POTENTIAL
MULTIPLE
FOR JOINT MANAGEMENT
AND
USE OF NYUNGWE FOREST, RWANDA
BY
Jean Damascene NDA YAMBAJE
Thesis presented in partial fulfilment of the requirements
for the
degree of Master of Science in Forestry Sciences at
the University of Stellenbosch, South Africa.
Supervisor:
Dr. Isla M. GRUNDY
March 2002
University of Stellenbosch
.'
'1·,
..-- ~
~
Stellenbosch University http://scholar.sun.ac.za
DECLARATION
I, the undersigned,
hereby declare that the work contained
in this
thesis is my own original work and has not previously in its entirety
or in part been submitted at any university for a degree.
Signature
Date
Stellenbosch University
II http://scholar.sun.ac.za
ABSTRACT
Nyungwe Forest in the South West of Rwanda is surrounded by a dense human
population (350 people/km') which depends on subsistence agriculture and livestock.
Previous reports on the Forest have shown that its survival is threatened by illegal use
of its resources. Participatory rural appraisal was conducted in five Communes next
to the Forest to study the community's perceptions, uses and interests with respect to
access and management of the Forest. Respondents perceived the Forest as a source
of 18 categories of products of which only 14 were said to be of value for the
respondents'livelihoods.
Land, timber and minerals (gold) were the priority resources
preferred by more than 55% of the respondents. Other important resources included
fodder/pastures, wooden goods, fuelwood and poles. Fourteen different species were
most preferred for timber, 16 for poles, 45 for medicinal materials and six for wooden
goods. However, the majority of these species were preferred for multiple uses as
timber, poles and medicines.
A forest resource assessment was carried out to identify the stock of the preferred tree
species. Not all woody resources mentioned as being preferred were available, with
some tree species not found at all in the Forest. In order to identify tree species that
can be used sustainably, different criteria including abundance (density), dominance,
diameter size-class distribution and regeneration, were used in combination. Of the
12 tree species recorded in the Forest and most preferred for timber, only three
species were present in sufficient abundance and sizes to allow sustainable utilisation.
Of the 12 pole species identified, six were not vulnerable to exploitation, and of the
preferred medicinal tree species only five were considered to fit into this category.
Low resource availability and the need to address the interests of adjacent
communities necessitated management options which enable access to some resources
and benefits the communities as well as conservation measures to protect the
biodiversity. A range of joint forest management options is discussed in order to
assess the feasibility of a collaborative approach in the management of Nyungwe
Forest. Some recommendations are made with respect to access to resources, the use
of substitutes and areas for future research.
Stellenbosch University
III http://scholar.sun.ac.za
OPSOMMING
Nyungwe Bos in die suidweste van Rwanda word omring deur 'n digte bevolking van
350 mense/km- wat van bestaanslandbou en lewende hawe afhanklik is. Vorige
verslae het aangedui dat die voortbestaan van die Bos bedreig word deur die
onwettige gebruik van sy hulpbronne. 'n Evaluering is in vyf gemeenskappe
(Communes) langs die Bos gedoen om die gemeenskap se persepsies oor, gebruike
van en belangstelling in die toegang tot die Bos, sowel as die bestuur daarvan te
bestudeer. Die mense van hierdie plattelandse gebied is by die evaluering betrek.
Respondente sien die Bos as 'n bron van produkte wat in 18 kategorieë ingedeel kan
word, maar waarvan slegs 14 kategorieë van waarde is vir die respondente om 'n
bestaan te kan maak. Meer as 55% van die respondente het grond, hout en minerale
(goud) as die belangrikste hulpbronne aangedui. Ander belangrike hulpbronne sluit in
veevoer of weivelde, houtartikels, brandhout en pale. Die respondente het 14
verskillende houtspesies verkies as timmerhout, 16 vir pale, 45 vir medisinale
gebruike, en ses vir houtartikels. Die meerderheid van hierdie spesies is egter verkies
vir veelvuldige gebruike soos timmerhout, pale en medisyne.
'n Evaluering van die hulpbronne in die bos is uitgevoer om te bepaal hoeveel bome
van die gunstelingspesies daar in die bos is. Nie al die houthulpbronne wat deur die
respondente verkies is, was beskikbaar nie, en sommige boomspesies is glad nie in die
Bos gevind nie. Ten einde boomspesies te identifiseer wat vir lewensmiddele gebuik
kan word, is 'n kombinasie van verskillende kriteria gebruik, insluitende die
hoeveelheid bome (digtheid), dominansie, die verspreiding van die verskillende klasse
deursneegroottes, en regenerasie. Van die 12 boomspesies in die Bos wat na
aanleiding van hierdie evaluering opgeteken is en wat voorheen as gunstelingspesies
vir timmerhout aangedui is, is daar slegs drie wat volop en groot genoeg is om vir
lewensmiddele gebruik te word. Van die 12 spesies wat vir pale geïdentifiseer is, is
ses teen oorontginning bestand, en van die gunsteling medisinale boomspesies val
slegs vyf in laasgenoemde kategorie.
Lae hulpbronbeskikbaarheid en die behoefte om die aangrensende gemeenskappe se
belange aan te spreek, het bestuuropsies genoodsaak wat toegang tot sommige
hulpbronne moontlik maak en wat die gemeenskappe bevoordeel, sowel as
bewaringsmaatreëls ten einde die biodiversiteit te beskerm. 'n Verskeidenheid
gesamentlike bosbestuuropsies is bespreek ten einde die uitvoerbaarheid van 'n
benadering van samewerking in die bestuur van die Nyungwe Bos te ondersoek.
Enkele aanbevelings is gemaak wat betref toegang tot hulpbronne, die gebruik van
plaasvervangers, en studiegebiede vir verdere navorsing.
Stellenbosch University
IV http://scholar.sun.ac.za
ACKNOWLEDGEMENTS
I wish to express my sincere gratitude to the following people and organisations:
•
The Rwanda Government, through the Ministry of Education and the Rwanda
Embassy in South Africa, for covering my scholarship.
•
The Rwanda Agricultural Research Institute, my employer, for supporting my
application
for a scholarship and for giving me permission to undertake my
postgraduate studies.
•
Dr. Isla M. Grundy, my supervisor, for her guidance, comments and constructive
ideas and interests shown throughout the study.
•
The Nyungwe Forest Conservation Project and its staff, for the provision of
information on Nyungwe Forest and for access to their mini-library.
•
The leaders in the Communes surveyed, for facilitating household interviews and
group discussions.
•
Mrs. A. Sadie, a lecturer at the University of Stellenbosch, for her assistance in
data analysis and interpretation.
•
Prof. D. Ward, a lecturer at the same university, for the provision of ideas for site
selection and sampling design.
•
Mr. J. P. Mwitwa, a Ph.D. student, for encouragement,
assistance
III
data
organisation and analysis.
•
Miss C. Riginos, for checking and correcting grammatical errors and for useful
comments on the presentation of the results.
•
To all my fellow students in Community Forestry, for the exchange of ideas in
data analysis and editing of the initial draft.
•
The Rwandan students in Western Cape, particularly those in the University of
Stellenbosch, for their encouragement during my studies.
•
To Mr. D. Niyibaho, who took me around the Forest and showed patience during
my fieldwork.
•
Finally but not least, I wish to thank my wife and sons who supported my absence
during the course of my studies.
Stellenbosch University
v http://scholar.sun.ac.za
DEDICATION
This thesis is dedicated to my dear wife, Mrs Assumpta, to my sons
Jean Yves, Jean Hubert, Ghislain and Joel Clint, and to my sister,
Donatilla.
Stellenbosch University
VI http://scholar.sun.ac.za
TABLE OF CONTENTS
DECLARA.TI ON
1
11
ABSTRA.CT
OPSOMMING
III
ACKNOWLEDGEMENTS
IV
DEDICATION
V
TABLE OF CONTENTS
VI
LIST OF TABLES
X
LIST 0 F FIGURES
XI
LIST OF APPENDICES
1
INTRODUCTION
XIII
1
1.1
RWANDA:
1.2
POPULATION .........................................................................................•..........
2
1.3
STATUS OF FORESTRY IN RWANDA
3
OVERVIEW
1
1.3.1
Forest plantations
3
1.3.2
Montane forests and savanna woodlands
.4
1.4
FOREST PRODUCTS
6
1.4.1
Wood products
6
1.4.2
Non-wood products
7
1.4.2.1
Medicinal plants
7
1.4.2.2
Honey
8
1.4.2.3
Wildfruit andforestry seeds
8
1.4.2.4
Fodder plants and grasses
9
1.4.2.5
Wildjlowers
9
1.4.2.6
Essential oil plants
10
VII http://scholar.sun.ac.za
Stellenbosch University
2
1.4.2.7
Mushrooms
10
1.4.2.8
Game meat
10
1.4.2.9
Ecotourism
10
1.4.2.10
Other goods
11
1.5
FOREST POLICY AND LEGISLATION
ll
1.6
PROBLEM STATEMENT
13
1.7
JUSTIFICATION OF THE STUDY
14
1.8
AIM AND OBJECTIVES OF THE STUDY
18
1.8.1
Overall aim of the study
18
1.8.2
Objectives of the study
18
1.9
POTENTIAL BENEFITS
19
1.10
THESIS STRUCTURE
19
STUDY AREA AND METHODS
2.1
STUDY AREA ..••••.......•••••••......•...•..........••.......••••.•.......•.................•.•.•.....••••••..
2.1.1
Description ofNyungwe Forest..
21
21
21
2.1.1.1
Size and location ofNyungwe Forest
21
2.1.1.2
Geology and drainage
22
2.1.1.3
Flora
23
2.1.1.4
Fauna
24
2.1.1.5
Management history and conservation interests
25
2.1.1.6
Adjacent communities
28
2.2
RESEARCH METHODS
29
2.2.1
Selection of the study sites
30
2.2.2
Participatory Rural appraisal
30
2.2.2.1
Selection of households
30
2.2.2.2
Selection of groups and key informants
32
2.2.2.3
Data collection
32
2.2.2.4
Data processing and analysis
33
2.2.3
Forest resource assessment
34
2.2.3.1
Sampling design and plot sizes
34
2.2.3.2
Data Collection
36
2.2.3.3
Data analysis
38
Stellenbosch University
VIII http://scholar.sun.ac.za
3
RESUL TS
3.1
PARTICIPATORY
42
RURAL APPRAISAL ••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••
42
3.1.1
Socio-demographic characteristics
42
3.1.2
Farm size and land ownership
.44
3.1.3
Livelihood activities and sources of income
.45
3.1.3.1
Crop farming
47
3.1.3.2
Other plant species grown
48
3.1.3.3
Livestock farming
49
3.1.3.4
Tree crops
51
3.1.4
Soil fertility status and management
54
3.1.5
Respondents' perception ofNyungwe Forest resources
54
3.1.6
Resource preferences and ranking
56
3.1.7
Tree species preferences and uses by the respondents in the
study area
59
3.1.7.1
Timber tree species
59
3.1.7.2
Building poles
60
3.1.7.3
Medicinal plants
61
3.1.8
Uses of wild foods
63
3.1.9
Respondents' knowledge of wildlife species
64
3.1.10
Current resource access and use
66
3.1.11
Respondents' perception of conflicts and conflict resolution
66
3.1.12
Respondents' perception of past and present forest management..
