altitude ranges between 579 and 668 m above sea level. Merremia hadramautica, and Acacia campoptila are the dominant species. In addition Tephrosia nubica, Indigofera spinosa and Fagonia indica occur at a high frequency. In areas that are subjected to severe flooding (see plate 5.10), the number of plant species does not exceed 7. Merremia hadramautica only occurs in this association. Site subjected to severe flooding Plate 5.10. Association 5.5.7. Foreground Tephrosia nubica and Merremia hadramautica. Background trees of Acacia campoptila. Very few species are found in sites that are subjected to severe flooding, the number of plant species here does not exceed 7, while dense cover is found at the edge of the wadi bed. 196 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 5.5.8. Forsskaolea tenacissima - Boerhavia elegans association The structure of this association is grassland and includes sparse grassland found on mountain slopes facing the main wadis of site 1. Stipagrostis hirtigluma is the dominant species. In addition, the following species also occur at a high frequency: Aerva javanica, Farsetia linearis and Fagonia indica. The terrain is eroded, moderate to steep very rocky shallow slopes, with rill and gully erosion, and soil of loam and sandy loam texture. The CeC is 5.2, O.C% is 0.2, CaCO3 is 35.7%, EC is 1%, pH is 8.2% and moisture content is 1.6%.The altitude ranges between 690 and 827 m above sea level. The vegetation cover is very poor and ranges between 1% and 5%, the average tree cover is less than 1% and the average herbaceous cover is about 4%. The vegetation cover reaches 12% in more favourable sites where moisture and soil accumulate, such as drainage lines or depressions. (Plate 5.7). The smaller particles that accumulate in these sites with a cover of gravels, sands and silts are able to support quite good plant growth. Plate 5.11 . Association 5.5.8. Very eroded slope with vegetation concentrated on the drainage lines and depressions. Foreground plants are Reseda sphenocleoides; and background plants are Stipagrostis hirtigluma, Aerva javanica and Dichanthium insculptum. This slope presents a good barrier against the loss of a lot of particles such as gravels, sands and silts and in consequence supports quite good plant growth. 5.5.9. Fagonia paulayana - Boerhavia elegans - Cleome scaposa association 197 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen An association predominantly made up of sparse grassland and is confined to mountain slopes facing the main wadis of site 3. Fagonia paulayana and Cleome scaposa are the dominant species. In addition the following species also occur with a high frequency: Boerhavia elegans, and Cleome brachycarpa. The vegetation cover is very poor and mainly consists of herbaceous plants, and ranges between less than 1 % and 3%, with an average herbaceous cover of 3%. The terrain consists of very eroded, moderate to steep slopes, rocky with high soil loss, and soil of pale brown colour, loam and sandy loam texture. The CeC is 7.6, O.C% is 0.2, CaCO3 is 37.3%, EC is 0.57%, pH is 7.9% and moisture content is 7.4%. Rill and gully erosion are very obvious. The altitude ranges between 590 m and 685 m above sea level. Plate 5.12. Association 5.5.9. (sample plot 310). Very eroded rocky slope with high soil loss. The vegetation cover is less than 1%. 198 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Plate 5.13. Association 5.5.9. Eroded rocky steep slope with vegetation cover less than 2% in stone pavement. 5.5.10. Indigofera spinosa - Stipagrostis hirtigluma association This association is predominantly sparse grassland found on mountain slopes facing the main wadis of sites 1 and 2. Very similar to the previous association but different in having fewer species. The common species in the previous association such as Fagonia paulayana, Enneapogon desvauxii, Cleome scaposa and Tribulus arabicus are absent here. Indigofera spinosa, which is common here, was not found in the previous association. The terrain is much eroded, moderate to very steep, rocky, shallow dry slopes; with soil of sandy loam and loam texture; rill and gully erosion are seen in some locations. The CeC is 5.7, CaCO3 is 37%, O.C% is 0.2, EC is 0.42%, pH is 8.1% and moisture content is 3.2%. The altitude ranges between 692 and 763 m above sea level. Stipagrostis hirtigluma is the dominant species. The vegetation cover is very poor and mainly consists of herbaceous plants, ranging between 2% and 8%; the average herbaceous cover is about 3 %. The number of species in this habitat is very low and does not exceed 10. 199 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Plate 5.14. Association 5.5.10. Very eroded and unstable steep slopes with vegetation cover less than 2%. The vegetation is mainly found in runnels and rills. 5.5.11. Seetzenia lanata - Dichanthium insculptum association This association is widespread and forms a grassland and sparse grassland, and is found on plateaus and their slopes over different parts of the study sites. The terrain varies from almost flat to moderate steep slopes, and is very rocky, with shallow loamy soils. The CeC is 8, CaCO3 is 37%, O.C% is 0.34, EC is 1.6%, pH is 8.1% and moisture content is 3.3%. The altitude ranges between 910 and 995 m above sea level. Stipagrostis hirtigluma is the dominant species. Other important species of this association are Farsetia linearis and Dichanthium insculptum. Arnebia hispidissima, Portulaca oleracea, Seetzenia lanata and Launaea sp. are only seen on the slopes of this habitat. The vegetation cover is generally very poor and ranges between 1% and 9%. The average tree cover is less than 1%, the average shrub cover is less than 1%, and the average herbaceous cover is 13.2 %. However, the cover rises to 40% in places where surface stony sites have been cleared and soils been exposed (plate 5.15); the average herbaceous cover is about 14%. On wet slopes that are contaminated water from oil exploration work (plate 5.16); the cover can reach almost 91%. 200 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Plate 5.15. Association 5.5.11. Foreground, eroded surface with Stipagrostis hirtigluma and Farsetia linearis (sample plot 124), background undisturbed surface with very poor or no vegetation cover. With the removal of the surface stone layer, the exposed soil is able to support a high vegetation cover and reaches 40%. This new composition of plant species can be considered as a primary succession habitat. Plate 5.16. Association 5.5.11 (sample site 237) dominated by Dichanthium insculptum and Farsetia linearis. Compared with the surrounding dry gentle slopes, very dense vegetation cover can be seen in the foreground and the adjacent wadi. This is due to seepage of contaminated water from oil exploration work. 201 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 5.5.12. Chrysopogon aucheri- Stipagrostis hirtigluma association This is a widespread association, predominantly grassland, with a few sites covered by sparse shrubland, sparse grassland and dense grassland. This association is the richest with a number of plant species and is found on narrow secondary wadis and drainage lines over the plateaus, across all the study sites and on alluvial fans on the plateau of site 1. The vegetation cover ranges between 4 and 37%; in favourable moisture soil sites the cover reaches 86%. The average tree cover is 2.7%, the average shrub cover is less than 1%, the average dwarf shrub cover is 1.1% and the average herbaceous cover is 17.6 %. The characteristic landform consists of slightly steep slopes to almost flat, narrow secondary wadi drainage lines that cut the plateaus and alluvial fans. The predominant pavement rock covers 80 to 99 % of the site. The CeC is 6.1, O.C. is 0.2%, CaCO3 is 37%, EC is 6.9%, pH is 7.4% and moisture content is 3%. The altitude ranges from 858 to 979m. The dominant species are Dichanthium insculptum and Stipagrostis hirtigluma. The following species are found only in the wettest, highly saline (about 7 ms/cm) sites of this association: Abutilon bidentatum, Pluchea dioscoroides, Leptadenia arborea, Pulicaria undulata and Tamarix arabica. Other species that occur only in this association are: Moltkiopsis ciliate, Chrysopogon aucheri, Hermannia paniculata, Glossonema varians, and Abutilon bidentatum. Plate 5.17. Association 5.5.12 (sample site 130). Alluvial fan with good cover on the plateau of site 1. Foreground Cymbopogon schoenanthus and Acacia oerfota. Background Commiphora foliacea. 202 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Plate 5.18 Association 5.5.12. Narrow drainage line on the plateau, with predominantly pavement rock covering 80 to 99 % of the site. Foreground Dichanthium insculptum, Stipagrostis hirtigluma and Barleria aff bispinosa. Background Acacia hamulosa. Plate 5.19. Association 5.5.12. Very wet site with dense vegetation cover reaching 86%. Foreground, Chrysopogon aucheri, Dichanthium insculptum, Farsetia linearis, Aerva javanica and others. Background trees of Acacia oerfota, Acacia mellifera and Commiphora foliacea. Very dense vegetation cover can be seen due to ?seepage/watering of/by contaminated water from oil exploration work. 203 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 5.5.13. Maerua crassifolia - Jatropha spinosa - Stipagrostis hirtigluma association The structure of this association is predominantly sparse shrubland with a few locations covered by sparse grassland, woodland, dense grassland and grassland found in site 2. The terrain is moderate to steep rocky slopes (30-50%) and alluvial fans with rill erosion, and soils of sandy loam and loamy sand texture. The CeC is 4.3, O.C.% is 0.35%, CaCO3 is 36.8%, EC is 0.26%, pH is 8.1% and moisture content is 1.8%. The altitude ranges between 899 and 977 m. The vegetation of this association is quite rich predominantly herbaceous and the cover ranges between 3% and 61%. The average tree cover is 3%, the average shrub cover is 2%, the average dwarf shrub cover is 1% and the average herbaceous cover is 14%. Jatropha spinosa is the dominant species. In addition to the dominant species, the following species also occur in high frequency: Dichanthium insculptum, Stipagrostis hirtigluma, Zygophyllum decumbens and Commiphora foliacea. The following are only species that almost occur in this association: Maerua crassifolia, Grewia erythraea, Helichrysum pumilum, Seddera latifolia, Chascanum marrubifolium, Tarenna graveolens, Cadaba heterotricha, Arnebia hispidissima, Trichodesma calathiforme and Euphorbia rubriseminalis. The latter species was only seen once on a rocky, narrow wadi bed and adjacent foot slopes (alluvial fan). Plate 5.20. Association 5.5.13 (sample site 245). Dry, shallow, very eroded slope with vegetation cover less than 5%. Background sparse cover of Jatropha spinosa with Dichanthium insculptum and Farsetia linearis 204 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Plate 5.21. Association 5.5.13 (sample site 238). Runnels on slope, influencing vegetation cover. Foreground herbs dominated by Indigofera spinosa. Background trees mainly Acacia mellifera, Maerua crassifolia, and Jatropha spinosa. The richness of this slope is due to stone ring shape foundation on the plateau which diverts more water during the rainy season towards this slope (top picture) and also the v shape type of slope which directs the flow of the water towards the dense cover. (The arrows show the direction of water flow). 205 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 5.5.14- Cornulaca amblyacantha - Stipagrostis hirtigluma association This association is confined to site 3 and is predominantly sparse grassland with a few locations covered by sparse shrubland. The terrain is a moderately steep very rocky slope, with rill and gully erosion, and soils of a sandy loam texture. The CeC is 5%, O.C%. is 0.64%, CaCO3 is 37.4%, EC is 0.37%, pH is 8% and moisture content is 1%. The altitude ranges between 932m and 955 m above sea level. Stipagrostis hirtigluma is the dominant species. In addition to the dominant species, the species Jatropha spinosa also occur at a high frequency. The species Cornulaca amblyacantha only occurs in this association. The vegetation cover is very poor with only 13 plant species, and ranges between less than 1 % and 4%. The average tree cover is < 1, the average shrub cover is less than 1 %, the average dwarf shrub is 1% and the average herbaceous cover is 1%. Plate 5.22. Association 5.5.14 (sample site 326). Very eroded rocky slopes with vegetation cover <4%. Background Jatropha spinosa with Stipagrostis hirtigluma. 206 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 5.5.15. Zygophyllum decumbens - Dichanthium insculptum association The structure of this association is confined to site 2 and is predominantly sparse grassland with few locations covered by grassland. This association is very similar to association 10 but different in having fewer species. Some species that are found in association 10, such as Boerhavia elegans, Cleome brachycarpa and Indigofera spinosa, were absent here. On the other hand, the common species of this association (Zygophyllum decumbens) was not found in association 10. The terrain is a moderately steep slope (30-40%) very rocky, with rill erosion, high soil loss, and soil with a sandy loam texture. The CeC is 5.2%, O.C%. is 0.5 %, CaCO3 is 37%, EC is 0.5%, pH is 8% and moisture content is 14%. The altitude ranges between 922 and 949m. Dichanthium insculptum is the dominant species, the species Zygophyllum decumbens always occurs in this association. This association represents the degraded sites of association 13 that occur in the same study site and in the same type of land formation. The vegetation cover is very poor, with only 8 plant species that are mainly herbaceous, ranging between 2% and 12%, , with an average of 4% cover. Plate 5.23. Association 5.5.15 (sample site 224), very rocky slope with sparse vegetation cover dominated by Dichanthium insculptum and Zygophyllum decumbens. 207 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 5.6. Summary of the vegetation associations. No The vegetation association Sites Altitude 1 Alhagi graecorum - Phoenix dactylifera Flat fallow and cultivated lands,site1 670-722 2 Citrullus colocynthis - Zygophyllum album Flooded wadi bed, site 1 670 - 700 3 Convolvulus glomeratus – Merremia hadramautica – Crotalaria persica Old fallow land, site 2 675-677 4 Tephrosia dura - Tephrosia apollinea- Fagonia indica Acacia ehrenbergiana – Rhazia stricta – Acacia campoptila wadi beds and depressions, site 1 683-733 almost flat to undulating rocky, wadi beds and banks, site 1 & 2 624-686 a.s.l. (m) 5 6 Salsola imbricate - Blepharis edulis Wadi beds and adjacent banks, site 2. 632-692 7 Tephrosia nubica - Acacia campoptila - Merremia somalensis Wadi beds and adjacent banks, site 3 579-668 8 Forskohlea tenacissima - Boerhavia elegans moderate to steep rocky slopes and alluvial fans facing the main wadis of site 1. 690-827 9 Fagonia paulayana - Boerhavia elegans – Cleome scaposa very eroded moderate to steep rocky slope facing the 590-685 main wadis of sites 3. 10 Indigofera spinosa - Stipagrostis hirtigluma eroded moderate to steep rocky slope, facing the main wadis of sites 1 & 2 692- 763 11 Seetzenia lanata - Dichanthium insculptum plateaus and their slopes, all sites 910-995 12 Chrysopogon aucheri - Stipagrostis narrow wadis and drainage lines of the 858-979 hirtigluma plateaus in all study sites Maerua crassifolia - Jatropha spinosa – slope of plateau, site 2 899-977 13 Stipagrostis hirtigluma 14 Cornulaca amblyacantha - Stipagrostis hirtigluma Slope of plateau, site 3 932-955 15 Zygophyllum decumbens - Dichanthium insculptum Slope of plateau, site 2 922-949 Figure 5.9. Summary diagram of floristic vegetation associations across the main topographical units. 208 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Most of the vegetation associations are dominated by herbaceous plants Vegetation associations 1, 2, 4, 8 and 13 are confined to site 1; vegetation associations 3, 6, 13 and 15 are confined to site 2, while vegetation associations 7, 9 and 14 are confined to site 3. The vegetation associations of the main wadis are dominated by Acacia campoptila, Fagonia indica, Tephrosia apollinea and Cymbopogon schoenanthus. Trees on the main wadis are dominated by Acacia campoptila and are found over large areas with a sparse and scattered distribution, but in some locations, they form open woodland. Associations 8, 9 and 10 are confined to mountain slopes facing the main wadis and are dominated by Dichanthium insculptum, Boerhavia elegans, Cleome brachycarpa, Stipagrostis hirtigluma and Farsetia linearis. Associations 13, 14 and 15 are confined to the slopes of the plateau and dominated by Dichanthium insculptum Zygophyllum decumbens, Jatropha spinosa and Stipagrostis hirtigluma. Association 7 is considered to be the second richest association. This association is confined to wadi beds and adjacent banks of Wadi Athahab and its tributaries (site 3). Association 12 is the richest association and is found on narrow secondary wadis and drainage lines over the plateaus across all the study sites and on alluvial fans on the plateau of site 1. Associations 1 and 3 have the highest vegetation cover percentages. Association 1 is dominated by trees and shrubs namely Phoenix dactylifera, Prosopis farcta and Alhagi graecorum, while association 3 is dominated by dwarf shrub and herbaceous cover namely Merremia hadramautica, Crotalaria persica, and Heliotropium ramosissimum. The following Figures (5.10 to 5.12) illustrate the different land uses and topographical features of each study site. 209 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Figure 5.10. Main vegetation associations of site 1 210 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Figure 5.11. Main vegetation associations of site 2 211 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Figure 5.12. Main vegetation associations of site 3. 212 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 5.7. Similarity between the main vegetation associations In the previous sections, the main floristic associations were described in terms of vegetation composition, structure, locality and dominant plant species. The next stage has been the analysis of similarities between the various vegetation associations using Bray-Curtis similarity method with Group-Average clustering and presence and absence data. This method appears to give a reasonable grouping of clusters, which shows two main groups (Figure 5.13):Group A. Comprises the associations of main wadis and Group B. Comprises the associations of slopes facing the main wadis and the plateau zones. Associations 6 and 7 as identified in Figure 5.13 and Table 5.14 are considered as the most similar habitats within the study sites, with similarities of 66%, followed by associations 8 and 9 with a similarity of 62% (Table 5.14). The former two associations are found on the main wadi beds of sites 1 and 3 and the latter two are found on the mountain slopes that face the main wadis of sites 1 and 3. 213 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Figure 5.13. Similarity dendrogram produced by the Bray-Curtis similarity method using Group-Average clustering, showing the similarity between the main floristic associations in the 3 study sites. Associations 6 and 7 are the most similar habitats in the study sites with a similarity of almost 66%, then associations 8 and 9 with a similarity of almost 62% (Table 5.8). Associations 1 to 7 (group A) are confined to the main wadis, while the remaining associations (group B) are distributed over the slopes facing the main wadis (8-10) and the plateau zones (11-15). Table 5.14. Similarity matrix between the main floristic associations. This matrix contains numbers which represent the percentage of similarity between the main floristic associations . Associations 6 and 7 are the most similar habitats, with a similarity of nearly 66%. There are also high similarities between associations 8 and 9 (about 62%) and between associations 11 and 12 (about 57%). 214 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen The similarity between the vegetation associations produced using Bray-Curtis was based on the number of plant species occurring in these vegetation associations. For example, out of 41 plant species in associations 6 and 7, 20 were occurring in both associations and out of 29 plant species in associations 8 and 9, 15 are found in both associations. The similarity percentage between 2 associations increases with the increase of similar plant species in both associations. The cluster method also separated the associations of main wadis from the rest of the associations, this also based on the previous reasons. 5.8. The distribution of vegetation associations throughout the study sites As described earlier, 15 floristic vegetation associations have been distinguished according to their floristic composition using Braun-Blanquet and Twinspan classification methods (section 5.3). Table 5.15 shows the distribution of these vegetation associations throughout the study sites. Association 1 2 + + 3 4 5 6 + + + + 7 8 9 10 11 12 13 + + + + + + + + + 14 15 site 1 2 3 + + + + + + Table 5.15. The distribution of the floristic associations throughout the study sites. Associations 11 and 12 are distributed throughout the study sites. As we notice from Table 5.9, most floristic associations are confined to specific sites except associations 5 and 10, which occur in sites 1 and 2 and associations 11 and 12, which are distributed throughout all the study sites. Association 11 is predominantly grassland, found mainly on the plateaus and surrounding slopes and dominated by Stipagrostis hirtigluma and Dichanthium insculptum. This association is characteristic of the vegetation that covers all the plateaus and surrounding degraded slopes. Association 13, which confined to site 2 is predominantly sparse shrubland and dominated by Jatropha spinosa with high frequency of Dichanthium insculptum and Stipagrostis hirtigluma. This association can be a climax vegetation before being disturbed and degraded to secondary succession 215 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen (Kent & Coker, 1992); this succession represented by association 15, which is characterised by very few species, not more than 8, of mainly herbaceous plants. In order to use vegetation as an environmental indicator, the plant species were subdivided into sociological groups, or groups of plant species that behave similarly (Zonneveld, 1989) or have the most similar distribution across the sample plots (Martin et al., 2003). Figure 5.14 shows the vegetation associations (in the columns) versus sociological species groups (in the rows), illustrating the distribution patterns of the sociological species groups across the different landforms. The bar diagram shows that each floristic vegetation association is characterised by a specific combination of plantsociological groups, except for association 1 which has only sociological species group A. This association is confined to the cultivated fields, and has specific plant species which are absent in other environments. Some sociological species groups are confined to a particular association, for example groups A, B, G, I, N, T and V are found only in associations 1, 2, 4, 5, 6, 11, and 12 respectively. Most of the sociological species groups are present in more than one association, while sociological species group H occurs in almost all of the floristic vegetation associations. Some sociological species are represented at very low frequencies, as for example, groups S, W, X, AA, BB and CC; these groups are characterised mainly by very rare plant species. For more detail see Table 5.6 which describes the sociological species groups. 216 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Figure 5.14. A bar diagram of floristic vegetation associations 1-15 (in the columns) versus sociological species groups A-CC (in the rows). It can be seen that there are very clear separations between 3 groups that grow on main wadis (1 to 7), on mountain slopes facing main wadis (8 to 10) and on the plateau zones (11 to 15). Some sociological species groups, such as groups A and B, have very limited distributions; the first is confined to cultivated fields and the latter is confined to flooded sandy wadi beds of site 1; whereas other groups such as group H have variable distribution across a wide range of landforms, with high frequency on the plateaus and group E with high frequency in the wadi beds. Also distributions limited to particular landforms can be seen, for example plant species of groups U, V, W and X are mainly confined to secondary wadis of plateaus, while group Z is confined to slopes of plateaus of site 2 and 3. Group Z, Groups AA, BB and CC are rarely represented species. The distribution of vegetation associations (in the columns) varies from one site to another. For example, associations 1, 2 and 3, have limited distribution while associations 11 and 12 are distributed throughout the 3 study sites. Main floristic vegetation associations are: 1. Alhagi graecorum - Phoenix dactylifera, 2. Citrullus colocynthis - Zygophyllum album, 3. Convolvulus glomertaus - Merremia hadramautica - Crotalaria persica, 4. Tephrosia dura - Tephrosia apollinea- Fagonia indica, 5. Acacia ehrenbergiana - Rhazya stricta - Acacia campoptila, 6.Salsola imbricata - Blepharis edulis, 7. Tephrosia nubica - Acacia campoptila - Merremia hadramautica, 8. Forskohlea tenacissima - Boerhavia elegans, 9. Fagonia paulayana - Boerhavia elegans - Cleome scaposa, 10. Indigofera spinosa - Stipagrostis hirtigluma, 11. Seetzenia lanata - Dichanthium insculptum, 12. Chrysopogon aucheri- Stipagrostis hirtigluma, 13. Maerua crassifolia - Jatropha spinosa Stipagrostis hirtigluma, 14. Cornulaca amblyacantha - Stipagrostis hirtigluma, 15. Zygophyllum decumbens - Dichanthium insculptum 217 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen The plant species were subdivided into groups of species that behave similarly or are distributed similarly across the sample plots (i.e. sociological species groups). These groups were distributed characteristically across the different habitats. The distributions of the sociological groups depend on the environmental nature of the landforms. Some associations have limited distribution to specific habitats such as agricultural fields, wadi beds, rocky slopes or plateau sites and others have a wide range of distribution. For example, associations 1, 2 and 3, have limited distribution while associations 11 and 12 are distributed throughout the 3 study sites. Also plant species of groups U, V, W and X are mainly confined to secondary wadis of plateaus, while group Z is confined to slopes of plateaus of site 2 and 3. Group Z, Groups AA, BB and CC are rarely represented species. The distribution of vegetation associations (in the columns) varies from one site to another. Each vegetation association has its own sociological species group, some have one group and others have more than one group 5.9. Environmental data analysis 5.9.1. The sampling plots The sample plots of the study sites and environmental factors (Figure 5.15) have been analysed by Canonical Correspondence Analysis (CCA) using the software of the Multi-Variate Statistical Package (MVSP) version 3.1. Environmental data were available on altitude, surface stoniness, erosion and slope percentage. The points represent sample plots and arrows represent the environmental variables. The arrows point in the direction of maximum change of the environmental variable across the diagram. 218 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Figure 5.15. Canonical Correspondence Analysis (CCA) ordination biplot of sampling points on main landforms and environmental variables. Arrows represent the environmental data and point in the direction of maximum change of the environmental variable across the diagram. Altitude and stoniness are the most important environmental variables. All environmental variables in particular altitude and erosion are strongly correlated with ordination axis 1. There is a very clear segregation between two groups, the sample plots of the main wadis and cultivated areas (right) and the sample plots of the other landforms (left). The first group is strongly correlated with erosion from flooding, and the second group is strongly correlated with the remaining environmental variables. Altitude and surface stoniness are the most important environmental variables and increase along a gradient from the right to the left. The landform groups are clearly distinguished by the environmental variables. There is a strong relationship between the distribution of sample plots on the main wadis (right) and erosion from flooding, and between the distribution of sample plots of the mountain slopes and plateau zones (left) and the remaining variables. The sample plots of cultivated and fallow lands are negatively correlated with stoniness, while the sample plots of the main wadi beds are located at the low altitude and negatively correlated with slope percentage. 219 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 5.9.2. Plant species The CCA axis (Figure 5.16) shows that slope %, altitude and stoniness increase along a gradient from the right to left and are strongly correlated with ordination axis 1. The CCA axis represents the following orders: 1. Species more commonly associated with cultivated fields (top right, e.g. Prosopis farcta, Alhagi graecorum, Phoenix dactylifera, Cressa cretica, Ziziphus spina-christi and Senra incana), that are negatively correlated with increasing altitude, slope % and stoniness. 2. Species commonly associated with severely flooded wadi beds (middle right), e.g. Zygophyllum album, Prosopis juliflora, Tamarix aphylla, Citrullus colocynthis, Heliotropium rariflorum, Dipterygium glaucum, Calotropis procera, Merremia hadramautica, Tephrosia nubica and Panicum turgidum; these species consist of sociological groups B, C, N and O. 3. Species commonly associated with rocky wadi beds (bottom right), e.g. Cymbopogon schoenanthus, Ziziphus leucodermis, Zygophyllum coccineum, Tephrosia dura and Barleria aff bispinosa;, these species are positively correlated with stoniness and flood erosion and negatively correlated with increasing altitude and slope%, they were not found in the wadi beds where the floods are severe. 4. The species Glossonema varians, Euphorbia rubriseminalis, Cornulaca amblyacantha, Arnebia hispidissina, Hermannia paniculata, Moltkiopsis ciliate, Forskahlea tenacissima, Hermania paniculata, Moltkiopis ciliate, Tamarix arabica and Chascanum marrubifolium top left are strongly correlated with high altitude and moderately steep slope, secondary wadi and alluvial fans of the plateau in site 2, most of these species are found in association 12, chapter 5). These species are characterised as being rare. Also species of rocky slopes of the plateau e.g. Enneapogon desvauxii, Fagonia paulayana, Commiphora foliacea and Corallocarpus glomeruliflorus are positively correlated with moderately steep slope and high altitude. 5. The species bottom left e.g. Maerua crassifolia, Zygophyllum decumbens, Jatropha spinosa, Grewia erythraea, Helichrysum pumilum, Cornulaca amblyacantha and Hochstetteri schimperi, Commiphora kua are strongly 220 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen correlated with high altitude rocky steep slopes of association 13, sociological species groups Z and AA. Figure 5.16. Canonical Correspondence Analysis (CCA) ordination biplot of plant species and landform environmental variables. Arrows represent the environmental data. Altitude and stoniness are the most important environmental variables. All environmental variables except stoniness are strongly correlated with ordination axis 1. Species of cultivated fields (top right) are negatively correlated with increasing altitude, the species of rocky wadi beds (bottom right) are strongly correlated with flood erosion. The species top left are correlated with moderately steep slope, while the species bottom left are strongly correlated with slope % and stoniness. Erosion and in particular flooding plays an important part in changing and destroying the vegetation structure in many areas and even in causing the disappearance of some individual species that cannot stand the extreme floods characteristic of the main wadis. To get a broader understanding of the effect of erosion, the data were analysed taking only erosion as a variable, but the result was not satisfactory to interpret, therefore taking to consideration of almost uniformly altitude over the see level of wadis, altitude variable was added to erosion variable (Figure 5.17). The results show that the species such as Zygophyllum album, Dipterygium glaucum, Ochradenus 221 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen baccatus, Prosopis juliflora, Tamarix aphylla, Rhazya stricta, Tephrosia nubica, Calotropis procera, and Panicum turgidum are strongly negatively correlated with flood erosion, while species such as, Kohautia retrorsa, Acacia hamulosa, A. ehrenbergiana, Ziziphus leucodermis and Zygophyllum coccineum are sensitive to extreme flood erosion. Figure 5.17. Interpretation of the effect of erosion on the species using 2 variables (erosion and altitude). The species (bottom left e.g. Zygophyllum album, Dipterygium glaucum, Ochradenus baccatus, Prosopis juliflora, Tamarix aphylla, Rhazya strica, Tephrosia nubica, Calotropis procera and Blepharis edulis are strongly negatively correlated with flood erosion. The species Cymbopogon schoenanthus, Kohautia retrorsa, Acacia hamulosa, Barleria aff. bispinosa, Tephrosia dura, Ziziphus leucodermis and Zygophyllum coccineum are quite sensitive to extreme flood erosion. The plant species were further analysed using 4 soil variables (soil moisture, CaCO3, EC% and CeC). Figure 5.18a shows that weed species of agricultural fields (bottom right) such as Chloris barbata, Convolvulus arvensis, Cynodon dactylon, Ziziphus spina-christi and Withania somnifera and some species of secondary wadis and slopes on the plateau such as Acacia mellifera, Acacia oerfota, Tamarix arabica, Cometes abyssinica, 222 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Corallocarpus glomeruliflorus, Arnebia hispidissima, Pulicaria undulata, Zygophyllum decumbens and Halothamnus bottae are strongly correlated with moisture content and EC. Species such as Jatropha spinosa, Grewia erythraea, Helichrysum pumilum, Maerua crassifolia, Commiphora kua (top left) are negatively correlated with EC% and soil moisture, while species such as Indigofera spinifolia,, Chrozophora tinctoria, Dipterygium glaucum, Tephrosia apollinea, Aerva javanica and Iphiona anthemidifolia (top right) are positively correlated with EC% and negatively correlated with CaCO3. Figure 5.18a. Canonical Correspondence Analysis (CCA) ordination biplot of plant species and 4 environmental variables. Soil moisture and CaCO3 are the most important environmental variables. All environmental variables except CeC are strongly correlated with ordination axis 1. Weedy species of agricultural fields (bottom right) such as Chloris barbata, Convolvulus arvensis, Cynodon dactylon, Ziziphus spina-christi and Withania somnifera and some species of secondary wadis and slopes on the plateau such as Acacia mellifera, Acacia oerfota, Tamarix arabica, Cometes abyssinica, Corallocarpus glomeruliflorus, Arnebia hispidissima, Pulicaria undulata, Zygophyllum decumbens and Halothamnus bottae are strongly correlated with moisture content and EC. Species such as Jatropha spinosa, Grewia erythraea, Helichrysum pumilum, Maerua crassifolia, Commiphora kua (top left) are negatively correlated with EC% and soil moisture, while species such as Indigofera spinifolia,, Chrozophora tinctoria, Dipterygium glaucum, Tephrosia apollinea, Aerva javanica and Iphiona anthemidifolia (top right) are positively correlated with EC% and negatively correlated with CaCO3. 223 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Because no soil data from all sample points has been taken, the above result can be considered as first attempt to discuss the relationship between the plant species and the main environmental factors. Generally the plant species of the cultivated fields are negatively correlated with CaCO3, while the species of the main wadi bed, the plateau and the slope of the plateau are positively correlated with CaCO3 and negatively correlated with EC% and soil moisture (Figure 5.18b). Figure 5.18b. Canonical Correspondence Analysis (CCA) ordination biplot of main landforms and 4 environmental variables. The species of the cultivated fields are negatively correlated with CaCO3, while the species of the main wadi bed, the plateau and the slope of the plateau are strongly positively correlated with CaCO3 and negatively correlated with EC% and strongly with soil moisture. Some species such as Zygophyllum decumbens, Halothamnus bottae and Tamarix arabica resist the high content of EC%. Tephrosia apollinea, Dichanthium insculptum, Pulicaria undulate, Farsetia linearis, Portulaca oleracea and Corallocarpus glomeruliflorus are growing well on relatively flat moist sites. Chascanum marrubifolium, Cadaba heterotricha, Tarenna graveolens, Jatropha spinosa, Maerua crassifolia, Hochstetteria schimperi, Commiphora kua and Grewia erythraea grow well on dry stony steep slopes of the plateau, although the first 3 species are very rare. The 224 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Aristida triticoides grows well on dry stony steep north-east slopes. The vegetation associations on the slopes of the plateau appear to be dominated by Cadaba heterotricha, Tarenna graveolens, Maerua crassifolia, Aristida triticoides, Commiphora kua and Grewia erythraea and adjacent secondary wadis by Acacia mellifera, Acacia oerfota, Commiphora kua with Euphorbia rubriseminalis and Corallocarpus glomeruliflorus. Most of these species are no longer found in many sites. 5.9.3. Main vegetation associations The main vegetation associations of the study areas and their relationship to environmental factors have been analysed by Canonical Correspondence Analysis (CCA) using statistical software of MVSP version 3.1 (Figure 5.19). Environmental data were available on: moisture, CeC, O.C %, CaCO3, EC %, pH and soil depth. Figure 5.19.Ordination of main vegetation associations and environment yielded by CCA. Soil depth and CaCO3 are the most important environmental variables. CeC, OC%, pH and soil depth are strongly correlated with ordination axis 1, while soil moisture, CaCO3 and EC% are strongly correlated with ordination axis 2. The association of cultivated fields (1 and 3), the flooded wadi bed (2) are found to the right of the ordination, these associations are confined to site 1 and positively correlate with EC %, pH and soil depth, while the remaining associations are located to the left and positively correlate with caco3 and CeC. Numbers in the figure relate to the vegetation associations. 225 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen The most important variables in Figure 5.19 are soil depth and CaCO3. Vegetation association 1 of cultivated fields and 2 of sandy flooded wadis are strongly positively correlated with increasing pH, soil depth and EC%. Associations 1, 2 and 3 are negatively correlated with CaCO3. Associations 1 and 3 are characterised by being cultivated by both annual or perennial crops and wash out annually by floods. Association 7 of wadi bed of site 3 is strongly positively correlated with CeC and O.C%; this association characterised by species Merremia hadramautica, Tephrosia nubica, Ziziphus leucodermis and Panicum turgidum. CaCO3 variations between the associations are insignificant and generally range between 35.5 and 37.4, nevertheless the lowest record was found in association 3 (29.7) and the highest was in associations 14 (37.4), 9 (37.3) and 7 (37.2). The characteristic plant species of these associations are: Merremia hadramautica, Tephrosia nubica, Ziziphus leucodermis and Panicum turgidum, Boerhavia elegans, Dichanthium insculptum, Cleome scaposa, and Cleome brachycarpa. Organic Carbon (OC%) is low over most of the associations, especially on cultivated fields and places that are characterised by low vegetation cover such as slopes facing the main wadis. Onus et al. (2003) found that the amount of OC% was higher in natural sites than in cultivated fields due to presence of natural vegetation decomposing in the topsoil. 5.9.4. Vegetation structure and cover percentage The structure of the vegetation in the study areas varies from sparse grassland to dense woodland (Table 5.16). The main wadis are generally characterised by a low cover of herbaceous plants, while palm grove areas are characterised by high tree cover. The plateaus with their slopes and secondary wadis are characterised by a high cover percentage of herbaceous plants (Figures 5.20 and 5.21). There are variations of structure within the main wadis over the study sites and on the slopes of plateau of site 2. The difference in vegetation structure in the main wadis may be due to environmental and land use conditions including wood cutting, severe flooding and over-grazing. On the plateau of site 2, oil working activities have changed the vegetation structure in some parts from grassland to woodland and from sparse grassland to dense grassland. 226 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Vegetation structure types The stud Land fo Dense wood Wood-lan Open woodla Shrub-land Sparse shr site 6 2 land 1 3 4 1 w 2 3 3 ms pla sp wp 1 pal 1 1 11 w 2 2 3 ms pla sp 1 5 wp Dwarf shr land 9 - 111 3 w ms pla sp wp 1 3 1 1 Dense gra Grass-land Sparse gra land land 7 5 8 2 4 1 10 2 2 2 3 4 2 4 1 2 1 3 3 7 1 4 1 1 1 5 9 1 5 1 Total spec per plot 14 11 4 2 4 6 14 3 1 19 6 12 9 1 7 2 Table 5.16 .The distribution of the vegetation structure classes over the study sites. w = main wadi, pal = palm grove and fields, ms = slopes facing the main wadi, pla = plateau, sp = slope on plateau, wp = secondary wadi. Slopes facing the main wadi, secondary wadis and plateaus are dominated by grassland structures; as can been seen most sample plots are located under the sparse grassland. There are variations of structure within the main wadis over the 3 study sites and on the slopes of plateau of site 2. Figure 5.20. Number of sample plots and vegetation structure over the different landforms. w= wadi bed, f= cultivated fields including the palm areas, ms= slope facing the main wadi, pla= plateau, wp= secondary wadi, sp= slope on plateau. Grassland and sparse grassland are the dominant vegetation structures. Variation of structures is found on the wadi beds, slopes of the plateaus and secondary wadis; this can be a reflection of human activities on these landforms. 