68
3.1.13
Respondents' views of appropriate institutions for forest
management
3.2
FOREST RESOURCE ASSESSMENT
68
71
3.2.1
Identification of tree species
71
3.2.2
Tree species abundance and density analysis for the study area
72
3.2.3
Diameter size-class distribution of trees among and within the
study sites
76
3.2.4
Presence of preferred tree species in the study area
77
3.2.5
Abundance of the most preferred tree species in the sample area
80
3.2.6
Dominance of the most preferred tree species
82
Stellenbosch University
IX http://scholar.sun.ac.za
3.2.7
3.2.8
4
Size-class distribution and regeneration of most abundant
preferred tree species
83
Quality of harvest able tree species for timber and poles
86
DISCUSSION
4.1
RESOURCE
88
PREFERENCES,
USES AND AVAILABILITY IN
NYUNGWE
FOREST ••••••••••••••.••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••.••••••••••••••••
88
4.1.1
Wood products
88
4.1.1.1
Timber
88
4.1.1.2
Building poles
90
4.1.1.3
Fue/wood and charcoal
90
4.1.1.4
Wooden items
91
4.1.1.5
Bean stakes
92
4.1.2
Non-wood products
93
4.1.2.1
Land
93
4.1.2.2
Gold mining
94
4.1.2.3
Wildlife and ecotourism
95
4.1.2.4
Medicinal plants
96
4.1.2.5
Fodder plants andpastures
98
4.1.2.6
Beekeeping.
99
4.1.2. 7
Wildfoods
100
4.2
CHARACTERISTICS
4.3
POTENTIAL
OF THE PREFERRED TREE SPECIES •••••••••••••••••••••••••••••••I0l
SUPPLY AND VULNERABILITY
OF THE PREFERRED
TREE SPECIES TO EXPLOITATION
4.4
5
POTENTIAL
104
FOR JOINT MANAGEMENT OF NYUNGWE FOREST ••••••••••••••••••107
CONCLUSION AND RECOMMENDATIONS
114
REFERENCES
119
APPENDICES
130
x http://scholar.sun.ac.za
Stellenbosch University
LIST OF TABLES
Table 1.1. Distribution of forest plantations in Rwanda by tree species and
ownership in 1990
4
Table 1.2. Protected forests' cover change in Rwanda between 1960
and 1999
6
Table 2.1. Latin and scientific names of some primates found in
Nyungwe Forest.
Table 3.1. Education level of the respondents in the study area
25
.44
Table 3.2. Important food and cash crops grown by the respondents in
the study area
47
Table 3.3. Estimates of crop harvests by the respondents in the study area
.48
Table 3.4. Numbers and distribution oflivestock types in the study area
.49
Table 3.5. Causes of soil fertility decline listed by the respondents in the
study area
54
Table 3.6. List of important resources for community's livelihoods in the
study area
55
Table 3.7. Assessment of the availability of resources in Nyungwe Forest by
groups in the study area
Table 3.8. Ranking of resource categories in the study area
56
57
Table 3.9. Tree species most preferred for timber by the survey respondents around
Nyungwe Forest
60
Table 3.10. Tree species whose stems are preferred for building poles in the study
area
Table 3.11. Plant species used for medicinal purposes in the survey area
61
62
Table 3.12. Wild food resources found in Nyungwe Forest, as mentioned by the
respondents in the study area
63
Table 3.13. Respondents' knowledge of locally extinct wild animals in Nyungwe
Forest.
Table 3.14. Nature of conflicts listed by the respondents in the study area
66
67
Table 3.15. Importance of conflicts over resource use and access, as rated by the
respondents in the study area
Table 3.16. Proposed institutions for management by the respondents
67
68
Stellenbosch University
XI http://scholar.sun.ac.za
Table 3.17. Individual responsibility of the respondents for sustainable forest
management in the study area
70
Table 3.18. Categories of trees used during the analysis
71
Table 3.19. Analysis of variance of the densities of trees among tree categories
72
Table 3.20. Occurrence of the 10 most dominant tree species, by site and tree
category
73
Table 3.21. Diameter size-class distribution oflarge trees among the study sites
76
Table 3.22. Kolmogorov - Smimov tests comparing distributions of trees
of different sizes in pairs of sites
76
Table 3.23. Chi-square analysis to investigate the distribution of species
among the study sites and within diameter size classes
77
Table 3.24a. List of preferred timber and pole species and their occurrence
categories in the study area
79
Table 3.24b. List of preferred medicinal species and their occurrence categories in
the study area
79
Table 3.25a. Relative densities of the preferred timber tree species
80
Table 3.25b. Relative densities of the preferred pole tree species
81
Table 3.25c. Relative densities of the preferred medicinal tree species
81
Table 4.1 a. Classification of the preferred timber and pole species
into 'vulnerable' and 'non-vulnerable'
categories
l05
Table 4.1 b. Classification of the preferred medicinal tree species
into 'vulnerable' and 'non-vulnerable'
categories
l05
LIST OF FIGURES
Figure 1.1. Geographical location of Rwanda
Figure 2.1. Nyungwe Forest, showing altitudes inside the Forest.
2
22
Figure 2.2. Communes neighbouring Nyungwe Forest, showing the survey
sample distribution
31
Figure 2.3. Subdivision and size of the sample units used for forest resource
assessment, N yungwe Forest.
35
Figure 2.4. Different levels of stem quality used in the resource assessment,
Nyungwe Forest.
Figure 3.1. Age classes of the respondents in the studyarea
38
.43
Stellenbosch University
XII http://scholar.sun.ac.za
Figure 3.2. Family size of the respondents in the study area
.43
Figure 3.3. Farm sizes of the respondents in the study area
.44
Figure 3.4. Processes ofland acquisition by the respondents in the study area
45
Figure 3.5. Priority enterprises of the respondents in the study area
46
Figure 3.6. Sources of fodder during the dry season in the study area
50
Figure 3.7. Plantation areas owned by the respondents in the study area
51
Figure 3.8. Tree species planted by the respondents in the study area
51
Figure 3.9. Annual income from the sale of charcoal by the respondents in the
study area
52
Figure 3.10. Annual income from the sale oflumber by the respondents in the
study area
53
Figure 3.11. Annual income from the sale of building poles by the respondents
in the study. area
53
Figure 3.12a. Stem densities per hectare of the 10 most dominant tree species
represented by small trees
74
Figure 3.12b. Stem densities per hectare of the 10 most dominant tree species
represented by medium trees
74
Figure 3.12c. Stem densities per hectare of the 10 most dominant tree species
represented by large trees
Figure 3.13.
74
Comparison of stem densities per hectare of the 10 most dominant
species between tree categories, by study site
Figure 3.14. Occurrence categories of preferred tree species in the sample area
75
78
Figure 3.15. Size-class distribution of the most abundant and dominant trees
preferred for timber, poles and medicines
84
Figure 3.16. Relative abundance of seedlings and saplings of the abundant and
dominant tree species across the five sites
85
Figure 3.17. Proportion of trees with stem suitable for timber.
87
Figure 3.18. Proportion of trees with stem suitable for building poles
87
XIII http://scholar.sun.ac.za
Stellenbosch University
LIST OF APPENDICES
Appendix 1. Questionnaire for household interviews
Appendix 2. List of on-farm medicinal plants mentioned by the respondents in the
survey area.
Appendix 3. Palatable tree and grass species mentioned by 5 or more respondents in
the survey area
Appendix 4. List of preferred timber tree species by the respondents in the study area.
Appendix 5. List of preferred medicinal plants found in Nyungwe Forest, as
mentioned by the respondents.
Appendix 6. List of scientific and vernacular names of tree species recorded in all
study sites.
Appendix 7. Relative frequencies of trees recorded within the large tree category.
Appendix 8. Relative frequencies of trees recorded within the medium tree category
Appendix 9. Relative frequencies of small trees recorded within the small tree
category.
Appendix 10. Basal areas of tree species within the large tree category in the sample
area.
Stellenbosch University http://scholar.sun.ac.za
1
1.1
INTRODUCTION
RWANDA:
OVERVIEW
Rwanda is a small (26 338 km') land-locked country in Central Africa between
latitude 1°05' and 2°50' South and longitude 28°50' and 30°25' East.
It borders
Burundi in the South, Uganda in the North, Tanzania in the East and the Democratic
Republic of Congo in the West (Figure 1.1). It is hilly with altitudes less than 1 500
m in the eastern plateau but rising to between 1 500 and 2000 m in the central plateau
area and higher in the West and North.
The following geographic regions are
distinguished, moving from West to East (MINETERRE, 2000):
•
The north-south ridge of the watershed between the Congo and the Nile, with
peaks rising to between 2 400 mand 3 000 m. The remnants of Rwanda's natural
forests are found on this ridge;
•
The highlands that are an extension of those in Uganda and form the centre of the
country between the Congo-Nile Crete in the West and the eastern zone, and
decrease in altitude from West to East. Within the altitudinal range of 1500 mand
2000 m, the centre of the country is characterised by a mosaic of hills with
rounded tops, separated by large swamps which cover almost half of the country's
area. A ridge of five volcanoes in the far North-West dominates the Congo Nile
Crete.
From the West and moving eastwards are found the peaks of Kalisimbi
(4507 m), Bushokoro (3 711 m), Sabyinyo (3 634 m), Gahinga (3 474 m) and
Muhabura (4 127 m).
•
A large eastern zone of hills and swampy lakes, located from Kigali to the border
with the Republic of Tanzania, occurs at altitudes between 1 000 and 1 500 m. A
large section of this zone lies within Akagera National Park.
Rwanda has a sub-tropical climate with an average annual temperature of 18°C. The
average rainfall is 1250 mm, falling in two seasons separated by two dry seasons.
2
Stellenbosch University
http://scholar.sun.ac.za
UGANDA
.~
BURUNDI
'_Rusumo
._
0
... 0
•
"'''''
30mi
Figure 1.1. Geographical location of Rwanda
1.2
POPULATION
The population of Rwanda was estimated to be 7.733 million in 2000 with a natural
increase rate of 2.1 % and a density of 320 inhabitants per km2 (ADB, 2001). The
rural population density of Rwanda is one of the highest in Africa (MINAGRI, 1998).
In 1991, 94% of the population lived in rural areas and the remaining 6% in urban
areas. A socio-demographic survey by MINECOFIN (1998) found that 91.1% of the
active population is in the agricultural sector, 1.7% in industry and 7.2% in the
services sector. At national level, the average size of the household is 5 persons and
the farm size less than one hectare. The current demographic features lead to mixed
farming where diversified cropping and livestock farming are practised. The farming
techniques are exclusively manual (MINAGRI, 1998).
About 70% of 1.3 million ha of land is cultivated, the majority of it on family farms.
The latter are found on all types of land, including land of marginal quality for
agricultural production (MINAGRI, 1998). A large number of these farms are not
profitable and households live on the threshold of rural poverty. In 1997, the GDP
per capita was US$ 240 (MINECOFIN, 1998), which was less than half of the most
Stellenbosch University
3 http://scholar.sun.ac.za
recent estimate for low income Sub Saharan Africa of US$ 490 (World Bank., 1998).
The main constraints to production include a strong pressure on natural resources, the
natural land infertility and the lack of use of agricultural inputs (MINAGRI, 1998).
Land shortage is particularly prevalent in high altitude zones (MINAGRI, 1997).
1.3
ST ATUS OF FORESTRY IN RWANDA
1.3.1 Forest plantations
Forest plantations have been established in Rwanda since 1920 with the introduction
of fast growing exotic tree species, mainly Eucalyptus spp. (Eucalyptus tereticornis
and E. saligna), for the purpose of protecting high mountainous areas from erosion
and for the supply of fuelwood.
been established.
By 1967, about 20 000 ha of forest plantations had
Wide scale plantation
forestry
started
in 1967 with
the
commencement of the first donor-funded Forest Project, Projet Pilote Forestier (PPF).
From 1967 to 1975, PPF established 5500 ha in Kibuye Prefecture.
The year 1975
was a turning point for plantations with the start of a tree planting campaign and large
scale development
organisations
(Habiyambere,
projects,
abroad.
1997).
supported
financially
and technically
by several
The total area planted during that year was 25 500 ha
In 1990, 247 500 ha of forest plantations
were created
(Habiyambere, 1997). Between 1993 and 1995, all silvicultural activities stopped due
to war and genocide in Rwanda.
In 1995, these activities started again, including
establishment and rehabilitation of damaged plantations.
The total forest plantations
were estimated at 247500 ha in 1991 (Habiyambere, 1997) and at 256300 ha in 1997
(MINITERRE, 2000).
Apart from Eucalyptus spp., other new exotic tree species in plantations
Grevillea
robusta,
Pinus patula,
Cupressus
lusitanica,
Callitris
include
spp., Acacia
melanoxylon and Casuarina spp. These are established largely on government lands,
as only a small proportion of plantation land is privately owned (FAO, 1999; Mihigo,
1999). A few local tree species were planted, particularly in buffer zones around
indigenous forest reserves (Habiyambere, 1997). Beside plantation forestry, trees on
farmer's fields are an integral part of the production systems.
tree species with agroforestry potentialities
Both local and exotic
are mixed or inter-planted
with food
Stellenbosch University
4 http://scholar.sun.ac.za
crops.
They are normally established around the homesteads, along erosion control
ditches, scattered on farms or used to demarcate farms.
In 1989, trees on farms
covered an estimated area of 87 200 ha (FAO, 1999). Effective inventory data for
forest plantations are reported to be unavailable (FAO, 1999).
Based on ownership, forest plantations in Rwanda are divided into governmental,
communal and private categories.
The area of forest plantations belonging to the
second category was the largest in 1990 (Table 1.1).
Table 1.1. Distribution
ownership in 1990.
of forest plantations
in Rwanda by tree species and
70
Total
Area
(ha)
161 010
4360
5
32630
13
8
8720
10
21 550
9
6940
7
-
-
Il 220
5
3
2970
3
-
-
4800
2
-
-
-
-
4360
5
4360
2
Casuarina spp.
1 230
2
1 980
2
-
-
3210
1
Others
-
-
-
-
8720
10
8720
3
Total
61200
100
99 100
100
87200
100
247500
100
Species
Eucalyptus spp.
Government
Area
%
(ha)
30600
50
Communal
%
Area
(ha)
69370
70
Private
Area
(ha)
61 040
Pinus patuia
18360
30
9910
10
Cupressus lusitanica
4900
8
7930
Acacia menaloxylon
4280
7
Callitris spp.
1 830
Grevillea robusta
%
%
65
..
Source: Mihigo (1999), MINAGRI (1997) and Murererehe (1999)
The plantation
Habiyambere
forest
cover declined
following
the
1994 cnsis
In
Rwanda.
(1997) reported that the large displacement of people and cutting of
forest plantations for fuel wood and shelter needs resulted in a loss of 15 000 ha and
extensive damage to 25 000 ha.
1.3.2
Montane forests and savanna woodlands
Rwanda's protected forests consist of closed forests that are generally submontane or
montane, concentrated in the west of the country, and in the mountains bordering
Kivu Lake (FAO, 1999).
deciduous
The submontane forests are mainly broad-leaved,
forests and are characterised
by species such as Syderoxylon
moist
adolfi-
friederici and Entandrophragma excelsum. These are largely secondary growth, with
Stellenbosch University
http://scholar.sun.ac.za
5
most of the original forest having been cleared and only small patches of primary
forest remaining in inaccessible areas. Above 2 000 m, the predominant forest type is
broadleaved moist evergreen forest, with characteristic species including Syzygium
parvifolium, Xymalos monospora and Psychotria spp. Stands of Arundinaria alpina
are found in the mountains of the Zaire-Nile ridge above 2 300 m to 2 400 m in
almost pure condition or mixed with Podocarpus
latifolius and Polyscias fulva
(Bosshard, 1971). Arundinaria stands are also found in the Birunga National Park
bordering Uganda and the Democratic Republic of Congo.
Major montane forests in Rwanda include three natural forests, namely Nyungwe,
Mukura
and Gishwati. These forests belong to the afromontane
forests of the
Albertine Rift and are characterised by high biodiversity (MINITERRE, 2000). The
Volcano National Park, in the northern mountainous border of Rwanda and the
Democratic Republic of Congo and Uganda, is a remnant montane forest set aside as
a forest reserve since 1933 (Spinage, 1972). The east of the country has extensive
areas of savanna woodland, with Acacia seyal and Combretum binderanum found on
deep soils and Pappea ugandensis on lateritic soils (Hecq and Froment, 1961). Open
sclerophyllous
forests of Croton dichogamus and Euphorbia dawei cover part of
Akagera National Park (FAO, 1999). Savanna woodland and gallery forests make up
about 163 ha and are home to a variety of plant species used as traditional medicines,
foods and for other human needs (MINITERRE, 2000).
Pressure for agricultural land, settlement and forest products has led to a reduction in
size and to an irreversible loss of biodiversity of protected forests (MINAGRI, 1998).
Table 1.2 shows the reduction in size of these areas between
1960 and 1999.
Nyungwe Forest has undergone a significant rate of deforestation.
lts current area is
89 150 ha (Gapusi, 1999), much bigger than that of the second largest existing natural
forest, Mukura (1600 ha).
remaining
The Gorilla sanctuary, Birunga National Park, has a
area of 12 760 ha from an initial cover of 30 000 ha in 1924
(Twarabamenye,
1999). There is no precise estimate of the cover of Gishwati Forest
and savanna woodlands.
Parts of Gishwati Forest and Akagera National Park have
been declassified for human settlement (Mihigo, 1999; Williams and Ntayombya,
1999).
6 http://scholar.sun.ac.za
Stellenbosch University
Table 1.2. Protected forests' cover change in Rwanda between 1960 and 1999
(MINAGRI, 1999).
Cover
Forest cover (ha)
Protected forest
1960
1970
1980
1999
1996
1990
change, %
89150
21.8
1600
1600
46.7
14000
12760
12760
62.5
241 000
220000
90000
66.3
Nyungwe
114025
108800
97000
97000
94500
Gishwati
28000
28000
23000
8800
3800
Mukura
3000
3000
2000
2000
Birunga
34000
16000
15000
Akagera
267000
267000
267000
Other*
150000
150000
90000
50000
20000
Total
596025
572 800
494000
412800
352660
86.7
193 510
* Gallery and savanna woodlands.
1.4
1.4.1
FOREST PRODUCTS
Wood products
The major products from exotic and indigenous tree species include timber, poles,
crafts and fuelwood.
Quantitative data are disparate and only imprecise estimates
exist for these products. Eucalyptus poles are used chiefly for construction purposes.
The annual consumption of construction wood in the country's
capital, Kigali, is
about 15 000 to 20 000 m3 per year, while it averages 40 000 m3 in rural areas
(Habiyambere,
are Eucalyptus
1999). Timber is entirely produced by pitsawyers.
Preferred species
spp. (45%), Cupressus spp., and Grevillea robusta.
consumption of timber is estimated at 36 000 rrr' (MINECOFIN,
from a standing volume of about 72 000 m3 (Habiyambere,
The annual
1998), processed
1999).
Eucalyptus
represents 45% of the market for timber; Cypress and Grevillea both 23% and Pinus
4.5%.
The remaining 4.5% of this market is made up by imported wood from
neighbouring
countries
Entandrophragma
and
consists
of
high
value
tree
species
such
as
excelsum.
MINECOFIN (1998) reported that forestry supplies 96.2% of the energy consumed in
the country, of which 90.8% is firewood and 5.4% charcoal.
In 1997, fuelwood
consumption averaged 7.1 million nr', 5% of which was used by small industries. For
charcoal, which is traditionally produced and mainly consumed in urban areas, the
Stellenbosch University
7 http://scholar.sun.ac.za
annual production amounted to 46 000 tonnes, most of which (80%) was sold in the
country's capital, Kigali (Habiyambere, 1999).
In 1996, the charcoal deficit was approximately 4.5 million rrr' per year (FAO, 1996).
The result is the use of crop residues and other dead materials by rural people in place
of fuelwood and charcoal (Niang and Styger, 1990). In the absence of alternative
sources of energy, the forestry sector will continue to be the major source of energy
supply. However, the sustainability criteria of woodlands and protected forests will
not be met due to encroachment through fuel wood gathering and charcoal for sale in
towns (Habiyambere, 1999).