227 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 91 W 67 F 10 33 17 21 8 MS PLA WP 17 10 9 10 8 10 23 33 48 67 67 SP 3 10 17 5 100 90 80 70 60 50 40 30 20 10 0 17 % Figure 5.21a. Canonical Correspondence Analysis (CCA) of the percentage cover of vegetation structure and different landforms. Slopes facing the main wadi, secondary wadis and slopes on the plateau are characterised by a high percentage herbaceous cover, while main wadis are characterised by woody species cover. Herbaceous cover is most dominant vegetation over the different landforms in the study areas. Numbers refer to the sample plots Dense Woodland Open Shrubland Sparse Dw arf Dense Grassland Sparse w oodland w oodland shrubland shrubland grassland grassland vegetation structure Figure 5.21b, The percentage of each vegetation structural type over the different landforms. W= main wadi, F= fields, MS= slope facing the main wadi, PLA= plateau, WP= secondary wadi on plateau, SP= slope on plateau. Grassland and sparse grassland are the dominant structural types especially on slopes facing the main wadis and on the plateaus. The percentage of vegetation structural types over the main wadis varies with relatively high percentages of structural types dominated by woody shrubs. The vegetation of secondary wadis on plateaus is dominated by grassland structural types, while the vegetation of the slopes on the plateau is dominated by sparse grassland and sparse shrubland. 228 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Grassland and sparse grassland is the dominant structural types especially on the slopes facing the main wadis and on the plateaus. The percentage of vegetation structural types over the main wadis is varied with relatively high percentages of structural types dominated by woody shrubs. The vegetation of secondary wadis on plateaus is dominated by grassland, while the vegetation of the slopes on the plateau is dominated by sparse grassland and sparse shrubland. The mountain slopes facing main wadis and the plateaus are characterised by homogeneous vegetation composed mainly of sparse grassland. The abundance of each species within different landform units was analysed using the BioDiversity software. The output of cover percentage against log species rank is shown in Table 5.14 and Figure 5.22. Table 5.17 and Figure 5.22 show that each landform is dominated by a few species. Steep curves indicates that the landform is only dominated by few species, for example, pla3 and pla2 have less than 3 species and are almost dominated by one species namely Stipagrostis hirtigluma. The vegetation cover percentage of this species ranges from 1 to 10% per 100 m². Acacia campoptila and Fagonia indica are the dominant species on the main wadis, Stipagrostis hirtigluma is the dominant on the plateaus while Dichanthium insculptum and Stipagrostis hirtigluma are the dominant species on the slopes and secondary wadis of the plateaus. 229 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Table 5.17. Species covering more than 1% per 100 m² for each landform unit. w = wadi, ms = mountain slopes facing main wadis, wp = secondary wadi, sp = slope on plateaus, pla = plateau. The plateaus the mountain slopes are represented by a few species, while the main wadis and secondary wadis on plateaus have a relatively high number of species. 230 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Figure 5.22. Average vegetation cover percentage per 100 m² against log. species rank. w = wadi, ms = mountain slopes facing main wadis, cliff = very steep cliffs facing the main wadis, wp = secondary wadi, sp = slope on plateaus, pla = plateau. pla3 and pla2 at the bottom have less than 3 species and dominated by almost one species namely Stipagrostis hirtiglu;, wp2 and wp3 at the top are dominated by more than on plant species. Acacia campoptila and Fagonia indica are the dominant species on the main wadis, Stipagrostis hirtigluma is the dominant species on the plateaus while Dichanthium insculptum and Stipagrostis hirtigluma are the dominant species on the slopes and secondary wadis of the plateaus. Generally, the vegetation cover percentage on the plateaus and mountain slopes facing main wadis is very low and ranges between 1 and 9%. The main wadis and secondary wadis on plateaus have a high cover percentage ranging from 13 to 23% in the first case and from 8 to 29% in the second case (Figure 5.23). The vegetation cover percentage per 100 m² (Figures 5.23 and 5.24) varies across the landforms and is almost very similar to the number of species per hectare. The vegetation percentage (Figure 5.24) over the plateaus increases steadily from the exposed flat surfaces to the slopes then to the bottom of the secondary wadis. In the main wadis the number of species for the study sites increases from the slopes facing the main wadis to the main wadi beds. The peak in the cover percentage (29% per 100 m²) is found at the bottom of the secondary wadis that cut the plateaus of site 2. The lowest is found on the plateau surfaces. 231 w3, 13 ms3, 2 ms2, 3 ms1, 2 pla1, 9 pla3, 1 pla2, 2 wp1, 12 wp2, 29 w2, 14 w1, 23 sp3, 3 sp2, 17 sp1, 13 wp3, 8 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Average vegetation cover % Figure 5.23. Average vegetation covers percentage per 100 m² for each landform. w = wadi, ms = mountain slopes facing main wadis, wp = secondary wadi, sp = slope on plateaus, pla = plateau. The average vegetation cover percentage varies across the landforms. The avarage cover on the main wadis is relatively high and ranges between 13 and 23%. The average cover in the secondary wadis of the plateaus is also high and ranges between 8 and 29%. The average cover of the plateaus and mountain slopes facing main wadis are very low and ranges between 1 and 9%. pla2 and pla3 have the lowest value. 35 30 25 site 1 20 site 2 15 site 3 10 5 0 w ms pla sp wp Land form Figure 5.24. Average vegetation cover percentage per 100 m² over the different landforms. The vegetation percentage over the plateaus increases steadily from exposed flat surfaces, to the lopes then to the bottom of the secondary wadis. In the main wadis the percentage increases from the slopes facing these wadis to the main wadi beds. High percentage cover is found on main wadis and on the secondary wadis of the plateaus. The low percentage of cover at secondary wadis of sites 1 and 3 is due to the landform characteristic of being relatively narrow and shallow with steep short sides; in addition, the surface of these secondary wadis have solid pavement with a soil cover of less than 2%. 232 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen It is noticed from previous section that the structure of the vegetation in the study areas varies from sparse grassland to dense woodland. The main wadis are generally characterised by a low cover of herbaceous plants. The plateaus with their slopes and secondary wadis are characterised by a high cover percentage of herbaceous plants. There are variations of structure within the main wadis over the study sites. The difference in vegetation structure in the main wadis may be due to environmental and land use conditions including wood cutting, severe flooding and over-grazing or to the variation in soil properties. Grassland and sparse grassland is the dominant structural types especially on the slopes facing the main wadis and on the plateaus. The percentage of vegetation structural types over the main wadis is varied with relatively high percentages of structural types dominated by woody shrubs. The vegetation of secondary wadis on plateaus is dominated by grassland, while the vegetation of the slopes on the plateau is dominated by sparse grassland and sparse shrubland. The vegetation cover percentage on the plateaus and mountain slopes facing the main wadis is very low while that of the main wadis and secondary wadis is high. 5.10. Summary of main findings Using the Braun-Blanquet and Neighbour-Joining methods and Modelling Patterns in Environmental Data (MOPED) software (see Table 5.6), the plant species that behave similarly (Zonneveld, 1989) or are distributed similarly across the sample plots (Martin et al., 2003) were clustered in groups. These similar groups are named to as Sociological species groups. The species of sociological group E and H occur throughout the entire area, in particular on mountain slopes facing the main wadis, the plateaus and their slopes and secondary wadis. These species are: Acacia campoptila, Fagonia indica, Tephrosia apollinea, Cymbopogon schoenanthus, Dichanthium insculptum and Boerhavia elegans. With the exception of Tephrosia apollinea, they are not seen on the sandy, strong, flooded wadi beds and cultivated areas. Acacia campoptila is near-endemic, and the remaining species are also found in wadis in many parts of the tropical Sahara and Arabia regions. 233 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen A total of 116 sample plots in different landforms and ecologies throughout the study sites were combined according to the similarity in their species composition, following the classification methods of the Braun-Blanquet and Twinspan, to form 15 groups of floristic vegetation associations. Most of the vegetation associations are dominated by herbaceous plants Vegetation associations 1, 2, 4, 8 and 13 are confined to site 1; vegetation associations 3, 6, 13 and 15 are confined to site 2, while vegetation associations 7, 9 and 14 are confined to site 3. The vegetation associations of the main wadis are dominated by Acacia campoptila, Fagonia indica, Tephrosia apollinea and Cymbopogon schoenanthus. Associations 8, 9 and 10 are dominated by Dichanthium insculptum, Boerhavia elegans, Cleome brachycarpa, Stipagrostis hirtigluma and Farsetia linearis. Associations 13, 14 and 15 are dominated by Dichanthium insculptum Zygophyllum decumbens, Jatropha spinosa and Stipagrostis hirtigluma. The community dominated by Acacia campoptila, with association of Panicum turgidum, Aerva javanica, Indigofera spinosa, Fagonia indica, Ochradenus baccatus, Dipterygium glaucum Calotropis procera and Rhazya stricta is widespread over the wadi beds of arid areas of Yemen such as Marib and Rada’ in Yemen (Al Khulaidi, 1989; Westinga and Thalen, 1980; Al Hubaishi and Muller-Hohenstein,1984). Acacia spp. (e.g. A. tortilis and A. hamulosa) form savanna-like communities in association with Panicum turgidum, Leptadenia pyrotechnica, Aerva javanica, Calotropis procera and Ziziphus spina-christi. They are widespread throughout the tropical Sahara region and in the wadis of Sudan, Egypt, Palestine and Arabia (Zohary, 1973; White, 1983). The species Calotropis procera and Dipterygium glaucum are the most characteristic species of wadi borders, sand dunes and sands (Wood, 1997) throughout the coastal areas and desert. Association 2 reflects the transitional zone between arid and hyper-arid climate and is very similar to the associations that are common further west on the sand dunes of the Ramalat Al Sabaten desert area (chapter 3). Grass root systems on sandy soils extended to depth of 1.4 m or more likely to be an adaptation to capture more soil water from frequent small rainfall events (Gibbens and Lenz, 2001). Some species of this habitat (e.g. Panicum turgidum and Zygophyllum album) have very 234 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen long roots which can spread over the surface as well as in depth utilizing both the available surface moisture and water that has infiltrated deep layers. Additionally the sandy soils of this habitat may offer better water supplies (Monger, 2002). Wadi vegetation is found in hydrological disturbance areas, including annual floods, erosion and alluvial deposition. The active surface receives intense flooding events. Well-established vegetation associations are only found on parts of the flood plain that generally do not receive intensive flood waters. The associations on the active surface are adapted to rapid colonization of disturbed and flooded surfaces. Extreme floods are common on main wadis. Floods have significant effect on the structure of vegetation of wadis. The wadis have varied relationships with surface and ground water. Some associations develop under conditions where their deep root systems can utilize the water from basal gravels and lower, silty alluvial layers (Coode Blizard Ltd, 1997). In these habitats, woody perennial trees such as Acacia campoptila, Ziziphus leucodermis, the date palm Phoenix dactylifera and the invasive Prosopis juliflora form their own distinctive associations. Savanna-like vegetation on wadi beds and runnels in several locations has been transformed to woodland due to the encroachment and establishment of the invasive tree Prosopis juliflora. The association dominated by Zygophyllum album and Dipterygium glaucum was only found on relatively wide wadis that are subjected to severe flooding, with loamy to loamy sand soils and high proportions of sand (86-88%). Other species?/included in this association are Citrullus colocynthis, Heliotropium rariflorum, Dipterygium glaucum, Panicum turgidum, Prosopis juliflora, and Tamarix aphylla. Most of these species have deep roots, which probably offer good mechanical support against the relative high water velocity (Wittmanna et al., 2004) during flood ?episodes/periods. Other species showing adaptations to extreme floods are Acacia campoptila, Fagonia indica, Tephrosia apollinea subsp. longistipulata and Cymbopogon schoenanthus. As a result of severe floods, woody vegetation dominated by Ziziphus leucodermis, Acacia ehrenbergiana, Acacia hamulosa, Tephrosia dura and date palm trees has disappeared or shrunk and has been replaced by other communities more resistant to the severefloods, such as Prosopis juliflora, Zygophyllum album, Acacia campoptila and Tephrosia apollinea. 235 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Association 12 (Figure 5.25) of secondary wadis of plateaus and association 7 of the main wadi beds of site 3 are the most diverse. The number of species in these associations range between 34 and 44 and the individuals range between 468 and 537. Associations 2, 10, 14 and 15 are characterised by a low number of species and individuals. Associations 1 and 3 have the highest vegetation cover percentages, the former is dominated by trees and shrubs namely Phoenix dactylifera, Prosopis farcta and Alhagi graecorum, while the latter is dominated by dwarf shrub and herbaceous cover namely Merremia hadramautica, Crotalaria persica, and Heliotropium ramosissimum. no of individuals 600 80 no of species and total cover % 70 500 60 400 50 300 40 30 200 20 100 10 0 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Association number of species average vegetation cover % number of individuals Association number of species average vegetation cover % number of individuals 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 13 19 24 27 24 34 27 16 11 36 44 31 14 7 41 196 16 122 70 91 23 301 14 285 12 310 13 537 4 211 1.9 284 3 58 13 437 20 468 19 3 453 75 4 93 Figure 5.25. Numbers of species, individuals, and the total vegetation cover percentages of the main floristic associations. Associations 7 and 12 are the most diverse and associations 2, 10, 14 and 15 are the least diverse. 236 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen There are distinct variations in vegetation structure on the main wadis over the study sites and on the slopes of plateau of site 2. These are indicative of human and natural activities that are significant on these landforms. In the first case, floods play an important part in changing the vegetation structure from woodland to grassland. In the second case, oil-exploration plays a role in changing the structure from shrubland or/and grassland to sparse grassland. Table (5.18) shows the relationship between vegetation association and structure. As can be observed from this table, two or more sample plots that belong to the same vegetation association may have considerable differences in vegetation structure. For example vegetation associations 5, 12 & 13 have 5 different structure classes. The difference in vegetation structure are likely to result from environmental and land use conditions, including wood cutting, severe flooding and over-grazing. In the study area cutting of wood, road construction and grazing are common in the study sites. vegetation association vegetation structure 1 1 2 3 4 1 1 3 3 4 3 4 6 1 6 7 2 2 5 5 2 8 9 10 9 3 13 14 15 3 2 1 1 2 2 2 12 1 1 1 7 8 11 1 2 6 1 6 6 8 6 2 5 3 8 2 1 1 1 8 2 1 1 1 3 4 Table 5.18. The relationship between vegetation association and structure. 1= woodland, 2= open woodland, 3= shrubland, 4= sparse shrubland, 5= grassland, 6= dense woodland, 7= dense grassland 8= sparse grassland, 9= dwarf shrubland. As can be observed from this table, two or more sample plots that belong to the same vegetation association may have considerable differences in vegetation structure. For example vegetation associations 5, 12 & 13 have 4- 5 different structure classes. The difference in vegetation structure are likely to result from environmental and land use conditions, including wood cutting, severe flooding and over-grazing. 237 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen EC % pH soil depth altitude slope % surface stoniness rock outcrops 0.4 0.2 0.5 0.4 0.3 0.5 7 0.2 0.2 0.2 0.3 0.2 0.4 0.6 0.5 35.5 37.1 29.7 37.1 37 37 37.2 35.7 37.3 37 37 37 36.8 37.4 37.2 3 0.3 0.5 7.4 3.8 0.2 1.4 1 0.6 0.4 1.6 6.9 0.3 0.4 0.5 7.9 8.7 8.1 8.3 7.9 7.6 7.6 8.2 7.9 8.1 8.1 7.4 8.1 8 8 50 30 30 20 20 20 20 20 20 20 20 30 20 10 10 724 700 676 733 686 681 668 757 685 763 995 979 977 955 949 1 2 2 3 2 2 2 30 36 42 12 8 30 29 36 1 1.6 1.5 82 45 64 77 84 93 90 89 85 85 97 95 0 0 0 27 0 1.7 2.3 2.3 8.2 13 0.8 3.7 3.7 4.5 2.2 2 5 4 4 4 4 4 3 3 3 3 4 2 3 2 24 1 0 2.9 8.2 1 2.2 1 0 0 1 2.5 2.7 1 0 0 0.3 0 12 2.2 0.3 1 0 0 0 1 0.2 1.8 1 0 14 0 60 1 1.4 5.3 9 0 0 0 0 1 0.5 1 0 2.6 16 10 7.7 2.8 3.8 2.5 3.6 2.5 2.9 14 17 14 1.3 4 Table 5.19. Main floristic vegetation associations with characteristic soil, strata and landform data.. Finally, it is understood from this chapter that the study sites have various vegetation associations and structure. Oil working activities and floods have affected vegetation associations, composition and structure. The vegetation cover percentage also varies from one landform to another. Previously the main vegetation associations and the vegetation structure types, their environmental characteristics and similarities were described. In the next chapter the vegetation associations and the distribution of main plant species in the three study sites together with the distribution of some important endemic, near-endemic and rare plant species of Hadhramaut region will be analysed. 238 total vegetation cover CaCO3 % 6.2 1.7 6.1 5.3 5.5 6.2 10 5.2 7.6 5.7 8 6.1 4.3 5 5.2 herbs O.C. % 3 1.3 5.5 2.1 1 2.8 3.1 1.6 7.4 3.2 3.3 3 1.8 1 3 Dwarf shrub CeC 10 8 8 8 5 7.5 8.7 10 13 9.7 13 4 8.7 12 6 shrub moisture 40 5 32 37 36 35 36 40 43 39 35 36 35 31 38 trees clay % 49 87 60 55 59 58 54 50 45 52 52 60 56 57 56 erosion silt % 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 sand % Table 5.19 shows a summary of environmental characteristic of each vegetation association. 41 16 70 23 14 12 13 3.6 1.9 2.8 14 19 19 2.7 4 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Chapter 6. Distribution and mapping of the vegetation and land cover. 6.1. Introduction With the availability of new GIS techniques such as ERDAS Imagine and ArcView, the development of vegetation mapping and the plotting of the distribution patterns of plant taxa have taken rapid steps forward. An understanding of natural distribution patterns and the environmental restrictions on natural plant species helps in the formulation of conservation strategies (Ferrier, 2002).Vegetation and land use maps support conservation activities and, consequently, mapping plant distribution patterns provides a valuable framework and rapid visual for land management strategies. In addition, vegetation and plant distribution maps provide essential information on setting conservation priorities; for example, this study has shown that the south-facing mountain slopes of the Hadhramaut region are an important centre for endemic and near-endemic plants. Vegetation and distribution maps are also basic sources of information for the regional and national mapping of plant species. Finally, they furnish a baseline database for vegetation, landscape and conservation biology studies in the future. In previous chapters, the plant diversity and vegetation associations of the study sites were described in detail. The aims of this chapter are to:    describe the distribution and the dynamics of the main plant species of the Hadhramaut region by plotting them on maps, outline the main vegetation associations on large scale vegetation maps, provide information on the names and geographic distribution of plant species in the Hadhramaut region – concentrating on those of special status, for instance, those that are rare, endemic or near-endemic, or of economic importance, In this section, distribution maps of 60 plant species from the study sites and 24 from the Hadhramaut region are presented. These species are distributed across all landforms and habitats, including wadi beds, rocky mountain slopes and plateaus. 239 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 6.2. Methods of study: The accuracy of the vegetation map depends on the spatial and spectral resolution and the seasonal variability in vegetation cover types and soil moisture conditions (Congalton, 1988). One of the main problems when generating land cover maps from digital images is the confusion of spectral responses from different features (Eiumnoh and Shrestha, 1997). When the vegetation classification using ERDAS IMAGINE was carried out, different features with similar spectral behaviour were grouped into the same class. For example, the cover of the mountain slopes and the slopes of the plateau were grouped with agriculture fields in the main wadis (see Figure 6.1). This confusion between vegetation areas, with greater soil moistures, geology materials and bare areas in shadow or steep sloped, limited the degree of differentiation possible in an automatic classification. As in other arid or semi-arid areas (Miller and Cope, 1996), the study sites are characterized by sparse vegetation cover with most of the land surfaces, in particular on the plateaus, being almost bare. Consequently, the spectral response in the area essentially represents the geology rather than the vegetation cover. 240 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Landsat TM image of 1998 Slopes on the plateau Land cover map produced by unsupervised classification using ERDAS IMAGINE 8.4 Mountain slope & shadow fields Different land cover units were grouped into same class. Here the shadow, the cover on mountain slopes and slopes on the plateau classified grouped with the agriculture fields of main wadi. the the are the Figure 6.1. Vegetation and land cover map using ERDAS IMAGINE. The cover of the mountain slopes and the slopes of the plateau were grouped with agriculture fields in the main wadis. In the study area, each landform has a combination of vegetation associations (for instance, associations 1 to 7 in the main wadis and 8 to 10 on the mountain slopes) and because of the difficulties of mapping each vegetation association separately, the main landform units of the study sites (see chapter 1) were used to create the vegetation association maps. The different landforms were identified by interpreting the satellite imagery and delineating their boundaries. Mapping the secondary wadis as a separate unit was not possible due to limited resolution of the image. Therefore, a combination of landforms was used to describe the vegetation associations of these units. The resulting geomorphic maps of the study sites were generated using ArcView 3.2a. This was done by entering the available Landsat 241 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen imageries of 1989 (pixel size 25X25) and of 2001 (pixel size 15X15) as data layer. Then, using the irregular polygon button, lines were drawn around the different landforms. The species distribution maps were plotted using ERDAS IMAGINE and DIVA-GIS software. The spot data for each species are based on the geographical coordinates of known localities obtained during fieldwork (using a handheld GPS) and from reliable literature sources (Lavranos, and Al-Gifri 1999; Lavranos, and Mies 2001; Lavranos, 1996; Thulin, 2001 and 2002; Thulin, and AI-Gifri 19953 and 1995; Kilian, et al., 2002). ArcView 3.2a ArcView is made by Environmental Systems Research Institute (ESRI). Over the past 20 years ArcView has proved to be a powerful tool that brings geographic information to the desktop and gives the power to visualize, explore, query and analyze land cover and land form data. ERDAS IMAGINE 8.4 This is software designed for image mapping and visualization, image processing and advanced remote sensing. DIVA-GIS DIVA-GIS is a free mapping software package for biodiversity and ecological research that allows the user to analyze plant species distributions. DIVA-GIS 4 is developed by Robert J. Hijmans, Luigi Guarino, Prem Mathur, Rachel O'Brien and Andrew Jarvis and can be downloaded from . http://www.diva-gis.org/ 6.3. Result The initials stage of the analysis was depiction of the topography and wadis in the 3 study sites (as described in chapter 1). The results are shown in Figures 6.2 to 6.4. 6.3.1- Relief and drainage 242 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Figure 6.2. Topographic features of study site 1. The altitude ranges between 660 and 1020 m asl. The study site is located between 15º 45’ and 15º 58’ latitude and 48º 34’ and 48º 35’ longitude. The study site is located on the main Wadi Hadhramaut, the tributary wadis are: W. Nia’m and W. Ssuhaybah (north), W. Al Khashamer, she’b Sultan and W. Sar (south). The main villages are Jubah and Huwaylah (north), Khashamer and Khumur (south). 243 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Figure 6.3. Topographic features of study site 2. The altitude ranges between 640 and 960 m asl. The study site is located between 15 52’ and 15 48 latitude and 48 53’ and 48 59’ longitude. The main wadi is W. Adim, the tributary wadis are; W. Bayut and W. mushghula (west), W. Hikma, W. Al Quful and W. Utaeir (south), Shei’b Assalaq and W. Suna (east), W. Annier, W. Al Auass and W. Sharyuf (north). The main village is Hikma. 244 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Figure 6.4. Topographic features of study site 3. The altitude ranges between 630 and 1009 m asl. The study site is located between 16 03’ and 16 05 latitude and 48 55’ and 48 59’ longitude. The site is located on Wadi Athahab, the tributary wadis are: W. Thibi (north), Shie’b Al Aeim and W. Shib (east), W. Nasma (west). The main village is Al Marabedh. 245 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 6.3.2. The vegetation associations of the study sites The next stage was the evaluating of the main vegetation associations (as outlined in chapter 5). These associations are overlain onto the relief and drainage maps. Table 6.1 shows the allocation of the vegetation associations across the 3 study sites, Figures 6.5 to 6.7 illustrate their distributions across the landscape. Landform units were used to plot each association. Association 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Site 1 + + Site 2 Site 3 + + + + + + + + + + + + + + + + + + + Table 6.1. Distribution of the vegetation associations across the study sites: associations 1,2,4,8 and 15 are confined to site 1; associations 3, 6 and 13 are confined to site 2; associations 7, 9 and 14 are confined to site 3; associations 5 and 10 are confined to sites 1 and 2, while associations 11 and 12 are distributed across all the study sites. The vegetation associations are described in detail in chapter 5. 246 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Figure 6.5. Main vegetation associations of site 1. vegetation structure: 1- woodland, 2- open woodland, 3shrubland, 4- sparse shrubland, 5- grassland, 6- dense woodland, 7- dense grassland, 8- sparse grassland, 9-dwarf shrubland. sl= sandy loam, l= loam, ls= loamy sand. 247 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Figure 6.6. Main vegetation associations of site 2. Vegetation structure: 1- woodland, 2- open woodland, 3shrubland, 4- sparse shrubland, 5- grassland, 6- dense woodland, 7- dense grassland, 8- sparse grassland, 9-dwarf shrubland. sl= sandy loam, sil= silt, ls= loamy sand. 248 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Figure 6.7. Main vegetation associations of site 3. Vegetation structure: 1- woodland, 2- open woodland, 3- shrubland, 4- sparse shrubland, 5- grassland, 6- dense woodland, 7- dense grassland, 8- sparse grassland, 9--dwarf shrubland. sl= sandy loam, l= loam, ls= loamy sand, sil= silt, nd= no data. 249 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen For each land unit depicted in the previous maps, the surface areas have been calculated as shown in Tables 6.2 to 6.4. The total area of the 3 study sites is about 2541 hectares. The class Association 2 Associations 8 and 10 Associations 11 and 12 Association 4 Association 5 Cultivated field (association 1) Bare land total Area/ha 508 1006 2108 160 122 1877 304 6085 percentage 8.3 16.5 34.6 2.6 2.0 30.8 5.0 Table 6.2. The area in hectares of each association and other land forms of site 1. The area of each land unit is calculated in hectare as following: The class Association 3 Association 5 and 6 Association 10 Association 11 Association 11, 12, 13 and 15 Irrigated field Rainfed field Bare land Total Area/ha 11 714 2496 9258 2689 238 229 70 15705 Percentage 0.1 4.5 15.9 58.9 17.1 1.5 1.5 0.4 Table 6.3. The area in hectares of each association and other land forms of site 2. The class Association 7 Association 9 Associations 11, 12 , and 14 Renfed field Bare land total Area/ha 359 1098 1542 237 380 3616 Percentage 9.9 30.4 42.6 6.6 10.5 Table 6.4. The area in hectares of each association and other land forms of site 3. 250 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Because of the complexity of landforms in the study sites and because of difficulties of mapping each vegetation association separately, combinations of associations were mapped together. For example, the landform and the vegetation associations of the secondary wadis of the plateaus and the adjacent slopes could not be satisfactorily separated. So, for instance, associations 11 and 12 in site 1 (Figure 6.8), 11, 12, 13 and 15 in site 2 (Figure 6.10) and 11, 12 and 14 in site 3 (Figure 6.11) were mapped together and presented on the previous vegetation maps as a group of associations. On the other hand, a specific landform can have more than one association, for example the mountain slopes of site 1 have a combination of associations 8 and 9 (Figure 6.9) and the slopes of the plateau of site 2 have a combination of associations 11, 13 and 15 (Figure 6.10). Because of the difficulties in mapping the vegetation associations as separate units, the cover percentage of each association in these landforms was estimated by taking into consideration the number of the sample plots that formed each association. The following figures show the estimated percentages of each association based on the number of sample plots. However, the percentage cover of each association in reality does not reflect the actual cover because each land form has a considerable number of bare sample plots. In particular, the mountain slopes facing the main wadis, the plateau surfaces and the slopes of the plateau have a large number of bare plots. As an illustration, 3 out of 4 sample plots on the plateau surface of site 2 were bare, which indicates that 75% of this landform is without vegetation. Similarly, 5 out of 21 sample plots on the slopes of the plateau in site 2 were bare, which indicates that 24 % of this landform surface is also without initially devoid of vegetation. 251 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Site 1 Association 11 covers most of the plateau areas; this association is restricted to the exposed plateau surface and adjacent slopes while association 12 is confined to secondary wadis of the plateau. Association 11 12 Land form Plateau Slope on plateau Secondary wadi Sample plot 4 2 5 Percentage area cover 36 % 18 % 45 % Figure 6.8. Percentage covers of associations 11 and 12 on the plateau of site 1 Figure 6.8 shows the percentage covers of associations 11 and 12 in each land form for the complex of topographical units on the plateau of site 1. . Association Land form 8 10 Mountain slope Sample plot 7 4 Percentage area cover 64 % 36 % Figure 6.9. Percentage cover of associations 8 and 10 on the mountain slope facing the main wadi. Association 8 covers most of the mountain slopes. Figure 6.9 shows the percentages of associations 9 and 10 in the mountain slope facing the main wadi of site 1. Association 8 covers 64 % of the slopes, while association 10 covers only 36 %. 252 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Site 2 Association 11 12 11 13 15 Land form Plateau Secondary wadi Slope on plateau Sample plot 1 4 1 10 5 Percentage 5% 19 % 5% 48 % 24 % Figure 6.10. Percentage cover of associations 11, 12, 13 and 15 on the plateau area of site 2. Association 13 covers most of the slopes of the plateau; the plateau surface is composed mainly of association 11, while the secondary wadis are composed only of association 12. The slopes of the plateau have a combination of associations 11, 13 and 15 in which association 13 covers almost half of these land forms. Associations 11 found on 2 different landforms (plateau surface and slope of the plateau). Site 3 Association 11 12 14 Land form Plateau Slope on plateau Secondary wadi on plateau Slope on plateau Sample plot 1 3 2 4 Percentage 10 % 30 % 20 % 40 % Figure 6.11. Percentage cover of associations 11, 12, and 14 on the plateau area of site 3. Association 14 covers most of the slope of the plateau; the plateau surface is composed mainly of association 11, while the secondary wadis are composed of association 12. 253 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Association Land form 11 14 Slope on plateau Sample plot 3 4 Percentage area cover 43 % 57 % Figure 6.12. Percentage cover of associations 11 and 14 on the slopes of the plateau of site 3. The slopes on the plateau have a combination of associations 11 and 14 in which association 14 covers 57 % and association 11 43% of the land surface of these land forms. 6.3.3. Agricultural fields The agricultural fields are mainly located in the main wadi. Most of the wadi bottom at site 1 is heavily cultivated, while few areas in sites 2 and 3 are cultivated. The main crops in the region are: 5. Cereals (wheat, sorghum and sesame), 6. Vegetables (onion, potato, tomato, garlic), 7. Fodder (alfalfa – Medicago sativa) Another type of cultivated area is represented by palm-groves. These are composed of pure stands of Phoenix dactylifera; sometimes mixed with annual crops such as Alfalfa, vegetables (e.g. onions, potatoes, tomatoes, garlic and okra) and sesame. Alfalfa is a major fodder crop in Wadi Hadhramaut. The irrigated fields are mainly cultivated with palms, sorghum, wheat, alfalfa (Medicago sativa) and other vegetables. The cultivated areas of the study sites were measured using ArcView and found to be as following (see Figures 6.13):Site 1: 1877 ha Site 2: 467 ha (irrigated = 238 ha and rainfed = 229 ha) Site 3: 237 ha 254 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Site 1 Site 3 Site 2 Figure 6.13. Maps of the cultivated fields of the 3 study sites. The areas are as following:Site 1: irrigated fields total 1877 ha. Site 2: irrigated fields total 238 ha and rainfed fields total 229 ha Site 3; rainfed fields total 237 ha 255 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 6.3.4- Plant species distribution within the study sites Distribution maps of the main plant species in the 3 study sites were generated using ERDAS Imagine 8.4. Each point on each map represents the GPS location of a plant population. As expected, distribution patterns differ from one plant to another, however, some common patterns have emerged. Certain species occur across the 3 study sites whilst others are limited to only one or two study sites. Furthermore, distributions differ depending on landform; some species are restricted to the main wadis (e.g. Prosopis farcta, Alhagi graecorum, Zygophyllum album, Rhazya stricta and Ochradenus baccatus), others prefer the dry rocky slopes of the plateaus (e.g. Jatropha spinosa, Maerua crassifolia, Hochstetteria schimperi, Commiphora kua, Grewia erythraea, Helichrysum pumilum and Cornulaca amblyacantha), and whilst yet others grow well on sandy deep soil (e.g. Dipterygium glaucum and Panicum turgidum). Many species in the study sites show altitudinal preferences (Table 6.5), but further surveys are needed over the whole study area to get a complete picture of the altitudinal range of all the species. 256 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 980 990 970 975 960 965 950 955 855 900 800 850 765 770 755 760 700 750 690 695 680 685 670 675 660 665 650 655 640 645 630 635 620 Plant species / Altitude 625 Table 6.5. The distribution of the plant species along an elevation gradient. Boerhavia elegans Cleome brachycarpa Fagonia paulayana Tephrosia apollinea Aristida triticoides Farsetia linearis Cymbopogon schoenanthus Fagonia indica Senna italica Cleome droserifolia Indigofera spinosa Acacia campoptila Merremia somalensis Ziziphus leucodermis Aerva javanica Blepharis edulis Kohautia retrorsa Heliotropium ramosissimum Rhazya stricta Ochradenus baccatus Panicum turgidum Tephrosia nubica Indigofera oblongifolia Dichanthium insculptum Plicosepalus curviflorus Calotropis procera Acacia ehrenbergiana Pulicaria undulata Senna holosericea Tribulus arabicus Salsola imbricata Cleome scaposa Continued overleaf 257 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Plant species / Altitude Barleria aff bispinosa Halothamnus bottae Seddera latifolia Dipterygium glaucum Tephrosia dura Heliotropium rariflorum Tamarix aphylla Corchorus depressus Lasiurus scindicus Stipagrostis hirtigluma Enneapogon desvauxii Leptadenia arborea Citrullus colocynthis Cressa cretica Phoenix dactylifera Ziziphus spina-christi Zygophyllum album Zygophyllum coccineum Cynodon dactylon Convolvulus arvensis Chloris barbata Prosopis juliflora Zygophyllum album Crotalaria persica Chrozophora tinctoria Convolvulus glomeratus Alhagi graecorum Indigofera spinifolia Senra incana Prosopis farcta Rhynchosia memnonia Zygophyllum simplex 620 625 630 635 640 645 650 655 660 665 670 675 680 685 690 695 700 750 755 760 765 770 800 850 855 900 950 955 960 965 970 975 980 990 995 Table 6.5. Continued 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 Continued overleaf 258 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Plant species / Altitude Acacia hamulosa Capparis spinosa Cometes abyssinica Salvadora persica Cadaba farinosa Iphionia scabra Forskohlea tenacissima Capparis cartilaginea Cyperus conglomerates Maerua crassifolia Iphionia anthemidifolia Reseda sphenocleoides Lindenbergia indica Anticharis glandulosa Aerva artemisioides Commiphora kua Heliotropium longiflorum Chrysopogon aucheri Commiphora foliacea Hochstetteri schimperi Jatropha spinosa Zygophyllum decumbens Cornulaca amblyacantha Acacia oerfota Corallocarpus glomeruliflorus Seetzenia lanata Crotalaria saltiana Acacia mellifera Grewia erythraea Helichrysum pumilum Arnebia hispidissima Tamarix arabica. Cucumis canoxyi Euphorbia rubriseminalis Glossonema varians Hermannia paniculata Tarenna graveolens Chascanum marrubifolium Moltkiopsis ciliata 620 625 630 635 640 645 650 655 660 665 670 675 680 685 690 695 700 750 755 760 765 770 800 850 855 900 950 955 960 965 970 975 980 990 995 Table 6.5. Continued 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 259 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen The altitudinal setting of plant species is usually regarded as a reliable marker of their ecological statute (Zohary, 1973), but still other factors such as human activities and climate, may be involved in this distribution. It can be noticed in Table 6.5 that some species occur over a wide range of altitude; examples being Boerhavia elegans and Cleome brachycarpa, while others are restricted to low altitude areas, for instance, some species of the main wadi, namely Ochradenus baccatus, Panicum turgidum and Tephrosia nubica; other species restricted to high altitude areas, for instance, Commiphora foliacea, Hochstetteri schimperi, Jatropha spinosa, Zygophyllum decumbens, Cornulaca amblyacantha, Acacia oerfota and Corallocarpus glomeruliforus. A number of species had a distribution that was clearly and always linked with altitude. For example, species only found below 700 m include Tamarix aphylla, Lasiurus scindicus, Zygophyllum album, Alhagi graecorum, Prosopis farcta, Ochradenus baccatus, Tephrosia nubica, Rhazya stricta and Panicum turgidum. Almost all of these species are Saharo-Sindian elements, while species restricted to higher altitude areas (over 800 m) include: Commiphora kua, Commiphora foliacea, Grewia erythraea, Jatropha spinosa, Tarenna graveolens, Hochstetteri schimperi, Zygophyllum decumbensis, Euphorbia rubriseminalis, Chascanum marrubifolium, Moltkiopsis ciliata, Acacia mellifera and A. oerfota. Other environmental factors should also be taken to consideration. For example, some factors may influence plant species distribution differently over altitude (such as moisture and human activities). However, these factors may contribute to the decline in plant species that is often observed with a wide range of altitude. However, there is an overall decline in plant species with altitude as shown by Lawton et al, (1987). For example Stipagrostis hirtigluma is a widespread at high altitude but it often disappears or becomes rare at much lower altitudes, probably because of its vulnerability to over-grazing or due to other human or climate factors. Similarly, Maerua crassifolia, which is palatable in particular for camels and is common at high altitudes, is rare down at low altitude. Other example is Jatropha spinosa, which was not observed at low altitude in the study area, but is common at low altitude along the coastal plain of Yemen (Al Hubaishi & Muller-Hohenstein, 1984; Al Khulaidi, 1989; Wood, 1997). The absence of this species at a similar altitude in the main wadi of the study area may be 260 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen because of the humidity that characterises the coastal area or due to the difference in timing of rainfall from the north (Wadi Hadhramaut) to the south (coastal area). The rainfall in the north falls in spring and summer, while in the coastal area falls mainly in spring and winter. Therefore, within the range of altitude of a vegetation type, smaller-scale variations in distribution may be controlled by smaller-scale features of environment such as soil type, moisture, human activities or topography (Woodward, 1987). The following section comprises distribution maps of important species from the study sites and a selection from the wider Hadhramaut. The distribution is also described in the caption of each map. The maps are arranged to illustrate the following distribution patterns:-: 1. Widespread species found on the different land forms of the collated study sites in particular in the main wadi beds, occasionally found on the slope of the plateaus. 