1.4.2
Non-wood products
Non-wood forests products (NWFP) are still considered as secondary forest products
in Rwanda (Murekezi, 1999). For many years, forests were considered as source of
wood products, namely fuelwood, timber, building poles and woodcraft.
Today, the
interest in NWFP is gaining popularity as responses to national and international
concern over the conservation of biodiversity and the multiple roles ofNWFP in local
livelihoods (FAO, 1992). In Rwanda, NWFP are found in managed forest plantations,
in indigenous forests and woodlands.
Commonly used NWFP include medicinal
plants for traditional human and veterinary medicine, wild ornamental plants, honey,
wild fruit, tree seeds, essential oil plants, mushrooms, game meat, and ecotourism
(Murekezi, 1999).
1.4.2.1
Medicinal plants
Medicinal plants used in traditional human and veterinary medicine are increasingly
in demand for research interests and traditional medicinal purposes.
Mbarubukeye
(1992) found that 80 000 patients
For instance,
have visited the Centre for
Traditional Medicine of Bare in south-eastern Rwanda, since its creation in 1978
(Kayinamura, 1986).
Additionally, the Pharmaceutical
Centre (CURPHAMETRA)
within the Rwanda Institute of Scientific and Technological Research (IRST), based
in Butare Province in the South of the country, trains traditional healers' groups in
collaboration with medical doctors in order to enhance the value of medicinal herbs.
Stellenbosch University
8 http://scholar.sun.ac.za
In veterinary medicine, 536 healers were identified in seven provinces of Rwanda
(Mbarubukeye, 1992).
Woody plants whose parts are used in traditional human and veterinary medicine are
found in natural forests, in savanna woodlands or in forest plantations.
Mbarubukeye
and Niang (1996) outlined a list of 59 tree species used in traditional human medicine
and 39 tree species for veterinary medicine.
Despite their value, these species are
disappearing
through
due
to population
pressure
deforestation
and
burning
(Habiyambere, 1999).
1.4.2.2 Honey
Forests and woodlands play an important role in beekeeping.
Habiyambere (1999)
reported a list of 27 plant species associated with honey production in Nyungwe
Natural Forest and 14 plant species outside this Forest.
The potential for honey
production in natural forests and forest plantations is not fully exploited (Nyilimana,
1997). Because harvesting methods are still traditional and include the use of fire,
forests are at high risk of burning.
At National level, honey production declined between 1990 and 1994 due to the death
of beekeepers during the genocide and following massive destruction of hives and
other bee-keeping
materials and equipment (Murekezi,
1999).
In 1989, honey
production reached as high as 441 365 kg (Murekezi,
1999).
Since 1994, the
production has been well below average. Available data on honey production in 1998
at the Rwanda Association for the Promotion of Integrated Development
(ARDI)
indicated a production level of 23 000 kg (ARDI, 1998).
1.4.2.3 Wild/ruit and/orestry seeds
According to Murekezi (1999), wild fruits are present in the forests but they receive
little interest from the people, who prefer to consume fruits from agroforestry
orchards.
Wild fruits are generally eaten by children and by adults only during the
periods of food deficit.
available.
Quantitative information on wild fruit production is not
In Nyungwe Forest, woody plants producing edible wild fruits were
9 http://scholar.sun.ac.za
Stellenbosch University
identified
as Clerodendron
(Murekezi,
1999).
fuscum,
Parinari
excelsa,
and Myrianthus
holstii
In savanna woodlands, edible fruits are produced by Rumex
maderensis, R. usambarensis, Cajanus cajan, Carica edulis, Rhus vulgaris, Solanum
taitente and other Solanum spp. (Dessouter, 1991).
Local collection of tree seeds in natural stands and plantations is carried out by the
Tree Seed Centre (TSC) within the Rwanda Agricultural Research Institute (ISAR).
The seeds are supplied for tree planting and afforestation
country.
activities all over the
Before 1994, the annual delivery of tree seeds by TSC was nine tonnes,
purchased by both tree planting projects and private growers (CGF, 1993).
1.4.2.4
Fodder plants and grasses
Natural forests and savannas are valuable sources of animal feeds. However grazing
is not allowed within state's forests.
In the case of Nyungwe Forest, silvopastoral
practices were permitted in new forest plantations in the buffer zone. About 20 grass
species were grazed by livestock on the ground layer of these plantations (Gasana,
1988).
The pressure of pastoralists on Gishwati Natural Forest resulted in the
degazettement of 1200 ha for grazing between 1983 and 1993 (Murekezi, 1999). The
Akagera National Park in the eastern region of the country contains palatable grass
species, the most important of which are Themeda triandra and Panicum maximum
(Mvukiyurnwami, 1987).
1.4.2.5
Wildflowers
Many wild plant species are invaluable sources of horticultural material. Fischer and
Hinkel (1992) produced a list of some species producing spectacular
individual
blooms or mass displays of colour and interesting forms. Unfortunately, the export
market of cut flowers is not developed.
Netherlands
and Belgium.
Potential markets have been identified in the
A local co-operative,
Highland Flower Association,
exports about 1500 kg of flowers per week; but exports are limited by the number of
flights to Kanombe International Airport (Kigali) where only one airline makes two
trips per week (Murekezi, 1999).
10 http://scholar.sun.ac.za
Stellenbosch University
1.4.2.6 Essential oilplants
Many local plant species are used to produce medicines and perfumes in Rwanda.
The Rwanda Institute for Scientific Research and Technology (IRST) extracts various
oils which are used for pharmaceutical purposes. Some of the plant species producing
valuable oils are Plantago lanceolata, Calendula officinalis, Datura stramonium,
Eucalyptus globulus, E. smithii, Capsicum frutescens, Neorautanenia mitis, Thymus
vulgaris,
Pentas
longiflora,
Pelarganium
graveolens,
P. raduta,
Cymbopogon
afronardus, C. winterianus, C. citratus and C. jlexanomus (MICOMART, 1998).
1.4.2.7 Mushrooms
Wild mushrooms
occur in most of Rwanda's
forests and are associated
with
Termitomyces (Hayes, 1991). Artificial production of these mushrooms is not yet
possible due to difficulties in collecting the micro-organisms (Murekezi, 1999). Thus
the potential productivity of wild mushrooms is not sufficiently enhanced.
women and children collect and eat the mushrooms.
mushrooms
Only
However, the demand for
increases in towns as well as in rural areas.
At the national level,
artificial production of mushrooms in 1991 was about six tonnes (Hayes, 1991). The
potential for mushroom production is high and the export market is available either in
Europe or in Africa (Murekezi, 1999).
1.4.2.8 Game meat
In the past in Rwanda, game meat was obtained from buffalo, antelopes, zebra, guinea
fowl, and duikers among others, through hunting.
Today, illegal hunting
for
bushrneat is still underway in protected forests. As a result, a few mammals and birds
are present in both Akagera National Park (Ndayambaje,
1999) and in Nyungwe
Forests (Gapusi, 1999) as a result of human disturbances.
1.4.2.9 Ecotourism
Rwanda's forests and savannas are diverse.
The country has a number of endemic
plant and animal species with high ecotourism potential.
Akagera National Park,
Stellenbosch University
11 http://scholar.sun.ac.za
Nyungwe Natural Forests and Birunga National Park are important wild lands that,
when their maximum exploitation for ecotourism is attained, will contribute a large
share towards
the availability
of foreign
exchange
in the national
economy
(Habiyambere, 1999). For instance in 1989, the turnover from ecotourism was about
US$ 1 078000 of which US$ 475 500 accrued from Birunga National Park (ORTPN,
1989).
This income takes account of direct revenues from entry fees and is
considered to be low in relation to the ecotourism potential of the country (Murekezi,
1999). Forest reserves are the major focus of the ecotourism industry (MINAGRl.
1998). The challenge, however, is to satisfy the interests of the visitors and at the
same time avoid social conflict and pressure on the environment.
The best way is to
ensure that both the country and the communities share the benefits from conservation
of ecotourism resources.
1.4.2.10 Other goods
Various other goods are manufactured
stands.
from plants obtained mainly from natural
These are used to make various crafts and curios. Mbarubukeye and Niang
(1996) illustrated a number of plant species whose parts are used to manufacture
various objects, including baskets, mats, beehives and calabashes.
1.5
FOREST POLICY AND LEGISLATION
Rwanda lacks a formal forest policy and legislation.
The first piece of legislation
providing for protected areas dates back to 1925 when the country, together with
Burundi and the Democratic Republic of Congo, were under Belgian colonial rule
(Spinage, 1972). Official reserves were created by the Belgian authorities in 1933
(Vedder and Weber, 1984). Since Rwanda achieved its independence in 1962, Law
Ordinances have provided for establishment of managed areas. The Law Ordinance
of 18 June 1973 established rules applicable to national parks, hunting reserves, and
strict nature reserves.
December
This law has been slightly modified by the Decree Law of
18, 1973, establishing by presidential decree special reserves, national
parks and hunting reserves.
12 http://scholar.sun.ac.za
Stellenbosch University
In 1988, Rwanda reshaped its forestry legislation, with the adoption of Law No.
47/1988 (JORR, 1989). This law was based on an old decree of 18 December 1930
that provided, among other things, for the creation of forest reserves
and the
protection of forest species. Dating from the colonial mandate, it applied to felling
carried out by, or for, European settlers, leaving the Rwandan population to use the
forest in conformity with their customary rights.
The rapid population
growth
resulted in severe deforestation that the old legislation was unable to control (FAO,
1989).
In response, the revised forest legislation defined a legislative framework for action
and intervention in forestry.
community
It highlighted aspects of forest ownership
and private forest land), forest conservation
(public,
and protection,
forest
establishment and tending, silviculture and management, harvesting, funding of forest
programmes,
national
and communal
forest planning,
as well as forest
law
enforcement.
Generally, the State reserves the right of ownership and control of all
forest management operations (MINAGRI, 1984; JORR, 1989). In 1984, because of
the conservation importance of natural forests, the Rwandan Government adopted an
Action Plan aimed at maintaining essential ecological processes and at preserving
genetic diversity of flora and fauna as well as satisfying national demand for direct
and indirect benefits from forests (MINAGRI, 1984). Major programmes within the
plan focussed on:
•
The establishment of integral natural reserves of 59 340 ha, of which 15 000 ha
are in Birunga National Park, 5 000 ha in Gishwati and 39 840 ha in Nyungwe
forests;
•
The demarcation of 7 500 ha of forests for restricted use;
•
The establishment of buffer zones between the different forests and the adjacent
communities.