2. Widespread species found over the different land forms throughout all the study sites but occasionally or not common on the main wadi beds. 3. Widespread species found on all land forms throughout the study sites. 4. Species which are not abundant, but which are found in almost all the study sites. 5. Species found only in site 1. 6. Species occurring only in site 2. 7. Species found in sites 1and 2. 8. Species of sites 1 and 3. 9. Species of sites 2 and 3. 261 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen I. Widespread species found in a variety of land forms across the study sites but always in the main wadi beds and usually on the slopes of the plateaus. 1- Acacia campoptila Grows mainly in the wadi beds over the study sites between 600 and 770 m asl, but rarely found at the base of the mountain slopes facing the main wadis and on secondary wadis of the plateau between 890 and 920 m asl. 3- Tephrosia apollinea subsp longistipulata Mainly grows abundantly in the wadi beds between 600 and 730 m asl, but also occurs on sandy flooded sites and as a weed on cultivated fields and fallow lands. This species is also found in moist areas on the sides of rocky mountain slopes and plateau areas between 750 and 980 m asl. 2- Fagonia indica Grows on almost all landforms, but particularly in the wadi beds between 620 and 760 m asl; Found in less frequency on other landforms (between 800 and 970m asl. 4- Cymbopogon schoenanthus Grows mainly in rocky wadi beds, more or less throughout the study sites between 620 and 750 m asl; also found in relatively moist sites in the rocky drainage lines that cut the plateaus between 900 and 980 m asl. Fagonia indica, Tephrosia apollinea, Cymbopogon schoenanthus and Acacia campoptila are widespread species found in a variety of land forms across the study sites but grow well in the main wadi beds. The first three species are widespread in the arid regions of east Africa, Arabia, Pakistan and India. The last one is near-endemic but similar to Sudano-Zambezian (Wood, 1997). The species of this group considered as the dominant plant species of the vegetation associations of the main wadi beds. 262 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen II. Widespread species found over a variety of land forms throughout the study sites but only occasionally or absent from the main wadi beds 1- Stipagrostis hirtigluma Abundant on the mountain slopes facing the main wadis, on the plateaus, as well as the slopes and narrow rocky secondary wadis of the plateaus (between 670 and 995 m asl). Rarely found in the main wadi beds where it is restricted to wide, very rocky hamada surfaces and to colluvial sites. 2- Aristida triticoides A species of the stony mountain slopes facing the main wadis, of the plateaus, and drainage lines, as well as the slopes and narrow rocky secondary wadis of the plateaus (between 620 and 990 m asl). Rarely found on the main wadi beds where it is restricted to wide, very rocky hamada surfaces in site 1 and to wide rocky wadi beds near the bottom of the adjacent slopes at 620 m asl in site 3. 3- Farsetia linearis 4- Fagonia paulayana Abundant on the stony mountain slopes facing the main wadis, on the plateaus, as well as the slopes and narrow rocky secondary wadis of the plateaus (between 680 and 980 m asl). It is not found in the main wadi beds A species of the stony mountain slopes facing the main wadis, as well as on the slopes and narrow rocky secondary wadis of the plateaus. Not found in the main wadi beds. The species of this group are widespread across all the land forms of the 3 study sites, except the main wadis. They are mainly plants of the stony mountain slopes facing the main wadis, the plateaus, and the slopes and narrow rocky secondary wadis of the plateaus. They are rarely found in the main wadi beds. Stipagrostis hirtigluma and Aristida triticoides are widespread from Africa, Asia to India (Clayton et al., 2005) 263 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen III - Widespread species found on all land forms throughout the study sites. 1- Dichanthium insculptum Widespread perennial grass, found between 600 995 m asl, in varied different landforms. It is not seen on cultivated or fallow lands and in less abundant in the wadi beds and grows well on the rocky and stony slopes. Where the floods are higher, it is found in less flooded tributary wadis and dry habitats such as rocky slopes and plateau. 2- Boerhavia elegans subsp. elegans Widespread annual or perennial herb, found between 600 995 m asl, in varied different landforms. It is not seen on cultivated or fallow land and in less abundant in the wadi beds and grows well on the rocky and stony slopes. The two species in this section are found on all the land forms throughout the study sites. They grow well on rocky and stony sites and are both important palatable plants which are sought out by livestock and collected by pastoralists during good seasons; as a result they are under some pressure and are sensitive to over-utilisation by pastoralists, Dichanthium insculptum is a widespread species in Africa, Yemen and Italy, while Boerhavia elegans is endemic to Arabia. 264 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen IV. Species which are not abundant, but which are found in almost all the study sites. 1- Acacia hamulosa 2- Acacia ehrenbergiana Species grows occasionally on the relatively moist wadi bed of site 1(between 685 and 750 m asl, and on the alluvial fans and drainage lines that cut the plateaus between 850 and 980 m asl. Occasionally found in moist sites along the wide flat sandy wadi beds and along river banks and on depressions near cultivated fields (between 625 and 750 m. Not often found on the plateau. 3- Acacia mellifera 4- Acacia oerfota This tree occasionally occurs on the moderately steep slope drainage lines and rocky dry steep slopes of the plateaus (between 950 and 960 m asl). A rare species occurring in the secondary wadis and on the rocky dry steep slopes of the plateaus of sites 1 and 2 (between 959 and 980 m asl). There are 5 species of Acacia in the study area (A. campoptila, A. ehrenbergiana, A. hamulosa, A. mellifera and A. oerfota), 4 of them are also found in NE. tropical Africa one (A. campoptila) is endemic to Yemen. Acacia tortilis a drought-resistant species that occurs in the coastal area of Hadhramaut reaches its northern limit of distribution in the W. Daua’n (about 1000 m.). It is a plant of wadi beds, dry stony flat plains and depressions and dominates some communities in the coastal areas. Acacia ehrenbergiana, A. hamulosa, A. mellifera and A. oerfota occur across a range of land forms, but are not as common as A. campoptila (see section I) which is often abundant where it occurs. Acacia oerfota was not found in site 3. 265 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 5- Rhazya stricta Occasionally found in wadi beds. it is restricted to remote wadi beds probably because it is collected for use in folk medicine by local people. This species is widespread and often abundant on sand dunes in the western part of the study area. 6- Iphiona scabra A species of the stony moderately steep, mountain slopes facing the main wadis, the plateaus, and in the narrow rocky secondary wadis of the plateaus (between 700 and 980 m asl). Rarely found on the slopes of the plateaus. Rhazya stricta and Iphiona scabra are Saharo-Arabian species grow in hot desert areas from Egypt to Pakistan. The first is mainly grows in the main wadi beds and sand dune areas, while the second grows on the rocky surface of the plateau and secondary wadis in particular site 3. 7- Ziziphus leucodermis A rare near-endemic species occurring scattered on flat to undulating rocky wadi beds, river banks (between 550 m and 680 m asl), and occasionally in the drainage lines or narrow wadis that cut the plateaus. 8- Merremia hadramautica Abundant on the flooded wadi bed of site 3, and occasionally found near cultivated fields, fallow lands and disturbed wet sites between 550 and 675 m asl. In sites 1 and 2 it is restricted to relatively wet sites such as cultivated fields or fallow sites. It is very rare on the plateau, found only once in a secondary wadi of site 3 at an altitude of 915 m asl. Ziziphus leucodermis and Merremia hadramautica are very palatable species and vulnerable to overgrazing, mainly found in the wadi beds. Where the grazing is less intensive such as site 3, they occur in high frequency. The first is near endemic. The second is endemic to Hadhramaut and adapted to rapidly colonise disturbed and flooded areas. 266 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 9- Blepharis edulis 10- Indigofera spinosa Found on rocky wadi beds particularly in sites 2 and 3 (between 620 and 750 m asl). It is abundant on the flat rocky Hamada surfaces and colluvial sites. On the plateau it is restricted to SW-facing steep rocky slopes between 850 and 960 m asl. A species mainly of very rocky wide wadis (between 550 and 690 m asl), but which extends on to the adjacent rocky, moist slopes and on to the slopes of the plateau (over 700m above sea level). 11- Commiphora foliacea A species mainly of the rocky dry steep slopes of the plateaus at site 2 (between 900 and 965 m asl), but occasionally also found in secondary wadis and the adjacent slopes of the plateaus in sites 1 (between 975 and 979 m asl) and 3 (at 950 m asl). 13- Maerua crassifolia Restricted to rocky slopes on the plateaus in sites 2 and 3 (between 900 and 975 m asl). Occasionally found on the tributaries of the main wadis and adjacent to or on cultivated fields in sites 1 and 2. 12- Ziziphus spina-christi A species of fertile cultivated field margins (between 670 and 750 m asl. In particular in site 3. 14- Dipterygium glaucum Most abundant on sandy and sand dune habitats between 645 and 750 m asl, but occasionally found in wet fallow lands and disturbed tributary wadi beds. The species from 9 to 14 are African elements. Blepharis edulis and Indigofera spinosa are of very rocky areas, mainly in the main wadis and colluvial sites with high frequency in sites 2 and 3. Commiphora foliacea and Maerua crassifolia are mainly of the plateau with high frequency in site 2. Ziziphus spina-christi is mainly of agriculture fields, found with high frequency in site 3.Dipterygium glaucum prefer the sand dune and fallow lands. 267 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 15- Cleome brachycarpa A species with wide broad distribution, particularly in the northern part of the study area (between 615 and 995 m asl). It mainly occurs on the rocky mountain slopes facing the main wadis of sites 1 and 3, it is also occasionally found in wadi beds, in particular in site 3 and on disturbed sites on the plateau in site 1. It is very rare in site 2. 16- Cleome scaposa Found on almost all flat moist wadi beds (between 600 and 700 m asl). Where annual rainfall is less than 50 mm, for instance in site 3, it is restricted to the relatively moist very rocky mountain slopes. The 3 Cleome species are African elements grow on the arid and semiarid areas from North Africa to Pakistan. They grow in high frequency in site 3 and prefer the rocky mountain slopes facing the main wadis. 17- Cleome droserifolia a species of the stony, moderately steep, mountain slopes facing the main wadis, the plateaus, and of the narrow rocky secondary wadis and alluvial fans of the plateaus (between 620 and 970 m asl). Rarely found in the main wadis. 268 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen V. Species found only in site 1 1- Prosopis farcta 2- Alhagi graecorum A perennial shrub which is often dominant near cultivated fields and fallow lands at altitudes between 680 to 750 m asl in the western part of the study area. A perennial dwarf shrub which is often dominant near cultivated lands, waste, moist sites and fallow lands at altitudes between 675 and 750. 3- Zygophyllum album 4- Zygophyllum coccineum A halophytic succulent herb found mainly in the western part of the study area. It grows mostly on sandy, flood plains at altitudes below 700 m, but is occasionally found in tributary wadis of site 1, in particular at the bottom of mountain slopes. A species of slightly sloping but very rocky wadi beds in site 1, at altitudes between 700 and 740 m asl. It also extends on to north- and southwest-facing mountain slopes, between 667 and 690 m asl. 5- Moltkiopsis ciliata 6- Iphiona anthemidifolia A very rare species occurring on the shallow drainage lines that cut the plateau in site 1 (between 970-980 m asl). A very rare species found only once on rocky wadi beds at 750 m asl in site 1. 269 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 7- Prosopis juliflora A fast-growing invasive tree. So far not found in sites 2 or 3. However, in many places the vegetation of the main wadi beds has been transformed into Prosopis juliflora woodland beneath which nothing else grows. It is spreading along the main wadis and represents a serious threat to the natural vegetation of the wadis. The species of this group are restricted to site 1. They are occasionally dominant or co-dominant in some plant association. Perhaps the most interesting species in this group is Zygophyllum album, which grows under the most desertic conditions further west. It forms a shrubland community with Prosopis juliflora, which is spreading further east and invading large areas of the main Wadi Hadhramaut. Some of these species are weeds of cultivated lands and others are very rare being found in only one location. Alhagi graecorum and Prosopis farcta are Irano-Turanian species probably they have been migrated from their origin areas in the past to study area and adapted to the arid with annual rainfall less than 40 mm that characterised site 1. 270 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen VI. Species occurring only in site 2 1- Zygophyllum decumbens 2- Hochstetteri schimperi A species mainly occurring on rocky, dry steep slopes on the plateaus (between 900 and 975 m asl). Also seen once in a moist secondary wadi on the plateau. Restricted to dry rocky slopes on the plateaus (between 900 and 975 m asl). 3- Cadaba farinosa 4- Grewia erythraea A very rare species found at the foot of a mountain slope in a tributary wadi (at 692m asl). A rare species restricted to moist rocky slopes on the plateaus (between 950 and 960 m asl). 5- Chascanum marrubifolium, 6- Euphorbia rubriseminalis Very rare species, only found on northeastfacing, relatively wet rocky slopes on the plateaus (at 950 m asl). A very rare species, only found on northeast facing relatively wet secondary wadi of the plateaus (at 920 m asl). 271 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 7- Tarenna graveolens Very rare species, only found on northeastfacing, relatively wet rocky slopes on the plateaus (at 950 m asl). 8- Cadaba heterotricha Very rare species, only found on northeastfacing, relatively wet rocky slopes on the plateaus (at 950 m asl). 9- Arnebia hispidissima A rare species, only found in site 2 on northeast facing, relatively wet rocky slopes on the plateaus (at 950 m asl) and in an adjacent secondary wadi (at 920 m). Most of the species confined to site 2 are of palaeotropical affinity, for example Tarenna graveolens is found in Kenya, Tanzania, Uganda and Cadaba heterotricha in Ethiopia , Kenya , Pakistan , Somalia, Oman (Apioi and Wronski, 2005). The exception is Chascanum marrubifolium, which is found in Saudi Arabia and India (Atiqur Rahman , et al., 2002; Al-Turki, 2004 ). 272 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen VII. Species found in sites 1and 2 1- Citrullus colocynthis 2- Cometes abyssinica A species of the flood plains and sandy wadi beds between 670-700 m asl, mainly in site 1. Also occurs on fallow lands and cultivated fields. A rare species preferring moist sites on very rocky alluvial fans and the moister slopes as well as eroded surface sites on the plateaus in sites 1 and 2 (between 690 and 950 m asl). The species of this group are rare. The first species prefer the sand dune and fallow lands. The last 2 species prefer the disturbed rocky wet sites. Generally they occur in high frequency in site 1 and in low frequency in site 2. 3- Seetzenia lanata A rare species found only in moist sites such as on eroded surfaces on the plateaus and the very rocky slopes of the plateaus in sites 1 and 2 (between 950 and 990 m asl). 273 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen VIII. Species of sites 1 and 3 1- Tephrosia dura A species of slightly sloping, very rocky tributary wadi beds, at altitudes between 655 and 750 m asl. Outside Arabia, this species is only found in Somalia . 2- Heliotropium rariflorum A species of flood plains and sandy wadi beds (between 655-700 m asl). 3- Panicum turgidum 4- Tephrosia nubica A species of flood plains and sandy and rocky wadi beds. The distribution of this plant is limited to the north western part of the study area where the annual rainfall is less than 50 mm at altitude between 620 and 680 m asl. Abundant in the wadi beds of site 3 (between 550 and 670m asl) and occasionally in the tributary rocky wadi beds of site 1 at 690 m asl. It is palatable species and vulnerable to overgrazing. The species of this group which are confined to the northern plateaus and represent the Saharo elements. The low cover and the absence of some of these species from site 2, can be due to human activities. Panicum turgidum and Tephrosia nubica are palatable species and vulnerable to over-grazing. Tephrosia dura is Hadhramaut-Somalia connection and was also recorded as one single species by DOVE (2001) from colluvial fan of tributary wadi in site 2. 274 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen IX. Species of sites 2 and 3 1- Anticharis glandulosa 2- Corallocarpus glomeruliflorus A rare species found on moister sites on the moderately rocky slopes of the plateaus, between 850 and 955 m asl. A rare species of the moderately steep slopes of secondary wadis and the rocky, dry, steep slopes of the plateaus (between 900 and 980 m asl). 3- Chrysopogon aucheri Restricted to the secondary rocky wadis of the plateau (between 855 and 950 m asl). 5- Jatropha spinosa Restricted to the dry rocky slopes of the plateau (between 900 and 975 m asl). 4- Ochradenus baccatus Occasionally found in wadi beds, in particular along wadi sides between 620 and 690 m asl. 6- Cornulaca amblyacantha A species mainly of the upper rocky slopes of the plateau in site 3. In site 2, it is restricted to relatively wet secondary wadis (between 900 and 955 m asl). 275 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 7- Aerva artemisioides subsp. artemisioides A very rare endemic species found only on rocky surfaces on the plateau (between 850 and 950 m asl). 8- Commiphora kua Restricted to rocky slopes on the plateaus in sites 2 and 3 (between 850 and 975 m asl). 9- Cyperus conglomeratus Mainly found in secondary wadis on the plateau in site 2 (between 855 and 950 m asl)), but also occasionally seen on the slopes facing the main wadi, at 703 m asl (site 2) and on the rocky surface of the plateau at 903 m asl (site 3). The species of this group are mixture of Saharo and Sudanian elements. Mainly found in site 2 but are penetrated the main wadi Hadhramaut and dispersed to the northern plateau of site 3. 276 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 6.3.5. The distribution of endemic, near-endemic and rare plant species in the Hadhramaut region Using the coordinates of each plant species and GIS desktop software such as DIVA-GIS and Arc View, it was possible to plot the distribution of the endemic, near-endemic and rare species of Hadhramaut region. The results indicate that the most important areas for endemics in the Hadhramaut region lie in the mountain areas above al Mukalla (see chapter 3 and Figure 6. 14a). Figure 6.14a The distribution of endemic, near-endemic and rare species in Hadhramaut region, including the 3 study sites. Figure 6.14a shows that from the reconnaissance surveys and detailed survey from the study sites, there is a concentration of species in the southern foothills and south- and southeastfacing slopes of the mountains of Hadhramaut region. 277 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen A selection of 25 rare, near-endemic or endemic species is mapped in the following figures. These maps illustrate common patterns of distribution. It is intended that these maps will provide an important tool for future plant biodiversity management. 1- Ochradenus arabicus and Ochradenus baccatus show interesting patterns of distribution. The first species is distributed from the coastal mountains and extends to altitudes of about 1200m in the north; the second species has a limited distribution from the eastern part of W. Hadhramaut towards the north. Ochradenus arabicus is a near-endemic only known from Oman, central and south Saudi Arabia, the United Arab Emirate and Yemen (Figure 6.14b). Whereas O. baccatus is a widely distributed Saharo-Sindian region. There are other endemic Ochradenus (O. spartioides) to Hadhramaut, it is found on the southern summit above al Mukala (see Figure 6.14a) 2- Limoniastrum arabicum and Aerva artemisioides are species endemic to the Hadhramaut region, the first is restricted to the coastal area of Hadhramaut, and the latter is distributed from the southern mountain slopes to the plateau surface of the study sites 1 and 3. The main factors affecting the distributions of former 4 species (1 & 2) appears to be climate and topography. Ochradenus baccatus is found in the northern part of the Hadhramaut region (the study area), which is characterised by dry hot temperatures with spring and summer rainfall, while Ochradenus arabicus is found in the southern part of the Hadhramaut region which is characterised by spring and winter rainfall and subject to frequent fogs. Outside the region Ochradenus baccatus is widely distributed and has a typical SaharoSindian distribution (Miller, 1984) being found in the deserts of eastern north Africa, 278 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Somalia, Sudan, Egypt, the Gulf countries, Iran and Pakistan (Miller, 1984), The species was seen flowering in spring and winter. In the deserts of SW Asia with winter rainfall, the flowering period is associated with the winter rains (Wolf and Burns, 2001; Zohary and Orshan, 1956). The near-endemic species Ochradenus arabicus is a spiny perennial shrub, found in Oman, Saudi Arabia and UAE, the distribution patterns of this species is associated with fogaffected areas. There are several local endemics of Ochradenus in southern Arabia including O. spartioides endemic to Hadhramaut, O. aucheri subsp. aucheri from UAE and Oman, O. harsusiticus and O. gifrii from Oman (Ghazanfar, 2004; Miller, 1984) and al Mahara, Yemen (Thulin, et al., 2001) (see Figure 6.14a). Figure 6.14b. The distribution of 4 species of Ochradenus in Arabian Peninsula. (After Miller, 1984). Limoniastrum arabicum is an endemic species to Hadhramaut which grows only on sandy plains near the coast. Aerva artemisioides subsp artemisioides has wide range distribution, but with very low frequency. For example in the study area it was only found in 2 279 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen locations. There are 2 sub species of Aerva artemisioides: subsp. artemisioides, which is endemic to Hadhramaut and subsp. batharitica, which is endemic to central Oman. 3- Cryptolepis yemenensis 4- Caralluma foulcheri-delboscii var. greenbergiana Cryptolepis yemenensis, Caralluma foulcheri-delboscii var. greenbergiana and Caralluma lavrani are endemic species to Yemen. They are common on mist-affected southern mountain summits of Yemen, the first extends farther north to W. Dawa’n (1100m). Cryptolepis yemenensis is a new endemic species to Yemen ((Venter and Verhoeven, 1999). 5- Caralluma lavrani 6- Euphorbia quaitensis is endemic to the Hadhramaut region, found only on the summits of the southern plateau. 7- Euphorbia meuleniana is endemic to the Hadhramaut region, found only on the summits of the southern plateau (between 600 and 850 m asl) 280 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 8- Echidnopsis globosa is endemic to the Hadhramaut region (Thulin & Hjertson, 1995) and Al Mahara (Kilian et al., 2002) 10- Fagonia hadramautica is endemic to the Hadhramaut region; it is found on the northern plateau (between 90 and 1100 m asl) 12- Huernia hadhramautica is endemic to the Hadhramaut region and is found on the summits above al Mukala. 9- Echidnopsis seibanica is endemic species to Hadhramaut region, It is found on the southern summits and further north. 11- Tephrosia hadramautica is endemic to the Hadhramaut region; it is found on the southern hill slopes near to the coast. 13- Cryptolepis yemenensis is endemic to the Hadhramaut, (Venter and Verhoeven, 1999). 281 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 14- Euphorbia hadramautica is a nearendemic, found on the southern summit of the plateau (between 1300 and 1500 m asl) 16- Indigofera rubromarginata is a nearendemic found in the coastal plain and on the summit above al Rayan and coastal areas of Oman. 15- Euphorbia seibanica is endemic to the Hadhramaut region and is, found on the southern summit above Al Mukala. 17- Saltia papposa is endemic to Yemen found and is found on the southern summit above al Rayan. Whidespread on the western coastal area of Yemen. 18- Cucumis canoxyi is endemic to Hadhramaut. It has a wide distribution from the southern mountains at 615 m asl to study site 2 at 950 m asl. 282 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 19- Cadaba baccarinii is a shrub or small tree found only in Yemen, Oman and Somalia (Miller & Cope, 1996; Barkhadle et al., 1994). This species is restricted to the southern rocky mountain slopes (between 400 and 600 m asl) of Hadhramaut region. There are also some species that, outside Arabia are only found in Somalia; for example Aristolochia rigida and Tephrosia dura. The first was reported from Oman by (Nadaf, et al., 2004). 19- Cadaba baccarinii and Aristolochia rigida . 20- Euphorbia balsamifera is a plant of fog-affected areas. It grows only on rolling limestone plateau summits between 1300 and 1600 m asl. There are 2 sub species: subsp adenensi which occurs in tropical NW Africa and SW Arabia and subsp balsamifera which occurs in NW Africa and Macaronesian. There are several examples of taxa which show this “Tethyan” pattern of distribution; for instance species in the genera Dracaena and Campylanthus Euphorbia balsamifera with an African/ArabianMacaronesian distribution (after Ghazanfar and Fisher, 1998) Rhus glutinosa subsp neoglutinosa is a near-endemic found at Kor Seiban on the southern sumit above al Mukala (between 1850 and 200 m asl). It was considered as endemic to Ethiopia (Kilian et al., 2002), but found in the Hadhramaut making it an outstanding refugium of a palaeo-African flora in the region. 21- 283 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 22- Dracaena serrulata (syn D. ombet) is a very rare tree, found on rocky slopes above narrow wadis on the plateau (between 750 and 1400 m asl). A slow-growing species of steep cliffs and mountain plateaus in different parts of Yemen – apparently usually restricted to fog-affected areas. D. serrulata is found in Djibouti, Egypt, Ethiopia, Saudi Arabia, Somalia, Sudan and Oman. It is considered as in endangered species in the 2003 IUCN Red List of Threatened Species (Abuzinada, 2001). This plant is becoming rare or endangered over most of its range. The populations of this species on the Red Sea Hills and Jebel Elba in Sudan and Egypt are particularly threatened. Other smaller populations have been discovered in recent years in southeastern Egypt near the Sudanese border at Gebel Shindeeb, and it may also occur on Gebel Shinodai (El Azzouni 2003). In northern Sudan it appears that the populations of D. ombet (= D. serrulata) have completely vanished (El Azzouni 2003). In Hadramaut it also is rare with scattered individuals found mainly in inaccessible areas with little sign of regeneration. It is clearly endangered. It is likely that it was once was a widespread tree on the fog-affected slopes of the Hadhramaut mountains – it is now threatened by overgrazing preventing regeneration, harvesting of the leaves for rope and possibly by a gradual drying out of the region (see notes on Dracaena cinnabari in the Ethnoflora of the Soqotra Archipelago, Miller and Miranda, 2004). Another related species of Dracaena is Dracaena draco in the Canary Islands and Dracaena draco subsp. ajgal, in the Anezi region of the Anti-Atlas Mts. in western Morocco (IUCN, 2004). So the Dracaena spp may now only exist as scattered relict population. 284 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen The distribution of plant species across the study sites is infected by the considerable variation of climatic and ecological factors (e.g. edaphic) moisture and topography) across the region. Generally, there are many factors other than climate that play as an important part in determining species distributions and the dynamics of distribution changes. Such factors include biotic interactions, evolutionary change and dispersal ability (Pearson and Dawson, 2003). Floristically, a few species are confined to particular habitats, such as those species restricted to the main wadis and not found on the plateaus, for example, Rhazya stricta, Alhagi graecorum, Panicum turgidum, Zygophyllum album, Dipterygium glaucum, Acacia ehrenbergiana, Phoenix dactylifera, Capparis spinosa, Salvadora persica, Balanites aegyptiaca, Cassia italica, Forskohlea tenacissima, Ochradenus baccatus, Tamarix aphylla, Cadaba farinosa, Withania somnifera, Ziziphus spina-christi Tephrosia dura and Tribulus arabicus. Most of these species represent Saharo-Sindian floristic elements (Zohary, 1973). Important African elements in the main wadis are: Blepharis edulis, Calotropis procera, Cenchrus ciliaris and Dichanthium insculptum. Most dominant species of the study sites, in particular the plateau complex zone, are grasses, notably Dichanthium insculptum and Stipagrostis hirtigluma. These two grasses are important palatable grasses of arid and semi-arid areas. Plant species, which germinate quickly and uncommon or short-lived species that are selectively grazed or collected by local people during very good seasons are more likely to decline, example, being Boerhavia elegans, Corallocarpus glomeruliflorus, Lasiurus scindicus, Commiphora kua, Commiphora foliacea and Tribulus arabicus, Overgrazing has affected the abundance of some species, in particula: Merremia hadramautica, Boerhavia elegans and Tribulus arabicus. The first species is abundant on wadi beds of site 3 and is found in low cover near cultivated and fallow lands in sites 1 and 2. The abundance of these species in site 3 is due to control of grazing by local people in the past. Blepharis edulis and Boerhavia elegans may represent the grazing-adapted species. Collecting some plants for medicinal or other purposes has an impact on certain species such as Boerhavia elegans, Corallocarpus glomeruliflorus, and Rhazya stricta. According to local people, Boerhavia elegans was abundant 15 years ago, the farmers are used to harvest about 3 285 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen sacks seeds of this plant. As a result of reduction of precipitation and over-grazing, specially uncontrolled grazing by local and Bedouin people, the cover of this species has dramatically decreased. The grazing was practiced mainly on the river beds and was not allowed on the slopes at specific times of the year, such as during cultivation and before harvesting of Boerhavia elegans. The Hadhramaut flora has strong floristic relationship with adjacent parts of Africa and Arabia but some species and genera have interesting distributions. For example, Dracaena serrulata is a Tertiary relict with related species of Dracaena cinnabari, in Soqotra island and with Dracaena draco in the Canary Islands; Species of Euphorbia balsamifera are also found in Spain (Canary Islands), Mauritania, Senegal, Mali, Burkina Faso and Niger. Some species show strong links with Oman and Saudi Arabia, for instance Aloe mahraensis, Atractylis kentrophylloides, Boerhavia eleganas, Boscia arabica, Calligonum crinitum subsp. arabicum, Caralluma adenensis, Caralluma arabica, Caralluma flava, Caralluma luntii, and Dracaena serrulata, Fagonia (F. lahovari and F. luntii) have remarkable distributions between southern Arabia and the horn of Africa (for more see table 3.2, chapter 3). There are few species that, outside Arabia only found in Africa (e.g. Somali, Ethiopia and north Africa) example are: Cadaba baccarinii, Aristolochia rigida and Tephrosia dura. 286 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Figure 615. Remarkable disjunctions between 2 species of Fagonia (F lahovari and F. luntii) between southern Arabia and the horn of Africa (After Beier, 2005). This west-east migration took place during the Tertiary period, when Arabia and Africa were connected by land bridge that allowed the migration of African and Asian taxa along the coast of the Red sea. 6.8. Summary A total of 15 vegetation associations were distinguished in the study sites. The three principal vegetation associations, Seetzenia lanata - Dichanthium insculptum of the plateau (association 11), Tephrosia nubica - Acacia campoptila - Merremia hadramautica of the main wadis (association 7) and Maerua crassifolia - Jatropha spinosa – Stipagrostis hirtigluma of the plateau slopes (association 13) are the dominant ones and cover 11%, 10% and 9% of the landscapes respectively. The vegetation on the plateau surfaces and on the secondary wadis is more uniform, while the vegetation of the main wadis, the mountain slopes and the slopes of the plateaus is not homogeneous. This is considered mainly to be the result of the impact of human activities in the past, such as grazing and woodcutting or other natural factors such as floods. Because of complexity of landforms of the plateau areas in the study sites, it was not possible to map some of the vegetation associations separately; therefore, these associations were mapped together and presented in the vegetation maps as groups of associations. Association 11 covers most of the plateau areas in the all study sites; this association is restricted to exposed plateau surfaces and adjacent slopes. 287 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Phytogeography attempts to divide the global into natural floristic units (Miller and Cope, 1996). In terms of the phytogeographical classification of White (1983), the ecosystem of the main wadis and the northern part of the plateau is regarded as a part of the Saharan regional transition zone. According to Zohary (1973), the Hadhramaut region falls within Saharo-Arabian (covering the northern part) and Nubo-Sindian province of Sudanian-region (covering the southern part). Most of the Saharo-Arabian elements are found in the main wadis, while most of the Sudanian-region elements are found on the plateau. The northern limit of the Sudanian-region in the Hadhramaut region can be drawn with latitude ranging from 15° 49’ to 15° 50’ on the plateau of site 2 (see Figure 7.1b). Most of the plant species of the study sites are desert-adapted species which botanists have classified as Saharo-Sindian, or Saharo-Arabian (Zohary, 1973). Some of these species are of African origin. These species have adapted to the current harsh climate. As a result, many of these plants cannot now live outside the arid and semi-arid landscapes that characterise the study area. A total of 81 plant species were plotted on maps, 59 over the study sites and 25 over Hadhramaut region, in which 17 plant species were endemic and 7 near-endemic. It can be observed from the study that the distribution patterns show certain common features, some species are confined to the main wadis while others grow well on the rocky slopes of the plateau. Generally the species groups with the similar ecological features grow within similar ranges of altitude. For example, Alhagi graecorum, Cressa cretica and Prosopis farcta grow well in the cultivated and wet sites at low altitude areas of the main wadis. Panicum turgidum, Zygophyllum album grow, Heliotropium rariflorum, Dipterygium glaucum and Citrullus colocynthis prefer and grow well at low altitude in sandy wadi beds, while Jatropha spinosa, Maerua crassifolia Commiphora kua and Hochstetteri schimperi grow well at high altitudes on the steep rocky slopes in the plateau areas. Grazing can be also considered a disturbance at the level of the individual but it may or may not be a disturbance at the level of the population (Milchunas et al., 1989). Grazing has contributed in reducing the distribution of some plant species and examples includes: 288 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Boerhavia elegans, Panicum turgidum, Lasiurus scindicus, Stipagrostis hirtigluma, Aristida triticoides, Maerua crassifolia and others. The results of this chapter provide a better understanding of the distribution pattern of some plant species for Hadhramaut region. The findings also emphasise that most of the previous plant collections were obtained from the areas that were accessible or easy to reach, leaving vast areas in this remarkable region without any exploration. Therefore, there are still many places to be visited in order to obtain a comprehensive idea about the population of the plant species. This is particularly true of endemic, near-endemic and rare species, so that they can be cited properly in the red list data of the Yemen. With the help of data on distribution patterns of the species, the causes for decline in number and rarity may be identified. Identification of interesting and rich plant biological zones also helps in the move towards formulating conservation strategy. It is hoped that this work will contribute to the planning of priorities for the conservation of biodiversity in Hadhramaut region, particularly in the selection of areas suitable for the establishment of new protected areas as mentioned previously in chapter 3. The next chapter discusses these issues further and explores the main findings of the study. 289 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Chapter 7. Discussion and conclusion 7.1. Introduction Studies on the vegetation of Hadhramaut go back to the last two decades of the nineteenth century but the present work is the first systematic survey of the vegetation of the region. As indicated before, previous studies concentrated on compiling floristic checklists or on general descriptions of the vegetation communities found in selected habitats without following any consistent scientific methodology. The study area is still the least known part of the whole of Yemen. It is intended that the present research should form the basis for any future management plants for the region. Human activities and climate change have altered the natural environment over time; in particular they have tipped the ecological balance resulting in a fragmentation of habitats. Above all, human activities have contributed indirectly to the extinction of species by altering habitats in a way that favour certain species. For example, the increases of agriculture and over-grazing have allowed some species, such as Prosopis juliflora, to spread rapidly with a resultant detrimental effect on indigenous species. This chapter presents the main findings of the study and explores some of the resulting suggestions. It discusses the distribution and dynamics of the plants of the study area, and attempts to evaluate the role of climate and human activities on the vegetation by: 1- evaluating the types of vegetation, the structure of plant communities and their distribution, the plant species composition, plant biodiversity, and areas with the greatest plant diversity, 2- determining the physical and anthropogenic factors affecting contemporary plant distribution and composition, 3- determining the changes in plant communities over the past few decades and the processes affecting vegetation change, 4- identifying the potential of the study to contribute to conservation and management issues. 290 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 7.2. Floristic and phytogeography 7.2.1. Floristic The total number of plant species collected and recorded from the literature of the Hadhramaut region is 469 (see Appendix 3), of which about 23% (107 taxa) are either endemic or near-endemic (with distribution centred in Hadhramaut but also occurring just outside in Oman, Saudi Arabia and UAE). In total 134 species belonging to 41 families (about 30% of the known flora of Hadhramaut region) were recorded in the study area (Table 4.1) of these, 5 are endemic to Hadhramaut region, 7 species are endemic to Yemen and five are endemic to the Arabian Peninsula. A total of 117 plant species were collected in the sample plots (of which 4 have not yet been identified): 74 species in site 1; 83 species in site 2; and 61 species in site 3, with 39 plant species recorded in all 3 study sites. At each site, the detailed survey of the representative transects, is estimated as follows 0.46 ha (46 x 100 m² plots), 0.53 ha (53 x 100 m² plots) and 0.34 ha (34 x 100 m² plots). The following endemic and nearendemic taxa for the Hadhramaut were found in the study sites: Aerva artemisioides subsp artemisioides, Iphiona anthemidifolia, Cucumis canoxyi, Fagonia hadramautica, Merremia hadramautica and Acacia campoptila.  The largest families in terms of species in the study sites are:  Poaceae (Gramineae) with 13 species  Zygophyllaceae with 11 species  Mimosaceae with 8 species  Fabaceae with 12 species  Capparaceae with 9 species Asteraceae (Compositae) with 7 species 17 families were recorded containing only one species (Table 4.1). 291 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 7.2.2. Phytogeography Phytogeography attempts to divide the globe into natural floristic units (Miller and Cope, 1996). A number of phytogeographical classifications have been proposed for the Middle East. Below I have set out a summary of the most important of these (Table 7.1). Authority Kingdom The region Zohary Takhtajan White Arabian regional Holarctic province Egyptian-Arabian subzone of the Holarctic Northern part of of the Holarctic- Province of the Holarctic-Sindian origin the Hadhramaut Sindian region Holarctic region regional zone Nubo-Sindian Paleotropical Southern part of province of the the Hadhramaut Paleotropicalregion. origin South Arabian Province of the The Somalia-Masai Sudano-Zambesian regional centre of Region. endemism Table 7.1. Phytogeographical classification for the Middle East. 1. Zohary (1973). According to Zohary, the northern part of the Hadhramaut region is of Holarctic origin falling within the Holarctic province of the Holarctic-Sindian region whilst the southern part of the region is of Paleotropical origin falling within the Nubo-Sindian province of the Paleotropical-region. 2. Takhtajan, (1986). Taktajhan splits the Middle East into two floristic Kindoms: the Holarctic in the north and the Paleotropical in the south. The boundary between these floristic Kingdoms falls somewhere across the southern part of the Arabian Peninsula. The map on the end page of his book suggests that in fact this boundary runs across the middle of Hadhramaut. According to him, the northern part of Hadhramaut falls within the Egyptian-Arabian Province of the Holarctic region whilst the southern part falls within the South Arabian Province of the Sudano-Zambesian Region. 292 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 3. White and Léonard (1991). White (1983) proposed a new phytogeographical system for Africa based on “regional centres of endemism” separated by “regional transition zones”. This system was later extended to cover SW Asia (White and Léonard, 1991). Using this system Arabia is subdivided into three main phytogeographical areas: the Holarctic-Sindian regional zone; the Somalia-Masai regional centre of endemism and the Afromontane archipelago-like regional centre of endemism. The Holarctic-Sindian regional zone is further divided into the Arabian regional subzone and the Nubo-Sindian local centre of endemism. It can thus be seen that according to White and Léonard (1991), the northern part of Hadhramaut Region falls into the Arabian regional subzone of the Holarctic-Sindian regional zone and the southern part into the Somalia-Masai regional centre of endemism. These three systems are all fundamentally similar (see Table 7.1). They each place the northern part of the region in the Holarctic floristic Kingdom and the southern part in the Paleotropical floristic Kingdom. They differ only in the terms used to describe them. In the literature there is frequently confusion and mixing of these terms. To avoid this and for the sake of simplicity in the following discussion I have used the term “Holarctic” in the sense of the Holarctic-Sindian of Zohary, the Holarctic of Takhtajan and White and the term “Paleotropical” in the sense of the Nubo-Sindian or Paleotropical of Zohary, the SudanoZambesian of Takhtajan, the Somalia-Masai regional centre of endemism of White and Léonard or the more general term Paleotropical or tropical Paleotropical widely used in the literature. Figure 7.1a shows the general limits of the main phytogeographical regions of Africa and south west Asian. 293 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Figure 7.1a. General limits of the main phytogeographical regions of Africa and south west Asian; IT, Irano-Turanian regional centre of endemism; MS, Mediterranean-Sahara regional transition zone; SS, Saharo-Sindian regional zone further divided into the Arabian regional sub-zone (SS1), the Nubo-Sindian local centre of endemism (SS2); SM, Somalia-Masai regional centre of endemism; shaded black, Afromontane archipelago-like regional centre of endemism. After White and Leonard (1991) and Miller and Morris (2004). The placing of species into different phytogeographical elements can be problematic, however, the following sources were found to be useful: Zohary (1973) Zohary and FeinbrunDothams (1966-1986), White (1983) and Takhtajan, (1986). From these it was possible to assign floristic elements to a large number of taxa within the region. For instance, Holarctic elements include: the endemic genus Saltia papposa, Schouwia purpurea, and species of Echiochilon, Fagonia, Zygophyllum and Aristida. Paleotropical elements tend to be more numerous in the Hadhramaut region. According to Zohary (1973) they are largely confined to families and genera that are absent elsewhere in Asia or large parts of the Middle East. These include: Acacia hamulosa, A. laeta, A. mellifera, A. oerfota, Aloe spp., Balanites aegyptiaca, Cadaba baccarinii, C. farinosa, C. heterotricha, C. rotundifolia, Caesalpinia erianthera, Caralluma spp., Celtis africana, Cassia senna, Commiphora foliacea, Commiphora gileadensis, Commiphora habessinica, Commiphora kua, C. playfairii, Crotalaria aegyptiaca, Dodonaea viscosa, succulent Euphorbia spp., Ficus 294 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen cordata subsp. salicifolia, Grewia erythraea, Forsskaolea tenacissima, F. viridis, Jatropha pelargoniifolia, J. spinosa, Maerua crassifolia, Moringa peregrina, Olea europaea subsp. cuspidata, Psiadia punctulata, Sarcostemma vanlessenii, S. viminale, Tephrosia apollinea, T. nubica Apart from the species which can be identified as belonging to the two main elements described above, two species of Irano-Turanian origin (Zohary and Feinbrun-Dothan, 19661986) were also found. These are Alhagi graecorum and Prosopis farcta. Both species were found on or near cultivated and fallow lands in site 1 and both are probably introductions which have adapted to the arid conditions, with annual rainfall (that characterises the site) less than 40 mm. Identifying the boundary between the northern, holarctic and southern paleotropical flora is problematic because the vegetation consist of a mixture of a paleotropical and holarctic elements. For instance, holarctic elements are common in the ground flora of the main wadis, whilst paleotropical elements are common on the plateau. However, the dominant trees in the main wadis are typically species of Acacia, a genus of paleaotropical origin. Based on an analysis of the distributions of some holarctic and palaeotropical elements, I have here suggested a boundary between the two phytogeographic regions (see fig 7.1b). The northern limit of the paleotropical region in the Hadhramaut region runs between latitude 15° 22’ in Wadi Daua’n west to 15° 35’ in Wadi Sah east. However, considerable numbers of paleotropical elements are found further north between 15° 49’ to 15° 50’ on the plateau of site 2 (see Figure 7.1b). 295 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Figure 7.1b. The Phytogeography of the Hadhramaut region. The proposed (red line) boundary is based on the distribution patterns of some Holarctic and Paleotropical elements, such as palm cultivated fields, Rhazya stricta, Ochradenus baccatus, and O. arabicus.The boundary runs between latitude 15° 22’ in Wadi Daua’n west to 15° 35’ in Wadi Sah east. The occurrence of paleotropical elements within the main wadis, which are largely dominated by holarctic elements, may be due to the large climatic fluctuation that took place in the Sahara over the entire Quaternary period (Takhtajan, 1986), which allowed paleotropical elements (e.g. Acacias) to migrate along the wadis during periods of higher rainfall (Mandaville, 1990). The wadi systems also provided channels for the migration of paleotropical vegetation into the future Arabian Peninsula, even into the present holarctic region, during late Pliocene or early Pleistocene times (Mandaville, 1984). After this the African-Arabian Shiele was divided by the Red sea and overland migration was no longer possible (Miller and Morris, 1988; Mandaville, 1990). It is now widely accepted that the Arabian flora underwent its main floristic colonisation events during the Tertiary. This was the main period of invasion, especially for the extra-tropical Holarctic element that developed under the increasing aridity of the mid to late Miocene. According to Zohary (1973) the 296 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen northern border of the paleotropical element is marked in Arabia by open thorn woodland dominated by various Acacia spp. and with a strong decline in the abundance and diversity of annual plants. Therefore, is seems likely that paleotropical species such as Acacia campoptila, A. hamulosa, A. ehrenbergiana, Moringa peregrina, Balanites aegyptiaca, Ziziphus leucodermis, Salvadora persica, Ziziphus spina-christi, Tamarix aphylla, Tephrosia apollinea, Tephrosia nubica in the main wadis are relicts of a “mesic Paleotropical vegetation” that once occupied the area. Due to the change of the climate from wet to the current arid and semi-arid, most of these species either survived in more mesic niches or adapted to the current hostile climate before Arabia broke away from Africa some nine million years ago. For example, the dominant species of the main wadi beds, Tephrosia apollinea. subsp. longistipulata, Boerhavia elegans and Acacia campoptila have strong (vicariant) links with closely related Paleotropical species. The above argument is supported by fossil pollen of Miocene horizons which indicate that a Paleotropical flora existed in Arabia during the Mid Miocene (Mandaville, 1984) when conditions were much more mesic than today. It is largely due to the late separation of the Arabian Peninsula from the Paleotropical continent in the early Miocene, that the present flora of the southern part of Arabia exhibits close relationship with the Somalia-Masai regional zone. Most Paleotropical taxa reflect Tertiary migration routes dating from before the formation of the Red Sea (Mandaville, 1990). Mesic-Paleotropical relicts in the Hadhramaut region are: Gnidia somalensis the common species of the southern part of the Hadhramaut which has a link with the Gnidia glauca of afromontane taxa of Ethiopia (Takhtajan, 1986), other example of refugium of palaeo-African in the isolated areas of southern summits are Rhus glutinosa subsp. neoglutinosa and Cadaba baccarinii (see Table 3.2). There are numerous examples of taxa of paleotropical origin that have their closest relatives in Africa and a speciation centre in southern Arabia and which also extend eastwards across the Arabian Gulf, for example, genera such as Anogeissus (A. leiocarpa in Africa, A. dhofarica and A. bentii in Arabia), Dracaena (D. draco in Canary island, D. draco subsp. ajgal in Morocco D. cinnabaria in Soqotra and D. serrulata in Yemen, Oman and Africa), Ephedra (E. major and 297 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen E. intermedia in Iran, E. alata and E. aphylla in the northwest Africa desert, E. foliate in Arabia, Iran and north east Africa and E. milleri in Arabia), Fagonia (F. acerosa in Iran, F. burguieri in North Africa, Arabia and East Asia, F. mollis in Egypt, Palestine and Israel and Arabia, F. arabica and F. scabra in North Africa, F. glutinosa in Arabia, North Africa and Palestine, F. luntii in Arabia and Horn of Africa, F. lahovarii in Yemen and Horn of Africa and F. Hadramautica in Yemen (Beier, 2005; Bell et al., 2004), F. paulayana in North and East Africa, Pakistan and Arabia. Farsetia (F. longisiliqua and F. stylosa in Africa and Arabia , F. Linearis, and F. dhofarica in Arabia), Gymnocarpos (G. decandrus in Canary Islands, North Africa, and in Israel, Jordan, Syria, Saudi Arabia, Oman, Iran, Afghanistan and Pakistan (Brandes, 2004; Hassan, 2002), G. parvibravtus in Africa, G. rotundifolius in Arabia, Ochradenus (O. randonioides in East Africa, O. arabicus, O. harsusiticus and O. spartioides, O. gifrii in Arabia (Miller, 1984; Thulin, 1994), O. aucheri subsp. aucheri in Arabia (Miller, 1984; ICARDA, 1998) and subsp. ochradenii in Iran, subsp. rechingeri in and Iran Pakistan (Perveen and Qaiser, 2001) and O. baccatus in North and East Africa (Miller, 1984; Abd El-Fattah and Dahmash, 2002), Arabia and Pakistan), Schweinfurthia (S. pedicellata in Africa, S. pterosperma in Africa, Arabia and Asia, S. papillionaceae in Arabia and Asia, S. latifolia, S. pedicellata and S. spinosa in Arabia. Vernonia (V. arabica in Arabia and Africa, V. areysiana in Yemen, V. spathulata in Yemen and Somalia (Wood, 1997). Endemism and Speciation There are 105 endemic and near-endemic plant taxa recorded from the Hadhramaut region (Table 3.2), of these 67 are endemic to Yemen (40 are confined to the Hadhramaut region). The most endemic-rich family in the Hadhramaut region is Apocynaceae with 18 endemic taxa, and the most endemic-rich genera are Caralluma with 10 endemic taxa and Aloe with 9 endemic taxa. The relatively large number of endemics is largely due to the topography and climate of the region; many endemics are found on isolated, relatively mesic mountains, often with markedly different climatic regimes. The mountains of southern Arabia behave in many ways like an island archipelago - the sea in this case represented by hyper-arid deserts. 298 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Adaptive radiation and speciation by geographical separation has occurred in many genera. For instance, each mountain range along the southern and western escarpments of the Arabian Peninsula has its own endemic species (Ghazanfar and Fisher, 1998). The southern mountainous region of the Hadhramaut follows this pattern. It is in effect an island, isolated from other mountains by large tracts of desert. Furthermore, the presence of frequent low cloud, during the summer or SW monsoon, results in relatively good conditions for plant growth. This isolation and presence of mesic conditions have given rise to refugia which have become local centres of endemism (Plate 7.1). The flora of the southern region of Hadhramaut shows every sign of having been linked at some point with the African flora. These endemics are mainly of paleotropical origin – their presence explained by their migration into the region thousands years ago when the climate was very different from that of today. The continued survival of these plant species can be explained by the winds of the SW monsoon which, during the summer months, blow dense fogs inland. These fogs blanket the seaward-facing escarpments and frequently spill-over onto the summit plateau and bringing important precipitation for the ecosystem. Similar fog-affected seaward-facing mountain slopes occur in the mountains above Hauf in the Al Mahra Governorate of Yemen and in Dhofar in the southern region of Oman. Here, the slopes support dense deciduous woodland dominated by the regional endemic tree Anogeissus dhofarica, together with Acacia senegal and Commiphora spp., (Miller & Morris, 1988). As the influence of fog decreases further north, grassland is replaced by low succulent shrubland dominated by Euphorbia balsamifera. Beyond this, the vegetation becomes sparser and finally gives way to open desert (Miller, 1994). 299 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Plate 7.1. Dense fog on the southern mountains of Hadhramaut region (above the arid coastal plain).This fog is influenced by the southwest monsoon. This dense fog supports an important moisture contribution and as a result high diversity of flora and fauna and brings the moisture necessary to sustain the wildlife and plant species found here. This habitat has very rich endemic near-endemic and rare plants. The high density of plant species, dominated by Dodonaea viscosa and Dracaena serrulata at about 1350 m inland, is probably reflecting the maximum inland extent of fog moisture (see chapter 3) . The fog-affected escarpments of Dhofar have the highest number of endemic species and some of the most species-rich habitats in Oman. About 900 vascular plants have been recorded from the fog oasis of Oman and Yemen, of which 60 are endemic, including two endemic genera, Cibirhiza and Dhofaria (Miller & Morris, 1988). The volcanic mountain of Jebel al Areys, which lies west of Hadhramaut some 150 km east of Aden on the southern coast of Yemen, is another good example of a “fog oasis”. Its upper seaward-facing slopes are covered in dense woodland and the summit by a succulent shrubland dominated by Euphorbia balsamifera with Dracaena serrulata. These monsoon-affected mountains of the southern Arabian Peninsula harbour relict vegetation elements of paleotropical origin with some plant species such as the near-endemic Justicia areysiana showing a disjunct geographical distribution pattern with populations in the south Yemeni fog oases in Hadhramaut and in the Hawf /Dhofar region in east Yemen and 300 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen western Oman. Past oscillations between humid and arid periods connected with glacial and interglacial episodes in the Pleistocene and Holocene are considered responsible for observed patterns of genetic variation between populations in the different refugia (Meister et al., 2005). These woodland refugia of the southern mountains of the Arabian Peninsula (including the Hadhramaut) show strong affinities to the Somalia-Masai phytochorion of Africa. (BIOLO, 2004). Adaptations Most of the plant species of the study sites are desert-adapted species belonging to the Holarctic (Zohary, 1973). These plants can be considered to be xerophytes, that is they have morphological adaptations that help them to survive the arid climate of the region (Batanouny, 2000; Western, 1988). The rainfall in the study area mainly falls in summer when the temperature is very high; the rate of evaporation exceeds rainfall throughout the year, in such environmental conditions only the vegetation that is adapted to a dry environment can survive. Generally, as in other desert areas in the world, the important limiting factor in the study area is the quantity of moisture available to the plants. In order to survive in the extreme desert, a plant must, on one hand, utilize the little rainwater available to it, and on the other hand, survive the very long dry periods that characterise the region. The lack of rainwater is not the only problem in the study area, but also the rainfall does not arrive regularly. There are many examples of plants from the study area which exhibit xerophytic adoptions, for instance, plants with succulent leaves such as Aloe spp. and Zygophyllum spp. or plants with succulent stems such as Adenium obesum, the cacti form Euphorbia spp. and Caralluma spp. Other plants have very small or scale-like leaves which reduce the surface area of the plant and consequently reduce moisture loss, examples being Ochradenus baccatus in main wadi and Periploca visciformis in the plateau. Some grass species have extremely large and complex root systems that enable them to collect water over a wide area. Examples are Panicum turgidum and Dipterygium glaucum. Typically trees have deep root systems which can tap the ground water, for examples, certain trees which are abundant in the wadi beds including Acacia campoptila, Prosopis juliflora, Tamarix sp, Ziziphus spina-christi and the date palm Phoenix dactylifera. The ability to grow deep roots, has allowed some species 301 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen such as Prosopis juliflora to migrate and survive in the region and become the dominant species in many parts of main wadis, especially in neglected areas. Prosopis juliflora has both an extensive lateral root system that collects surface water after rain and long taproots that penetrate to depth of 24 m and tap the ground water supplies (FAO, 1993). Some species such as Panicum turgidum and Dipterygium glaucum have shallow but wide-reaching root systems for greater water-gathering ability (Western, 1988). Other species such as Salsola imbricata, and Zygophyllum coccineum, Z. album, Tamarix arabica and T. aphylla are much more salt-tolerant and grow well on some saline and alkaline sites of wadis (associations 2 and 4). These species were found evergreen and fleshy even during the dry season (during the field-work). Vegetation of the Study sites Some stands of typically paleotropical vegetation are found in sheltered slopes on the plateau of site 2. Typical paleotropical elements in this vegetation include, Maerua crassifolia, Cadaba heterotricha, Acacia mellifera, Tarenna graveolens, Grewia erythraea Commiphora kua, Commiphora foliacea, and Aristida triticoides. This vegetation possibly represents a relict of vegetation type that was previously widespread in this region. The present vegetation in the surrounding area is dominated by Jatropha spinosa with Hochstetteria schimperi, Zygophyllum decumbens, Stipagrostis hirtigluma, Dichanthium insculptum and Indigofera spinosa. This finding suggests that formerly the vegetation of the plateau was more favourable to tree life, which was then replaced gradually by vegetation dominated by herbaceous plants. In the main wadis a few stands of typically paleotropical vegetation with Salvadora persica, and Tamarix aphylla also gives an impression of what the previous climax vegetation might have been like when the climate was wet and before it was replaced by a vegetation association dominated by Acacia campoptila with Fagonia indica, Tephrosia apollinea, Cymbopogon schoenanthus, Rhazya stricta, Ochradenus baccatus and others that can survive the present arid climate. 302 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Paleotropical elements such as Anogeissus bentii, Maerua crassifolia, Ziziphus leucodermis, Acacia ehrenbergiana, A. hamulosa and Moringa peregrina were probably dominant in the main wadis in the past, but as the climate became drier in the region or as a result of human impact, their populations declined to their present levels. So now they are very rare and generally restricted to more favourable places such as areas of deep alluvial soil in wadis and around cultivated fields. These species are an important sources of browse for camels and fire wood in Arabia and so over-utilisation may also account of their present, limited, distribution (Zaroug, 1984). Ziziphus leucodermis, Acacia hamulosa, and A. ehrenbergiana are still abundant species in some wadis of Hadhramaut. According to Kenneth et al., (2001) the first two species have been present since 5239 yr B.P. but the third was not present until 2159 yr B.P. He suggests that the late Holocene appearance of Acacia ehrenbergiana, which is widespread throughout Yemen today, could indicate a return to moister climates following an extremely arid period in the late-middle Holocene. Acacia ehrenbergiana is now restricted to deep alluvial soils in the wadis and near or on old cultivated fields. Recent plant migration is a very real phenomenon with evidence both from prehistoric times and from present observations (Pitelka, 1997). Human activities or changes in climatic conditions may force plant species to adapt or migrate and occupy new territory via dispersal and reproduction. In the study area, humans have contributed to plant migration through the introduction of exotic plants such as Prosopis juliflora and Cuscuta campestris. Generally, there are many factors other than climate that play an important part in determining species distributions and the dynamics of distribution changes. Such factors include biotic interactions, evolutionary change and dispersal ability (Pearson and Dawson, 2003). In the coming sections, and based on the main findings in chapter 3 to 6, the main abiotic and biotic factors that determine vegetation distribution in Hadhramaut region will be discussed. 303 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Vegetation associations The vegetation of the main Wadis of Hadhramaut (associations 1 to 7) comprises a mixture of two phtogeographical elements: that is holarctic species mixed with a few paleotropical plants. Zohary (1973) referred to this type of vegetation as pseudo-savanna. He considered it to be confined to depressions and wadis in eastern Egypt, Sinai, S. Palestine and Israel. It differs from the vegetation found in these areas because of the presence of Acacia campoptila and Boerhavia elegans but the occurrence of Ochradenus baccatus, Panicum turgidum, Zygophyllum coccineum, Citrullus colocynthis, Blepharis edulis, Cleome droserifolia, Tephrosia apollinea, Senna italica and Fagonia indica in both the Hadhramaut and these others regions is significant. The Acacia campoptila wadi communities with Panicum turgidum, Aerva javanica, Indigofera spinosa, Fagonia indica, Ochradenus baccatus, Dipterygium glaucum, Calotropis procera and Rhazya stricta is widespread over the wadi beds of arid areas Marib and Rada’ in Yemen (Al Khulaidi, 1989; Westinga and Thalen, 1980, Al Hubaishi and Muller-Hohenstein, 1979). Species of Acacia. (e.g. A. tortilis and A. hamulosa) form savanna-like communities in association with Panicum turgidum, Leptadenia pyrotechnica, Aerva javanica, Calotropis procera and Ziziphus spina-christi. Similar communities are widespread throughout the tropical Sahara region and in the wadis of Sudan, Egypt, Palestine and Arabia (Zohary 1973, White, 1983). Therefore, it can be seen that the vegetation of the main wadis is dominated by holarctic elements with a few paleotropical elements and is characterised by being adapted to arid and semi-arid conditions. Woodland in the main wadis is dominated by Acacia campoptila. This woodland is found over large areas but is thin and sparsely distributed. In areas where water is relatively abundant, such as in depressions and near cultivated fields, the vegetation is rich with species such as: Acacia ehrenbergiana, Acacia hamulosa, Barleria aff. bispinosa, Boerhavia elegans, Dichanthium insculptum, Kohautia retrorsa, and Zygophyllum simplex. Association 2 (Citrullus colocynthis - Zygophyllum album) reflects the transitional zone between arid and hyper-arid climates and is very similar to the associations that are common further west on the sand dunes of the Ramalat Al Saba’tein desert area (see chapter 3) and the Aerva javanica, Calotropis procera and Dipterygium glaucum associations that are found on 304 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen the wadi borders, sand dunes and sandy plains throughout the coastal areas and deserts of Yemen (Wood, 1997). Some species in this habitat (e.g. Panicum turgidum, Dipterygium glaucum and Zygophyllum album) have very long roots which spread over the surface as well as deep roots; they can thus utilise surface moisture and as well as water that has infiltrated to deep layer (Plate 7.2). This is typical of the root systems of grasses on sandy soils which can extend to depths of 1.4 m or more and are likely to be adaptation to capture more soil water from the frequent small rainfall events (Gibbens and Lenz, 2001). Additionally, the sandy soils of this habitat may offer better water supplies (Monger, 2002). This vegetation association is currently confined to the western part of the study area (site 1) and represents the hyper-arid climate with sand dune landform type, but due to the frequent floods, this type of vegetation association is expected to spread further towards the eastern part of site. Plate 7.2. Methods of tackling water shortage. Some species have large root systems to improve the intake of water. The illustration is of the long root of Dipterygium glaucum, which can spread over the surface as well as into the soil utilizing the available surface moisture as well as water that has infiltrated deeper layers. The length of the root shown is about 100 cm representing the upper thick part of the root system. 305 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Association 11 (Seetzenia lanata - Dichanthium insculptum) is predominantly grassland, dominated by Dichanthium insculptum and Stipagrostis hirtigluma. It is found mainly on the plateaus and surrounding slopes. This association is characteristic of the vegetation that covers the plateaus and surrounding degraded slopes. Association 12 (Chrysopogon aucheri- Stipagrostis hirtigluma) of the secondary wadis of plateaus and association 7 of the main wadi beds of site 3 are the most diverse in the study sites. The number of species in these associations ranges between 34 and 44 and the individuals range between 468 and 537. Associations 2, 10, 14 and 15 are characterised by low number of species and individuals. Associations 1 and 3 have the highest vegetation cover percentages, the former is dominated by trees and shrubs, namely Phoenix dactylifera, Prosopis farcta and Alhagi graecorum, while the latter is dominated by dwarf shrub and herbaceous cover namely Merremia hadramautica, Crotalaria persica, and Heliotropium ramosissimum. Association 13 (Maerua crassifolia - Jatropha spinosa - Stipagrostis hirtigluma) is confined to the slopes of the plateau in site 2. It consists predominantly of sparse shrubland, dominated by Jatropha spinosa with Stipagrostis hirtigluma and Dichanthium insculptum also frequently present. This association forms a climax vegetation type. When this association is disturbed and degraded a secondary succession occurs (Kent and Coker 1992). This degraded succession is represented by association 15, which is characterised by a species-poor association (up to 8 species) of herbaceous plants. The climax vegetation type of this area was probably woodland dominated in the past by Cadaba heterotricha, Tarenna graveolens, Maerua crassifolia, Aristida triticoides, Commiphora kua and Grewia erythraea. Compared to other ecological zones, the vegetation cover of the wadis is generally rich with a relatively a extensive cover of plant species. All the rainfall from the slopes and drainage lines that cut the plateau runs into these wadis; they thus hold enough water in the rooting zone to support a variety of plant species, as well as being good for date cultivation. 306 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 7.3. The effect of biotic and abiotic factors on plant species Human activities and climate change have altered the natural environment over time; they directly cause destruction of habitats and extinction of species resulting in fragmentation of habitats. Human activities also contribute indirectly to the extinction of species by destroying the balance of ecosystems in favour of species that are adaptable to changing conditions. For example, increasing human activity in Hadhramaut has allowed certain species such as Prosopis juliflora to invade natural habitats. Different plant species have different environmental requirements; a better understanding of the demands of different species is important in determining what measurable variations affect their distributions. 7.3. 1. Climate Vegetation depends on a combination of variables such as rainfall, temperature, evaporation, radiation and relative humidity (Kutiel et al., 2000). Climatic variables, such as changes in average air temperature and precipitation are a significant aspect of global environmental evolution (Anderson, 1997). Several studies have shown that there is a lack of clear correlation between rainfall and species diversity (Kutiel et al., 2000). Although there are marked relationships between mean annual precipitation (MAP) and vegetation growth (for example woody cover in African savannas, Sankaran et al., 2005). Other studies have shown that temperature and precipitation are the main variables determining the distribution of plants (Woodward 1987, Huntley et al., 1995). Both of these variables are being changed by increasing levels of greenhouse gases emitted mainly through the anthropogenic emission of fossil fuels (Todd, 1995). Over the next 100 years, these changes in both the macro and microclimate will alter the distributions of many plant species (Overpeck and Bartlein, 1989). Therefore, to understand the importance of these changes, the influence that environmental factors have on the floristic composition of plant associations need to be examined (Danin et al., 1975). 307 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen As in other arid regions, the study area consists of a considerable extent of barren lands with a scarcity of water for plant growth (Isani, 1999; Cook & Warren, 1973). Therefore, available moisture is the main factor influencing plant species in the study area. High temperatures in summer (mean monthly maximum of 34-36˚C in coastal areas and of 4042˚C in the northern part of the region), result in evapotranspiration of 3-5 mm/day in coastal areas and 3-6.5mm/day in the mountainous north of the region. The highest levels of evapotranspiration at 8 mm/day occur in the deserts in northern part of the region. The yearly distribution of rainfall (nearly 80% of the rain falls from January to June) reduces the moisture available in the soil. Yet, landscape units have micro habitats or niches within them, such as secondary wadis on the plateaus and cracks in cliffs which capture run-off and increase the moisture available for plant growth. The plateau and the large wadis which dissect them are the main topographic features of the Hadhramaut region. The Wadi Hadhramaut acts as a major ecological barrier dividing both the climate and the plant species of the northern plateau from those of southern plateau. The best example of this division, which is essentially climatic, is shown by species which are confined to the northern plateaus (corresponding to the 50 mm isohyet or less), such as Glossonema varians, Hermannia paniculata, Crotalaria saltiana, Forskohlea tenacissima and Moltkiopsis ciliata and those that are confined to the southern plateau such as Arnebia hispidissima, Abutilon bidentatum, Cadaba heterotricha, Chascanum marrubifolium, Euphorbia rubriseminalis, Grewia erythraea, Helichrysum pumilum, Hochstetteria schimperi, Leptadenia arborea, Pluchea dioscoroides, Portulaca oleracea, Seddera latifolia, Tamarix arabica, Tarenna graveolens, Trichodesma calathiforme, and Zygophyllum decumbens. The 40 mm isohyet in Wadi Hadhramaut also corresponds to the eastern limits of species such as Zygophyllum album and Alhagi graecorum, Prosopis farcta and Heliotropium rariflorum (Association 2). In the study area, with no obvious macro-climatic variations, the influence of landscape variation is of major importance. Each landscape feature represents a distinct ecological zone with its own plant species. Some species were found in almost all ecological zones, whilst others had narrow ecological ranges and were confined to specific or favoured sites. The dominant species of the 3 study sites, in particular on the plateau were grasses, notably 308 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Dichanthium insculptum and Stipagrostis hirtigluma. These two grasses are important palatable grasses of arid and semi-arid areas of Australia, East Africa, the Arabian Peninsula and India. The first is an important grass in the arid areas of NE Australia (Gupta et al., 1986), and the latter is found in the Somalia-Masai Region (White, 1983), and in the dry savanna areas of Namibia (Becker and Getzin, 2000). Some species are only found in sites with high moisture content (e.g. adjacent to fields or fallow lands) these include Senna incana, Cressa cretica and Alhagi graecorum, whilst others are found on rocky wet plateaus such as the endemic Cucumis canoxyi. Other species that grow well in relatively wet sites are Acacia oerfota, A. hamulosa, A. ehrenbergiana, Pulicaria undulata, Crotalaria saltiana and Farsetia linearis. The vegetation in the region is concentrated in areas where moisture is available, such as: drainage lines, depressions, furrows, man-made channels, on the escarpments of the plateaus, near springs and in wadis. Over the plateau there is hardly any moisture, and consequently the vegetation is very poor or absent. Construction of the service tracks on the plateau (for oil exploration-companies) have created small linear runoff catchments or depressions along their sides, which trap soil and moisture and create conditions in which certain species such as Farsetia linearis and Zygophyllum simplex germinate and grow successfully. Other species such as Acacia oerfota, Pulicaria undulata, and Crotalaria saltiana also flourish in rain water reservoir on the plateau (Plate 7.3). Other species are tolerant of continuously saturated soils. Particular examples are the species of sociological group D (Pulicaria undulata, Crotalaria persica, Corchorus depressus, Convolvulus glomeratus, Chrozophora tinctoria, Indigofera spinifolia and Merremia hadramautica). These species are common in the wet depressions that were in former times cultivated after rain and then left fallow. 309 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Plate 7.3. Vegetation growing along the rain water reservoir on the plateau; species include Acacia oerfota, Pulicaria undulata, Aerva javanica, Crotalaria saltiana and Stipagrostis hirtigluma A steady increase in moisture over just a few months can result in the establishment of lush vegetation. This is dramatically illustrated by the rapid growth of plants, including Dichanthium insculptum, Farsetia linearis, Chrysopogon aucheriand, Pulicaria undulata, in a wadi on the plateau in site 2, which has been regularly watered with large amounts of contaminated water from oil-related activities. Almost half of the plant species of the plateau in site 2 were confined to this wadi. From this example it is evident that the addition of water can rapidly alter plant cover and composition. Although plateau habitats from all three sites are almost identical, the southern plateau where site 2 is located has a much richer flora and vegetation cover than those of sites 1 and 3. As indicate earlier, the southern plateau receives relatively higher rainfall than the northern plateau (SOGREAH 1979). However, there are some differences, Indigofera spinosa, Jatropha spinosa, Maerua crassifolia, Commiphora foliacea Zygophyllum decumbens, Farsetia linearis, and Aristida triticoides are abundant on the slopes and wadis on the plateau in site 2 but are very rare or absent in the same habitat in sites 1 and 3. The abundance of these species in the plateau of site 2 may support the theory of SOGREAH (1979), which suggests, using the evidence of the abundant floods in Wadi Adim, that the annual rainfall of this region approaches 300 mm. Therefore, these species, which require higher rainfall, in the order of 310 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 100-300 mm, become rare or absent in areas towards the north which receive less than 100 mm precipitation. Other species, such as Indigofera spinosa, Jatropha spinosa and Maerua crassifolia are also species which show a preference for areas of higher rainfall. These form an open woodland and grassland, at altitudes between 500 and 1900m, on the southern mountains and on some isolated mountains on the coastal plains (Al Khulaidi, 2000). These areas have a relatively higher annual rainfall than Wadi Hadhramaut (Figures 3.8, 7.2). Similar vegetation is also found in other arid areas where moisture is the main limiting factor for plant growth. For example, some species that are common in the study area such as Panicum turgidum, Fagonia indica, Rhazya stricta and Aerva javanica are also restricted to wadi beds and the sands of the north-eastern and eastern desert areas of Yemen (e.g. Marib and Rada’) and the deserts of Oman (Ghazanfar, 2004). It has also been reported that at lower altitude in the Hormozgan province in Iran a savanna-like community, of Acacia tortilis, A. ehrenbergiana and Rhazya stricta, is found along wadis (Zohary, 1973). The Acacia campoptila community association with Panicum turgidum, Aerva javanica, Indigofera spinosa, Fagonia indica, Ochradenus baccatus, Dipterygium glaucum, Calotropis procera and Rhazya stricta is also widespread in wadi beds in the arid areas of Marib and Rada’, in western Yemen (Al Khulaidi, 1989, Wistinga et al., 1980, Al Hubaishi et al., 1979). Marib area (Figure 7.2) is characterised by more rainfall than the study area. Some characteristic species of wadi beds in these areas, such as Leptadenia pyrotechnica, Lycium shawii, Jatropha spinosa and Acacia oerfota, were absent in similar habitats in the study sites. These last two species were confined to wet slopes on the plateaus and were absent on the wadi beds in the study sites. Some characteristic species of the study area such as Tephrosia apollinea subsp. longistipulata, Ziziphus leucodermis, and Merremia hadramautica were not recorded in either Marib or Rada’. Within Yemen, these species are only recorded from Hadhramaut. Some species, for example Jatropha spinosa and Acacia oerfota, which are now restricted to areas of relatively higher rainfall on the plateaus may in the past, have also been found in the wadis. However, with decreasing moisture in the wadis, they are no longer found there. 311 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 35 30 mm 25 AlKod 20 Sayun 15 Marib 10 AlHudayda AlRayan 5 0 J F M A M J J A S O N D Figure 7.2. Average monthly rainfall of 6 areas in the coastal and northern and eastern dry zones, Yemen. It can be observed that the rainfall in the western coastal areas (alHudayda) is more than the southern coastal areas (al Kod and al Rayan). The rainfall in the western dry areas (Marib) is greater than eastern dry areas (Sayun). (Source FAO, 1997) Places outside Yemen with a vegetation similar to the wadi vegetation of the study area, that is Acacia woodland or savanna-like communities in association with Panicum turgidum, Leptadenia pyrotechnica, Aerva javanica, Calotropis procera and Ziziphus spina-christi, are widespread throughout the Sahara-Sindian region and particularly in the wadis of Sudan, Egypt, Palestine and Arabia (Zohary, 1973; White 1983). Ochradenus baccatus corresponds to the southern limit of the northern zone of Hadhramaut where the rain falls mainly during the hot season, with 2 maxima, in spring and summer. This may correspond with the southern limit of Sahara-Sindian region (White, 1983). Other characteristic species of the main wadis, such as Cressa cretica inhabit saline fallow land in Egypt, inland salt marshes and Alhagi graecorum with Tamarix nilotica were found on sandy plains in Egypt (Abdul El-Ghani, 2000). This may indicate that the vegetation of the main wadis in the study area is of paleotropical origin (in this case with affinities to the Somalia-Masai reginal centre of endemism of White, 1983) although it is geographically located within the Saharo-Arabian or Saharo-Sindian elements of holarctic origin. 