In 1987, Rwanda adopted a ten-year National Forest Plan (1987 - 1997). This plan
defined objectives, perspectives and a plan of activities in the forest sector.
It has
been a very useful document but its implementation has sometimes been constrained
by disparate objectives and means, very wide interpretation of its content and lack of
monitoring and evaluation for a regular review (AGRIFOR CONSULT, 1997).
13 http://scholar.sun.ac.za
Stellenbosch University
1.6
PROBLEM STATEMENT
Protected areas in Rwanda have undergone a tremendous rate of regression since
1960. Gapusi (1999) reported a regression rate of21.8% ofNyungwe Forest between
1960 and 1998.
The study by Muderevu (1974) estimated the annual rate of
regression of Nyungwe Forest at about 750 ha. He concluded that the Forest would
disappear completely within a period of 84 to 134 years if adequate strategies are not
adopted and implemented.
in Nyungwe
Many authors have indicated that the loss of biodiversity
Forest results from illegal use (Gapusi
1999; MINAGRI,
1998;
MINITERRE, 2000).
Protected
forests in Rwanda are under acute threat by adjacent
(MINITERRE, 2000).
communities
The underlying causes are reported to be high population
densities, overwhelming poverty and dependence on land and other natural resources
(MINAGRI,
1998).
Therefore, these factors result in great land and resource
degradation, and consequent pressure on forest reserves.
Both at national and local
levels, the capacity to manage forest reserves with respect to human and financial
resources is inadequate (WCS and ORTPN, 1995). As a result of these factors, the
legal status, mandate and management of these forests have suffered substantially.
Various strategies have been proposed by many authors and consultants (IUCN, 1987;
IUCN and WWF, 1985; Mackinon and Mackinon, 1986; Perlez, 1988; Wilson, 1991)
on how to improve the conservation, protection and management of forest reserves in
Rwanda. However, resource degradation through encroachment by local people has
been increasing over the years and culminated during the 1994 crisis in Rwanda.
Various studies from Rwanda have emphasised the importance of improving the
relationship between local communities and the management of the forest reserves
(Bahigiki and Vedder,
1987; Gapusi,
1999; IUCN,
1979; MINITERRE,
Renner, 1991; Renner, 1992; Vedder and Weber, 1984; Weber, 1992).
2000;
Constant
pressures on protected forests, however, have resulted in incremental loss of resources
from the reserves.
The survival of Rwanda's forest reserves depends on mechanisms that recognise the
need for access and use by local people, while simultaneously co-ordinating
local
14 http://scholar.sun.ac.za
Stellenbosch University
development and conservation
of forest reserves.
Following gazettement
of all
natural forests as forest reserves in 1984, these local communities received limited
access to some resources (Renner, 1991). At present, there is a tendency by relevant
state departments to seek ways of accommodating the needs and interests of local
communities, with increased focus on maintaining a balance between conservation
and utilisation of forest resources in protected areas (MINETERRE, 2000). Therefore
sustainable utilisation of forest resources in protected areas, including Nyungwe
forest, is a mechanism for involving communities in conservation and management of
these areas.
1.7
JUSTIFICATION
OF THE STUDY
Prior to the establishment of Nyungwe Forest as a Forest Reserve, the local people
around the Forest had access to wood and NWFP. From 1933, they were barred from
removing the forest products from the Forest (Mbonyintwari,
1989). Consequently,
conflict between the Forest Department and the local people began.
For example,
during the establishment of the buffer zone between the local communities and the
Reserve, the land that was previously exploited by local people was taken back by the
authorities, instilling disenchantment in the people (Renner, 1992). On gazettement
ofNyungwe Forest as a Forest Reserve, local households utilising the buffer zone for
settlement, agriculture
and livestock farming were settled outside this zone or
compensated for having to leave the Forest.
After the people left, conservation
measures were instituted involving the planting of exotic tree species.
This, it was
believed, would serve to mark the boundaries of the Nyungwe Forest as stated in the
1984 Action Plan for the conservation and management of natural forests on the
Congo Nile Crete, where Nyungwe Forest is located (MINAGRI, 1984).
With respect to conservation
of Nyungwe Forest, the Rwandan Government,
in
collaboration with funding agencies, adopted strategic actions aimed at reducing the
dependence of local communities on the Forest. Access to some forest resources was
also granted only in the buffer zone, including
fuelwood
gathering,
taungya
agroforestry and beekeeping (Barbier, 1992). Some of these rights to local people
were established following socio-economic studies and recommendations
by various
Stellenbosch University
15 http://scholar.sun.ac.za
consultants including Bahigiki and Vedder (1987), Barbier (1992), MINAGRI (1992),
Renner (1991), Stebier et al. (1984), and Weber (1992).
Socio-economic studies show the importance of the Forest to local communities in
terms of wood and NWFP (Bahigiki and Vedder, 1987; Renner, 1992). On the other
hand, Nyungwe Forest is also an important water catchment's
invaluable role in soil protection (Budowski, 1976; Sorg, 1978).
area and plays an
Despite all these
advantages and perceptions, Nyungwe Forest is threatened by illegal activities that
impact
negatively
on its size and biodiversity
(MINAGRI,
1998).
Human
disturbances and unlawful activities are the major causes of regression and losses of
plant and animal species (Gapusi, 1999; Muderevu, 1974).
At present, Nyungwe Forest is managed by the Nyungwe Conservation
Project
(PCFN) in partnership with the Rwanda Office of Tourism and National Parks
(ORTPN) with limited financial and human resources (Barakabuye, 2001). Activities
in Nyungwe Forest and in the buffer zone are limited to forest protection and public
awareness programmes,
ecological studies and monitoring of animal populations
(WCS and ORTPN, 1995). Tourism activities are not fully enhanced.
Against this
background, opportunities for income-generating activities to local people are too few
or lacking. Few local people are employed as forests guards for the protection of the
Forest. The revenue from tourism in Nyungwe Forest is too little and has no impact
on the livelihoods of the local communities (Weber, 1992). The major focus of the
management
of the Forest is to contain all illegal activities coming from the
neighbouring communities,
The Government of Rwanda, represented by the Ministry of Land and Environment,
has vested interests in the conservation
of Nyungwe
Forest, with its rare and
endangered flora and fauna and potential for sustainable ecotourism, for which only a
small area of the original size is available (MINAGRI,
poverty, local peoples'
1997).
Faced with rural
strategies include access to land through
deforestation,
agricultural intensification, migration, and off-farm income activities in rural areas
(Habimana, 1982; Weber, 1989).
Stellenbosch University
16 http://scholar.sun.ac.za
It may be futile to aim at conserving the biological diversity in Nyungwe Forest
without understanding the problems underlying its conservation.
Addressing the
needs of the neighbouring communities can partly alleviate threats to conservation.
In
addition to tangible benefits, intangible benefits that local people derive from the
Forest near to them should be considered.
zone are rich in resources.
Nyungwe Natural Forest and its buffer
It is important to identify activities of interest to local
people that could be allowed so as to utilise the resource sustainably
conservation purposes.
Due to dependence on the Forest, communities
organised at the local level to use the resources.
and for
could be
This approach could result in the
local people to utilise the resources in a sustainable way, leading to their conservation.
The adoption of such an approach would tend to answer questions
related to
community involvement in forest management and to the benefits this community
may obtain in return. In Nyungwe Forest, resource exploitation is not practised, but
there is a need to determine whether the local communities could use the resource
sustainably if legal rights for resource exploitation are given to them.
With all its boundaries surrounded by poor households, the long-term future of the
Nyungwe
Forest depends on a strong partnership
including local communities.
enhancing biodiversity
among various stakeholders,
Addressing the conflict over the Nyungwe Forest,
and improving
availability
of resources
units require a
management option which protects the Forest from the damaging use and supervises
the use of the Forest.
Joint Forest Management (JFM) or Collaborative Management is generally accepted
as a viable arrangement that benefits the forest and the major partners, namely the
government and the local communities (UFRIC, 2000). JFM emerged from India and
encompasses management schemes based on the sharing of responsibilities, benefits
and obligation from the forest between local people and usually the state (Wollenberg,
1999). The communities receive benefits from such involvement, which are clearly
spelt out at the start of the programme (Nhira and Matose, 1996).
JFM is most applicable to public/state land, or to collective property where resources
,
are managed under local management rules. For its effectiveness, there must exist a
decision-making authority, recognised and respected at the local level. In the case of
Stellenbosch University
17 http://scholar.sun.ac.za
conflict, the presence of a mediator capable of leading negotiations, is important
(Dubois, 1998). Monitoring of the community level forest management enables the
provision
of checks and balances
community-level
management.
as well as the identification
It also contributes
to the building
confidence, as has been the case for the community-managed
Pacific (Fisher, 1999).
of successful
of trust and
forests in Asia and the
Thus, this study takes the first step in assessing ways of
managing Nyungwe Forest in a sustainable manner with the participation of local
communities who lost access to forest resources.
Approaches to achieving this stem
from the understanding of the priority needs and interests of local communities in the
Nyungwe Forest.
Resources of interest to local people are scattered throughout the forest at different
intensities.
Hall and Bawa (1993) and Peters (1996) have proposed methods for
collecting information on the abundance and distribution of NWFP from tropical
forests. In some countries like Uganda, some work has been done on the density and
distribution of different plant species being harvested in Multiple Use Zones of
montane forest areas (Bitariho, 1999; ITFC-EMP, 1999; Kamatenesi, 1997; Muhwezi,
1997; Ogwal, 1998). In Nyungwe Forest, many plant species have been harvested
illegally and are still in high demand for different uses (Gapusi, 1999).
In the case of Nyungwe Forest, no work has been done on resource exploitation and
impact. This study aims at providing baseline data on forest resources of importance
to local communities.
Further, it evaluates the Forest in terms of the numbers and
sizes of trees and of what species may be harvested. The assessment of regeneration
potential of valuable plant species is also emphasised, since it provides valuable data
for resource managers (ITFC-EMP,
2001).
The general approach of the study
involves focussing on the preferred forest resources through the use of participatory
rural appraisal techniques.
The potential of the Nyungwe Forest to provide these
resources has been assessed by focused research on the density, size-class distribution
and regeneration potential of preferred tree species for timber, poles and medicinal
materials.
In line with the formulation of policies that involve the management and conservation
of natural resources, it is worthwhile to assess the possibilities of recognising local
people rights and responsibilities as well as partnerships in decision-making.