312 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Arid and semi-arid regions are often characterised by an abundant flora of annual plants which complete their life-cycle within a relatively short favourable growth period (Zohary, 1973; Kutiel et al., 2000). Short-lived life forms such as annuals grow effectively in drought periods (Brown, 2001). The population of annual plants in the study area is high and comprises about 45% of the entire flora. Unlike other life forms, annual species respond directly to annual rainfall fluctuations (Kutiel et al., 2000). Annual plant species were seen to be strongly affected by the availability of moisture. They produce an enormous numbers of seeds which survive during the dry period and then grow rapidly with the onset of any short period of rain. During the fieldwork, annual species such as Boerhavia elegans, Senna italica, Tribulus arabicus and Blepharis edulis were abundant on the recently flooded wadi beds. Generally, over different visits to the study sites, the species richness and abundance was similar in the study areas, except for a slight increase in the number of individuals of some annual plant species such as Boerhavia elegans, Senna italica and Blepharis edulis on the flood plains. An experimental enclosure, located in Wadi Al-Khun (Bataher, 2004), in the northern part of the study area, shows also that Tribulus arabicus, Cleome brachycarpa, C. scaposa, Cenchrus ciliaris, Dichanthium insculptum and Stipagrostis hirtigluma, appear at very high densities in the rainy season and fall to low or absent in the dry season. The amount of rainfall required to initiate the germination of annual plant species in arid areas has been studied by different scientists (Gutterman, 1993; Went, 1955). In the study sites, very light showers in spring were enough to create a large number of annual species. Competition between plants in low density cover is negligible (Brown, 2001). Because the vegetation of the study sites is generally sparse, the competition is likely to be low or absent. Generally the most abundant species were found to be more competitive (Goldberg and Estabrook, 1998; Brown 2001). In the study sites, the most abundant annual species was the endemic Boerhavia elegans, which germinates quickly and survives successfully in the harsh environment and in over-used areas. Because of possible toxic substances present in the oil, the problem of the extinction of plant species in certain areas is expected to be high in the long run. This is especially true for 313 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen annual plant species, which rely on regular germination to maintain populations (Brown, 2001). The seed banks in the sand under the soil are also expected to be reduced. At sites near to oil-contaminated areas, species with similar ecological preferences such as Stipagrostis hirtigluma, Maerua crassifolia, Hochstettera schimperi and Commiphora kua were abundant. Some species such as Farsetia linearis, Tamarix arabica, Dichanthium insculptum, Tephrosia apollinea, Boerhavia elegans and Dichanthium insculptum seem to be tolerating oil contamination, but toleration may not last for long. Several studies have highlighted that the vegetation density of desert areas is closely related to soil moisture availability, so the high cover of these species on this site may be due to large quantities of seed in the “seed bank”, and also because of the tendency for there to be more moisture available in the polluted areas. Generally, more studies and monitoring need to be done before a conclusion can be drawn on the long-term effects of soil pollution. It is clear that moisture is an important factor for plant growth in the study area. The woody and perennial vegetation in the area is concentrated, or contracted, in areas where moisture is available such as drainage lines, depressions, furrows, man-made channels. Changes in the proportion and abundance of woody and perennial species require a steady increase in moisture over a period of months whereas slight increases in rainfall may rapidly alter plant cover and composition due to the rapid appearance of annuals. 7.3.2. Topography Heterogeneity of the local topography, edaphic factors, microclimatic conditions and degree of slope lead to variation of the distribution of plant associations (Furley and Newey, 1983; Al Wadie, 2002). In Brazilian savanna, variations in topography and drainage have resulted in different forms of vegetation structure (Furley, 1999). As an analogous example, variations in microclimate with topography and elevation were a major factor of species distribution within a forest landscape in Hubei Province, China (Yu Hua, 2000). In the desert ecosystem, topography is an important related factor because of its effects on the runoff of water and potential or free water evaporation. Also rainfall is an important factor for determining the structure and composition of desert vegetation (David et al., 1993). 314 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen As elevation increases, precipitation increases and temperature decreases (Vandeven, 1999). Aspect, slope, and surrounding topography determine potential isolation so that in the northern hemisphere, south-facing slopes are warmer and north-facing slopes cooler. Environmental heterogeneity including topography is thought to have a significant influence on the dynamics and structure of plant communities (Morzaria-Luna, et al.,2004). The mountains can also offer maximum modification of the arid climate by reducing air temperature or by shade effects and air drainage among the peaks and valleys (Eyre, 1969). In the study sites, there are obvious macro-features of topographic heterogeneity, from the wadis to the plateaus. At the micro-climatic scale, the topographic forms contain depressions, narrow channels and rock outcrops (plate 7.4 to 7.8). These different features have dramatically affected the distribution of vegetation as well as the type of plant community. The low plant density on the slopes of plateaus and those facing the main wadis may also resulted from topography that is characterised as being very rocky, with shallow soil of low fertility. On the cliffs, where the slope is very steep, there are hardly any plant species, because of the steepness and vulnerability to soil erosion. The inter-plateau slopes have more vegetation than the slopes facing the main wadis. Main wadis have a relatively high vegetation cover and are mainly composed of dwarf shrubs and herbs with scattered trees, mainly Acacia campoptila. The density of vegetation varies from site to another, for example the vegetation on wadis facing south was high except on areas with rock outcrops. In general the number of species is high in wadis that have a high percentage of rock outcrops. These create microclimatic conditions for plants to grow such as Cleome brachycarpa, Fagonia indica, Zygophyllum simplex, Aerva artemisioides subsp. artemisioides, Zygophyllum coccineum, and Ficus cordata subsp. salicifolia. Altitudes and topography also play an important role in governing plant distributions. Small changes in elevation are influential, for example, Acacia oerfota, A. mellifera, Corallocarpus glomeruliflorus, Jatropha spinosa, Commiphora kua, Tarenna graveolens and Cadaba heterotricha are confined to plateau drainage lines and slopes (over 700 m). The absence of these species in the main wadis and adjacent slopes can be due to the low altitude of these habitats (less than 600 m) and past human activities such as over-grazing and fire wood collections or most likely these species prefer cooler and moister sites that are 315 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen characteristic of their current habitats on top on the plateau. Some of these species, especially Corallocarpus glomeruliflorus, Jatropha spinosa, Commiphora kua, have been and are still used by local people as folk medicines. Where sites have been affected by increased runoff or soil erosion, less common species such as Seetzenia lanata were rare or absent. Variation in vegetation pattern was also seen due to disturbances of surface soil, or due to presence of micro-sites such as furrows cutting the slope and depressions or fallen rocks and finer particles accumulating at the base of the mountain slopes. These micro-sites contribute to additional species by improving soil moisture in the soil. For example Periploca visciformis, Leptadenia arborea, and Ficus cordata subsp. salicifolia were seen only on isolated slopes near water ponds or very wet slope drainage lines. Plate 7.4. An example of microclimate change is stone ring on the plateau. As can be seen more vegetation cover is found around the site. This stone ring feature also plays an important factor in diverting water to the adjacent slopes and as a result,, the vegetation on that slope become denser than the rest of the slopes (see association 13). The origin of the stone ring is possibly archaeological site from the Stone age (McCorriston, 2000). 316 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Plate 7.5. The vegetation is concentrated only on gullies and runnels, where moisture is available. Top: the main wadi (site 2): Acacia campoptila and Rhazya stricta, bottom: the slope facing the main wadi (site 3): Acacia hamulosa with Stipagrostis hirtigluma and Farsetia linearis. Plate 7.6. The surface of some secondary wadi on the plateau is covered by bare pavement rocks with soil covering less than 3%. 317 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Plate 7.7. Stones and rock outcrops (Shieb Sultan, site 1), under the arid conditions of the study area, this type of landform which is located directly at the bottom of the plateau, where different size of rocks and fine soil materials are accumulated, is favourable habitat for plants. Most of runoff water from the plateau above is well preserved here and protected against evaporation. In addition, big rocks become favourable habitats for several species such as Acacia campoptila, Tephrosia apollinea, Tephrosia dura and Maerua crassifolia (see association 4). Plate 7.8. A desert pavement (a surface gravel deposit of tightly packed pebbles, devoid of vegetation) is found on wadi bed surfaces site 2, Pavements such as this occur in areas where the stream flow is restricted to relatively stable channels nearby (background). 318 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Desert plants are highly sensitive to disturbance (Brown, 2001). Roads created by oil working activities and trampling by animals have resulted in compaction of the soil surface preventing roots from penetrating the soil (Plates 7.9a, 7.9b and Figure 7.4). Brown (2001) found that slight depressions caused by passage tracks on the desert surface can be regarded as important micro-sites for desert annuals; in these depressions, seeds may accumulate in large quantities. Off-highway vehicles have also disturbed large areas of the Mojave and Colorado deserts. The effects of these vehicles destroy soil stabilizers, resulted in soil compaction, increased wind and water erosion (Busack and Bury, 1974), and the destruction of vegetation (Webb and Wilshire, 1984). Off-road vehicle and hiking activity also have greatly expanded direct human use both spatially and temporally (Belnap, 2002). On the disturbed sites of the plateau where the land surfaces have been cleared, the first species to colonise appear to be Zygophyllum simplex and Farsetia linearis. Zygophyllum simplex was the most common annual seen in the desert of Oman after rain, especially in sandy depressions. (Ghazanfer, 2004). 319 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Plate 7.9a. Roads created by oil working activities have resulted in compaction of the soil surface (red arrow). Plate 7.9b. Roads created by oil working activities have resulted in compaction of the soil surface. (photo by Henry Thompson). 320 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Figure 7.3. Off-road effect. Oil working activities have a high impact on soil and vegetation cover on the plateau. These satellite imageries show a negligible off-road impact in 1984 to very high impact in 2001. Soil cover and the sparse vegetation on the plateau surfaces have been altered by extensive off-road travel. Some damage will remain obvious for decades. The current oil exploration survey methods which incur a high density of trails and cover extensive areas are readily visible from the air. The potential for further damage still exists because of the large number of vehicles and facilities used for exploration. 321 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen It has been demonstrated here that one of the main factors related to vegetation variation in the study area can be linked to topographic position. Other variables such as soil moisture, soil type and pH are clearly a consequence of topographical changes. The influence of landscape changes was very dominant. Each ecological zone paralleling landscape change has its own plant species. Some species were found almost in all ecological zones, and others are confined to specific and favourite sites. 7.3.3. Floods Wadi vegetation results from hydrological disturbance, including periodic floods, erosion and alluvial deposition (Robert and Thomas, 1997). Extreme floods have been recorded in the main wadis of the study area by Lewcock, (1987); INTERCONSULT and MacDonald (1993); SOGREAH (1979). Sociological species groups showing adaptations to these processes were B, C and E, namely Zygophyllum album, Prosopis juliflora, Tamarix aphylla, Citrullus colocynthis, Heliotropium rariflorum, Dipterygium glaucum, Calotropis procera, Acacia campoptila, Fagonia indica, Tephrosia apollinea subsp. longistipulata and Cymbopogon schoenanthus. These species were most common on active wadi surfaces such as bare grounds gravel areas, flooded wadi beds and wet patches. There is a very obvious vegetation transition from extreme to moderately flooded areas. This can be observed in the transition from former date palm fields to vegetation association 2 (Chapter 5). In this association, species which have the capacity of withstand the force of peak stream flows or to ability to colonize bare stone substrates rapidly are common. These are species of groups B and C (Table 5.6), the structure of this habitat is characterised by being sparse grassland. Vast areas of these habitats are occupied by Prosopis juliflora. This species was once planted widely for sand dune stabilization, probably around 1990, but it has become a major problem over large areas of the Wadi Hadhramaut (see 7.2.4). 322 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen The distribution of some species is related to stability of the surface for example sociological groups F, G and M (namely Kohautia retrorsa, Acacia hamulosa, Barleria aff. bispinosa, Tephrosia dura, Zygophyllum coccineum and Ziziphus leucodermis). These species prefer relatively stable less flooded surfaces such as tributary wadis or the base of mountain slopes. Some species show adaptations to extreme floods; these are Zygophyllum album, Prosopis juliflora, Tamarix aphylla, Citrullus colocynthis, Heliotropium rariflorum, Dipterygium glaucum, Calotropis procera, Acacia campoptila, Fagonia indica, Cymbopogon schoenanthus and Tephrosia apollinea subsp. longistipulata. These kinds of habitat contribute to raising number of the plant species especially, the exotic, which may be brought in by floods from other areas. Well-established vegetation dominated by Acacia campoptila is found on parts of the flood plain that generally do not receive intensive flood waters (Plate 7.10). On the active surfaces a dwarf srubland consist of not more than 6 plant species dominated by Merremia hadramautica and Tephrosia nubica is common, since these plants are adapted to rapid colonization of disturbed and flooded surfaces. Plate 7.10. Wadi bed of site 3 (plot 305). The active surface on the right and another less active, higher on the left only receive water during the most intense flooding events. Well-established vegetation dominated by Acacia campoptila is found on parts of the flood plain that generally, do not receive intensive flood waters. The plants on the active surface is a dwarf shrubland dominated by Merremia hadramautica and Tephrosia nubica; these plants are adapted to rapid colonization of disturbed and flooded surfaces. 323 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Extreme floods are common in main wadis. Floods have a significant effect on the structure of vegetation of wadis which to have varied relationships with surface and ground water. Some associations develop under the conditions where their deep root systems can utilize the water from basal gravel and lower, silty alluvial layers (Coode Blizard Ltd 1997). In these habitats, woody perennial trees such as Acacia campoptila, Ziziphus leucodermis, the date palm Phoenix dactylifera and the invasive Prosopis juliflora form their own distinctive associations. Savanna-like vegetation or pseudo-savanna as defined by Zohary (1973), on wadi beds and runnels in several locations, has been transformed to Prosopis woodlands due to the encroachment and establishment of the invasive tree Prosopis juliflora. Pure stands of Prosopis juliflora woodlands are common now in the river beds of the main Wadi Hadhramaut The association dominated by Zygophyllum album and Dipterygium glaucum was only found on relatively wide wadis that are subjected to severe flooding, with loamy to loamy sand soil combining a high proportion of sand (86-88%). Other species with this association are Citrullus colocynthis, Heliotropium rariflorum, Dipterygium glaucum, Panicum turgidum, Prosopis juliflora, and Tamarix aphylla. Most of these species form deep roots, which probably offer good mechanical support against the relative high water velocity during the flood periods (Wittmanna et al., 2004), see Plates 7.2 and 7.11. 324 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Plate 7.11. Vegetation dominated by Zygophyllum album on extreme flooded wadi beds (site 1). A succession community from palm field cultivated lands and woody vegetation dominated by Ziziphus leucodermis, Acacia ehrenbergiana, Acacia hamulosa and Tephrosia dura. It is has been shown that extreme floods have a significant effect on the structure of vegetation of wadis. Floods, as an abiotic factor, also contribute largely in destroying and disappearing of some individuals trees. Floods also contribute to raising the number of plant species especially the exotic, which may be brought in by floods from other areas, example are Prosopis juliflora and Zygophyllum album. As a result of severe floods, the woody vegetation dominated by Ziziphus leucodermis, Acacia ehrenbergiana, Acacia hamulosa, Tephrosia dura and date palm trees has disappeared or shrunk in coverage and has been replaced by other communities more resistant to the severe-floods and disturbances. 325 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 7.3.4. Human activities The effect of man on the desert vegetation may be directed to the vegetation cover itself or indirectly through its influence on the other components of the ecosystem (Batanouny, 1983). Biodiversity is being affected in many parts of the world as a result of human activities, especially land degradation and the overuse of resources. As an example, in 1995, approximately 10% of species in the Himalayas were listed as threatened, and the number of extinctions has increased since then. (UNEP and WMO, 1996). Due to unrestricted cutting of wood of vegetation, over-grazing by livestock, and cultivation of unsuitable lands, many arid zones have an inadequate timber, fuelwood, and fodder resources (FAO, 1989). Human activities and overgrazing have resulted in the destruction of the natural vegetation and changing the vegetation structure in many parts of the world (Humphry, 1962; Eyre, 1969; Sutter and Ritchison, 2005). Over-grazing is considered as the major cause of desertification in arid and semi-arid areas of the world (Brown, 2001) and has a major effect on plant diversity and composition (Todd, and Hoffman, 1999) as well can causes soil compaction and contributes to erosion and decreases soil fertility, organic matter content and water infiltration and storage (FAO, 1997; Webb and Stielstra, 1979). In the study area, uncommon or short-lived species that are selectively grazed or collected by local people during very good seasons are more likely to decline; instance of these are Boerhavia elegans, Corallocarpus glomeruliflorus, Lasiurus scindicus, Commiphora kua, Commiphora foliacea and Tribulus arabicus, Over-grazing has selectively affected the abundance of some species, in particular Merremia hadramautica, Boerhavia elegans and Tribulus arabicus. The first species is abundant on wadi beds of site 3 and is found with low cover near cultivated and fallow lands in sites 1 and 2. The abundance of these species in site 3 is due to control of grazing by local people in the past. The people in the study area have used to have traditionally made sustainable use of their natural habitats and conserved biodiversity - for example through the protection of land above cultivated fields during cultivated season. However, this system is not any more found, as a result overgrazing and other human pressures, contributing to rarity of some species such as Boerhavia elegans, and probably have contributed the extinction of some native plants. 326 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Roads and other human activities, especially on the plateaus, have affected vegetation cover. Many sites have been eroded and some species on eroded slopes have disappeared due to road construction, for example species such as Seetzenia lanata, and Monsonia heterotropoides are rare or absent in such conditions. Generally, soil surface disturbance and destruction of established plants offer small gaps (Focht and Pillar, 2003), which allow such areas be dominated or invaded by species that were not abundant in the site. Examples are the replacement of community dominated by Dichanthium insculptum with the community dominated by Chrozophora tinctoria, Boerhavia elegans, Aerva javanica, Calotropis procera, Zygophyllum simplex and Fagonia indica on the disturbed sites on the plateau of site 1. The impact of waste water on vegetation seems more positive, since the release of water has produce vigorous germination and growth of plant species (Plate 7.12). Some of these species may be truly xerophytic and cannot tolerate high water levels, or they may intolerant of salts and chemicals in the waste water, as a result they may disappear from the site once the bank seeds in the soil have been exhausted. The surface disposal of water which flows over dry slopes can lead to rill or gully erosion and the concentration of salt in this water may kill sensitive plant species (Plate 7.13). According to Komex International Ltd (2001), the contaminated water impact can be very bad, because the water is highly mineralised, the total dissolved solids can reach around 2,000 - 3,500 mg/lt - of which chloride and sodium comprise around 75% and when the companies dispose of the water on the plateau, the water infiltrates into the Jeza formation and infiltrates through the Jeza clays to the top of the Umm er Radhuma. It then has to seep out into the small spur wadis and flow down toward the main wadis. On the way the water picks up more salts in the Jeza, and evaporates, so becomes hyper-saline. By the time the water reaches the small side wadis the salt content may be as high as 8,000 mg/lt. This water then runs into the small kareefs and wadis - and sits in the root zone of the small trees that line the side wadis which may die. On the plateau, dead trees were observed below the spray farms of surface water disposal systems of companies such as DNO, TOTAL, and DOVE. 327 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Plate 7.12. The impact of waste water on vegetation. Some of these species may be truly xerophytic and cannot tolerate high water levels, or they may intolerant of salts and chemicals in the waste water, as a result they may disappear from the site once the bank of seeds in the soil is finished. The vegetation here may be dominated in future time by salt-tolerant tree species such as Tamarix arabica. Plate 7.13. The impact of concentration of waste water on the vegetation cover. Completely bare lands due to concentration of waste water. 328 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Waste water which flows from the slope to the wadi bed and contaminates the vegetation, has caused the death of some sheep and goats. This is one of the reasons why Bedouin are reducing grazing on the plateau and in areas near oil compounds. As a result, the pressure on natural vegetation of the main wadis has increased, and the Bedouin have started moving to other territories, thereby breaking the traditional way of grazing. Examples of this process are found in site 3, where the residents have complained of grazing invasion from Bedouins. Many of the traditional Bedouins have now settled in the study area. Changes in lifestyle, mobility and dependence on modern transport have affected the way people use the rangelands. In the past, every tribe had its own region for grazing. In dry times, a Bedouin from one tribe could move to the region of another tribe with his camels and sheep or goats. Some of these traditional grazing systems have been neglected as Bedouins have come to depend on cheap supplements such as sorghum, alfalfa (Medicago sativa) and Prosopis juliflora. Grazing in the study area may be also an important factor but it was not evaluated completely in this study. The variations of vegetation structure found in each floristic vegetation association may be due to over grazing and woodcutting. Tephrosia dura in association 3 is found in some locations with a high ground cover, and its occurrence could be a sign of overuse. Over-grazing, has a direct impact on species richness in different parts of the study sites, especially on the species that grow close to the villages in localities such as wadi beds and surrounding mountain slopes. Some of the species of arid lands have adapted to survival by growing in open, overgrazed areas (Naveh and Whittaker, 1979), and Blepharis edulis and Boerhavia elegans may represent such grazing-adapted species. The collection of some plants for medicinal or other purposes also has an impact on certain species such as Boerhavia elegans, Corallocarpus glomeruliflorus, and Rhazya stricta. According to local people, Boerhavia elegans was abundant 15 years ago and the farmers are used to harvest about 3 sacks seeds of this plant. As a result of reduction of precipitation and over-grazing in particular uncontrolled grazing by local and Bedouin people, the cover of this species has dramatically decreased. The grazing was practiced mainly on the river beds and 329 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen was not allowed on the slopes at particular times, such as during cultivation and before harvesting of Boerhavia elegans. Intensive grazing was noticed in sites 1 and 2. However, with the arrival of nomads to the study sites, in particular during rainy seasons from different parts of the Hadhramaut region, (Figure 1.20) grazing may become more intensive in previously less affected areas such as site 3, and local people have complained about this problem. As a result, some species that are abundant in sites 3 and are rare or become rare in other part of the study sites will be easily targeted by goats and camels in future time. For example palatable species such as Merremia hadramautica, Tephrosia nubica and Panicum turgidum are abundant and form shrubland communities in different locations in site 3 (see Sociological species group O, Table 5.6). The intensive grazing co-incident with extreme floods, has inhibited the development and growth of this woody vegetation in the other study sites. As a result, the vegetation composition and the vegetation structure have been changed. It is a general observation that in Mediterranean ecosystems, grazing inhibits the development and growth of woody vegetation (Carme and Kadmon, 1999) and results in larger herbaceous cover and lower tree cover. In contrast, grazing may create niches for woody seedling establishment, by trampling and/or reducing the biomass of competing herbaceous vegetation (Carme and Kadmon, 1999; Sutter and Ritchison, 2005). As a further illustration, because of grazing over the last 100 years, the structure, composition, and dynamics of semi arid western, interior forests of United States have changed dramatically (Belsky and Blumenthal, 1997). Batanouny (1979) reported that overgrazing between Jeddah-Mecca road, Saudi Arabia resulted in severe destruction of vegetation dominated by Panicum turgidum and replaced by habitat dominated by Dipterygium glaucum. In the study sites Panicum turgidum was rare and the degraded sandy wadi beds were dominated by Dipterygium glaucum and Zygophyllum album. Generally, to assess the effect of grazing on the structure and the composition of the vegetation in the study area, long-term monitoring research on the effects of grazing on vegetation dynamics is needed. Land use varies across the region depending on its suitability for agricultural or grazing activities. For example, much of the main wadi beds, which run almost at the middle or near 330 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen the foot slopes of the main wadis, are covered by desert alluvial wadi vegetation of shrubland with scattered trees of Acacia campoptila associated with Fagonia indica, Tephrosia apollinea, Cymbopogon schoenanthus, Boerhavia elegans and Dichanthium insculptum (see associations 4 to 7). In contrast, much of the fertile land, such as the foot slopes at the bottom of the plateau, is privately held and intensively cultivated by palm trees and other annual crops such as sorghum and wheat. The rocky slopes facing the main wadis and the plateau surface are covered by stony and gravely desert vegetation type dominated by herbaceous plants namely Stipagrostis hirtigluma, Farsetia linearis, Aristida triticoides Fagonia paulayana, Boerhavia elegans and Dichanthium insculptum. The slope on the plateau and the secondary wadis are generally less affected by human activities due to inaccessibility. As a result, still quite dense vegetation can be seen in some favoured sites, the vegetation here comprises shrubland dominated by Jatropha spinosa with Zygophyllum decumbens, Commiphora foliacea, Commiphora kua, Maerua crassifolia. Cadaba heterotricha, Tarenna graveolens, Aristida triticoides and Grewia erythraea. Native people influenced the vegetation intensively through grazing and agricultural clearing across the region from at least the post-Neolithic second millennium BC (McCorriston, 2000) until the 1880s, when many of the Hadhramaut people emigrated to Indonesia and Arabia (Boxberger, 2002). Lithic studies of surface material suggest that the study area was re-occupied or re-used as early-mid-first millennium BC (McCorriston, 2000). During the most recent glacial period in the northern hemisphere, the climate of the Arabian Peninsula was one of extremes of heat and cold, but this has gradually changed over the past 12,000 years into climate that we are familiar with today (Mundy and Musallam, 2000). Clearing of the land for agriculture and grazing or woodcutting peaked in the early 1900s, and then declined dramatically because of intensive of migration to the Far East Asia and Arabian Peninsula, after which lands were abandoned or naturally regenerated to woodland. Many of these abandoned lands have turned now to bare bad land severely eroded by deep gullies. Today, about 80 percent of the landscape in the study area is covered by range and marginal lands and less than 20 percent is utilised as agriculture, mostly irrigated. 331 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen The species composition has changed from species established as the result of past disturbances to more tolerant species and is adapted the current dry environment as the result of the decreased disturbance regimes after 1880. However, the disturbance has picked again after the unification of the countries (north and south Yemen) in 1990, with the development of the land use by human activities. This rapid development has caused high damage to the past vegetation and has led to the almost total disappearance of trees and shrubs from many places of the study area. For example, the vegetation in the main wadi, which was dominated by Acacia campoptila, A. ehrenbergiana, A. hamulosa, Rhazya stricta, Ochradenus baccatus Ziziphus leucodermis, Tamarix aphylla and Salvadora persica has vanished from several locations. Association 5 is probably a relict of this type of past vegetation community. Most arid and semi-arid lands have sparse vegetation cover, low surface-soil organic matter content and significant nutrient loss (Heff et al., 2005). Due to human activities, there has been a long-term reduction of the vegetation over time in part of the world. The long-term depletion of soil nutrients by intensive pastoral activities has been proposed as a factor contributing to degradation in Australia and New Zealand (Harris, 2000). Long term extensive grazing in north-eastern Arabia and gradual depletion of soil resources have led to a transformation of what was once a very open savanna-like woodland, to steppe and perennial less palatable species (Brown, 2001). Figure 7.4 shows the degradation processes in the main wadis, due to human activities and climate change, the former vegetation type has been altered to different forms of associations. Due to human activities and other disturbances, the unique areas of the study area covered by native species maybe replaced by species that are widespread and adapted to disturbed habitats such as Prosopis juliflora, Calotropis procera and Zygophyllum album. As a result, the vegetation of the study area could become more homogenous. This is already noticed in the main wadis, where past vegetation dominated by Acacia campoptila, Ziziphus leucodermis, Tamarix aphylla and Salvadora persica has been replaced by a pure stand of Prosopis juliflora woodlands. Studies conducted in the United States (McKinney, 2004) concluded that human activities have significantly increased the amount of homogenization by increasing the percentage of non-native plant species among disturbed communities. In the 332 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen following figures (Figures 7.4 to 7.8) vegetation change processes are shown by changes in plant cover, biomass and biodiversity within time. In contrast to the study sites, the flat limestone plateaus of the southern Hadhramaut region (Figure 7.6 top right) have a higher cover of vegetation than the northern plateau (bottom right) dominated by shrubs and dwarf shrubs. This is because the southern plateau has more rainfall than the northern plateau. The increase of moisture will turn the vegetation of the slopes and secondary wadis on the plateau in the study sites to a dense grassland dominated by Dichanthium insculptum and Farsetia linearis (bottom left, Figure 7.7), which then leads to its climax condition which can be a woodland dominated by Acacia mellifera, Maerua crassifolia, Cadaba heterotricha, Tarenna graveolens, and Commiphora spp. (top left, Figure 7.7). 333 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Figure 7.4. Vegetation degradation processes in the main wadis. Figure 7.4 explains that the main wadi bed of site 2 (top left was dominated by Typha sp, and Pluchea arabica with Palm trees. Nowadays the river beds that used to be permanent have become seasonal, filled up after rain and remain dry during the rest of the year; as a result the vegetation cover has dramatically changed to the current savanna-like habitat dominated by Acacia campoptila and others that are adapted to arid habitats (middle left). With the extreme floods and human activities the vegetation in many parts turned to sparse cover with scattered of Acacia campoptila, Rhazya stricta, Indigofera spinosa and Blepharis edulis (bottom left and bottom right) or to sparse dwarf shrubland dominated by Zygophyllum album (top right) or to sand dune shrubland habitat dominated by Acacia ehrenbergiana and Dipterygium glaucum. (top middle). Generally in the main wadis, a few stands of typical paleotropical vegetation with Salvadora persica, and Tamarix aphylla also serve to give an impression of what the previous climax vegetation might have been like when the climate was wet and before it was replaced by a vegetation association dominated by Acacia campoptila with Fagonia indica, Tephrosia apollinea, Cymbopogon schoenanthus, Rhazya stricta, Ochradenus baccatus and others that can survive the present arid climate. 334 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Figure 7.5. Vegetation degradation processes on the mountain slopes facing the main wadis Figure 7.5 shows the degradation processes on the mountain slopes facing the main wadis. Due to over grazing, the slopes that were covered by dense grassland and were dominated by Boerhavia elegans and Stipagrostis hirtigluma have become degraded to bare lands or to sparse grassland with few Cleome droserifolia and Fagonia paulayana species. 335 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Figure 7.6. Vegetation degradation processes in the mountain slopes facing the main wadis The degradation processes on the plateau and in the secondary wadis are illustrated in Figure 7.6. The top 2 pictures show the current vegetation cover of the southern part of the Hadhramaut region. Probably the vegetation of the northern part of the region was similar to this type of vegetation, but due to climate change and human impact, the vegetation cover degraded to sparse cover with few scattered species that adapted to current climate. The vegetation of the plateau (top right) was more favourable to tree and shrub life and then replaced gradually by vegetation which is dominated by herbaceous plants. 336 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen . Figure 7.7. Vegetation degradation processes on the slope of the plateau. The degradation processes on the slopes of the plateau can be noticed in Figure 7.7. The picture top left show the climax vegetation which may represents the past vegetation of these slopes. This vegetation type is dominated by Acacia mellifera, Tarenna graveolens, Commiphora kua, Maerua crassifolia and Jatropha spinosa, but due to human activities and climate change, the vegetation cover of these slopes has degraded to a sparse cover with few scattered species that adapted to the current climate, such as Jatropha spinosa. This stage represents association 13 (Maerua crassifolia - Jatropha spinosa - Stipagrostis hirtigluma association) which is dominated by Jatropha spinosa with Maerua crassifolia, Commiphora 337 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen foliacea Zygophyllum decumbens, Stipagrostis hirtigluma, Dichanthium insculptum and others, then it has degraded again to sparse grassland with few plant species, such as Zygophyllum decumbens, Dichanthium insculptum, Boerhavia elegans, Cleome brachycarpa and Indigofera spinosa. This stage is represented by association 12 (Zygophyllum decumbens Dichanthium insculptum). Increasing moisture on the slopes can results in dense vegetation cover dominated by Dichanthium insculptum and Farsetia linearis. This was noticed on the slopes that been irrigated by waste water (bottom left, plate 7.7). The following figure shows a summary of the vegetation degradation processes on the slopes of the plateau. Increasing degradation Figure 7.8. Vegetation degradation processes on the slopes of the plateau. 338 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Prosopis juliflora (Mesquite) is an invasive species that was observed only in site 1, but it has recently invaded neglected agricultural land and wadi beds in many parts of Wadi Hadhramaut (Plate 7.14 and Figure 7.9). This can be as a result of the increase in modern agricultural activities in recent years (after 1800). In some locations, it forms a pure stand of woodland. Local people suggested that Prosopis juliflora reached Wadi Hadhramaut as recently as 1990. This species was not recorded from Wadi Hadhramaut in the report of Barkuda and Sanadiqi (1986). Mesquite is a fast-growing, deciduous tree native to the West Indies, Central America, and northern South America. It is an invader species that competes with native species in many parts of the world. The species grows in very hot regions with annual precipitation from 150 to 750 mm and from sea level to 1500 m (Tiedemann, 1973). The roots penetrate to great depths in the soil (more than 50 m) over a variety of soils, including saline and alkaline areas and in sandy and rocky conditions (Hultine, 2000). Prosopis juliflora avoids drought by developing root systems that can access deep groundwater that is unavailable to other plants. This phenomenon has allowed Prosopis juliflora to survive in arid regions that are generally not suitable for many plant species. As in a number of countries such as south western North America (Kevin, 2001) and in the arid and semi-arid parts of the India and Pakistan (FAO, 1993), the natural habitats of shrubland or grassland have been transformed to woodlands due to the encroachment and establishment of the Prosopis juliflora. Plate 7.14. Prosopis juliflora has recently invaded abandoned agricultural land and flooded wadi beds in many parts of Wadi Hadhramaut. In some locations it forms a pure stand of woodland. 339 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen The following 2 satellite images show the area occupied by Prosopis juliflora. Figure 7.9. The change in Prosopis area in the Wadi Hadhramaut from 1998 (top) to 2001 (bottom), This image was generated using ArcView and Landsat imageries of 1998 and 2001.The total area of this tree has been increased from about 1000 ha in 1998 to 1654 ha in 2001. The Prosopis area is intensive in the middle of Wadi Hadhramaut and south Tarim. These areas are located near large government farms where intensive agriculture based on introduced crops and animal farming are practiced, this may indicate that Prosopis juliflora has started expanding from this place towards east and west mainly by means of domestic animals. The Prosopis area was identified by visual interpretation combined with field observations on false colour landsat imageries, using bands 2, 3 and 4 (BGR). In the study area, pseudo-savanna-like vegetation in wadi beds and runnels in several locations, has been transformed to Prosopis woodlands due to the encroachment and establishment of the invasive tree Prosopis juliflora. Pure stands of Prosopis juliflora woodlands are common now in the river beds of the main Wadi Hadhramaut. Many fields are invaded by this tree (Plate 7.14), which forms thickets on marginal lands, sandy, fallow lands and along road sides in many locations along the main wadi Hadhramaut. Because of its ability to invade lands it reduce the number of local plant species of areas, inhabits and competes with all indigenous species and covers many parts. Generally this invasive plant threatens to decimate indigenous vegetation and to reduce the plant diversity and the populations of many species in the main wadis. 340 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen However, this species is now one of the major sources of fuelwood, bees forage and fodder. Local people feed their animals with the pods of the Prosopis juliflora and nomads come with their herds from areas like Al Masila (eastern part of the study area) or from southern coastal areas to Wadi Hadhramaut to graze on the Prosopis juliflora pods or even to buy these pods from the farmers who are owning the lands that invaded by this tree and taken them to their areas. Awareness of the people living in the study area is important in order to draw out strategies in controlling this exotic species It is understood from previous findings that plant biodiversity is being affected in many parts of the study area as a result of human activities, especially over-grazing, clearing for road construction, and plant collections. Human activities and overgrazing have resulted in the destruction of the natural vegetation and changing the vegetation structure in many parts of the study area. Over-grazing and plant collection selectively affected the abundance of some species, in particular Merremia hadramautica, Boerhavia elegans and Tribulus arabicus. Livestock grazing is the main factor of the spread of the exotic plant Prosopis juliflora. The vegetation of the study area includes some plant species which are used by local people as food, medicinal, firewood and forage. Many of these plants are target and collect almost daily. For example, the continuous gathering of Boerhavia elegans, Corallocarpus glomeruliflorus and Rhazya stricta make these plants vulnerable. The IUCN Red Data Book, presented information about threatened plant species from different countries, among these is Dracaena serrulata, which is rare in the Hadhramaut region and declining rapidly. This plant is becoming rare or endangered over most of its range. It is likely that once it was a widespread tree on the fog-affected slopes of the Hadhramaut mountains. It is now threatened by over-grazing, preventing regeneration, harvesting of the leaves for rope, mats, baskets and possibly by a gradual drying out of the region. Roads and other human activities especially on the plateaus have affected vegetation cover. Many sites have been eroded and some species on eroded slopes have disappeared due 341 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen to road constriction by oil working activities. Oil working activities in recent years resulted in increase of road construction and pipeline establishment. These activities caused the removal of soil surfaces and vegetation from vast areas in particular previous inaccessible areas and accelerated soil erosion and plant destruction. Examples of plants that may be affected are Aerva artemisioides, Monsonia hetotropoides, Fagonia hadramautica, Cucumis canoxyi and Cleome albescens. The impact of waste water from oil working activities on vegetation is also very obvious. Thus, as a result of human activities and other disturbances, the unique ecosystem of the study area that was covered by native species is being replaced by species that are widespread and adapted to disturbed habitats such as Prosopis juliflora, Calotropis procera and Zygophyllum album. 7.3.5. Soil Other factors, such as soil fertility, may also affect the patterns of species richness and plant growth along elevation gradient (Vetaas and Grytnes, 2002). In arid regions, however, the soil is largely chemically homogenous, usually without nutrient deficiencies. Soil development is slow due to the limited pace of weathering and leaching of the soil profile (FAO, 1989.). Physical properties on the other hand, notably soil depth, texture and water holding capacity, can be highly variable. Overall, the chemical composition and physical characteristics of the soil parent material can be just as important as precipitation, wind speed and other climate factors (Cooke and Warren, 1973). For example, soil texture especially in arid and semi-arid regions influences the water balance, evapotranspiration and soil moisture infiltration and consequently, affects vegetation distribution and growth (Hoekstra and Shachak, 1999). Because evapotranspiration and infiltration are higher in sandy soils, the plants in these habitats depend on their roots systems. Some species in these habitats (e.g. Panicum turgidum, Dipterygium glaucum and Zygophyllum album) have very long roots which can spread over the surface as well as into a depth utilizing the available surface moisture and water that has infiltrated to deep layers. 342 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen In arid regions, the quantity of rain is less important than the amount of water remaining in the soil after evaporation, and thus available to plants. Coarser soils hold more water, so unlike a humid environment, clay soils are actually the driest environments in arid regions. (FAO, 1989). Sandy textured soil and loamy fine sand have high permeability and coarse texture; this type of soil is easily removed by wind erosion especially when the vegetation cover is removed. Sandy loam and loamy textures are dominant in the study area, and substrate moisture and variations in soil (rooting) depth are major factors. Because no soil data from all sample points has been taken, the results can be considered as first attempt to discuss the relationship between the plant species and soil factors in this region. Generally, the species of the cultivated fields are negatively correlated with CaCO3, which is often present in excess for plant growth, while the native species of the main wadi bed, the plateau and the slope of the plateau are positively correlated with CaCO3 and negatively correlated with EC% and, perversely, soil moisture. Some species such as Zygophyllum decumbens, Halothamnus bottae and Tamarix arabica resist the high content of EC%. Tephrosia apollinea, Dichanthium insculptum, Pulicaria undulate, Farsetia linearis, Portulaca oleracea and Corallocarpus glomeruliflorus grow well on relatively moisture soil sites. Chascanum marrubifolium, Cadaba heterotricha, Tarenna graveolens, Jatropha spinosa, Maerua crassifolia, Hochstetteria schimperi, Commiphora kua and Grewia erythraea grow well on dry stony shallow steep slopes of the plateau. The species Aristida triticoides grows well on dry, stony, steep north-east slopes. Ficus cordata subsp. salicifolia, Cadaba heterotricha, Tarenna graveolens and Leptadenia arborea grow well on wet rocky shallow slopes and along watercourses. Acacia campoptila occurs on deep soil, especially in the main wadis, where Stipagrostis hirtigluma and Dichanthium insculptum abundant on stony slopes and plateaus. Some species prefer the colluvial, very rocky fans and footslopes, which are distinct from other landforms by having high moisture, organic carbon, and other minerals than the surrounding landforms (Furley, 1974), example of these species are Tephrosia apollinea, Cymbopogon schoenanthus, Barleria aff. bispinosa and Merremia hadramautica, or prefer sandy plain and sand dune areas, such as Acacia ehrenbergiana, Panicum turgidum and Dipterygium glaucum. 343 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen The vegetation cover is sparser on the plateaus compared with other landforms. This is probably due to the effect of the thin limestone plateau soils which dry rapidly and are affected by sweeping winds (Plate 7.15). Generally most of the plant species are well adapted to calcareous soils which occur widely in the study area, and this type of soil is also widespread in some limestone mountain desert areas of the world such as Mule Mountains of Arizona. (Wentworth, 1981). Plate 7.15. A close-up view of the plateau pavement (exposed flat surface) shows that the gaps between rock fragments are small or rarely visible. Wind and periodic rains keep the surface free of dust, and plants have a difficult time becoming established due to lack of soil. Most of the rock fragments shown here is limestone. 344 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen It is understood from previous sections that:(1) The plant species that settle in the study area through long distance dispersal are distributed in specific habitats because of their differential response to phytogeographical features of the ground, soil texture, micro-climate especially moisture and human destruction. (2) Some obvious phenomena of human impact on natural vegetation can also be seen in the study area for example: Decreasing areas of Acacia campoptila, Salvadora persica and Tamarix aphylla woodlands with the simultaneous increasing of Prosopis juliflora woodland,  Decreasing density and frequency of some species such as Boerhavia elegans and Rhazya stricta due to over grazing and intensive collection. It is suggested that future management activities should consider the above problems more seriously. (3) The extensive oil working activities beginning at mid-1980 have made an intensive impact on vegetation and landscape distribution, and the effect can be seen clearly on the plateau. (4) The introduction of modern agricultural techniques from 1980 were the primary causes of the significant variation for vegetation distribution pattern along the different topographic areas. (5) Floods and over grazing played and still play as an important role in destruction, changing the structure and the composition of the vegetation in the study area. 7.4. Conservation and management of biodiversity In the past the people of the Hadhramaut lived more or less in harmony with their environment, but with increasing population, changes in life expectations and oil activities, the ecosystem and its natural resources have come under threat. It is clear that the impact of climate change and human activities on plant species composition will increase in the coming decades. Many species will have difficulty in responding to the rapid changes in climate (too fast for adaptation or migration) and are likely to become more restricted in their distribution. Already, the development of the oil fields has had a rapid and fundamental impact on the study area. Direct effects include the removal of plant cover and soil erosion on the plateau. 345 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 7.4.1- Conservation: traditional knowledge and land ownership issues Knowledge about the local and traditional uses of plant species (for instance, their importance as a source of food, for fodder, medicine, essential oils etc.) is vitally important in planning conservation strategies. However, this knowledge, often linked to local languages and dialects, is rapidly disappearing. It is often only old people who still retain this knowledge. It is no longer being passed on to the younger generation. Furthermore, local people and landowners play a vital role in the management of natural resources by following traditional and management systems. Therefore, the full cooperation and involvement of local people, communities, decision makers, students and local leaders plays a vitally important role in any conservation management plans. Large areas of Hadhramaut region are privately owned and the rapid increase in population is creating an increased demand for land. It is clear that landowners must be involved in any future management plans for the conservation and protection of the ecosystems of the region. Local people also need technical and financial assistance and education programs on land management and the conservation of habitats and plant species is an important aspect of any conservation programme. Financial and human resources are the main problems facing Yemeni Authorities, such as the Environmental Protection Authority, which are currently responsible for protected area management. Issues over land ownership and tenure can be a problem; most of the proposed protected areas belong to farmers, local people or tribes. Other problems are the need for trained staff, funding, proper conservation management and currently, poor infrastructure. As a result of these and other problems the protected areas remain no more than a list on paper. 346 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 7.4.2- Types of Threats The main threats to the vegetation in Hadhramaut are from the direct and indirect impacts of human activities. In general, these relate to changes in the traditional land management practices which in the past have protected the vegetation. In particular the vegetation in the region has been directly and indirectly changed by the recent rapid economic growth. Direct threats of development include: 1 The taking of land for building; this is widespread on the coastal plain where it will potentially effect important coastal communities such as at the beach east of al Mukala. The vegetation here is dominated by Aristolochia rigida, Cleome macradenia and Limoniastrum arabicum – the latter two are endemic and near-endemic to this particular habitat and of very limited distribution. 1. The building of new roads and pipelines. These affect the vegetation in various ways: a. by changing watersheds and creating micro-niches along the roadsides (see Plates 7.9a and b an Figure7.3.). It would be interesting to see if in future these act as wildlife corridors. b. in some areas opening up areas previously inaccessible to grazing; in other areas, the oil companies, for security reasons, are excluding transhumant pastoralists from some of their traditional pastures. 2. The pollution of the soil near oil installations. 3. The spreading of waste water; this has in some areas led to an explosion of plants; initially with high diversity, but it is expected that this will reduce as the area is overtaken by rank species. Also, it is expected that the build up of salts from the irrigation will gradually have an effect (see Plate 7.13). Changes in traditional land management practices and traditions: In direct threats of development include: 347 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 1. Change in grazing and browsing practices leading to changes in the vegetation. There are various examples of this: a. In certain areas access to traditional rangelands has been prohibited, for security reasons, on the plateaus around the new oil installations. This has fragmented the plateau into grazed and un-grazed localities. The obvious effect of this is to take pressure off from certain rangelands therefore allowing them to recover and ultimately to change. However, the pastoralist still needs to graze his animals and will move them into new areas. The effect of this can be seen in study site 3 where transhumant pastoralists who traditionally moved over large areas are now settling for longer periods in areas in the main wadis. b. Children who at one time herded animals now go to school. There are direct and indirect consequences of this, for example:. i. Directly, there has been a change in the practice of grazing on the rangelands, which has resulted in selling the animals or stopping grazing outside. ii. Indirectly, and perhaps more importantly, children lose a connection with their environment –traditional practices will no longer be passed on to them from their parents and understanding of them will be rapidly lost. In other areas where this has happened they become separated from their natural environment, no longer place a value on it and finally see no point in conserving it (Miller & Morris, 1985). 2. Changes in traditional practices have opened up habitats which can then be rapidly exploited by exotic species. The best example of this is the case of Prosopis juliflora. This is rapidly invading the main wadis and coastal areas. It is a serious pest of cultivated land and is, in many places swamping the natural vegetation in the main wadis. However, it is not all bad: some people welcome it as a major source of nectar and pollen for bee keepers (one of the most profitable and famous industries in the region) and it also provides fuelwood. 3. As people become settled they place more pressure on their immediate environment; their livestock grazes near the home and gathering of wood for fuelwood and timber increases. 348 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen On the whole the effect of oil exploitation in the area is complex. The most immediate effects are from pollution and waste water; these are very local and pose no major threat to the vegetation. Next in importance are changes to the vegetation caused by the creation of new ecological niches; for instance along roadsides; these again are not too dramatic and it is to be seen whether they will act as wildlife corridors. Of more importance are the affects of restricting access to traditional rangelands – the effect of this are twofold: decreasing grazing in some areas and increasing pressure on others; both outcomes effect the vegetation. Finally, and perhaps most importantly, are the radical and complex ways in which the new wealth and development is changing the area – everything from children no longer taking livestock into the range but now going to school and potentially losing their links with the environment; to major changes in grazing practices as age old practices of transhumance breakdown as people settle. All the various scenarios described above have been observed and recorded during the course of this study; to unravel simple cause and effect in such a complex situation has proven impossible. So what are the observed changes in vegetation, again, it is difficult to separate anthropegenic and natural processes. The best example of this is the changes to the vegetation following the devastating floods which effect the main wadis from time to time. It can be seen that they are becoming more destructive. Many large trees and areas of woodland have been destroyed in recent times. This increase in the severity of the floods could be caused by changes in rainfall bought on by gradual climate change or global warming. However, other factors have to be taken into account:  Changes in land-use – including the removal of the protective layer of vegetation by  over-grazing is leading to flood plains becoming more vulnerable to erosion  flash floods  thus slowing the flow of the wadis Over-grazing on the water catchment of the area increasing the run-off and causing Changes in the practice of by placing barriers in the wadis to divert water to the fields An increase in the production of cement for building – this is removing lime-rich substrates in the wadis and changing the contours of the flood plains. 349 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen It can be seen that there is no simple cause for the increase in severity of the floods; it is a complex system involving many factors – both natural and anthropogenic. 7.4.3. National and International Framework for Protection of Yemen’s Flora and vegetation In Yemen, the main responsibility for wildlife conservation and environmental protection lies with the Environment Protection Authority (EPA). Miller (1994) reported that UNEP and IUCN had recommended that a network of reserves be created throughout Yemen. To date no such network has been established. A few sites have been identified as candidates for protected area status, but, until now, with the notable exceptions of Otuma, in the western mountains, Jabal Bura’, a species rich area of relict woodland in the western escarpment mountains and the monsoon woodlands of the Huf area bordering Oman, none of these sites have any legal protection. Most recently the National Biodiversity Strategy development process of Yemen (EPA, 2005) has been funded by the Global Environment Facility (GEF). The work of the technical groups involved in this are coordinated by the Environment Protection Authority (EPA) which is based in Sanaa. The National Biodiversity Strategy consists of 4 goals: 1. conservation of natural resources 2. sustainable use of natural resources 3. integrating biodiversity in sectoral development plans 4. and implementation of enabling mechanism. The conservation of natural resources covers the following areas:1. Protected areas 2. Endemic and endangered Species 3. Ex situ conservation 350 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 4. Alien invasive species 1. Protected Areas (in situ conservation) The conservation of species in their natural habitats or in situ conservation, has long been the conservation method of choice for wild species and ecosystems (OECD, 2003). It involves the process of protection and conservation of ecosystems and natural habitats and the maintenance and recovery of viable populations of species in their natural habitat (Moore and Tymowski, 2004). Criteria and priorities that attract funds are necessary for conservation (Spellerberg, 1992). The main goal of in situ protection is the conservation of eco-systems through developing and maintaining a comprehensive and adequate network of protected areas. Some important areas have already been proposed by scientists and at workshops (Al Khulaidi and El-Ghouri, 1996; Miller et al, 2006). Amongst these areas are the Jol plateau in the Hadhramaut – to date this has no protection. In 1999 Otuma, in the Western part of Yemen in the province of Dhamar, was the first locality to be declared a protected area in Yemen. Two more will be declared as protected areas shortly (UNEP, 2004), these area are Jabal Bura’ on the SW escarpment mountains and Huf, in the Al Mahara Governorate in the eastern part of the country. These protected areas fall within IUCN Protected Area Category II, under which the area is to be managed mainly for ecosystem protection and recreation (IUCN, 2004, Anon, 1994). Generally, most of the proposed protected areas are managed by local communities, village leaders or Sheikhs, this is done by using traditional management practices. The implementation and details of rules governing the management of these differs from one place to another. These protected areas represent the most successfully managed sites in Yemen. The National Biodiversity Strategy Development Process in Yemen has recently been strengthened by the Global Strategy for Plant Conservation (GSPC), a global initiative, which could have an important impact on conservation in Yemen. Countries, such as Yemen, which 351 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen are signatories to the Convention on Biological Diversity (CBD) are now also obliged to adhere to the GSPC. The ultimate and long-term objective is to halt the current and continuing loss of plant diversity. Under the GSPC, a series of global targets has to be met by the year year 2010. The present studies will directly help the government of Yemen in reaching some of these targets and thus fulfilling its international commitments. The following three targets are most directly relevant: Target (i) a widely accessible working list of known plant species. Target (ii) a preliminary assessment of the conservation status of all known plant species, at national, regional and international levels. Target (v) protection of 50% of the most important areas for plant conservation assured – the most Important Plant Areas (IPAs) for plant diversity are to be identified according to the criteria including endemism, species richness, and/or uniqueness of habitats, including relict ecosystems. The GSPC is being coordinated in the countries of the Arabian Peninsula by the Arabian Plant Specialist Group (APSG), a Specialist Group of the Species Survival Commission of the IUCN, and consisting of a group of botanists with an interest in the region. The results of this present study will feed directly into the GSPC process, through the APSG. 352 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 2. Endemic and Endangered Species Endemism is a key component of biodiversity that particularly interests biologists and plant taxonomists (Shevock, 1996; Kruckeberg and Rabinowitz, 1985). Plants can be endemic to all kinds of features and geographic areas, and also to physical attributes such as soils and rock types. Endemism is also an important concept in conservation biology. Endemic species rely exclusively for their long-term health and continued existence on the management of the geographical area to which they are restricted. Furthermore, endemism is one of the criteria used to set priorities for species conservation efforts. Endangered species are a widely accepted favourites for conservation attention both nationally and globally and are frequently afforded high priority (Heywood and Dulloo, 2005; Sutherland, 2001). Endemic and nearendemic species that occur in restricted areas present particular problems. Their very rarity (in term of their limited distribution) is part of their vulnerability and turns them into key species in the conservation of biodiversity. The flora of the Hadhramaut is very rich in endemic, near-endemic and rare plant species. The majority of endemic taxa in the region are associated with the southern mountainous area which provides a rich variety of ecological niches (see chapter 3). The endemic plants of this region include 103 endemic and near-endemic species. A summary of the status and number of endemic, near-endemic and rare plant species in Hadhramaut has been presented in Table 3, Chapter 3). The southern mountain summits of the Hadhramaut, especially above Al Mukala and Al Rayan, can be considered a local centre of endemism. This region is characterised by a unique ecosystem (the summits, for instance, obtaining most precipitation from fog) with a very low population, isolated geographical location, dramatic landforms and an extensive range of altitude and climatic conditions. There is an urgent need to protect this unique ecosystem and its rich biodiversity. However, not only biological factors have to be taken into consideration. It is of vital importance for the long-term protection of the ecosystem that the people, who for generations have lived and preserved it, should be consulted. Any management plan for the 353 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen area must take into account that, it is largely a man-made landscape, and that for millennia the region has provided a sustainable livelihood for local people. 3- Ex situ Conservation The conservation of species outside their natural habitats or ex situ conservation typically is found in zoos, captive breeding programmes, botanic gardens and seed banks (Williams et al., 2001; Anderson et al., 1996). The selection of threatened species for ex situ management is guided by a variety of criteria (Havens, et al., 200; Heywood and Dulloo, 2005), these are: a. Endemism (selecting taxa that represent a unique local or regional fancies) b. Economic (selecting taxa that provide local or economic or social resources, such as medicinal plants), c. Ecological (selecting taxa that have a role in maintaining ecological processes or supporting habitat restoration), and d. Characteristic (selecting threatened taxa that can be used as flagships for promoting landscape-and habitat-level conservation). Clearly the ideal for ex situ conservation is one where representatives from all ecological zones can be grown. In Hadhramaut a combination of the following sites would achieve this:  A site on the plateau above Al Mukalla; this area has a geographical location, topographical features and climate suitable for growing a wide range of species from Hadhramaut, including plants of both the tropical and arid zones.  The Agricultural Research and Extension Authority (AREA), regional station at Sayun. Here the plants of the Wadi Hadhramaut and surrounding areas can be grow successfully, as well as desert plants from the Ramalat Assaba’tain and Al Rub’ al Khali. 354 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 5- Alien invasive species Human activities in the study area, have allowed invasive species, such as Prosopis juliflora, to rapidly spread with a resultant detrimental effect on the abundance of indigenous species. This species becomes the dominant species in many parts of the main wadis forming a pure stand of woodland. Pure stands of Prosopis juliflora woodlands are common now in the river beds of the main Wadi Hadhramaut and neglected agricultural land. Other invasive species on cultivated areas include: Cuscuta campestris, and Leptadenia arborea. Currently, local people are complaining about the negative impact of these species. Therefore, the full cooperation and involvement of local people, communities, decision makers, students and local leaders plays an essentially important role in any future activities directed to solve and control this problem. 355 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 7.5. Conclusions Primarily, this study has improved the understanding of the distribution and ecology of plant taxa in the Hadhramaut. However, it also emphasises that vast areas in this remarkable region are still botanically un-explored. The research is the first detailed vegetation survey to be completed in the Hadhramaut and provides data which can be used as a baseline for monitoring change. The methods and protocols developed during the study can be used as a basis for carrying out similar studies in other parts of Arabia and for helping to devise management and conservation programmes. In the study area, the major vegetation associations, their composition and biodiversity were identified and vegetation and landuse maps generated. Endemic, near-endemic, rare and threatened plant species were identified and presented in maps. These data are important in understanding the direct and indirect effects of oil development on the vegetation and environment of the region. This research has also been important identifying significant vegetation associations, habitats and areas of endemism thus allowing Yemen to fulfil its international commitments under the Global Strategy for Plant Conservation and in particular in reaching targets I, II and V for the year 2010. Such information will also feed into IUCN assessments of threat to endemic and near endemics by supporting certain criteria (habitat change etc). The data is also supporting the development of management plans and is relevant in supporting ecotourism and consequently wealth creation within the region, thus providing alternative employment for pastoralists. Although soil data was not collected from all sampling points, the study has assessed topographic and micro-climatic factors and is the first attempt to discuss the relationship between the plant species and environmental factors in the region. The effects of human intervention on soil in the study area may not yet be visible in such a short time-scale, but studying the region has allowed us to draw certain conclusions about trends that will help in interpreting any further problems which may arise. 356 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen It has been shown that the distribution and structure of vegetation associations show a relationship to topographic position and related factors such as soil depth and available moisture. The result of Braun-Blanquet and Twinspan analysis revealed 15 vegetation associations and demonstrated several distinct patterns of species distribution: 1. species found over wide range of ecological sites 2. species distributed mainly on wadi beds 3. species only found on slopes 4. species only seen on plateaus Due to the varied topography over the study area, several important microhabitats were identified, including runnels, wet sites, and run-off areas at the bottom of slopes. Each of these microhabitats supports special types of vegetation, each with a characteristic floristic composition and distinct physiognomy. It has been shown that plant migration during past geological periods, long-term geological and climatic change, recent changes in climate and hydrology and the effects of man and his livestock have all played a part in the establishment and development of the present flora and vegetation of Hadhramaut. It is clear, because of the early development of agriculture and pastoralism in Hadhramaut that no area can be considered to be “natural”; all to a lesser or greater degree shows the impact of man and his livestock. Changes in vegetation have to be considered in this context. Until recently there was no way of directly examining this. However, the study of Hyrax middens (see chapter 2), although in its infancy is likely to revolutionise the understanding of recent vegetation change in the region. Already, from a study of findings from surrounding regions it has been shown that the vegetation and flora of the Hadhramaut has been subjected to alternating periods of high and low rainfall. During periods of high rainfall plants of paleotropical ancestry migrated north, using the main wadis as corridors. During dry periods “desert” elements, the desert transition 357 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen zone of the Saharo-Sindian (a mixture of holarctic and paleotropical elements which have previously adapted to the arid condition) have migrated back into the wadis and up on to the plateaus. With each alternation of conditions, palaeotropical and holarctic elements would have been “left behind” in refugia, now represented, for instance, by the relict woodlands found in gullies on the plateau. An analysis of the floristic elements in the Hadramaut region shows that the geographic boundary between the floras of tropical African origin to the south and the sub-tropical and temperate floras of the north lies somewhere across the middle of the study region. However, this boundary is blurred by the constant mixing of elements which has occurred in the region over the millennia. This explains the high number of endemics and the interesting floristic mix found in the area. The study of past climatic change allows us to predict how global warming will alter the composition and distribution of vegetation in the future. The results of the present study should help in planning of economic, agricultural and conservation strategies in an uncertain future. The southern mountains of Hadhramaut with their highly localized endemics are particularly vulnerable to the present rapid and uncontrolled development in the region. In particular oil-related activities, may have a serious impact on these species. On the basis of the findings of this study a recommendation will be made to the Environmental protection Authority that urgent action needs to be taken to protect this area and that priority should be given to the development of a conservation management plan. It is clear that Hadhramaut is an area of great beauty with dramatic natural landscapes and picturesque desert cities. It is one of the most popular tourist destinations in Yemen. At the moment there is little “natural” or eco-tourism which could bring extra income and jobs when the oil runs out. It is important that protection of the environment and biodiversity is put on the political and economic agenda as soon as possible. So has this research matched the objectives and answered the questions posed at the beginning of this thesis (chapter 1) 358 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 1. “The principal objectives of the research are to analyse the distribution and dynamics of the plants of the Hadhramaut region, and to evaluate the role of physical parameters and human actions on the vegetation. The area selected for detailed study is representative of the Hadhramaut as a whole and is further subdivided into 3 sites for intensive survey” (chapter 1). The main objectives of this research were substantially achieved. The present distributions of all plants in the region have been analysed and the dynamics of their distribution discussed, including the importance of ancient floristic links, Quaternary migrations, or the more recent climate change and breakdown in traditional land management practices. The results of this study will serve as baseline against which future changes in the distribution plants in the region can be measured. 1. “The number and abundance of individual plant species has varied over time, especially over the past two decades.” It has been clearly shown that the number and abundance of species has changed over the last two decades. However, the reasons for this are complex - with the main factors including: environmental change (e.g. global warming), economic development (e.g. pollution and changing niches along road sides) changes in traditional practices (e.g. changing in grazing patterns), human activities have allowed invasive species, such as Prosopis juliflora, to rapidly spread with a resultant detrimental effect on the abundance of indigenous species. This species becomes the dominant species in many parts of main wadis forming a pure stand of woodland. Due to intensive floods, there is a very obvious vegetation transition from one form to another. Floods have a significant effect on the structure and the abundance of vegetation of wadis. Roads on the plateaus have affected vegetation cover. Many sites have been eroded and some species on eroded slopes have disappeared due to road construction, for example, species such as Seetzenia lanata, and Monsonia heterotropoides become very rare in such conditions. 359 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 2. “Plant communities and vegetation structure have remained relatively constant over the same period, although the populations of different plant species and their local distributions have changed”. In some areas there have been changes in the abundance of certain species as a result of selective grazing etc. It is also clear that the area has seen innumerable migrations over the millennia (witnessed by the mix of tropical and holactic elements); however, evidence for recent change (over two decades) is not so clear. Remote sensing imagery has shown the changes brought on by the invasion of Prosopis juliflora and change in land use in the wadis. It also shows changes to the extent of natural vegetation. Analysis of the survey sites has also shown more subtle changes to the vegetation between the sites. The most clear changes have been seen close to oil installations, for instance, where waste water has been dumped or where land has been cleared. The impact of waste water on vegetation in some areas seems more positive, since the release of water has produced vigorous germination and growth of plant species and as a result the population of plant species has increased. Floods have a significant effect on the structure of wadi vegetation and contribute largely to the destruction of individuals and the reduction of vegetation cover, Furthermore this has resulted in the introduction of other plant species especially, exotics, which may be brought in by floods from other areas, examples are Prosopis juliflora and Zygophyllum album. As a result of severe floods, the woody vegetation dominated by Ziziphus leucodermis, Acacia ehrenbergiana, Acacia hamulosa, Tephrosia dura and date palm trees has disappeared or shrunk in coverage and has been replaced by other communities more resistant to the severe floods and other environmental disturbances. 3. “Human activity has exploited the plant resources of the entire area but has preferentially targeted particular types of vegetation.” The present changes to the area brought on by the recent changes in economic development have not caused any major changes to the vegetation of the region. The area has a long history of human occupation and certain vegetation types, for example, woodland is either long gone or has been carefully managed. There is clear evidence of overgrazing in the 360 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen wadis (but not in all study sites), the collection of firewood and timber from certain trees and the unsustainable exploitation of certain medicinal and culinary plants (eg Boerhavia elegans). There has been little direct targeting of the natural resources on the scale of, for instance, the clearing of the tropical rain or any seasonal forests. Generally, the adverse effects of human activity have been less direct – mainly through the breakdown of traditional practices. However, these changes although small are likely to have a considerable impact on the fragile desertic ecosystems across the region. Over grazing has a direct impact on species richness in different parts of the study sites. The occurrence of some species such as Tephrosia dura could be a sign of over-grazing. Human activities in the study area targeting particular plants, for example the collection of some plants for medicinal or other purposes also has an impact on certain species, such as Boerhavia elegans, Corallocarpus glomeruliflorus, and Rhazya stricta. These activities have resulted in reducing the abundance of these species. With the arrival of nomads, grazing become more intensive in areas previously less affected. As a result, some species that were abundant become rare, examples include Merremia hadramautica, Tephrosia nubica, Maerua crassifolia and Panicum turgidum. Future research Further study on the factors that influence the distribution of plant species of Hadhramaut region and the possible effects of climate change on the distribution patterns are needed. It would be helpful if vegetation surveys could be combined with data or abiotic factors such as altitude, aspect, soil, moisture and other environmental factors and biotic factors such as human and other living organisms to get a better and clear idea on the distribution, abundance and composition of plants in particular rare, endemic and nearendemic species. Detailed vegetation as well as topography and landuse mapping of the whole region of the Hadhramaut using the modern softwares such as ArcView and ERDAS is recommended. In addition, several lines of investigation would be helpful for policy formulation and management, these include: assess the available information on the existing endemic, near- 361 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen endemic and rare plant species; the conducting of future targeted vegetation surveys to estimate the conservation status, population size and area of occupation of each of the endemic and threatened plant species and threatened habitats in the region; the .preparation of an IUCN red list of threatened plant species of the region and a list of Important Plant Areas; the conservation and rehabilitation of key endangered species through law enforcement; the collection of ethnobotanical data – before it is too late and, finally, the preparation of well grounded Recommendations to government on protecting key areas. The research has provided the first quantitative and systematic survey of the vegetation of the Hadhramaut. It has shown that the area lies at the transition between northern (Holarctic) and south western (Paleotropical) floras and also contains very few elements of eastern Irano-Turanian vegetation zones. The area has been shown to possess a number of fascinating endemic and near-endemic plant species and, together with their distinctive landscape habitats, the data collected justify the proposal for protective status for parts of the region through the formation of a protected lands or national park or other type of locally and nationally accepted conservation management. 362 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen References Abd El-Fattah, R.I. and A.M. Dahmash. 