The
Stellenbosch University
18 http://scholar.sun.ac.za
government may pass certain user rights and responsibility to the local communities
as a strategy to overcome the conflict between the community
needs and the
conservation of the Nyungwe Forest. Therefore, it is important to provide guidelines
that may lead the government
and non-governmental
partners to develop
and
implement decisions over the management of Nyungwe Forest with the dual focus of
improving the local community livelihoods and enhancing proper management of
Nyungwe
community
Forest.
Legislative
gain in sustainable
and policy framework
utilisation
would take account
and conservation
of biodiversity
of
ill
Nyungwe Forest.
1.8
AIM AND OBJECTIVES
1.8.1
OF THE STUDY
Overall aim of the study
The overall aim of this study was to contribute to improved livelihoods of rural
communities by involving them in sustainable utilisation of forest reserves.
Focus
was placed on assessing the potential supply of wood and non-wood products for
adjacent community needs in Rwanda's forests. Thus, the backbone of this study was
to assess the current status of resource use and management ofNyungwe Forest and to
determine the potential for community involvement in management and sustainable
utilisation of wood and NWFP from the Forest.
1.8.2
Objectives of the study
In view of the above aim, the study of the potential for joint management and multiple
use ofNyungwe Forest has the following objectives:
1.
To identify local people's problems, needs, preferences and interests in access to
and use of resources in Nyungwe Forest.
11.
To identify the potential ofNyungwe Forest to provide the most preferred timber,
pole and medicinal species by local communities.
Stellenbosch University
19 http://scholar.sun.ac.za
111.
To inventory the forest resources in order to ascertain the abundance, the sizeclass distribution and the regeneration potential of the preferred tree species in the
periphery ofNyungwe Forest. .
IV.
To determine the tree species that can be used sustainably.
v.
To recommend
community
a management
involvement
option that enables access to resources and
in management
and utilisation
of Nyungwe
Forest
resources.
1.9
Potential benefits
The potential benefits of this study relate to the participation of local communities in
management
and conservation of Nyungwe Forest, particularly deriving increased
benefits from the sustainable utilisation and conservation of the Forest.
Within this
framework, local communities are expected to work for the protection of Nyungwe
Forest, thereby avoiding illegal use of the Forest.
The study helps to assess the
potential for sustainable utilisation ofthe Forest and provides recommendations which
decision-makers
may
use
in devolving
some
power
and
responsibility
for
collaborative management of state forests.
1.10 THESIS STRUCTURE
The thesis is structured in the following manner:
•
Chapter Two describes how the research was conducted.
It covers the methods
used both for the Rapid Rural Appraisal exercise and the N yungwe Forest
resource assessment. It is also designed to give an account of the methods used in
the analysis of the households survey and forest inventory data.
•
Chapter Three covers the findings of the study. The results of the participatory
assessment cover household surveys, group and key informant interviews.
Also,
within this chapter, forest inventory results are provided, with emphasis on
preferred tree resources identified during the surveys.
Stellenbosch University
20 http://scholar.sun.ac.za
•
In Chapter Four, the discussion of the various results is presented, based on the
findings of the study and supported by relevant literature where appropriate.
•
Chapter Five summanses
recommendations
this study's findings and provides suggestions
for the way forward.
and
Areas of future research related to
community utilisation and sustainability of Nyungwe Forest are proposed.
Stellenbosch University
21 http://scholar.sun.ac.za
2
STUDY AREA AND METHODS
2.1
STUDY AREA
2.1.1
Description of Nyungwe Forest
This section provides information about Nyungwe Forest by reporting its biophysical
characteristics and biological diversity.
An overview of the Forest is given, relating
to conservation interests and the characteristics of the adjacent population.
2.1.1.1 Size and location of Nyungwe Forest
Nyungwe Forest is located in the South West of Rwanda, near the southern shore of
Kivu Lake (Figure 2.1). Nyungwe Forest covers approximately 90 000 ha (Gapusi,
1999).
It is one of the largest blocks of lower montane forest in Africa (Weber,
1989). The majority of its area stretches within Gikongoro and Cyangugu Provinces,
while a very small proportion of the Forest is found in Kibuye Province (Sorg, 1978).
It also extends to Kibira National Park in Burundi (Budowski,
1978).
Nyungwe
Forest is found at latitude 2017'S - 2049'S and longitude 29°05' - 29°29' E (Sorg,
1978).
It occurs within the western part of the tectonic Rift along the Zaire-Nile
Divide (Weber, 1989). The altitudes within the Forest range from 1 600 m to more
than 2 700 m (Budowski, 1976). The mean annual rainfall is 1744 mm, typical of an
African rainforest (Sun et al., 1996).
The temperatures
in Nyungwe Forest are
generally cool, with an average minimum temperature of 10.90e and an average
maximum temperature of 19.60e (Sun et al., 1996). Temperatures rarely fall below
o'c in high altitude
zones of the Forest (Budowski, 1976).
22
Stellenbosch University
http://scholar.sun.ac.za
N
+
newafrics.com Il'n
ac.1e 1:44,5.
8
Legend
Information Cantra
@
Mountains
Campsite
/'v?
Main Road
/'V
Motorabie Track
Foot Path
...
.
'
.. ..
-
UND
International Boundary
/'V
River
Lake
.. Reserve
. ::~:~
~:~
:
BUR
BpS
,
Boundary/Extent
Source: Newafiica (2000).
Figure 2.1.
2.1.1.2
Nyungwe Forest, showing altitudes inside the Forest.
Geology and drainage
The area of Nyungwe Forest is mainly composed of very old Precambrian rocks
(Storz, 1983). Storz (1983) differentiates three geological parts of the Nyungwe
Forest. In the eastern part of the Forest, the substratum is derived from granites and
Stellenbosch University
23 http://scholar.sun.ac.za
granitic rock. The homogenous character of these rocks creates narrow steep-sided
valleys dominated by round hillcrests.
Sorg (1978) reported that soils derived from
such rocks are usually poor. In the western part of the Forest schist, quartzite and
dolerite are the main parent rocks: the soils derived from these rocks are fertile (Sorg,
1978). These rocks result in a topography characterised by very steep slopes and very
narrow valleys. Soils in the area of the Forest near Gisakura are basic, with contents
of Silica, Iron and Magnesium.
The whole area of Nyungwe Forest is a vital water catchment containing sources of
several perennial rivers that drain the Congo Basin and the Nile Basin (Barabwiliza,
1992). For the Congo Basin, five rivers found in Nyungwe Forest flow to Kivu Lake:
Koko, Ruhwa, Rubyiro, Kamiranzovu and Kilimbi.
The Nile Basin, on the other
hand, receives water from the Nyabarongo, Nshili, Akanyaru, Rubyiro, Rukarara,
Mbirurume and Mwogo Rivers.
2.1.1.3
Flora
Nyungwe Forest contains about 1 400 plant species, of which 400 are identified tree
species, hundreds of Orchidaceae and over 200 mosses and lichens (MINAGRI,
1997).
Its flora contains 42 endemic tree species, various medicinal plants and
valuable hardwoods (Habiyambere,
1999).
On the basis of altitude ranges, four
vegetation strata are identified and recognised by dominant tree species, as described
by Sorg (1978). Between 1 600 m and 2000 m, dominant tree species reach 35 to 40
m in height.
These include Parinari excelsa, Newtonia buchananii, Symphonia
globulifera, Entandrophragma
excelsum and Albizzia gummifera.
Orchidaceae and
ferns are abundant. This vegetation type covers 17% of the total area. At the altitude
range of 2 000 - 2 300 m, dominant canopy tree species are Entandrophragma
excelsum,
Parinari
excelsa,
Prunus
africana,
laurifolia and Chrysophyllum gorungosanum.
Ocotea
usambarensis,
Ficalhoa
The shrub stratum is less dense and
consists mainly of ferns. This vegetation represents 21 % of the Forest. Podocarpus
latifolius dominates at altitudes between 2 300 m and 2 500 m. This valuable tree
species reaches 15 to 20 m in height. At this altitude mosses and lichens also occur,
but the grass layer is poor and discontinuous.
This type of vegetation occupies the
largest area, estimated at 43%. At very high altitudes between 2500 mand 2700 m,
Stellenbosch University
24 http://scholar.sun.ac.za
the structure of the vegetation is dominated by shrub and herb growth forms. Tree
species found at these altitudes include Phillipia benguellensis, Agauria salicifolia,
Faurea saligna and Hagenia abyssinica.
Due to high humidity, mosses, lichens and
epiphytes are present. This vegetation is distributed over 19% of the total Forest area.
Kamiranzovu
wetland within Nyungwe Forest contains typical wetland vegetation
types over an area of more than 3 500 ha in the northwest part of the Forest at an
altitude of 1 860 m (Bouxin, 1974).
The other large swamp of the Forest is
Uwasenkoko.
Nyungwe Forest has relatively high species diversity.
Studies of species dynamics
found 62 plant species in an area of 12 500 m2, which is higher than the 45 plant
species per 10 000 m2 found in South American montane forests (Gapusi, 1999). At
an international scale, 13 species of endemic Orchidaceae are recognised (MINAGRI,
1998).
These include Disa eminii, D. robusta, Catyrium crassicaule,
kassneriana,
horsfalii,
Cynorkis
Disperis anthoceros, Calanthe sylvatica, Stolzia cupuligha, Eulophia
Polystachya
virginia,
P.
vulcanica,
Chamaeangis
sarchopylla,
Draphananthe burtii and Cyrtochis arcuata. Gapusi and Mugunga (1997) have listed
a number of tree species in Nyungwe Forest that are rare or in danger of extinction.
The major species are Habenaria macrantha, Disa robusta, Dispais anthoceros,
Cynorkis
kassneriana,
Calanthes
sylvatica,
Polystachia
virginia,
Chamiseangis
sarcophylla, Cyrthorchis arcuata, Aningeria altissima and Conyza volkensii.
2.1.1.4
Fauna
A broad array of fauna is present in Nyungwe Forest, made up of approximately 265
bird species, 25 to 30 large mammals, many other small mammals, reptiles and insects
(MINAGRI,
1998), including
126 species of Lepidoptera
(Habiyambere,
1999).
Mammals are dominated by primates that are unique to the area; a description of the
most common species is given by Storz (1983). Table 2.1. provides a list of Latin and
English names of some primate species found in the Forest.
25 http://scholar.sun.ac.za
Stellenbosch University
Table 2.1. Latin and scientific names of some primates found in Nyungwe Forest
(Storz, 1983).
Latin names
Pan troglodytes schweinfurthi
Papio anubis
Colobus angolensis adolfi-friederici
Cercopithecus lhoesti
Cercopithecus mittis doggetti
Cercopithecus mona denti
Cercocebus albigena johnstoni
Perodicticus potto
Galago spp.