2002. Plant and soil relationship in the nort-eastern desert, Egypt. Egyptian J. Desert Res., 52(1): 1-20. Abdul El-Ghani, M. 2000. Vegetation composition of Egyptian inland salt marshes. Botl. Bull. Acad. Sin. 41: 305-314. Abdul El-Ghani, M. 2000. Vegetation of Egyptian inland salt marshes. Bot. Bull. Acad. Sin. 41:305-314. AbuZinada A. H. 2001. National Report on the Convention on Biological Diversity. The National Commission for Wildlife Conservation and Development. Ackermanna, O., A. M. Maeira and H. J. Bruinsb 2004.Unique human-made catenary changes and their effect on soil and vegetation in the semi-arid Mediterranean zone: a case study on Sarcopoterium spinosum distribution near Tell Es-Sofi/Gath, Israel. Catena 57: 309-30. Adams J. 2002. African and Eurasia during the last 150000 years. Environmental Science Division, Oak Ridge National Laboratory, USA. Adler, P.B., D.A.Raff and W.K.Lauenroth 2001. The effect of grazimg on the spatial heterogeneity of vegetation. Occologia, 128:465-479. USA. Agwu, C.O.C. & H.J. Beug 1982 Palynological studies of marine sediments of the West African coast. Meteor-Forschungsergebnisse C 36: 1-30. AI-Gifri, A.N. & H. Kurschner 1996. First records of bryophytes from Hadramout and Abyan, Yemen. Nova Hedwigia 62:137-148. Al Gifri, A.N. and Gabali 1999. The present status of floristic knowledge in the Republic of Yemen. A paper submitted to the workshop on the conservation of the flora of the Arabian Peninsula. National Commission for wildlife conservation and development and IUCN. Al Hubaishi, A and K. Muller-Hohenstein 1984. An introduction to the vegetation of Yemen. GTZ. Al Khulaidi, A.A. & M. El-Ghouri 1996. Main Natural vegetation and protected areas in Yemen. A workshop on natural areas and biodivesity conservation in Yemen. EPC & MAWR, Sanaa, Yemen. Al Khulaidi, A.A. 1989. A comparative vegetation survey of four photographic regions in the Y.A.R. unpublished MSc thesis, ITC, Enschede, the Netherlands. Al Khulaidi, A.A. 1992: Ecological vegetation survey of three areas located Southern Tihama - Technical report 1990/91, AREA, regional station, Taiz, Yemen. Al Khulaidi, A.A. 1992: Vegetation of Taiz – Yemen. journal of Agriculture research university of Aden 2: 173-192. Yemen. Al Khulaidi, A.A. 1993: Natural vegetation of Kuhlan Affar area -IDRC/ICARDA and AREA, Sanaa, Yemen. Al Khulaidi, A.A. 1996. A vegetation survey of wadi Rimaa and Sharis catchments areas, Yemen. Land and water conservation project, forestry component, (UTF/YEM/023) (IDA Credit TEM), Sanaa, Yemen. Al Khulaidi, A.A. 2000. Flora of Yemen. (a check list). Sustainable Environmental Management Program (SEMP), YEM/97/100, sub-programm 11 and AREA, Sana’a, Yemen. 363 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Al Mashreki, M.H. 1999. The soils of the south-eastern coastal area. Part 2. The soil of upper Hajar Valley. Field document for GCP/YEM/021NET Project FAO, Yemen. Al Mashreki, M.H. 2005. The Soil of Yemen. RNRRC-AREA. Dhamar, Yemen. Al Wadie, H. 2002. Floristic composition and vegetation of Wadi Talha, Aseer mountains, south west Saudi Arabia. Online Journal of Biological Sciences 2(5): 285-288. Al-Turki, T.A. 2004. A prelude to the study of the flora of Jabal Fayfa in Saudi Arabia, Kuwait J. Sci. Eng. 31(2):77-145. Alwan, A.K. 1990. Production and economic efficiency of the agricultural resources in state farms at Wadi Hadhramaut. A thesis for Master of Science. Ain Shams University, Egypt. Anderson L S, C E Davies and D Moss (1996). The UN convention on biological diversity. Follow-up in EEA Member Countries. A report prepared under the supervision of U Pinborg Project Manager, European Environment Agency. Anderson, H. 1997. History of Climate change. UNEP. Angood, S. 2000. The place of ancient agricultural practices and techniques in Yemen today: problems and perspectives. Department of Plant Protection- Nasir's College of Agriculture- University of Aden. Anon, 1994; Guidelines for Protected Area Management Categories, IUCN and the World Conservation Monitoring Centre, Gland, Switzerland and Cambridge, UK. Anthony, K.R.M., A.M. Khaled and J.E.A. Ogborn 1963. Factors affecting crop production in the Wadi Hadhramaut. Empire Journal of Experimental Agriculture 31(122). Apio, A. and T. Wronski 2005. Foraging behaviour and diet composition of bushbuck (Tragelaphus scriptus Pallas, 1766) in Queen Elizabeth National Park, western Uganda. African Journal of Ecology, 43(3):225–232. AREA, 1999. Agro-Climatic Resources of Yemen. Agricultural Research and Extension Authority, Dhamar, Yemen. Atiqur Rahman M., M. S. Al-Said, J. S. Mossa, M. A. Al-Yahya and 1M.A. Fahad Al-Hema 2002. A Check List of Angiosperm Flora of Farasan Islands, Kingdom of Saudi Arabia. Pakistan Journal of Biological Sciences 5 (11): 1162-1166. Awadelkarem, H.A. 1997. Some guidelines for improved farm management in Wadi Hadhramout. FAO/W. Hadhramout Agricultural Development Project. UTF/YEM/019. Ayele, T. and S.A. Al Shadily 2000. Some of the Engineering Geological and Hydrogeological Problems and Conditions of Ethiopia and Yemen. Acta Geologica Universitatis Comenianae 55:51 – 62. Balba, M. 1995. Management of Problem Soils in Arid Ecosystems CRC Press. Inc. Ballantine, A. J., G. S. Okin, D. E. Prentiss and D. A. Roberts. 2005. Mapping North African landforms using continental scale unmixing of MODIS imagery. Remote Sensing of Environment 97: 470 – 483. Barkhadle A. M.I., L. Ongaro and S. Pignatti. 1994. Pastoralism and plant cover in the lower Shabelle region, Southern Somalia, Landscape Ecology 9(2): 79-88. Barkuda, Y and N. Sanadiqi (1986). Plant Genetic Resources of some plant species collected between 7 and 25 January 1985 from south Yemen. ACSAD A technical report no 47, Damascus. Bataher, A. S. 1998. The management of Ziziphus spina-christi as a componenet of integrated farming system in Wadi Hadhramout, Yemen. Unpublished Master thesis, Faculty of Forestry, university of Khartoum. 364 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Bataher, A. S. 2004. Vegetation survey of Wadi Al-Khun enclosure (Hadhramaut, Yemen ). ICARDA-APRP annual report 2003-2004 on Sustainable Management of Natural Resources and Improvement of Major production systems of Arabian Peninsula. Batanouny, K.H. 1979. Vegetation along the Jeddah-Mecca road: pattern and process as affected by human impact. Journal Arid Environments 2:21-30. Batanouny, K.H. (1981). Ecology and Flora of Qatar. Alden Press Ltd. Oxford. Batanouny, K.H. 1983. Human impact on desert vegetation. In Holzner, W., Werger, M.J.A. & Ikusima, I. (eds.) .Man’s impact on vegetation. Geobotany (1):139-149. Batanouny, K.H. 2000. Adaptation of Desert Organisms. Springer. pp. 93. Beadle N. C. W. 1953. The Edaphic Factor in Plant Ecology with a Special Note on Soil Phosphates, Ecology 34 (2): 426–428. Becker, T. and S. Getzin. 2000. The fairy circles of Kaokoland (North-West Namibia) origin, distribution, and characteristics. Basic and Applied Ecology 1: 149-159. Beier, B.-A. 2005. A revision of the desert shrub Fagonia (Zygophyllaceae). Systematics and Biodiversity 3(3):221-263. Bell, F. G. 2003.Geological Hazards: Their Assessment, Avoidance and Mitigation. Spon Press (UK). Bell, F. G., J.A.A. Nylander, M.W. Chase and M. Thulin. 2004. Phylogenetic relationships and biogeography of the desert plant genus Fagonia (Zygophyllaceae), inferred by parsimony and Bayesian model averaging. Molecular Phylogenetics and Evolution 33: 91–108. Belnap, J. 2002. Impacts of off-road vehicles on nitrogen cycles in biological soil crusts: resistance in different U.S. deserts. Journal of Arid Environments 52: 155-165. Belsky, A. J. & D. Blumenthal. 1997. Effects of Livestock Grazing on Stand Dynamics and Soils in Upland Forests of the Interior West. Conservation Biology 11(3). BIOLOG, (2004). Biodiversity Monitoring Transect Analysis in Africa. The influence of fragmentation and disturbance on the biodiversity of East African highland rain forests. BIOTA Final Report Phase I. Blatter, E. 1914-1916. Flora of Eden. Rec. Bot. Surv. India, 7(1-3): p. 418. Blatter, E. 1936. Flora Arabia. India: Calcutta, Super. Gov. Printing. Boulos, L. 1988. A contribution to the flora of South Yemen. Candollea 43(2): 549-585. Boxberger, L. 2002. On the Edge of Empire: Hadhramawt, Emigration, and the Indian Ocean, 1880S-1930s. State University of New York Press, 292 pp. Brandes, D. 2004. Gymnocarpos decandrus (Caryophyllaceae) on Fuerteventura. Working Group of Vegetation Ecology. Technical University Braunschweig. Brigham, C. A. and M. W Schwartz (2000). Population Viability in Plants- Conservation, Management, and Modelling of Rare Plants. Springer. Brown, G.M. 2001. Vegetation ecology and biodiversity of degraded desert areas in northeastern Arabia. Final report S0 073/Habilitatationsschriff. Kuwait University. Bryant, J.P. 2004. Biodiversity and conservation, a Hypertext Book. School of Biological Sciences, University of California. Busack, S. D., and R. B. Bury. 1974. Some effects of off-road vehicles and sheep grazing on lizard populations in the Mojave Desert. Biological Conservation 6:179-183. Canfield, R. 1941. Application of the line interception method in sampling range vegetation. Journal of Forestry 39:388-394. 365 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Carmel, Y. & R. Kadmon. 1999. Effects of grazing and topography on long-term vegetation changes in a Mediterranean ecosystem in Israel. Plant Ecology 145: 243–254. CBD, and UNEP 2002. Global Strategy for plant conservation. Secretariat of the Convention on Biological Diversity and Botanic Gardens Conservation International. Central Statistical Organisation. 2001. Hadhramaut in Figures 2000. Sanaa, Yemen. Chao, A., Robin L. C., Robert K. C. and T. Shen. 2000. A new statistical approach for assessing similarity of species composition with incidence and abundance data. Ecology Letters 8:148-159. Cheng, C.C. 2004. Statistical approaches on discriminatory spatial variation of species diversity. Bot. Bull Acad. Sin. 45:339-346. CLASP, (1995). The conservation of lands in Asia and the Pacific. A framework for action. FAO. Claussen M., C. Kubatzki, V. Brovkin and A. Ganopolski 1999. Simulation of an abrupt change in Saharan vegetation in the mid-Holocene. Geophysical Research Letters 1:14. Clayton, W.D., K.T. Harman and H. Williamson 2005. World Grass Species: Descriptions. Royal Botanic Garden, Kew, Version: 8th. Cloudsley, J.L. 1976. Man and the biology of Arid zone. English Language Book Society. Comrie, M.C. and K.M. Glazebrook. 1992. Hydrogeology of the NW Masila Block., Hadhramaut, Yemen, Congalton, R. G. 1988. Using spatial autocorrelation analysis to explore the errors in maps generated from remotely sensed data. Photogrammetric Engineering and Remote Sensing 54:587 - 592. Coode Blizard Ltd, 1997. Soil and land classification studies. MAWR, Wadi Hadhramaut Agricultural Development Project, Yemen. Cooke, R.U. and A. Warren. 1973. Geomorphology in deserts. B.T. Batsford Ltd. London. Cope, T.A. 1985. Studies in the Flora of Arabia: XX, Akey to the grasses of the Arabian Peninsula. Arab Gulf J. Sci. Res., Special Publ. 1: 1-82. Cottam, G. and J.T. Curtis. 1956. The Use of Distance Measures in Phytosociological Sampling. Ecology 37: 451 – 460. Cremaschi, M. & F. Negrino. 2005. Evidence for an abrupt climatic change at 8700 14C yr B.P. in rockshelters and caves of Gebel Qara (Dhofar-Oman): Palaeoenvironmental implications. Geoarchaeology 20(6):559 – 579. Dai, A., Meehl, G.A., Washington, W.A., Wigley, T.M. and Arblaster, J.M. 2001. Ensemble simulation of 21st Century climate changes, business as usual vs. CO2 stabilisation. Bulletin American Meteorological Society 82 (11), 2311-2388 Danin A., G. Orshan and M. Zohary 1997. The vegetation of the Northern Negev and the Judean Desert of Israel. Israel Journal of Botany 24: 118-172. Danin, A. (1983). Desert vegetation of Israel and Sinai. Cana Publishing House. Israel. Daubenmire, R.F. 1968. Plant communities: A textbook of plant synecology. Harper and Row, New York. 300 pp. David, W., O. W. Linda and L. Michael 1993. Vegetation-environment relationships in a Negev Deset erosion cirque. Journal of vegetation Science 4:83-94. Deflers, A.1896. Plantes de 1`Arabie meridionale recueillis pendant les annees 1889,1890,1893 et 1894. Bull. Soc. Bot. France 43: 321-332. 366 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Deil, U. & K. Muller-Hohenstein 1988. Euphorbias from Arabia Felix Habitats and distribution. Euphorbia journal 5: 108-120. Pasadena. Deil, U. 1986. Vom Menschen gestaltetes Relief in der Arabischen Republik Jemen und Konsequenzen fur vegetation skundliches Arbeiten. in Gehu, J.M. (Ed.) vegetation et Geomorphologie = Coll. Phytosoc, 13: 375-395. Berlin. Deil, U. and D. Rappenhoner 1988. Studies in phytosociology & fodder yield in the Y.A.R. - a comparison of methods & Results. Colloques phytosocioloques XVI, Paris. Deil, U. and Muller-Hohenstein, K. 1991. Euphorbias from Arabia Felix Part II: Succulent associations and Phytogeographical aspects. Euphorbia journal 7: 72-97. Pasadena. Deil, U. & A.N. Al Gifri 1998. Montane and Wadi vegetation of Arabian Peninsula 6: 125174. Kluwer Academic. Deil, U. 1991. Rock communitie in tropical Arabia. Flora et vegetatio Mundi IX:175-187. Deyn, G.B., C.E. Raaijmakers and W.H. van der Putten (2004). Plant community development is affected by nutrients and soil biota. Journal of Ecology 92: 824–834. Diccon, and Miller 1998. A draft of Ethnoflora of the Soqotra Archipelago. Diersing, V. E., R. B Shaw and D.J. Tazik. 1992. US Army Land Condition-Trend. Analysis (LCTA) program. Environmental Management 16:405-414. Dixon, J., A. Gulliver and D. Gibbon. 2001. Farming Systems and Poverty: Improving Farmers' Livelihoods in a Changing World. FAO & World Bank, Rome, Italy & Washington. DOVE, 2001. Environmental Impact Assessment of Sharyoof Field Development. Block 53, Hadhramaut, Yemen. Draz, O. 1969. The Hima system of range reserves in the Arabian Peninsula: its possibilities in range improvement and conservation projects in the Middle East. Food and Agriculture Organization (FAO), Rome. Eiumnoh, A. and R. P. Shrestha 1997. Can DEM Enhance the digital image classification. A report submitted to Asian conference on Remote Sensing (ACRS). El Azzouni, M. 2003. Conserving Dracaena ombet, Egypt's Dragon Tree. Plant Talk 34: 3839. Eldridge, D., S. Skinner & T. J..Entwisle 2003. Survey Guidelines for Non-Vascular plants, NSW Biodiversity Strategy Report: Botanic Gardens Trust Sydney. EPA, 2005. National Biodiversity Strategy and Action Plan of Republic of Yemen. Ministry of Water and Environment Protection Authority, UNDP/GEF/IUCN YEM/96/G31.Sanaa. Eyre, S.R. 1969. Vegetation and soils, a world picture, Journal of Ecology, 57(2):571-572. FAO, 1989. Arid zone forestry: A guide for field technicians. FAO Conservation Guides 20: pp 150, Rome. FAO, 1993. Prosopis Species in the Arid and Semi-Arid Zones of India. Proceedings of a conference held at the Central Arid Zone Research Institute, Jodhpur, Rajasthan, India, No 21-23. The Prosopis Society of India and the Henry Doubleday Research Association FAO, 1997. Agro-climatic resources of Yemen, part 1 Agro-climatic inventory. Agricultural Research and Extension Authority and Ministry of Agriculture and water resources, Yemen, pp 121. FAO, 1997. Livestock-Environment Interactions. Briefing notes production systems management. 367 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen FAO, 2003. Map of World Soil Resources, scale 1:25 000 000. FAO Rome. Ferrier, S. 2002. Mapping Spatial Pattern in Biodiversity for Regional Conservation Planning: Where to from here?. Systematic Biology 51 (2): 331 – 363. Findlay, A. M. 1996. Population and Environment in Arid Regions. Policy and Research Paper, 10:2-87. Fischer, R. A and N.C. Turner 1978. Plant Productivity in the Arid and Semiarid Zones. Annual Review of Plant Physiology 29: 277-317. Focht, T. and V. D Pillar, 2003. Spatial patterns and relations with site factors in a campos grassland under grazing. Brazilian Journal of Biology 63(3). Fossati, J. G. Pautou.and J.P Peltier 1999. Water as resource and disturbance for wadi vegetation in a hyperarid area (Wadi Sannur, Eastern Desert, Egypt). Journal of Arid Environments 43 (1): 63-77. Furley, P.A. 1974. Soil-slope-plant relationships in the northern Maya Mountains Belize, Central America. Journal of Biogeography 1:171-186. Furley, P.A. 1999. The nature and diversity of neotropical savanna vegetation with particular reference to the Brazilian cerrados. Global Ecology and Biogeography 8:223-241. Furley, P.A. and W.N. Newey 1983. Geography of the Biosphere. An Introduction to the Nature, Distribution and Evolution of the World’s Life Zones. Butterworths & Co. Gabali, S. A, and A. N. AI-Gifri 1991. A survey of the vegetation of Hadhramaut, Yemen. Frag. Flor. Geobotanica 36:127-134. Gabali, S. A. and A. N. AI-Gifri 1990. Flora of South Yemen - Angiosperm. A provisional checklist. Feddes Repertor 101:373-383. Gabali, S.A. & A.G Miller 1992. Provissional cheecklist of th southern region of Yemen (excluding Socotra). Checklist of south Yemen (Draft 10/1/1992). Gabali, S.A. 1998. Studies in the flora of Yemen 4. The endemic species. Cons. Et. Jor. Bot. 53 (1):1-73. Geneva . Gage, J.D 2002. Sieve size influence in studies of macrobenthos. Mar Ecol Prog Ser 225: 97107. Ghazanfar, S.A. 2003. Biology of the central desert of Oman. Turk Journal Botany 28: 6571. 368 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Ghazanfar, S.A. and A.A. Al Kiyumi. (1999). Status of the flora plant conservation in the Sultanate of Oman. A paper submitted to the workshop on the conservation of the flora of the Arabian Peninsula. National Commission for wildlife conservation and development and IUCN. Ghazanfar, S.A. and M. Fisher (1998). Vegetation of the Arabian Peninsula. Series: Geobotany 25. Kluwer Academic Publishers, pp 372. Gibbens, R.P., and J.M. Lenz, (2001). Root Systems of Some Chihuahuan Desert Plants. Journal of Arid Environments, 49(2): 221-263. Goebel, P.C., B.J. Palik and K.S. Pregitzer (2003). Plant diversity contributions of riparian areas in watersheds of the northern lake states, USA. Ecological Application, 13(6):1595-1 609. Goldsmith F. B., C. M. Harrison and A. J. Morton (1986). Description and analysis of vegetation. in P. D. Moore and S. B. Chapman, (editors) Methods in Plant Ecology (second edition). Blackwell Scientific Publications, Oxford. pp 437 - 524. Gupta, R., J.P. Srivastava B.S. Gupta and K.S. Dutta (1986). Palatability and nutritive value of expeller pressed kosum (Schleichera oleosa) cake. Indian Journal of Animal Health 25: 165-169. Harris, P. (1979). Flora in the UAE . Emirates Natural History Group Bulletin 7 Harris, P. S. (2000). Grassland resource assessment for pastoral systems. FAO plant production and protection paper 162. FAO, Rome. Hassan, L.M. (2002). Plant Life in the Digla Conserved Area, Hyperarid Desert, Egypt Online Journal of Biological Sciences 2(8):533-537. Havens, K.i, M. Maunder, E. O Guerrant (2004). Ex Situ Plant Conservation: Supporting Species Survival in the Wild. Island Press, pp 491. Havstad, K.M (1999). Improving Sustainability of arid rangelands. New Mexico Journal of Science, 39: 174-197. Hawley, D (1978). Courtesies in the Gulf area. A dictionary of colloquial phrase and usage. Stacey International, London, pp 96. Heathcote, R.L. (1983). The Arid lands their use and abuse. Longman, pp 323. Hepper, (1975). Checklist of north Yemen, un published. Herzog, M: (1993). The Woody Plant Species. Iconography of Yemen (centered on Jebel Bura'). FAO Sana'a. Heywood, V.H. and Dulloo M.E. (2005). In situ conservation of wild plant species: a criteria global review of good practices. IPGRI Technical Bulletin II, and FAO, Rome. Hill, M.O. (1973). Diversity and evenness: a unifying notation and its consequences. Ecology 54: 427–473. Hill, M.O. (1979). Decorana– A FORTRAN Program for Detrended Correspondence Analysis and Reciprocal Averaging. Ithaca, New York. Hoekstra, T W. and M. Shachak (1999). Arid Lands Management: Toward Ecological Sustainability. University of Illinois Press, pp 384. Holten, J. (2000). Vascular plant species richness along Elevation Mountains of different geomorphology, bedrock and macroclimates in the southern Scandes, Norway Terrestrial Ecology Research, N-7350 Buvika, Hopper, S.D. and B.R. Maslin (1978). Phytogeography of Acacia in Western Australia. Australian Journal of Botany 26(1): 63-78. 369 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Hultine K. (2000). The Evolutionary Relationship Between Drought Tolerance and Distribution of Mesquites. Aridus 12(1): 1-4. Humphry, R.R. (1962). Range Ecology. The range press company, New York. Hunter, M. L. (2002). Fundamentals of Conservation Biology. Blackwell Science, pp 576. Hunting Technical Services Limited (1992). National Land & Water Conservation Project woodland Resources Mapping Project. Technical manual for MAWR. Hunting Technical Services Limited, (1992). Woodland Resources Mapping Project (Ground Observation records). Technical report on behalf of the Land and Water Conservation Project (LWCP), Yemen. ICARDA, (1998). Collection of valuable indigenous plant species of the Arabian Peninsula Collection in the Sultanate of Oman. International Center for Agricultural. Research in the Dry Areas (ICARDA) and Arabian Peninsula Regional Program (APRP). INTERCONSULT, and MacDonald (1993). Shibam floods protection. Feasibility study, discipline report for the World Bank. IPCC. (1996). Climate Change 1995: The science of climate change. Contribution of working group I to the second assessment report of the Intergovernmental Panel on Climate Change. Cambridge University Press. IPCC. (2001). Climate Change 2001: The scientific basis. Summary for policymakers. Shanghai draft (21/01/01), Intergovernmental Panel on Climate Change, Cambridge University Press. Isani, G. B. (1999). Status paper on Situation of Arid Zones of Sindh. IUCN – The World Conservation Union Sindh Programme. IUCN, (1986). Plants in Danger, what do we know?. The international Union for Conservation of Nature and Natural Resources, Switzerland. IUCN, (1994). IUCN Red List Categories. Prepared by the IUCN Species Survival Commission. Switzerland. IUCN, (2004). 2004 IUCN Red List of Threatened Species. IUCN, (2004). Protected Areas and Biodiversity Report. Defining Protected Area Management Categories. Jallow, T. (2002). Climate Forecasting in Africa’s Dryland Areas. Columbia University. James, K.O. and R.A. Zielinski (2003). Production water and Hydrocarbon releases at the Osage-Skiatook Petroleum environmental research sites, Osage County, Oklahoma: Introduction and Geologic Setting. USGS, USA. John, R. D. and H.L. Gallegos (2001). Estuarine diversity of tintinnids (planktonic ciliates). Journal of Plankton Research 23(9):1009-1027. Jolly, D., C. Prentic, R. Bonnefille and A. Ballouche (1998). Biome reconstruction from pollen and plant macrofossil data for Africa and the Arabian Peninsula at 0 and 6000 years. Journal of Biogeography 25:1007-1027. Junior, M.C. (1995). Tree communities of the gallery forests of the IBGE Ecological reserve, Federal District, Brazill. Ph.D. Thesis in Geography, University of Edinburgh. Kaller, A. (2001). Vegetation-environment interactions on a boreonemoral forest in east central Sweden. Master thesis. Department of Environmental Assessment. Swedish University of Agricultural Sciences. Kanfoush, S.L (1998). Desert Geomorphology. USGS, Biology Resources. 370 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Kenneth, C., J. McCorriston and Miller A. (2001). Holocene Paleoenvironments of the southern Arabian highlands reconstructed using fossil hyrax middens.USGS. Kent, M. and P. Coker (1992). Vegetation description and analysis. A practical Approach. John Wiley and Sons, New York, pp 384. Kessler, J.J. (1987). A rangeland vegetation survey of Dhamar Montane plains, RLIP communication. no. 14. Agricultural Research Authority, Yemen and DHV Consulting Engineers, Amersfoort. The Netherlands. Kessler, J.J. (1988). Mahjour Areas: Traditional Rangeland Reserves in the Dhamar Montane plains. RLIP communication no 16. Agricultural Research Authority Yemen and DHV Consulting Engineers, Amersfoort. The Netherlands. Kevin, R.H. (2001). The evolutionary relationship between drought tolerance and distribution of Mesquites. School of Renewable Natural Resources 12(1). Kharaiti, L.M. (2000). The place of ancient agricultural practiced and techniques in Yemen today: problems and perspectives. First theme: origins of agriculture and its techniques. Sanaa, Yemen. Kigel, J., Y. Osem and A. Perevolotsky (2004). Similarity between seed-bank and ensuing vegetation is productivity dependent in a semi-arid Mediterranean annual plant community." Presented, Joint session MEDECOS 10th international conference and seed ecology 2004 international meeting. Rhodos. Kilian, N., P. Hein and M.A. Hubaishan (2002). New and noteworthy records for the flora of Yemen, chiefly of Hadhramaut and Al Mahara. Willdenowia 32:239-269. Koellner, T., A.M. Hersperger and T. Wohlgemuth (2004). Rarefaction method for assessing plant species diversity on a regional scale. Ecology 27:532-544. Komex International Ltd, (1999). Environmental Evaluation of oil Gas Exploration in Yemen. Block 43. Ocean Energy Inc. Houston. Koppen, W (1936). Das geographische System der Klimate. In: Koppen W. Geiger R (eds) Handbuch der Klimatologle. Band 1. Teil V. Berlin. Kovach, W.L. (1985). MVSP- A MultiVariate Statistical Package, version 3.2. Institute of Earth Science, University college of Wales, Aberystwyth. Krebs, C.J. (1999). Ecological Methodology, 2 nd edition. Addison Wesley Longman, Menlo Park, California, USA, pp 620. Krishnaswamy J. and D. D. Richterb (2002). Properties of Advanced Weathering-Stage Soils in Tropical Forests and Pastures. Soil Science Society of America Journal 66:244-253. Kruckeberg, A. R., and D. Rabinowitz. (1985). Biological aspects of endemism in higher plants. Annual Reviews of Ecological Systematics 16:447–479. Kurschner, H., A.N. Al gifri, M.Y. Al-Subai and A.K Rowaishad (1998). Vegetational Patterns within coastal salines in southern Yemen. Feddes Rep. 109:147-159. Kutiel, P., H. Kutiel and H. Lavee (2000). Vegetation response to possible scenarios of rainfall variations along a Mediterranean-extreme arid climate transect. Journal of Arid Environments 44: 277-290. Langstrom, E. (2002). Systematics of Echiochilon and Ogastemma (Boraginaceae), and the phylogeny of Boraginoideae. Comprehensive summaries of Uppsala dissertations from the Faculty of Science and Technology 693. Acta university upsaliensis, Uppsala. Last, W.M. and J.P. Smol (2005). Tracking Environmental Change Using Lake Sediments. Volume 1. Kluwer Academic Publishers. 371 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Lavranos, J.J. (1966): Notes on Succulent Flora of Southern Arabia. Cactus and Succulent Journal of America 39:3-7. Lavranos, J.J. (1971). Notes on the Succulent flora of North East Africa and Southern Arabia. Cactus and Succulent Journal of America 43(4):60-67. Lavranos, J.J. (1974). Notes on the Succulent flora of North East Africa and Southern Arabia .Cactus and Succulent Journal of America, 46:125-130. Lavranos, J.J. (1993a). Two new species of Huernia from south-western Arabia. Jou. South Africa Bot. 29: 97-99. Lavranos, J.J. (1993b).Three new species of Caralluma from south-western Arabia. Jou. South Africa Bot. 29: 103-110. Lavranos, J.J. and A. N. Al gifri (1999). A new spiny Euphorbia and an amplified description of a perennial woody species, both from the Hadramout, Yemen. Cactus and Succulent Journal 71(3): 135-142. Lavranos, J.J. and B. Mies (2001). A new species of Rhytidocaulon (ApocynaceaeAsclepiadaceae) from south eastern Yemen. Cactus and Succulent Journal 73(6): 299304. Lawton, J.H., M. MacGaevin and P.A. Heads (1987). Effects of altitude on the abundance and species richness of insect herbivores on Brachen. Journal of Animal Ecology 56(1):147-160. Lee, H. M. (2000). Natural Resources Development in the Sahel: The role of the United Nations System. The united nation university. Lewcock, R.B. (1987). Wadi Hadramaout and the walled city of Shibam. United Nations Educational. pp.135. Lind, E.M. and M.E. Morrison. (1974). East African Vegetation. Longman. pp257. Lisa, S. and G. Susan (2001). Review of Sampling Techniques used in Studies of Grassland Plant Communities. Oklahoma Natural Heritage Inventory, Oklahoma Biological Survey. Llewellyn, O.R. (2000). The WCPA regional action plan and project proposal for North Africa and the Middle East. PARKS 10(1). Lubomir, T. (2002). JUICE Software for vegetation classification. Journal of vegetation Science 13: 451-453. Mahdi, A. & A.A. Al Khulaidi, (1999). Vegetation degradation of mountain at Taiz, Yemen. Journal of Dalta 1: 75-95. Mandaville, J.P. (1984). Studies in the flora of Arabia XI: Some historical and Geographical Aspects of a Principal Floristic Fronties. Note RBG Edinb 42(1):1-15. Mandaville, J.P. (1986). Plant life in the Rub’ al-Khali (the Empty Quarter), south-central Arabia. Proceedings of the Royal Society of Edinburgh 89B:147-157. Mandaville, J.P. (1990). Flora of Eastern Saudi Arabia. Kegan Paul International, pp 482. Martin, K., C. Milan and T. Lubomír (2003). Formalized reproduction of an expert-based phytosociological classification: A case study of subalpine tall-forb vegetation. Journal of Vegetation Science 14: 601-610, 2003. MAW, (1996). A draft Geology and Hydrogeology Report. Wadi Hadhramout Agricultural Development Project-Phase3. Addendum Contract for Supplementary Studies. Coode Blizard Ltd. Purley, UK. 372 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen McAleeece, N. (1997). BioDiversity Professional Beta 1. Devised by P.J.D. Lambshead, G.L.J. Paterson and J.D. Gage. The natural history museum and the Scottish association for marine science. McCorriston, J. (2000). Early settlement in Hadhramawt: preliminary report on prehistoric occupation at Shi’b Munatder. Arabian archaeology and epigraphy 11 (2: 129-153. McKinney, M.L. (2004). Measuring floristic homogenization by non-native plants in North America. Global Ecology and Biogeography 13:47-53. Meister, J., M. A. Al Hubaishan, N. Kilian and C. Oberprieler (2005). Chloroplast DNA variation in the shrub Justicia areysiana (Acanthaceae) endemic to the monsoon affected coastal mountains of the southern Arabian Peninsula. Botanical Journal of the Linnean Society 148: 437 – 444. Meister, M, N. Kilian, M. A. Hubaishan & C. Oberprieler (2002Phylogeographical studies in monsoonal woodland refugia of the coastal mountains in the Southern Arabian Peninsula. A paper submitted to A 3-day international symposium from 13 to 15 November 2002 in Leiden, The Netherlands. Michael, L. R. (2003). Reconciliation ecology and the future of species diversity. Oryx, 37(2): 194-205. Mies, B.A. (1994). Checkliste der Gefasspflazen, Moose und Flechten und botaanische Bibliographie der Sokotra und des Sokotrinischen Aarchiples (Jemen.Indisscher Ozean). Senckenbergiana biological 74(1/2)213-2258, Fraankfurt, Germany. Miclhael, C.J., F.H. Warners and D. Warners (eds.) (2001). Atlas of the breeding birds of Arabia. Phoenix Index: 1-18, England. Milchunas, D. G., W. K. Lauenroth, P. L. Chapman and M. K. Kazempour (1989). Effects of grazing, topography, and precipitation on the structure of a semiarid grassland. Plant Ecology 80(1): 11-23. Mill, R.R. and A.G. Miller (1984). Studies in the flora of Arabia : IX. Asynopsis of Paracynoglossum (Boraginaceae). Notes RBG Edinb 41 (3): 473-482. Miller, A. G. (1984). A review of Ochradenus. Notes from the Royal Botanical Garden Edinburgh 41: 491–504. Miller, A. G. (1994). Highlands of south-western Arabia: Saudi Arabia and Yemen. Pages 317-319 in S. D. Davis, V. H. Heywood and A. C. Hamilton, editors. Centres of Plant Diversity, Vol.1. WWF, IUCN, Gland, Switzerland. Miller, A.G. & Nyberg, J.A. (1991). Patterns of endemism in Arabia. Flora et Vegetatio Mundi IX: 263-279. Miller, A.G. and A.A. Al Khulaidi (2004). A provisional list of endemic, near-endemic and non-endemic plant species of the Republic of Yemen. Environmental protection Authority (EPA). Miller, A.G. and M. Morris (1988). Plants of Dhofar. The southern region of Oman. Traditional economic and medicinal uses. Sultanate of Oman: Government of Oman. pp 388. Miller, A.G. and M. Morris (2004). Ethnoflora of the Soqotra Archipelago. RBGE. pp. 776., UK. Miller, A.G. and T.A. Cope (1996). Flora of the Arabian Peninsula and Socotra. vol.1, EUP, UK, pp 586. Miller, A.G., Hedge, I.C. & King, R.A. (1982). Studies in the flora of Arabia: 1, A botanical bibliography of Arabian Peninsula, Notes Roy. Bot. Gar. Edinburgh 40. 373 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Miller, A.G.., A.A. Al Khulaidi & P. Scholte (2006). Draft List of Important Plant Areas in Yemen (in press). Moehl, G.A. (1996). The South Asian Monson and the Tropospheric Biennial Oscillation. Journal of Climate 10(8): 1921-143. Monger, H.C. (2002). Arid Soil. Encyclopedia of Soil Science. p. 84-88. Marcel-Dekker, New York. Moore, G. and W. Tymowski (2004). Explanatory Guide to the International Treaty on Plant Genetic Resources for Food and Agriculture. IUCN. Moreno, C.E. and G. Halffter (2000). Assessing the completeness of bat biodiversity inventories using species accumulation curves. Journal of Applied Ecology 37:149-158. Morzaria-Luna, H., J.C. Callaway, G. Sullivan and J.B. Zedler (2004). Relationship between topographic heterogeneity and vegetation patterns in a Californian salt marsh. Journal of Vegetation Science 14:523-530. Mueller-Dombois, D. and G. Spatz. (1975). Application of the relevé method to insular tropical vegetation for an environmental impact study. Phytocoenologia 2: 417- 429. Mueller-Dombois, D. and H. Ellenberg (1974). Aims and methods of vegetation ecology. John Wiley and Sons, New York. pp 547. Muller-Hohenstein, K. & Rappenhoner, D. (1991). Vegetation mapping under different aspects of basic and applied vegetation science. Large scale examples from the Y.A.R. in Engel, T., Frey, W. & Kurschner,H. (Eds.). Contributiones selectae ad floram et vegetationem orientis. Flora et vegetatio Mundi 9:199-213. Berlin. Muller-Hohenstein, K. (1984). Bericht uber den Stand der vegetationskundlichen Arabeiten in der Arabeitschen Republik jemen am Ende des jahres 1982- in Kopp, H.& Schweizer, G. (Hrsg); Entwicklungsprozesse in der Arabischen Republik Jemen. Studien 1: 211-223. Wiesbaden. Muller-Hohenstein, K. (1986). Methodische Probleme Vegetation skundichen Arbeitens in semiariden Raumen am Beispiel des Nordjemen. Geomethodica 11: 109-143. Muller-Hohenstein, K. (1988). Zur Arealkunde der Arabischen Halbinsel (Chorological aspects of the Arabian penensula). DIERDE 119(5): 65-74. Muller-Hohenstein, K. (1989).Desertification in the mountainous region of Yemen. Resource Conservation and Desertification Control in the Near East, Report by Dieter Rappenhoner: 62-65. Mundy, M. and B. Musallam (2000). The transformation Nomadic society in the Arab East. Cambridge university press, UK. Murray, G. W. (1951). The Egyptian Climate: An Historical Outline. The Geographical Journal 117(4):422-434. Nadaf, S.K., S.F. Al Farsi, S.A. Al Hinai, A.N. Al Bakri and A.S. Al Harthy (2004). Establishment of field gene bank of indigenous pasture plant species of Oman. ICARDA-APRD, Sustainable management of natural resources and improvement of major production system of Arabia peninsula, annual report 2003-2004. Nautiyal, B.P., R. S. Chauhan, V. Prakash, H. Purohit and M. C. Nautiyal ( 2003). Population studies for the evaluation of germplasm and threat status of the alpine medicinal herb, Nardostachys jatamansi. Plant Genetic Resources IPGRI-FAO 136:3439. Naveh, Z. and R.H. Whittaker (1979). Structural and floristic diversity of shrublands and woodlands in northern Israel and other Mediterranean areas. Vegetation 41(3): 171-190. 374 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Neff, J. C., R. L. Reynolds, J. Belnap, and P. Lamothe (2005). Multi-decadal impacts of grazing on soil physical and biogeochemical properties in southeast Utah. Ecological Applications 15(1): 87–95. Nellemann, C., L. Kullerud, I. Vistnes (2001). Global Methodology for Mapping Human Impacts on the Biosphere UNEP & GLOBIO. Neumann, K. (1989). Holocene vegetation of the Eastern Sahara: charcoal from prehistoric sites. African Archaeological Review 7(1):97-116, The Netherlands. Norton, J. B., J. A. Sandor and C. S. White (2003). Hillslope Soils and Organic Matter Dynamics within a Native American Agroecosystem on the Colorado Plateau. Soil Science Society of America Journal 67:225-234. Novikova, N.M. 1970. Experience of compiling the preliminary vegetation map of Arabia. In: Geobotanical mapping. Leningrad. pp. 61-71. (In Russian). OECD, (2003). Agriculture and Biodiversity: Developing Indicators for Policy Analysis. Organization for Economic Co-operation and Development pp 280. Onus, A., S. Cattle and I. Odeh (2003). How do Lachlan valley cotton soils compare to northern NSW. Univ of Sydney and the Australia Cotton Cooperative Research Center 24(3):28. Otten, E. J. (2001). The effect of human population on Biodiversity. Biodiversity and human health. Overpeck, J.T. and P.J. Bartlein (1989). Assssing the response of vegetation to future climate change. The potential effects of global climate change on the United State. USA Environmental Protection Agency, Washington, pp1-32. Ozgul, M, & T. Oztas (1990). Overgrazing effect on rangeland soil properties. Ataturk University, Faculty of Agriculture, Department of Soil Science, Erzurum, Turkey. Pearson, R. G. and T. P. Dawson (2003). Predicting the impacts of climate change on the distribution of species: are bioclimate envelope models useful ?, Global Ecology & Biogeography 12(5): 361-371. Perveen, A. and M. Qaiser (2001). Pollen Flora of Pakistan - XXVIII: Resedaceae Turk J Bot, 25:39-42. Pitelka, L. F (1997). Plant migration and climate change. American Scientist 85(5):464-474. Pitkänen, S. (2000). Classification of vegetational diversity in managed boreal forests in eastern Finland. Plant ecology 146: 11-28. Powell, J. M., R. A. Pearson and P. H. Hiernaux (2004). Review and interaction Crop– Livestock Interactions in the West African. Dry lands Agronomy Journal 96:469-483. Qinfeng, G., J. H..Brown, and T. J. Valone (2002). Long-term dynamics of winter and summer annual communities in the Chihuahuan Desert. Journal of Vegetation Science 13: 565-574. Rajakaruna, N. (2004). The Edaphic factor in the Origin of Plant Species. International Geology Review 46(5):471-478(8). Rathjens, C. and Wissmann, H. (1934). Ergebnisse der 4. Suedarabienreise. Voelkerkundemuseum Hamburg. Zum Klima von Mittel- und Suedarabien. in: Annalen der Hydrologie, Berlin. Robert, J.P. and A.S. Thomas (1997). Distribution of herbs and shrubs in relation to landform and canopy cover in riparian forests of coastal Oregon. Can J. Bot. 76: 298-315. 375 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Rojas, N. E., M. A. Marins and O. Rocha (2001). The effect of Abiotic factor on the hatching of Moina micrura Kurz, (Crustacea: Cladocera) phippial Eggs. Braz. J. Biol. 61(3): 371-376. Salzmann, U. (2000). Are modern savannas degraded forests?-A Holocene pollen record from the Sudanian vegetation zone of NE Nigeria. Vegetation History and Archaeobotany. 9(1): 1-15. Sankaran, M. + 30 other authors (2005). Determinates of woody cover in African savannas. Nature 438:846-849. Sarnthein M. (1978). Sand deserts during glacial maximum and climatic optimum. Nature 272:43-46. Scheiner, S.M., S.B Cox and M. Willig (2000). Species richness, species-area curves and Simpson’s paradox. Evolutionary Ecology Research 2: 791-802. Scholte, P.T., Al Khulaidi, A.A. & Kessler, J.J. (1991). The vegetation of the Republic of Yemen (western part). EPC and ARA, Sanaa and DHV, Amasfort, The Netherlands. Schroder, B. (2002). The last Glacial Maximum (18 000 yr BP).Department of Physial Geography, University of Mannheim. Schwartz, O. (1939). Flora des tropischen Arabien. Mitteilungen aus dem Institut für allgemeine Botanik in Hamburg 10, pp 393. SELKHOZPROMEXPORT, (1984). Scheme of water and land resources development in Hadhramaut valley. Final report. Moscow and Aden. Shaw, K.M., P.D Lambshead and H.M Platt (1983). Detection of pollution-induced disturbance in marine benthic assemblages with special reference to nematodes. Mar. Ecol. Prog. Ser. 11: 195-202. Sheded, M. (2002). Vegetation in the southern eastern desert of Egypt. Online Journal of Biological Sciences 2(9):573-581. Shevock, J. R. (1996). Status of Rare and endemic Plants. Sierra Nevada Ecosystem Project: Final report to Congress, vol. II, Assessments and scientific basis for management options. Davis: University of California, Centers for Shmida, A. and J.A. Aronson (1986). Sudanian elements in the flora of Israel. Annals of the Missouri botanical garden 73(1):1-28. SOGREAH, (1979). Wadi Hadhramout Feasibility Study, Interim Report. SOGREAH, (1981).Wadi Hadramawt fesibility study, Specifice study report, 2-Soil survey and land classification. SOGREAH, (1981).Wadi Hadramawt fesibility study, Specifice study report, 4- Agronomy and water requirements. Spellerberg, I. F (1992). Evaluation and Assessment for Conservation: Ecological Guidelines for Determining Priorities for nature conservation Series: Vol. 4, pp 280. Chapman and Hall, London. Stark, F. (1936). The southern gate of Arabia. A journey in the Hadhramout. John Murray Ltd, London. Sulayem, M. and E. Joubert (1994). Management of protected areas in the Kingdom of Saudi Arabia. An international journal of the forestry and food industries, 176 (45). Sutherland, W.J. (2001). The Conservation Handbook, research, management and policy. Blackwell publishing. pp 278. 376 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Sutter, B. and G. Ritchison. (2005). Effects of grazing on vegetation structure, prey availability, and reproductive success of Grasshopper Sparrows. Journal of Field Ornithology, 76 (4):345-351. Takhtajan, A. (1986). Floristic region of the world. University of California Press, pp. 522. ter Braak, C. J. F. (1986). Canonical correspondence analysis: a new eigenvector technique for multivariate direct gradient analysis. Ecology 67: 1167–1179. ter Braak, C. J. F., and P. Smilauer. (1998). CANOCO Reference Manual and User's Guide to Canoco for Windows: Software for Canonical Community Ordination (version 4). Microcomputer Power (Ithaca, NY USA) pp 352. Thesiger, W. (1959). Arabian sands: Longmans, Green & Co., London. Thompson, G.G., P.C Withers, E.R. Pianka and S.A. Thompson (2003). Assessing biodiversity with species accumulation curves; inventories of small reptiles by pittrapping in Western Australia. Austral Ecology 28: 361–383. Thompson, R.S., K.H. Anderson and P.J. Bartlein (1997). Assessment of Potential Future Vegetation Changes in the Southern United State. Impacts of Climate Change on Life and Ecosystems. USGS, USA. Thulin, M. (1994). A new species of Ochradenus (Resedaceae) from southern Arabia. Nord. J. Bot. 14:383-384. Thulin, M. (2001). Pentzia (Asteraceae-Anthemideae) in the Horn of Africa region. Nordic journal of Botany 21(3):249-252. Thulin, M. (2002). Cleome socotrana (Capparaceae) and allied species in the Horn of Africa region. Nordic journal of Botany 22(2):215-218. Thulin, M. and A. N. AI-Gifri (1995). Euphorbia applanata sp. nov. (Euphorbiaceae) from Yemen, with a note on E. quaitensis, Nordic journal of Botany 15:193-195. Thulin, M. and A. N. Al- Gifri (1993). Cucumis canoxyi (Cucrbitaceae), a new species from Yemen. Nordic journal of Botany 14:315-317. Thulin, M., A.N. Al gifri, S.A. Gabali and M. A. Hussein (2001). Addition to Yemen Flora. In Friis & O. Ryding (ed.). Biodiversity Research in the Horn of Africa Region 54: 137153. Tiedemann, R.J.O. (1973). Effect of Mesquite on Physical and Chemical Properties of the Soil. Journal of Range Management, 26 (1). Tilman, T.D. (2000). Causes, consequences and ethics of biodiversity. Nature 405: 208 – 211. Todd, J.R. (1999). Modeling plant distributions through global climate change: new directions. The University of North Carolina. Todd, S.W and M.T. Hoffman (1999). A fence-line contrast reveals effects of heavy grazing on plant diversity and community composition in Namaqualand. South Africa. Plant Ecology 142(1-2):169-178. Toy, T.J., G.R. Foster and K.G Renard. (2002). Soil Erosion: Processes, Prediction, Measurement, and Control. John Wiley & Sons, Inc., New York. UNEP and WMO, (1996) Climate Change 2001: Working Group II: Impacts, Adaptation and Vulnerability. Intergovernmental panel on climate change. UNEP, (1992). Protected areas of the World: A review of national systems. Republic of Yemen. UNEP, (2004). The environmental in the news. Communications and Public Information. 377 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen UNESCO, (1979). Map of the world distribution of arid regions and explanatory notes. Tech. Note 7, Man and the Biosphere. Paris, France. Pp 54. USDA, (2002). Vegetation and Ecology Characteristic of Mixed Conifer and Red forests at the Teakettle Experimental forest. USDA forest Service Gen. Technical report PSWGTR. 186. USGS, (2002). Past, recent and twenty-first century vegetation change in the Arid Southwest. Forest and rangeland ecosystem science center, Colorado Plateau field station. USGS, (2005). Effect of climatic variability and land use on American dry lands: land use effects on ecosystems. Van, der M.l (2004). Vegetation Ecology. Blackwell Publishers, Oxford. Van, Gils H., W. van Wijngaanden and I.S. Zonneveld (1985), Vegetation and rangeland survey, ITC lecture note N-7, 5th edition, Enschede, The Netherlands. Vandeven, C. (1999). Predicting arid vegetation species distribution using AVIRIS, Topography and Geology in the white mountains eastern California. Environmental Science Stanford University, Bldg 320, Stanford. Venter, H.J.T. and R.L. Verhoeven (1999). A new species of Cryptolepis (Periplocoideae, Apocynaceae) from Arabia. Botanical Journal of the Linnean Society 131(4):417-422. Verba, M. P. (1995).Impact of irrigation on micromorphological features of desert soils of Hadremout Valley (Yemen). Eurasian Soil Science 27/9:108-124. USA Vetaas, O and J. Grytnes (2002). Distribution of vascular plant species richness and endemic richness along the Himalayan elevation gradient in Nepal. Global Ecology and Biogeography 11:291–301. Vitousek, P. M., H. A. Mooney, J. Lubchenco and J. M. Melillo (1997). Human domination of earth’s ecosystems. Science 277: 494-499. Vogt, B. and A. Sedov (1997). Surveys and resource excavations in the Hadhramawt Governorate, Republic of Yemen. Roy Mimeo. Walker, A.S. (1998). Deserts: Geology and Resources. USGS, Biology Resources. Walker, S. and G. L. William (2002). Alluvial grasslands of Canterbury and Marlborough, eastern South Island, New Zealand: vegetation patterns and long-term change. Journal of the Royal Society of New Zealand 32 (1):113–147. Webb R. H. and H. H. Wilshire (1984). Environmental Effects of Off-Road Vehicles: Impacts and Management in Arid Regions. Journal of Ecology 72(3):1101-1102. Webb, R.H. and S.S. Stielstra (1979). Sheep grazing effects on Mojave Desert vegetation and soils. Environmental Management 3(6):517-529. Wentworth, T.R. (1981). Vegetation on Limestone and Granite in the Mule Mountains, Arizona. Ecology 62 (2): 469-482. Western, R.A. (1985). A Botanical Reconnaissance of the Northern Emirates. Natural History Group (ENHG) Bulletin 27. Western, R.A. (1988). Adaptation of Plants to a Desert Environment. Emirates Natural History Group (ENHG) Bulletin 36:17 - 23. Westinga, E. and D.C. Thalen (1980). A survey and problem analysis of the rangelands in the Rada district RIRDP technical note 5. Ministry of Agriculture and Forestry, Yemen. White, F (1983). The vegetation of Africa. A descriptive memoir to accompany the Unesco/Aetfat/UNSO vegetation map of Africa. UNESCO, Paris. 378 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Wickens, G.E. (1982). Studies in the Flora of Arabia: III. A biographical index of plant collectors in the Arabian Peninsula (including Socotra). Notes from the Royal Botanic Garden Edinburgh 40(2):301-330. Wickens, G.E. (1995). Role of Acacia species in the rural economy of dry Africa and the Near East..FAO. Wilkinson, T. J. (2005). Soil erosion and valley fills in the Yemen highlands and southern Turkey: Integrating settlement, geoarchaeology, and climate change. Geoarchaeology 20(2):169 –192. Williams, J., C. Read, T. Norton, S. Dovers, M. Burgman, W. Proctor, H. Anderson and G. Whatman (2001). Biodiversity. Australia State of the Environment Report (Theme Report), CSIRO. Publishing on behalf of the Department of the Environment and Heritage, Canberra, pp 217. Willott, S. J. (2001). Species accumulation curves and the measure of sampling effort. Journal of Applied Ecology 38: 484-486. Wilsey, B.J. and C. Potvin (2000). Biodiversity and ecosystem functioning: the importance of species evenness in an old field. Ecology 81(4): 887-892. Wissmann, H. Von (1968). Zur Archäologie und antiken Geographie von Südarabien: Hadramaut, Qataban und das ‘Aden-Gebiet in der Antike [On the Archaeology and Ancient Geography of South Arabia: Hadhramaut, Qataban and the Aden Area in Ancient Times], pp119. Wittmanna, F., J.J. Wolfgang and T. F. Piedadeb (2004). The várzea forests in Amazonia: flooding and the highly dynamic geomorphology interact with natural forest succession. Forest Ecology and Management. 196 (2-3): 199-212. Wolfe, L.M. and A. Shmida. (1997). The ecology of sex expression in a gynodioecious Israeli desert shrub (Ochradenus baccatus). Ecology 78(1):101-110. Wolfe, L.M. and J.L. Burns (2001). A rare continual flowering strategy and its influence on offspring quality in a gynodioecious plant. American Journal of Botany 88:1419-1423. Wood. J.R.I. (1997). A Handbook of the Yemen Flora. Royal Botanic Gardens, Kew. UK. pp 434. Woodward, F.I. (1987). Climate and plant distribution (Cambridge studies in Ecology). Cambridge University Press. WRAY, (1995). The water resources of Yemen. A summery and digest of available information. MOMR, Yemen and TNO, The Netherlands WRI, (World Resources Institute). (1992). World Resources 1992-1993. Oxford University Press, New York WWF and IUCN, 1994. Centres of plant diversity: A guide and strategy for their conservation. 3 Volumes. IUCN Publications Unit, Cambridge, UK Yousif, K. and K.O. James (2003). .Environmental Impacts of Petroleum Production: Initial Results from the Osage-Skiatook Petroleum Environmental Research Sites, Osage County, Oklahoma. USGS, USA. Yu Hua, (2000). Distribution of Plant Species Richness along Elevation Gradient in Hubei Province, China. International Institute for Earth System Science (ESSI), Nanjing University. Zaroug, M. (1984 ). Importance of fodder trees and shrubs in the productivity of rangelands and agriculture systems in the Near East. A paper presented at the International Round Table on Prosopis tamarugo,. ARICA, CHILE. FAO. 379 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Zohary, M. (1962). Plant life of Palestine. The Ronald Press Company. New York. Zohary, M. (1973). Geobotanical foundations of the Middle East. Gustav Fischer Verlag volum 1&2., Stuttgart, W. Germany. Zohary, M. and N. Feinbrun-Dothan. (1966-1986). Flora Palaestina. The Israel academy of science and humanities. Zohary, M. and G. Orshan (1956). Ecological studies in the vegetation of the Near East Deserts II Wadi Araba with 12 figures, 2 photographs and 9 tables. Plant Ecology 7(1): 15-37. Zonneveld, I.S. (1986). Lectures on vegetation science. Lecture notes no. 5. International Institute for Aerospace Survey and Earth Sciences (ITC) and Landbouw University (LU). The Netherlands. Zonneveld, I.S. (1989). The land unit - A fundamental concept in landscape ecology, and its applications. Landscape Ecology 3 (2): 67-86. Zousou, R. and P.A. Furley (1975). The nature and formation of a toposequence of arid soils over the terraces of the Balick River, Northern Syria. Research Discussion paper no. 7. Department of Geography University of Edinburgh. 380 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Appendexes 1 Rainfall data Rainfall data in mm from 1981 to 2002 for Sayun months year J F M A M J J A S O N D 1981 0 0 50 0 0 2.9 2.6 2 0 0 0 0 57 1982 33 2.2 16 1.1 0 0 4.3 68 0 0 0 0 125 1983 6.5 12 0.3 61 0 0 0 30 0 0 0 0 109 1984 0 0 0 0 16 0 1.8 0 0 0 4.5 0 22.3 1985 6 0 0 0.2 0 0 0.1 2.6 0.1 0 0 0 9 1986 0 12 0 19 0 23 1.8 13 0 0 0 0 68 1987 0 0 61 24 0 1.5 0 22 0 0 0 2.5 110 1988 0 0 0 0.5 0 0 5.7 1.5 0.7 0 0 0 8.4 1989 0 0 112 60 0 1 2.2 1.2 0 0 0 0 176 1990 0 57 0 4.1 0 0 11 0.5 0 0 0 0 72.6 1991 0 0 12 0 0 0 1 18 0 13 0.5 0 44 1992 4 0 0 44 6.2 0 8.5 7.2 1.5 11 0 0 81.7 1993 4.5 48 0 14 9 0 2.2 1.2 0 0 0 0 78.7 1994 0 0 0.6 0.8 0 0 0 29 0 0 0 0 30.3 1995 0 0 68 1.4 0 1.2 39 18 0.1 0 0 0 128 1996 0 0 1 0 1.2 66 21 0.7 0 0 0 0 89.4 1997 0 0 45 0.6 0 1.1 19 0.8 0 17 2.7 0 86.2 1998 0 0 0 0 0 0 4.4 35 0 0.4 0 0 39.6 1999 0 0 0 0 0 0 29 44 0 39 0 0 111 2000 0 0 0 0 0 0 0.5 2.7 7.5 12 0 0 22.7 2001 0 0 7.9 0 38 0 18 20 0 0 0 0 84.4 2002 0 0 0 39 0 0 0.2 6.3 34 0 0 0 79.6 average 2.5 5.9 17 12 3.2 4.4 7.8 15 2 4.1 0.4 0.1 74.2 total Appendix 1a . Annual rainfall recorded in Sayun weather station. Most of the rainfall occurs in March, April and August A variation in both amount and time of occurrence from year-to-year and from month to month is obvious. The average monthly rainfall ranges from .1 mm in December to 17 mm in March. The maximum monthly rainfall took place in March 1989 with about 112 mm. The high figure of rainfall in March 1989 is due to heavy rain that last for about 9 hours and caused the most major floods in the this century. 381 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Appendix 2. Rainfall data between 1996 and 1999 of 22 rainfall stations located along Wadi Hadhramaut. . ND no data collected. The highest rainfall was in 1989. STATION 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 average QASM 6.7 83 173 25 223 59 20 20 51 7 83 133 72 20 72 32 67.4 ALRUDOOD 0 45 66 42 112 21 80 85 44 21 108 121 48 42 12 25 54.5 SAH 7.5 66 72 13 92 26 10 77 59 0 ND 83 31 26 139 44 49.5 SHIBAM 0.7 89 99 20 11 43 34 129 68 0 8 ND ND ND 1 21 11 Al Qatan 2.2 20 85 0 106 44 14 55 56 0 ND ND ND ND ND ND 38.1 Bahran 3.5 3.5 52 2.6 101 28 0 63 23 0 ND ND ND ND ND ND 27.6 Al Mashhad 2.3 37 50 16 14 43 9.8 67 62 0 ND ND ND ND ND ND 30.2 ALUGUBIEH 0 70 95 65 125 41 14 81 0 0 0 32 7 21 20 32 37.6 Qaren BAMSAOOD 6.1 45 53 53 107 86 64 161 93 30 3.2 21 18 11 14 0 47.8 AL HALEH 19 79 83 78 161 92 15 198 142 83 69 178 133 36 100 98 92..3 GAIDUN ND 81 67 79 130 3.3 0 159 102 12 ND 179 12 ND ND 19 52.1 ALKORIBEH 0 23 135 81 139 67 25 223 135 0 0 66 160 106 149 87 87.3 SOBEIKH ND ND ND 44 123 56 43 92 125 67 38 127 97 44 38 665 120 ALHAJRAIN ND ND ND 12 92 23 7.8 121 96 0 0 0 44 36 ND 49 40 AMD ND ND ND 44 69 76 18 94 87 41 24 0 31 34 419 55 76.1 SHARJ ALSHARIF ND ND ND 42 171 51 35 107 145 0 0 86 85 57 ND 38 68 GheilBin Yamain ND ND ND ND 47 8.1 9.4 33 70 0 ND ND ND ND ND 0 27.8 Alsalasil ND ND ND ND 90 34 13 33 99 0 0 0 0 24 90 0 31.9 Katbeh ND ND ND ND 0 44 4.5 15 0 0 ND ND ND ND ND 30 10.5 AL JAWADEH ND ND ND ND 121 26 18 79 58 14 47 28 84 32 66 51 51.9 AL MINSAF ND ND ND ND 127 19 22 15 70 0 0 210 0 31 65 11 47.5 Abd Gharieb ND 79 95 54 134 0 ND ND ND ND ND ND ND ND ND ND 72.2 WADI BIN ALI ND ND ND ND ND 24 21 93 69 0 138 148 88 20 65 27 63 YABHOOD ND ND ND ND ND 9.9 27 72 87 0 106 0 0 7.4 ND 46 35.6 HAINAN ND ND ND ND ND 17 3.5 33 45 31 56 37 0 0 76 41 30.8 BODHAH ND ND ND ND ND 160 165 237 0 23 486 ND 120 167 83 160 TARIM ( DAMOON ) ND ND ND ND ND ND ND ND ND ND ND 127 60 35 91 ND 78.2 JUEIMEH ND ND ND ND ND ND 38 ND ND ND ND 56 94 28 142 45 67 382 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Appendix 3. Daily rainfall (mm) between 1981 and 2002 from Sayun year month / day 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 1981 March 1982 1983 1984 1985 1986 1987 June July August January February march April July August January February March April August May July November January April July August September February April June July August March April June August December 2.5 22.4 5.4 19.2 1.9 1.0 1.5 0.8 0.3 2.0 33.0 0.6 1.6 15.0 1.3 1.1 4.3 68.0 0.1 6.5 5.0 6.5 0.25 0.20 16 25 10 7 1.0 1.25 0.25 24.0 0.25 5.0 5.0 0.5 0.25 16 0.25 1.5 4.5 6 0.2 0.1 0.1 1.5 1.0 0.1 4.10 7.5 9.1 5.5 3.3 0.2 0.4 23.0 1.8 1.0 4.1 8.0 3.0 13.6 31.0 4.5 15.5 6.5 9.9 1.50 16.8 0.4 1.0 4.1 2.5 383 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Appendix 3. Continue year 1988 1989 1990 1991 1992 1993 month / day April July August September March April June July August February April July August March July August October November January April May July August September October January February April May July August 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 0.5 0.1 0.70 1.2 2.4 2.0 0.8 0.7 2.5 6.2 35.6 36.0 7.4 45.0 0.2 5.5 4.5 8.0 20.0 0.5 0.8 1.2 1.0 0.8 0.4 38.0 1.1 0.2 1.7 0.4 0.2 3.0 14.5 3.0 9.0 1.0 0.5 2.2 10.0 1.0 17.5 12.8 0.5 4.0 2.5 6.3 35.0 6.2 8.5 5.2 1.0 1.0 1.5 1.5 9.0 0.5 14.7 0.5 21.3 10.5 0.7 4.0 0.5 1.5 2.6 2.6 6.9 4.5 2.5 2.2 1.2 384 2.0 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Appendix 3. Continue year 1994 1995 1996 1997 1998 1999 2000 2001 2002 month / day March April August march April June July August September March May June July August March April June July August October November July August October July August October July August September October March May July August April July August September 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 0.6 0.3 13.9 25.6 7.4 1.2 27.5 0.2 0.7 1.3 0.6 0.3 0.3 1.1 17.5 1.4 1.2 0.5 0.1 0.2 8.7 0.4 2.2 2.4 6.3 36.0 0.1 1.0 0.3 5.2 2.0 0.2 26.0 2.5 1.0 2.0 19.0 6.7 6.6 1.9 0.4 3.0 0.9 10.0 0.5 1.0 0.2 2.4 41.5 0.1 0.1 21.6 7.2 0.6 0.7 0.1 0.3 12.6 0.1 6.8 0.7 0.3 16.2 2.7 1.0 0.1 3.3 0.4 0.6 5.0 0.4 14.5 0.4 7.7 4.7 5.5 0.7 0.7 0.2 33.0 0.5 1.0 42.5 0.5 1.0 1.7 7.0 12.0 0.5 5.2 34.4 0.4 2.7 3.0 0.3 2.6 0.4 3.7 5.3 0.9 33.3 0.3 0.9 1.0 7.9 3.5 5.7 11.9 0.2 3.3 3.4 26.5 0.5 3.7 385 3.0 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Appendix 4. List of plant species from Hadhramaut region scientific name 1. Abutilon bidentatum Hochst. ex A.Rich. 2. Abutilon fruticosum Guill. & Perr. 3. Abutilon pannosum (Forrsk.) Schlechl. 4. Acacia campoptila Schweinf. ** 5. Acacia ehrenbergiana Hayne. 6. Acacia hamulosa Benth. 7. Acacia laeta R.Br. 8. Acacia mellifera (Vahl) Benth. 9. Acacia oerfota (Forssk.) Schweinf. 10. Actiniopteris semiflabellata Pichi-Ser. 11. Adenia venenata Forssk. 12. Adenium obesum (Forssk.) Roem. & Schult. 13. Aeluropus lagopoides (L.) Trin. ex Thev. 14. Aerva artemisioides Vierh. & Schwartz subsp. artemisioides. * 15. Aerva javanica (Borm.f.) Juss. ex Schult 16. Aerva lanata (L.) Juss. ex J.A. Schult. 17. Agrostis viridis Gouan 18. Aizoon canariensis L. 19. Alhagi graecorum Boiss. 20. Aloe luntii Baker * 21. Aloe abyssicola Lavr. & Bilaidi * 22. Aloe doei Lavr. * 23. Aloe eremophila Lavr. * 24. Aloe fleurentinorum Lavr. & Newton ** 25. Aloe inermis Forssk. * 26. Aloe luntii Baker * 27. Aloe mahraensis Lavr.& McCoy ** 28. Aloe mccoyi Lavr.& Mies * 29. Aloe serriyensis Lavr. * 30. Amaranthus ascendens Lois. 31. Amaranthus graecizans subsp. graecizans L. 32. Anabasis setifera Moq. 33. Anagallis arvensis L. var. caerulea 34. Aneilema aequinoctiale P.B.Kunth. 35. Anisotes trisulcus (Forssk.) Nees. 36. Anogeissus bentii E.G.Baker. * 37. Anticharis glandulosa Asch 38. Anticharis linearis (Benth.) Hochst. ex Asch. ** 39. Aptosimum pumilum (Hochst.)Benth. 40. Areca catechu L. 41. Argyrolobium roseum (Camb.) Jaub. & Spach 42. Argyrolobium uniflorum (Decne.) Jaub. & Spach 43. Aristida triticoides Henrard 44. Aristolochia rigida Duch. 45. Arnebia hispidissima (Lehm.) DC. 46. Arthrocnemum fruticosum (L.) Moq. family Malvaceae Malvaceae Malvaceae Mimosoideae Mimosoideae Mimosoideae Mimosoideae Mimosoideae Mimosoideae Actiniopteraceae Passifloraceae Apocynaceae Poaceae (Gramineae) Amaranthaceae Amaranthaceae Amaranthaceae Poaceae (Gramineae) Aizoaceae Fabaceae (Papilionoideae) Aloeaceae Aloeaceae Aloeaceae Aloeaceae Aloeaceae Aloeaceae Aloeaceae Aloeaceae Aloeaceae Aloeaceae Amaranthaceae Amaranthaceae Chinopodiaceae Primulaceae Commelinaceae Acanthaceae Combretaceae Scrophulariaceae Scrophulariaceae Scrophulariaceae Arecaceae (Palmae) Fabaceae (Papilionoideae) Fabaceae (Papilionoideae) Poaceae (Gramineae) Aristolochiaceae Boraginaceae Chinopodiaceae local name Sumur Qatad Dhubyan Orfut Aden Ra Ra Wadha' Sammana 386 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. 61. 62. 63. Arthrocnemum macrostachyum (Moric.) K.Koch. Asepalum eriantherum (Vatke) Marais Asphodelus fistulosus L. Asystasia petalidiodes Defl. * Atractylis kentrophylloides (Bak.) F.G.Davis ** Atriplex coriacea Forssk. Atriplex griffithii Moq. Atriplex leucoclada Boiss. Balanites aegyptiaca Del. Barleria acanthoides Vahl Barleria aff bispinosa (Forssk.) Vahl Barleria candida Nees Barleria farinosa Defl. * Barleria hochstetteri Nees Barleria proxima Lind. Barleria trispinosa (Forssk.) Vahl. Bauhinia ellenbeckii Harms. Chinopodiaceae Cyclocheilaceae Liliaceae Acanthaceae Asteraceae (Compositae) Chinopodiaceae Chinopodiaceae Chinopodiaceae Balanitaceae Acacnthaceae Acacnthaceae Acanthaceae Acanthaceae Acanthaceae Acanthaceae Acanthaceae Caesalpiniaceae 64. 65. 66. 67. 68. 69. 70. 71. 72. 73. 74. 75. 76. 77. 78. 79. 80. 81. 82. 83. 84. 85. 86. 87. 88. 89. 90. 91. 92. 93. 94. Blepharis ciliaris (L.) B.L.Burtt Boerhavia elegans Choisy ssp elegans ** Boscia arabica Pestalozzi ** Bosnanthe berberoides (Chiov.) W.j.de Willde. Cadaba baccarinii Chiov. Cadaba farinosa Forssk. Cadaba heterotricha Stocks ex Hook. Cadaba rotundifolia Forssk. Cadia purpurea (Picc.) Ait. Caesalpinia erianthera Choiv. var. pubescens Brenan Calligonum crinitum Boiss. subsp. arabicum ** Calotropis procera (Ait.) Ait.f. Capparis cartilaginea Decne. Campylanthus antonii Thulin Campylanthus junceus Edgeweirth Campylanthus pungens Schwartz ** Capparis spinosa L. Caralluma adenensis (Defl.) Burg. ** Caralluma arabica N.E.Br. Caralluma dolichocarpa Schwartz * Caralluma flava N.E.Br. ** Caralluma foulcheri-delboscii var. greenbergiana Lavr. * Caralluma lavrani Rauh & Wertel * Caralluma penicellata (Defl.) N.E.Br. ** Caralluma quadrangula (Forssk.) N.E.Br. ** Caralluma shadhabana Lav. & Newton ** Caralluma subulata (Forssk.) Decne. ** Cassia senna L. Caylusea hexagyna (Forssk.) M.L.Green Celtis africana Burm.f. Cenchrus ciliaris L. Acanthaceae Nyctaginaceae Capparaceae Passifloraceae Capparaceae Capparaceae (Capparidaceae) Capparaceae (Capparidaceae) Capparaceae (Capparidaceae) Fabaceae (Papilionoideae) Caesalpiniaceae Polygonaceae Apocynaceae (Asclepiadaceae) Capparaceae (Capparidaceae) scrophulariaceae scrophulariaceae scrophulariaceae Capparaceae (Capparidaceae) Apocynaceae (Asclepiadaceae) Apocynaceae (Asclepiadaceae) Apocynaceae (Asclepiadaceae) Apocynaceae (Asclepiadaceae) Apocynaceae (Asclepiadaceae) Apocynaceae (Asclepiadaceae) Apocynaceae (Asclepiadaceae) Apocynaceae (Asclepiadaceae) Apocynaceae (Asclepiadaceae) Apocynaceae (Asclepiadaceae) Caesalpiniaceae Resedceae Ulmaceae Poaceae (Gramineae) Halaj, Ssur Shakhadh Shawk Adhib, Balma Haydawan 387 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 95. Ceratonia oreothauma Hillc., Lewis & Verdc. ** 96. Ceropegia botrys K.Schuhmann. 97. Ceropegia subaphylla K. Schuhmann. 98. Chascanum marrubifolium Fenzl ex Walp. 99. Chloris barbata Sw. 100. Chrozophora plicata (Vahl) Juss. 101. Chrozophora tinctoria A. juss. 102. Chrysopogon aucheri (Boiss.) Stapf var. quinqueplumis 103. Chrysopogon plumulosus Hochst. 104. Cienfuegosia welshii (T.Anders.) Garcke 105. Cistanche rosea E.G.Baker Caesalpiniaceae Apocynaceae (Asclepiadaceae) Apocynaceae Verbenaceae Poaceae (Gramineae) Euphorbiaceae Euphorbiaceae Poaceae (Gramineae) Poaceae (Gramineae) Malvaceae 106. Citrullus colocynthis (L.) Schrad. 107. Citrullus schimperi (Naud.) Hook.f. 108. Cladostigma dioicum Radlin. Orchidaceae Cucurbitaceae Cucurbitaceae Convolvulaceae 109. Cleome brachycarpa Vahl. ex DC. 110. Cleome albescens Franch. 111. Cleome droserifolia Del. 112. Cleome hadramautica Thulin * 113. Cleome macradenia Schweinf. * 114. Cleome nocam Boiss. 115. Cleome pruinosa T.Anders. ** 116. Cleome scaposa DC. 117. Cometes abyssinica (R.Br.) Wallich. 118. Commelina boissierana C.B.Clk. 119. Commicarpus mistus Thulin 120. Commicarpus stenocarpus (Chiov .) Cuf. 121. Commiphora foliacea Sprague. 122. Commiphora gileadensis (L.) Christ. 123. Commiphora habessinica (O.Berg.) Engl. 124. Commiphora kua (J.F.Royale) Vollesen 125. Commiphora playfairii (Hook.f. ex Oliv.) Engl. 126. Conocarpus lancifolius Engl. 127. Convolvulus arvensis L 128. Convolvulus glomeratus Choisy 129. Convolvulus littoralis Vatke 130. Convolvulus sericophyllos T. Anders. * 131. Corallocarpus glomeruliflorus Schweinf. 132. Corchorus depressus (L.) Christ 133. Cordia nervillii Alston 134. Cornulaca amblyacantha Bunge 135. Cornulaca ehrenbergii Aschers 136. Cornulaca monacantha Del. 137. Cressa cretica L. 138. Crotalaria aegyptiaca Benth. 139. Crotalaria dumosa Franch. 140. Crotalaria persica (Burm.f.) Merr. 141. Crotalaria retusa L. 142. Crotalaria saltiana Andr. Cleomaceae Cleomaceae Cleomaceae Cleomaceae Cleomaceae Cleomaceae Cleomaceae Cleomaceae Caryophyllaceae (Illecebraceae) Commelinaceae Nyctaginaceae Nyctaginaceae Burseraceae Burseraceae Burseraceae Burseraceae Burseraceae Combretaceae Convolvulaceae Convolvulaceae Convolvulaceae Convolvulaceae Cucurbitaceae Tiliaceae Boraginaceae Chinopodiaceae Chinopodiaceae Chinopodiaceae Convolvulaceae Fabaceae (Papilionoideae) Fabaceae (Papilionoideae) Fabaceae (Papilionoideae) Fabaceae (Papilionoideae) Fabaceae (Papilionoideae) Tinwab Quthaylaba, Khuwayma Busham Qafal Mudrika Shuwayla Nazaa' 388 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 143. Cryptolepis yemenensis Venter & R.L.Verh. * 144. Cucumis canoxyi Thulin & Gifri * 145. Cucumis prophetarum L. 146. Cuscuta campestris Yunker 147. Cuscuta kotschyana Boiss. 148. Cyamopsis tetragonoloba (L.) Traub 149. Cyclocheilon somalensis Oliv. 150. Cymbopogon schoenanthus (L.) Spreng. subsp. schoenanthus 151. Cynodon dactylon (L.) Pers. 152. Cyperus conglomeratus Rottb. 153. Cystostemon kissenioides (Delf.) A.Miller & H.Riedl. * 154. Dactyloctenium aegyptium (L.) P.Willd. 155. Datura innoxia Mill 156. Delonix elata (L.) Gamble. 157. Desmostachya bipinnata (L.) Stapf 158. Dichanthium foveolatum (Del.) Roberty 159. Dichanthium insculptum (A.Rich.) Clayton 160. Digera muricata (L.) Mart. subsp. muricata 161. Digitaria ciliaris (Retz.) Koeler 162. Digitaria nodosa Parl. 163. Diplotaxis harra (Forssk.) Boiss. 164. Dipterygium glaucum Decne. 165. Dodonaea viscosa ( L.) Jacq. 166. Dracaena serrulata Baker 167. Ecbolium strictum Schwartz * 168. Echidnopsis bentii N.E.Br. * 169. Echidnopsis globosa Thulin & Hjertson * 170. Echidnopsis seibanica Lavr. * 171. Echiochilon arabicum (Schwartz) I.M.Johns. * 172. Echiochilon colona (L.) Link 173. Echiochilon longiflorum Benth 174. Echiochilon strigosum (Defl.) I.M.Johns. 175. Eleocharis geniculata (L.) Roem. & Schult. 176. Enicostema axillare (Lam.) A. Raynal 177. Enneapogon desvauxii J.E.Smith 178. Enneapogon lophotrichus Chiov. ex Scholz 179. Ephedra foliata Boiss. ex C.A.Mey 180. Ephedra milleri Freitag & Maier-Stolte. ** 181. Eragrostis maharana Schweinf. 182. Eriochloa fatmensis (Hochst. & Steudel) W.D.Clayton 183. Eruca sativa Miller 184. Erucastrum arabicum Fisch. & Mey. 185. Euphorbia applanata Thulin & Gifri * 186. Euphorbia arabica Hochst. & Steud. ex Boiss. 187. Euphorbia balsamifera Ait. 188. Euphorbia cactus Ehrenb. 189. Euphorbia fodhliana Defl. * 190. Euphorbia granulata Forssk. Apocynaceae Cucurbitaceae Cucurbitaceae Cuscutaceae (Convolvulaceae) Cuscutaceae (Convolvulaceae) Fabaceae (Papilionoideae) Cyclocheilaceae Poaceae (Gramineae) Poaceae (Gramineae) Cyperaceae Boraginaceae Poaceae (Gramineae) Solaniaceae Caesalpiniaceae Poaceae (Gramineae) Poaceae (Gramineae) Poaceae (Gramineae) Amaranthaceae Poaceae (Gramineae) Poaceae (Gramineae) Brassicaceae (Cruciferae) Capparaceae (Capparidaceae) Sapindaceae Agavaceae Acanthaceae Apocynaceae (Asclepiadaceae) Apocynaceae (Asclepiadaceae) Apocynaceae (Asclepiadaceae) Boraginaceae Poaceae (Gramineae) Boraginaceae Boraginaceae Cyperaceae Gentiaceae Poaceae (Gramineae) Poaceae (Gramineae) Ephedraceae Ephedraceae Poaceae (Gramineae) Poaceae (Gramineae) Brassicaceae (Cruciferae) Brassicaceae (Cruciferae) Euphorbiaceae Euphorbiaceae Euphorbiaceae Euphorbiaceae Euphorbiaceae Euphorbiaceae Shahber, Shinan Rumam Halta A'lqa Kitha Tuhilba 389 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 191. Euphorbia hadramautica E.G.Baker. 192. Euphorbia meuleniana O. Schwartz. * 193. Euphorbia noxia Pax. 194. Euphorbia quaitensis S.Carter * 195. Euphorbia riebeckii Pax. ** 196. Euphorbia rubriseminalis S.Carter * 197. Euphorbia schimperi Presl 198. Euphorbia seclusa N.E.Br. 199. Euphorbia seibanica Lavr. & Gifri. * 200. Euryops arabicus Steud. 201. Fagonia arabica L. 202. Fagonia bruguieri DC. 203. Fagonia hadramautica Beier & Thulin * 204. Fagonia indica Burm.f. 205. Fagonia lahovarii Volkens & Schweinf. 206. Fagonia latifolia Del. 207. Fagonia luntii Bak. 208. Fagonia paulayana Wagner & Vierh. 209. Farsetia dhofarica Jonsell & Miller ** 210. Farsetia linearis Decne Ex Boiss. ** 211. Farsetia longisiliqua Decne. 212. Farsetia stylosa R.Br. 213. Ficus cordata L. subsp. salicifolia 214. Fimbristylis polytrichoides R.Br. 215. Flaveria trinervia (Spreng.) Mohr. 216. Forsskaolea tenacissima L. 217. Forsskaolea viridis Ehrenb. 218. Gaillonia jolana Thulin * 219. Gaillonia yemenensis Thulin 220. Glossonema varians (Stocks) Benth. ex Hook. 221. Gomphocarpus fruticosus (L.) Ait.f. var. setosus (Forssk.) Schwartz 222. Grewia erythraea Schweinf. 223. Gymnocarpos decandrus Forssk. 224. Gymnocarpos rotundifolius Petruss & Thulin ** 225. Gypsophila montana Balf. f. 226. Halopeplis perfoliata (Forssk.) Bunge ex Asch. & Schweinf. 227. Halopyrum mucronatum (L.) Stapf. 228. Halothamnus bottae Jaub. & Spach ** 229. Helichrysum pumilum (Klatt.) Moes. 230. Helinus integrifolius (Lam.) Kuntze 231. Heliotropium lasiocarpum Fischer & C.A.Meyer 232. Heliotropium adenense Gurke. 233. Heliotropium arbainense Fresen. 234. Heliotropium bacciferum Forssk. 235. Heliotropium bottae Defl. * 236. Heliotropium fartakense Schwartz ** 237. Heliotropium lignosum Vatke Euphorbiaceae Euphorbiaceae Euphorbiaceae Euphorbiaceae Euphorbiaceae Euphorbiaceae Euphorbiaceae Euphorbiaceae Euphorbiaceae Asteraceae (Compositae) Zygophyllaceae Zygophyllaceae Zygophyllaceae Zygophyllaceae Zygophyllaceae Zygophyllaceae Zygophyllaceae Zygophyllaceae Brassicaceae (Cruciferae) Brassicaceae (Cruciferae) Brassicaceae (Cruciferae) Brassicaceae (Cruciferae) Moraceae Cyperaceae Asteraceae (Compositae) Urticaceae Urticaceae Rubiaceae Rubiaceae Apocynaceae (Asclepiadaceae) Durayma Durayma Durayma Sabt'ah Apocynaceae Tiliaceae Caryophyllaceae (Illecebraceae) Caryophyllaceae (Illecebraceae) Caryophyllaceae (Illecebraceae) Chinopodiaceae Poaceae (Gramineae) Chinopodiaceae Asteraceae (Compositae) Rhamnaceae Boraginaceae Boraginaceae Boraginaceae Boraginaceae Boraginaceae Boraginaceae Boraginaceae 390 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 238. Heliotropium makallense Schwartz * 239. Heliotropium ophioglossum Stocks ex Ait. 240. Heliotropium paradoxum Vatke * 241. Heliotropium pterocarpum (DC.) Steud. & Hochst. ex Bunge 242. Heliotropium ramosissimum (Lehm.) Sieb. ex DC. 243. Heliotropium rariflorum Stocks 244. Heliotropium steudneri Vatke 245. Heliotropium strigosum Willd. 246. Heliotropium wagneri Vierh. * 247. Heliotropium wissmannii Schwartz * 248. Hermannia paniculata Franch 249. Hibiscus micranthus L.f. 250. Hochstettera schimperi DC 251. Huernia hadhramautica Lavr. * 252. Hyoscyamus albus L. 253. Hypoestes forskalei (Vahl) Sol. ex Roem. & Schult. 254. Indigofera articulata Gouan 255. Indigofera coerulea Roxb. 256. Indigofera nephrocarpoides J.B.Gillett * 257. Indigofera oblongifolia Forssk. 258. Indigofera semitrijuga Forssk. 259. Indigofera spinosa Forssk. 260. Iphiona anthemidifolia (Bak.) A.Anderb. * 261. Iphiona scabra DC. 262. Iphiona senecionoides (Bak.) A..Anderb. ** 263. Iphiona teretefolia A.Anderb. * 264. Iphiona vierhapperi Schwartz 265. Ipomoea pes-caprae (L.) R. Br. 266. Jatropha pelargoniifolia Courb. 267. Jatropha spinosa (Forssk.) Vahl 268. Justicia areysiana Defl. ** 269. Justicia flava (Vahl) Vahl 270. Kelleronia gillettiae Baker.f. var. gillettiae 271. Kickxia pseudoscoparia D.A.Sutton 272. Kleinia deflersii Defl. * 273. Kleinia odora (Forssk.) A.Berger ** 274. Kleinia squarrosa Cufod. 275. Kohautia retrorsa (Boiss.) Bremek. 276. Lasiurus scindicus Henrard 277. Launaea angustifolia (Desf.) O.Kuntze 278. Launaea capitata (Sprenq.) Dandy 279. Launaea castanosperma F.G.Davies ** 280. Launaea crassifolia (Balf.f.) C. Jeffrey 281. Launaea hafunensis Chiov. 282. Launaea intybacea (Jacq.) Beauverd 283. Launaea massauensis (Fresen.) Sch. Bip. ex Kuntze. 284. Launaea spinosa (Forssk.) Sch. Bip. ex Kuntze 285. Launaea procumbens (Roxb.) Ramayya & Rajgopal. Boraginaceae Boraginaceae Boraginaceae Boraginaceae Boraginaceae Boraginaceae Boraginaceae Boraginaceae Boraginaceae Boraginaceae Sterculiaceae Malvaceae Asteraceae (Compositae) Apocynaceae (Asclepiadaceae) Solanaceae Acanthaceae Fabaceae (Papilionoideae) Fabaceae (Papilionoideae) Fabaceae (Papilionoideae) Fabaceae (Papilionoideae) Fabaceae (Papilionoideae) Fabaceae (Papilionoideae) Asteraceae (Compositae) Asteraceae (Compositae) Asteraceae (Compositae) Asteraceae (Compositae) Asteraceae (Compositae) Convolvulaceae Euphorbiaceae Euphorbiaceae Acacnthaceae Acanthaceae Zygophyllaceae Scrophulariaceae Asteraceae (Compositae) Asteraceae (Compositae) Compositae Rubiaceae Poaceae (Gramineae) Asteraceae (Compositae) Asteraceae (Compositae) Asteraceae (Compositae) Asteraceae (Compositae) Asteraceae (Compositae) Asteraceae (Compositae) Asteraceae (Compositae) Asteraceae (Compositae) Asteraceae (Compositae) Rimram Hasar Kindush, Raqma Dhueila Dumum Harjal, Harajraj 391 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 286. Lavandula macra E.G.Baker. 287. Lavandula subnuda Benth. ** 288. Leptadenia arborea (Forssk.) Schweinf. 289. Leptadenia pyrotechnica (Forssk.) Decne. 290. Leptotherim senegalense (Kunth) W.D.Clayton 291. Limoniastrum arabicum J. Edmondson * 292. Limonium cylindrifolium (Forssk.) Verdc 293. Lindenbergia indica (L.) Kuntze 294. Livistonia carinensis (Chiov.) Dransfield & N.Uhl 295. Lochia bracteata Balf.f. subsp. bracteata 296. Lycium schweinfurthii Dammer 297. Lycium shawii Roem. & Schult 298. Maerua angolensis DC. 299. Maerua crassifolia Forssk. 300. Maerua thomsonii T. Anders. 301. Medicago sativa L. 302. Megalochlamys linifolia (Lindau) Lindau 303. Megastoma pusillum Coss. & Durs. 304. Merremia hadramautica (Baker) R.R.Mill * 305. Moltkiopsis ciliata (Forssk.) I.M.John. 306. Monsonia heliotropoides (Cav.) Boiss. 307. Moringa peregrina (Forssk.) Fiori 308. Nannorrhops ritchieana (Griffith) Ait. 309. Nigella sativa L. 310. Nogalia drepanophylla (E.G.Baker) Verdc. 311. Ochradenus arabicus Chaud. Hillcoat & Miller ** Lamiaceae (Labiatae) Lamiaceae (Labiatae) Apocynaceae (Asclepiadaceae) Apocynaceae (Asclepiadaceae) Poaceae (Gramineae) Plumbaginaceae Plumbaginaceae Scrophulariaceae Arecaceae (Palmae) Caryophyllaceae (Illecebraceae) Solaniaceae Solaniaceae Capparaceae Capparaceae (Capparidaceae) Capparaceae (Capparidaceae) Fabaceae (Papilionoideae) Acacnthaceae Boraginaceae Convolvulaceae Boraginaceae Geraniaceae Moringaceae Arecaceae (Palmae) Ranunculaceae Boraginaceae Resedaceae 312. Ochradenus baccatus Del. 313. Ochradenus gifrii Thulin ** 314. Ochradenus spartioides (Schwartz) Abdullah * 315. Olea europaea L. subsp. cuspidata 316. Ophioglossum polyphyllum A.Br. 317. Orbea luntii (N.E. Brown) Bruyns ** 318. Otostegia fruticosa (Forssk.) Briq. subsp. fruticosa 319. Otostegia fruticosa (Forssk.) Briq. subsp schimperi 320. Pancratium tortuosum Herbert 321. Pancratum maximu Forssk. Resedaceae Resedaceae Resedaceae Oleaceae Ophioglossaceae Apocynaceae (Asclepiadaceae) Lamiaceae (Labiatae) Labiatae Amaryllidaceae Amaryllidaceae 322. Panicum turgidum Forssk. 323. Pappea capensis Eckl. & Zeyh. 324. Pavonia subaphylla Schwartz * 325. Pegolettia senegalensis Cass. 326. Pentatropis nivalis (Gmel.) Field & Wood 327. Pentzia arabica Thulin * 328. Pergularia tomentosa L. 329. Periploca somalensis Browicz 330. Periploca visciformis (Vatke) K.Schum. 331. Peristrophe paniculata (Forssk.) Brummitt 332. Perrolderia coronopifolia Cass. Poaceae (Gramineae) Sapindaceae Malvaceae Asteraceae (Compositae) Apocynaceae (Asclepiadaceae) Asteraceae (Compositae) Apocynaceae (Asclepiadaceae) Apocynaceae Apocynaceae (Asclepiadaceae) Acanthaceae Asteraceae (Compositae) Sarh, Sarha Matka Qardhiya, Dhaayan Thummam, Mahsham 392 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 333. Pistacia falcata Becc. ex Martelli 334. Phagnalon hypoleucum Schultz-Bip. 335. Phoenix dactylifera L. 336. Phragmites australis (Cav.) Trin. ex Steud. 337. Phyllanthus maderaspatensis L. 338. Piper betle L. 339. Pithecellobium dulce (Roxb.) Benth. 340. Pituranthos tortuosus Benth. & Hook. 341. Plicosepalus curviflorus (Benth. ex Oliv.) Tiegh. 342. Pluchea arabica (Boiss.) Qaiser & Lack ** 343. Pluchea dioscorides (L.) DC. 344. Pluchea indica (L.) Less. 345. Polygala abyssinica R.Br. ex Fresen. var. abyssinica 346. Polygala mascatensis Boiss. 347. Polygala senensis Klotzsch 348. Polygala thurmanniana Chodat 349. Portulaca oleracea L. 350. Potamogeton nodosus Poir. 351. Primula verticillata Forssk. 352. Prosopis cineraria (L.) Druce 353. Prosopis farcta (Banks. & Sol.) Mc Bride 354. Prosopis juliflora (S.W.) DC. 355. Psiadia punctulata (DC.) Vatke 356. Psilotrichum virgatum C.C.Towns. 357. Pteris vittata L. 358. Pulicaria argyrophylla Franchet 359. Pulicaria cylindrica (Bak.) O. Schwartz 360. Pulicaria inuloides (Poir.) DC. 361. Pulicaria lancifolia Schwartz * 362. Pulicaria nivea Schwartz * 363. Pulicaria rauhii Gamal-Eldin * 364. Pulicaria somaliensis O.Hoffm. 365. Pulicaria undulata (L.) C.A.Mey. 366. Reseda sphenocleoides Defl. ** 367. Rhazya stricta Decne 368. Rhus flexicaulis Baker * 369. Rhus glutinosa A. Rich subsp neoglutinosa Gilbert 370. Rhus natalensis Bernh. ex Krauss 371. Rhynchosia memnonia (Del.) DC. 372. Rhytidocaulon mccoyi Lavr. * 373. Rumex vesicarius L. 374. Salsola cyclophylla Baker 375. Salsola imbricata Forssk. 376. Saltia papposa (Forssk.) Moq. * 377. Salvadora persica L. 378. Samolus valerandii L. 379. Sarcostemma vanlessenii Lavr. 380. Sarcostemma viminale (L.) R.Br. subsp. stipitaceum (Forssk.) Meve & Liede Anacardiaceae Asteraceae (Compositae) Arecaceae (Palmae) Poaceae (Gramineae) Euphorbiaceae Piperaceae Mimosoideae Apiaceae (Umbelliferae) Loranthaceae Asteraceae (Compositae) Asteraceae (Compositae) Asteraceae (Compositae) Polygalaceae Polygalaceae Polygalaceae Polygalaceae Portulacaceae Potamogetonaceae Primulaceae Mimosoideae Mimosoideae Mimosoideae Asteraceae (Compositae) Amaranthaceae Pteridophyta Asteraceae (Compositae) Asteraceae (Compositae) Asteraceae (Compositae) Asteraceae (Compositae) Asteraceae (Compositae) Asteraceae (Compositae) Asteraceae (Compositae) Asteraceae (Compositae) Resedaceae Apocynaceae Anacardiaceae Anacardiaceae Anacardiaceae Fabaceae (Papilionoideae) Apocynaceae (Asclepiadaceae) Polygonaceae Chinopodiaceae Chinopodiaceae Amaranthaceae Salvadoraceae Primulaceae Apocynaceae (Asclepiadaceae) Jithjath Amhid Raqma Apocynaceae (Asclepiadaceae) 393 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 381. Schouwia purpurea (Forssk.) Schweinf. 382. Schweinfurthia latifolia (Baker) Oliver * 383. Schweinfurthia papillionacea (L.) Boiss. 384. Schweinfurthia pedicellata (T.Anders.) Balf.f. 385. Schweinfurthia spinosa Miller, Sutton & Short ** 386. Scirpus corymbosus Heyne & Roth. 387. Sclerocephalus arabicus Boiss. 388. Seddera arabica (Forssk.) Choisy 389. Seddera hadramautica R.R.Mill 390. Seddera latifolia Hochst. & Steud. 391. Seetzenia lanata (Willd.) Bullock 392. Segeretia thea (Osb.) M.S.John 393. Selaginella imbricata (Forssk.) Spring 394. Senna holosericea (Fres.) Greuter 395. Senna italica Miller 396. Senra incana (Cav.) DC. 397. Sesamum indicum L. 398. Setaria verticillata (L.) P. Beauv. 399. Setaria viridis (L.) P.Beauv. 400. Sideroxylon mascatense (A.D.C.) T.D. Penn. * 401. Solanum forskalii Dunal 402. Solanum nigrum L. 403. Solanum pubescens Willd. 404. Solanum schimperianum Hochst ex A. Rich 405. Sonchus oleraceus L. 406. Sporobolus helvolus (Trin.) T.Duran. & Schinz 407. Sporobolus ioclados (Nees ex Trin.) Nees . 408. Sporobolus ruspolianus Chiov. 409. Sporobolus spicatus (Vahl) Kuntze. 410. Sporobolus tourneuxii Coss. 411. Stachys yemenensis Hedge * 412. Sterculia africana (Lovr.) Fiori 413. Stipa parviflora Desf. 414. Stipa tigrensis Chiov. 415. Stipagrostis ciliata (Desf.) de Winter 416. Stipagrostis hirtigluma (Steud. ex Trin. & Rupr.) de Winter 417. Striga asiatica (L.) Kuntze 418. Strobopetalum bentii N.E.Br. 419. Stultitia araysiana Lavr. & Bilaidi * 420. Suaeda aegyptiaca (Hasselq.) Zoh. 421. Suaeda pruinosa Lange 422. Suaeda vermiculata Forssk. 423. Tagetes minuta L. 424. Tamarix aphylla (L.) H.Karst. 425. Tamarix arabica Bunge 426. Tarchonanthus camphoratus L. 427. Tarenna graveolens (S.Moore) Bremek. 428. Taverniera aegyptiaca Boiss. Brassicaceae (Cruciferae) Scrophulariaceae Scrophulariaceae Scrophulariaceae Scrophulariaceae Cyperaceae Caryophyllaceae (Illecebraceae) Convolvulaceae Convolvulaceae Convolvulaceae Zygophyllaceae Rhamnaceae Pteridophyta Caesalpiniaceae Caesalpiniaceae Malvaceae Pedaliaceae Poaceae (Gramineae) Poaceae (Gramineae) Sapotaceae Solaniaceae Solaniaceae Solaniaceae Solaniaceae Asteraceae (Compositae) Poaceae (Gramineae) Poaceae (Gramineae) Poaceae (Gramineae) Poaceae (Gramineae) Poaceae (Gramineae) Labiatae Sterculiaceae Poaceae (Gramineae) Poaceae (Gramineae) Poaceae (Gramineae) Poaceae (Gramineae) Scrophulariales Apocynaceae (Asclepiadaceae) Apocynaceae (Asclepiadaceae) Chinopodiaceae Chinopodiaceae Chinopodiaceae Euphorbiaceae Tamaricaceae Tamaricaceae Asteraceae (Compositae) Rubiaceae Fabaceae (Papilionoideae) Eshriq Afar, Enhaq T'afh Muzawila 394 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen 429. Taverniera glauca Edgeworth * 430. Taverniera multinoda Thulin * 431. Taverniera schimperi Jaub. & Spach * 432. Tephrosia apollinea (Del.) DC. subsp. longistipulata Vierh. 433. Tephrosia dura Baker 434. Tephrosia hadramautica Thulin * 435. Tephrosia heterophylla Varke 436. Tephrosia nubica (Boiss.) Baker subsp. arabica (Boiss.) Gillett 437. Tephrosia purpurea (L.) Pers. subsp. leptostachya (DC.) Brumm. 438. Tephrosia subtriflora Hochst. ex Baker. 439. Tetrachaete elionuroides Chiov. 440. Tetrapogon villosus Desf. 441. Teucrium eximium Schwartz * 442. Teucrium leucocladum Boiss. 443. Teucrium rhodocalyx Schwartz * 444. Tragus berteronianus Schult. 445. Tragus racemosus (L.) All. 446. Trianthema salsoloides Olive. 447. Tribulus arabicus H.Hosn. 448. Tribulus terrestris L. 449. Trichodesma calathiforme Hochst. 450. Turraea parvifolia Defl. 451. Typha sp. 452. Verbascum luntii E.G.Baker. * 453. Vernonia arabica F.G.Davis 454. Vernonia areysiana Defl. * 455. Vernonia spathulata (Forssk.) Sch. Bip. 456. Viscum schimperi Engl. 457. Withania somnifera (L.) Dun 458. Xerotia arabica Oliver. ** 459. Zehneria anomala C. Jeffrey 460. Ziziphus leucodermis (E.G.Baker) Schwartz ** 461. Ziziphus lotus (L.) Lam. 462. Ziziphus spina -christi (L.) Willd. 463. Zygophyllum album L.var. amblyocarpum 464. Zygophyllum coccineum L. 465. Zygophyllum decumbens Del. var. decumbens 466. Zygophyllum decumbens Del. var. megacarpum 467. Zygophyllum fontanesii Webb. 468. Zygophyllum hamiense Schweinf. 469. Zygophyllum simplex L.  * = endemic to Yemen  ** = endemic to Arabia Fabaceae (Papilionoideae) Fabaceae (Papilionoideae) Fabaceae (Papilionoideae) Fabaceae (Papilionoideae) Fabaceae (Papilionoideae) Fabaceae (Papilionoideae) Fabaceae (Papilionoideae) Fabaceae (Papilionoideae) Fabaceae (Papilionoideae) Fabaceae (Papilionoideae) Poaceae (Gramineae) Poaceae (Gramineae) Lamiaceae (Labiatae) Lamiaceae (Labiatae) Lamiaceae (Labiatae) Poaceae (Gramineae) Poaceae (Gramineae) Aizoaceae Zygophyllaceae Zygophyllaceae Boraginaceae Meliaceae Typhaceae scrophulariaceae Asteraceae (Compositae) Asteraceae (Compositae) Asteraceae (Compositae) Viscaceae Solaniaceae Caryophyllaceae (Illecebraceae) Cucurbitaceae Rhamnaceae Rhamnaceae Rhamnaceae Zygophyllaceae Zygophyllaceae Zygophyllaceae Zygophyllaceae Zygophyllaceae Zygophyllaceae Zygophyllaceae Khudhayrah Yaa'bur Dhubiyah Zahr Allal Habdhah Rabl Rabl 395 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Appendix 5. Name of local people contributed to the study Site 1: Site 2 Site 3 Abdul Lah Saed Al Naser Abdul Lah Saed bin Hadh Abdul Majid Saed Amer Ssaleh Al Ssua’yri Aydhah Ssaleh Al Ssu’yri Aynan Saed hashem Al Kuthayri Emad Jaa’far Jaber Husayn Abdul Lah Al Ssu’yri Jamal Abdul Lah Juma’an Masser Al Shan Juma’an Mohammed Badhawi Khaled Ahmed Zabir Khamis Juma’an Mohammed Ahmed Aqil Mohammed Omar banaduh Mubarak Saed Hadi Muhsin Mohammed Badhawi Nasser Juma’n Al Aa’wash Omar Ssaleh Al Qua’yti Saed A’wadh hassn Salah Mubarak bin Hassn Saleh Al Abd Attamimi Saleh Yaslem Shaikh Rabie’ Omar Al Kuthairi Subayt Bajandue Yaslem Awadh Jaber Abud Karama Azubaydi A’ref Rizq Nassr Abud Salem Saed Al Kuthayri Abdul Lah AZzubaydi Karama Abdul Lah bin Ebrahea Ahmed Sirri Saed Al Jabiri Hadi Abdul Lah Al Kharaz Saed Hamtut Al Ali Hasan Al A’miri Saed Saleh Al Ali Jaa’far Salem Saleam Umar Ba Wazir Muhammed Ba Shua’yb Saleh Abdul Lah Ba Wazir Muhsen Al A’miri Saleh Attumar Al Uwaythaly Saed Badr Al Kuthairi Saleh bin Ali Saed Rizq Bin Nassr Yaser Abdul Lah Malss Salem A’ydhah Al A’miri Salem Bin A’aydhah Al A’amiri Salem Bin Mubarak Assibae’i Salem Rizq Nassr Salem Ssaleh A’bidi Shaykh Ubaid Ba Shua’yb Ssaleh Bin Ubayd Al A’amiri Ssaleh Maqt’un Ubaid Muhammed Ba Shua’yb Umar Ubaid Abdul Lah Uthman Attamimi 396 Environmental and human determinates of vegetation distribution in the Hadhramaut region, Yemen Species-area curve of the m ountain slope 20 Number of species Number of species Species-area curve for the m ain w adi 15 10 5 0 1 10 100 225 8 6 4 2 0 1 10 Quadrat size 20 15 10 5 0 10 225 Species-area curve for the slope on the plateau Number of species Number of species Species-area curve of the secondary w adi 1 100 Quadrat size 100 225 15 10 5 0 1 Quadrat size 10 100 225 Quadrat size Number of species Species-area curve for the plateau 12 10 8 6 4 2 0 1 10 100 225 Quadrat size Appendix 6. Progressive doubling of quadrat size for minimal area of species-area curves for the different landforms in the study sites. Generally the method showed that 10 x 10 m quadrat (100m²) was adequate for sampling most of the plant species recorded in each landform. Quadrat size is in square metres. 397

RELATED PAPERS

RELATED TOPICS

Log In

Environmental and human determinates of vegetation distribution in the Hadhramaut region, Republic of Yemen

Environmental and human determinates of vegetation distribution in the Hadhramaut region, Republic of Yemen

Related Papers

RELATED PAPERS

RELATED TOPICS