Cercopithecus ascanius schmidti
Cercopithecus aethiops
English names
Eastern chimpanzee
Olive baboon
Angolan black-and-white colobus
L'Hoest monkey, mountain monkey
Sliver monkey
Mona monkey
Grey-checked mangabey
Potto gibbon
Bush baby
Red-tail monkey
Grivet monkey
A number of animal species is reported to be rare or endangered.
include the forest hog Hylochoerus meinertzhageni,
The most rare ones
the blue duiker Cephalophus
monticola, the bushbuck Tragelophus scriptus, the small-scaled tree pangolin Manis
tricuspis, the giant pangolin M gigontea, the spotted hyena Crouta crocuta, and the
leopard
Panthera
pardus.
MINAGRI
(1998)
reported
characterising afromontane forests that are protected by CITES.
36
animal
species
Examples of these
animals include the lturi Forest chameleon Chamaeleo adolfi frederici,
the flap-
necked chameleon C. dilepis idjwensis, the Ruwenzori three-homed chameleon C.
johnstonii, Rhampholeon boulengeri and the mountain monkey Cercopithecus lhoesti.
2.1.1.5
Management
history and conservation interests
Nyungwe Forest was gazetted as a Forest Reserve under Legal Notice No. 83 bis
Agris of December
iz",
1933 (PCFN, 1989).
prevent the total disappearance of the Forest.
The Colonial authority aimed to
In 1934, the status was modified
slightly by the Legal Notice No. 33/Agri of May 24th, 1934 to include a special
exploitation regime of the Forest (Mbonyintwari, 1989).
Within the framework of management and conservation of natural forests on the
Congo Nile Crest, the 1984 Action Plan aimed at conserving Nyungwe Natural Forest
qualitatively
and quantitatively
(Renner,
1991).
responsible for its management and conservation.
The Forest Department
was
To facilitate this, four management
zones of Nyungwe Forest were delineated as stated in the 1984 Action Plan.
In
Stellenbosch University
26 http://scholar.sun.ac.za
partnership with the Swiss Government, the World Bank, and the Central Fund for
Economic
Co-operation,
management
the Rwandan
of the Forest.
Various
Government
developmental
promoted
activities
conservation
and
integrated
rural
development with conservation.
'The Projet Pilote Forestier (PPF)' was the first project that operated in Nyungwe
Forest and the buffer zone since 1970 (MINAGRI, 1984). This project established
exotic plantations with the objective of marking the boundaries of Nyungwe Forest
and keeping
it from encroachment
by neighbouring
communities
(Grison
and
Braesco, 1984). The project initiated a series of other activities consisting of rural
afforestation so as to secure fuelwood and pole supplies to local communities
adjacent Communes.
in
At Rangiro (Cyangugu) and Gisovu (Kibuye), a large number
of pit sawyers who encroached on the Forest were used as project workers.
In 1982, the 'Projet Crête Zaire Nil' (CZN) was opened in the North-East
Nyungwe.
This project integrated forestry, agriculture and social activities.
of
It also
created a buffer zone and managed pastures (MINAGRI, 1984; PCFN, 1989). Both
CZN and PPF did not cover the whole buffer strip between Nyungwe Forest and the
local people.
The southeast and southwest borders of Nyungwe Forest lacked
adequate protection.
Thus forest degradation continued, particularly due to illegal
logging and gold mining (PCFN, 1989).
Since 1982, based on geographical zones, Nyungwe Forest has been managed by four
projects named "Unité de Gestion de la Zone" and referred to as UGZ 1 to UGZ 4
respectively (UGZ1, 1987; Weber, 1989), under the financial assistance of many
bilateral and multilateral organisations.
In the North-West, Nyungwe was managed
by UGZI under financial support of the Swiss Co-operation since 1986. Two projects
operated in the northeast: UGZ2, funded by European Development Fund (EDF) since
1982, and Mudasomwa Project, assisted by Belgium. The latter project concentrated
its action in Mudasomwa Commune and conducted afforestation activities in rural
areas.
The southeast of Nyungwe Forest has been managed by UGZ3 since 1989,
using funds provided by the Rwandan Government and the World Bank. Since 1987,
the southwest of the Forest was under the responsibility of UGZ4, which was funded
by the French Government (MINAGRI, 1991). The various UGZ realised different
Stellenbosch University
27 http://scholar.sun.ac.za
activities, including the establishment of 8 300 ha of forest plantations.
In 1991, the
various projects established about 10 000 ha of forest plantations of predominantly
exotic tree species (MINAGRI, 1991).
Other important roles played by the UGZ
include (MINAGRI, 1991):
•
The construction and maintenance of infrastructures: offices, staff accommodation
houses, roads and trucks, forest roads, schools, water adduction, dispensaries, etc.
•
Timber harvesting from thinning and marketing of timber and timber products;
•
Inventory
and research
in Nyungwe
Forest in collaboration
with research
institutions;
•
Protection of the Forest from encroachment through regular patrols inside and
outside the Forest;
•
Establishment
of forest management plans to guide interventions in the buffer
zones;
•
Public awareness and education programmes on the protection and conservation
ofNyungwe Forest.
Due to its high biological diversity, PCFN in conjunction with ORTPN aimed at
preserving Nyungwe Forest with increased focus on ecotourism, research and public
awareness of conservation (ORTPN, 1991). PCFN was funded by USAID, African
Wildlife Fund, Digit Fund and Wildlife Conservation International (WCI).
Funds
were used to improve technical and management capabilities, to increase the level of
on-going biological, physical and socio-economic studies, to raise public awareness
on the importance of biodiversity conservation and to provide various forms of
administrative support to the project (PCFN, 1989). Activities of the various projects
operating in Nyungwe Forest were planned to be co-ordinated by a Conservation
Centre that had not become operational before 1994. Offices and staff houses were
established at Kitabi (Mudasomwa Commune) but they were damaged during the
genocide in 1994 (Gapusi, 1999).
Stellenbosch University
28 http://scholar.sun.ac.za
2.1.1.6 Adjacent communities
On an administrative basis, the population of the Republic of Rwanda is organised
into Prefectures and the latter into Communes. Depending on its size, each Prefecture
comprises a number of Communes, which in turn are subdivided into administrative
Sectors. A Sector is further broken up into administrative Cells, each being made up
of at least 50 households. At each level, the leaders are elected locally.
The population living within the boundaries of Nyungwe Forest was organised and
unevenly distributed over 11 Communes (today, two or more Communes have been
fused together to form a District). This population was estimated at 500 000 people in
1997 (MINAGRI, 1997). MINAGRI (1984) distinguished two types of communities
living around the Nyungwe Forest.
The eastern ridge comprised less populated
Communes within the agricultural region of Congo Nile Crest, characterised by poor
soils on steep topography, relatively higher farm size than at national level, low
population growth, relatively low quality of life and food insecurity.
The other
characteristic of this region was the high decrease in farm size that made agricultural
intensification by use of agricultural inputs difficult.
Land in the border strip of
Nyungwe Forest was of particular interest due to its high organic matter content.
MINAGRl (1997) defined the western ridge of Nyungwe Forest as falling within the
Communes of the "Impala and Kivu Lake Borders" agricultural
farming conditions were favourable.
region, where
The average farm size was very small (about
0.39 ha per household) but sufficiently productive.
A large area of Nyungwe forest
was cleared through expansion of arable land in this locality.
Besides agriculture, the population as a whole derives a livelihood from activities in
the Forest and in the buffer zone. In the buffer zone, livestock farming is not well
developed and is dominated by sheep, goat, and pig rather than cattle (MINAGRI,
1984).
Within the forest of Nyungwe, activities are illegal and consist of gold
mining, logging, hunting, bee-keeping, agriculture and collection of medicinal herbs
(MINAGRl, 1984; PCFN, 1989, Gapusi, 1999).
29 http://scholar.sun.ac.za
Stellenbosch University
2.2
RESEARCH
METHODS
This chapter describes the methods employed in the collection and analysis of
quantitative and qualitative data in order to achieve the study objectives outlined in
section 1.8.2: The methods used for the study have two components:
•
Participatory
rural appraisal (PRA) techniques
were used to determine
the
problems, needs, priorities and interests of local communities in Nyungwe Forest.
Interviews were conducted in the surrounding areas of the Forest.
Interviews
within individual households were in the form of open-ended questions.
Further,
data were obtained through in-depth discussion on general issues with small
groups and key informants.
•
Forest resource assessment was carried out as the focal point of the study in order
to find out how well stocked Nyungwe Forest is to guarantee a sustainable supply
of resources that are mostly preferred by local communities.
In addition, it aimed
at collecting information on which to base recommendations
for resource uses.
Because of time and financial constraints, it was not possible to cover more than a
small part of forest area adjacent to communities, and so the survey focussed only
on the highly preferred plant species identified during the PRA exercise.
The
survey was carried out to achieve the following specific objectives: -
1.
To estimate the abundance (presence or absence) of the forest resources that
are believed to be important to the livelihoods of local communities;
11.
To determine the abundance
of each species present and the size-class
distribution of individuals at different diameter classes;
111.
To assess
the regeneration
potential
of useful
plant
species
to local
communities;
IV.
To recommend the resources for which the communities could have access.
30 http://scholar.sun.ac.za
Stellenbosch University
For the purpose of this study, basic information on species types, density, frequencies,
diameter-class
distribution,
species dominance
and regeneration
were collected.
Some of these parameters determine sustainable levels of resource use as described by
Peters (1994).
2.2.1
Selection of the study sites
The study area was stratified into two sections, based on agro-ecological
zones
defined by Niang and Styger (1990) using biophysical characteristics and crop types.
The Bufundu and Bushiru highlands in the eastern and northern part of the forest
comprise eight ex-Communes of similar population densities in the range of 200 to
250 persons per krrr'. The second area, the Impala Ferralsols Region, stretches over
three ex-Communes (Karengera, Kagano, Busozo) in the western ridge of the Forest
with a population density of250 to 350 inhabitants per krrr'.
A further subdivision of the region was based on past project management zones.
Within each zone, one Commune was selected to represent the project area.
Other
bases for selection of sample sites included the interface of the Communes with the
Reserve, the degree of deforestation at the forest edge and altitude ranges within the
Forest.
The latter criteria were used because of their potential impact on the
availability and diversity of resources.
Based on geographical contiguity with the
Forest, a total of five Communes were selected for this study. The five Communes
were Kagano and Busozo at the western ridge of the Forest, Nshili in the south-east,
Mudasomwa and Musebeya in the eastern side of the Forest (Figure 2.2).
2.2.2
2.2.2.1
Participatory Rural appraisal
Selection of households
The households of each Commune are not equally close to the Forest. Administrative
Sectors extending into Nyungwe Forest were identified and further stratification
aimed at selecting a variety of households at "Cell" level, using the criteria of the
closest location to the Forest. At most, two Sectors next to the Forest were considered
for each Commune and the respondents randomly selected in two Cells.
31
Stellenbosch University http://scholar.sun.ac.za
Source:
Figure 2.2. Communes neighbouring Nyungwe Forest, showing the survey
sample distribution.
Within each Cell, the number of households interviewed was 10, representing a fifth
of 50 households per Cell. Therefore, a total number of 200 households spread over
five Communes in different Sectors and Cells in the vicinity of the Forest were
interviewed.
A sample of this size is much larger than the sample size of 70
households used by Wambugu (1999) in Embu district of Kenya and recommended
by Palm et al. (1993) as appropriate for making inferences about the larger
population. Due to heterogeneity and the large adjacent population, the sample size
of200 households was thought to represent the conditions around Nyungwe Forest.
Stellenbosch University
32 http://scholar.sun.ac.za
2.2.2.2
Selection of groups and key informants
In the
communities
neighbouring
Nyungwe
homogenous groups using the forest resources.
Forest,
there
are
no
identified
Groups of five to 10 people living
near the Nyungwe Forest were randomly identified with the aid of the head of each
Cell or members of the local administrative committee at Cell levels, since they are
living with and leading the people living close to the Forest. Individuals known to be
the heaviest users of the Forest and forest products were selected for group discussion.
Various key informants were approached to identify the current challenges and
opportunities facing managers of the Forest.
These respondents were important
stakeholders in the future conservation and management of Nyungwe Forest.
They
included the administrative authority of each Commune being surveyed and staff of a
number of relevant bodies, including PCFN, ORTPN, the Department of Environment
in the Ministry of Land and Environment, various NGOs and Co-operatives,
the
Regional Agricultural Services, tea factories and private enterprises.
2.2.2.3
Data collection
Household and group interviews provided information on resource availability within
Nyungwe Forest as well as species preferences and their uses, land tenure and landuse, socio-economics and general information on agriculture and livestock keeping,
and perceptions of institutions that might shoulder the responsibility for sustainable
utilisation and conservation of Nyungwe Forest resources (Appendix 1). Interviews
with groups and key informants focussed on resource preferences and ranking.
The
discussions also focused on utilisation and conservation strategies of the Forest with
active participation and increased benefits to local communities.
Information that relates to resource abundance and distribution were explored through
subjective assessment based on the interviewee's knowledge of a particular resource.
The abundance of a given resource was measured on a scale of 1 to 5, corresponding
to the DAFOR (Dominant = 5, Abundant = 4, Frequent = 3, Occasional = 2 and Rare
=
1) scale (Smith et al. (1985). From a list of available resources in the Forest, three
important resources to communities were prioritised and appraisals were made to
determine the uses of the resources.
In order to identify habitats of the Forest that
Stellenbosch University
33 http://scholar.sun.ac.za
might reeerve intensive use by the local communities,
information
on resource
abundance and distribution were sought in relation to landscape, i.e. whether the
resource occurs on hilltops, on slopes, in the bottomlands
or in the swamps.
Nyungwe Forest resources were categorised based on household use of the resources.
Subjective quantity estimates of marketed products were investigated, focussing on
the season of resource harvest and market location. Harvesting of some parts of plant
resources such as roots and bark has an impact on the availability of the forest
resources and their response to disturbance (Peters, 1996; Wild and Mutebi, 1996).
Consequently, respondents were asked to identify parts of plants used for different
purposes.
For wildlife resources, emphasis was put on the game used to provide
various game products to local communities.
Methods and time of hunting were also
emphasised as they were thought to influence the wildlife population dynamics.
Land tenure, land-use, and socio-economic status of local communities were obtained
through semi-structured interviews with groups and individuals and were collected by
means of an open-ended questionnaire and discussions.
Data on socio-economic
status concerned the personal economic and financial situations in terms of the level
of education, the enterprises undertaken by the household and the sources of income.
Information on land tenure and land use aimed to understand the process of land
acquisition and the management of soil fertility. These were thought to influence the
dependence of people on forest resources.
The questionnaire
also included questions on institutional
arrangements
that are
believed by households to enable the conservation of the resources for the benefit of
rural communities.
Questions focused on mechanisms at local level that enabled
access to resources by local communities.
Finding strategies to coordinate local needs
and the conservation of Nyungwe Forest and linking the management of the Forest to
community involvement, sustainable utilisation, and governance was emphasised.
2.2.2.4 Data processing and analysis
Prior to analysis, the interview data were coded and organised
into categories
following the procedure described by Babbie and Halley (1998), and Neuman (2000).
Numerical codes were assigned to semi-structured
responses on the background
Stellenbosch University
34 http://scholar.sun.ac.za
information of the respondents. Responses were all recorded on paper and, depending
on the frequency of the answers, broad categories were created.
The survey data were entered and organised into Statistical Package for Social
Sciences (SPSS) (Babbie and Halley, 1998) data format. The reeode procedure was
used to transform
all pre-existing
variables
into different
variables
either by
discarding the "don't know" and "no answer" responses or by dividing a pre-existing
variable into different categories and coding each category differently.
The latter
procedure has been applied to questions that related to family size, level of education,
farm size, quantities of crops at harvest, and others.
In some instances, categorical
data were transformed into continuous values without contorting the respondents'
responses.
As the interview was flexible and allowed detailed and anticipated
information, some responses from some local people could not be coded and therefore
the information was used to support the discussion of the results. For the survey data,
limited quantitative analysis was performed, focusing mainly on frequencies and
associated histograms so as to provide visual illustrations and to support descriptive
results.
2.2.3 Forest resource assessment
2.2.3.1 Sampling design and plot sizes
The size of plots depends on the characteristics of the community being sampled.
Peters (1996) recommended plot sizes of 16 to 100 m2 for shrubs and understorey
vegetation.
For the study of small trees and shrubs, Stanley and de Olieveira Castro
(1959) recommended the use of small plots of 4 x 4m or 2 x 2m, always located in the
same corners of the larger plots. However, small plots produce density estimates with
a large error term. In contrast, Kruger et al. (1997) recommended using plot sizes that
are scaled to the canopy height of a particular forest. In their study in South African
forests, they used 20 m by 20 m plots, randomly located in a forest with a canopy
height of 20 m.
For this study, plots were established with the aid of a prismatic compass, tape
measure, machete, and ranging poles. They were placed 1 km inside Nyungwe Forest
35
Stellenbosch University http://scholar.sun.ac.za
from each side of the forest edge community being surveyed. In each part of the
forest area, two plots each measuring 2500 m2 (i.e. 50 m x 50 m) in size (Sample unit
A) with the corresponding nested areas (Sample units Band
C) were established
(Figure 2.3). The survey procedure was based on the method used by Fedlmeier
(1998) for the study of the development of secondary forest in Costa Rica. Over
sample unit A, trees greater or equal to 10 cm diameter at breast height (OBH) were
recorded.
Sample unit A: 50 x 50 m
Sample unit B: 707 m2
ample unit C: 12 x 4 m2
Figure 2.3. Subdivision and size of the sample units used for forest resource
assessment, Nyungwe Forest.
Sample unit B was measured off from sample unit A for appropriate assessment of
abundance of small trees greater than 1.30 m height and DBH of 2.5-9.9 cm. A
circular shape of 15 m radius (i.e. 707 m2) was selected. Seedlings and saplings were
recorded in twelve 2 m x 2 m nested sub-units (Sample unit C) within sample unit B.
The sub-units were delineated following the diagonals of Sample unit A and placed at
3 m intervals. Individuals ofO.10 m to 0.30 m height and DBH ofless or equal to 2.5
cm were recorded. Peters (1996) scaled small trees of this size to tree seedlings or
saplings. Therefore, regeneration assessments were done on an area of 48 m2, lying in
the range of plot sizes recommended by Peters (1996). For the study, a total number
of 10 sample units with 10 corresponding nested subunits and 60 sub-subunits were
used.
36 http://scholar.sun.ac.za
Stellenbosch University
2.2.3.2
Data Collection
Quantitative and qualitative information was recorded on tree species found in the
sample plots. The aim was to gather information on the availability and abundance of
the preferred tree species in relation to other tree species present.
Preferred tree
species were defined as those that were identified during the participatory
rural
appraisal as mostly being sought after by local communities for use as timber, poles
and medicinal
materials.
They were also referred to those species that were
frequently used in the past by the local people before the Forest was gazetted a forest
reserve. The following parameters were recorded during the forest assessment: tree
species composition (local names), stem densities, diameter at breast height (DBH),
and stem quality of the most preferred timber and pole species.
Plant identification and species composition
Within each sample unit, tree species were identified in the field by a team of five
local persons including two forest rangers who provided vernacular names.
Local
people were included in the survey team because they were knowledgeable
of the
local names of the tree species. Scientific names were identified by making reference
to the work on Rwanda Flora by Troupin (1985, 1983, 1978), Troupin and Bridson
(1982) and Troupin and Donneaux (1982).
Individual plants of every species were
also recorded.
Density, frequency and diameter size-class distribution
Density, or the count of individuals of species within the plot (Davis and Johnson,
1987; Kent and Coker, 1992), is the ecological parameter of interest to ethnobotanists
(Peters, 1996).
Hall and Bawa (1993) and Peters (1994) reported that density and
size-class distribution are indicators of the impact of exploiting a given population.
Density describes not only the degree to which the site is being utilised, but also the
intensity of competition between trees.
In this study, density measures that combined tree sizes and the number of individuals
were used. Individual trees were counted in each of the sample plots and recorded in
37 http://scholar.sun.ac.za
Stellenbosch University
the appropriate diameter class. Frequency as a measure of species abundance (Kent
and Coker, 1992) was used to provide the proportion of individuals of the most
preferred tree species in relation the number of individuals of all other tree species
present in the sample plots.
Tree species abundance was quantified by tallying the
number of stems of each species in different DBH size classes.
Large canopy tree
species or tree species greater than 10cm DBH were grouped into 10 cm diameter
class.
Species dominance
Species dominance was used as a measure of the contribution of each individual
species to the forest standing biomass (Owiunji, 1997). The assessment of standing
biomass was measured by the basal area, measured at breast height, of individuals of a
particular species. Species dominance was used as an index to ecological dominance
and an indirect measure of biomass (Cunningham, 2000; Peters, 1994).
Assessment of the regeneration of the most preferred tree species
Regeneration
assessment was used for quantifying the density of seedlings and
saplings in the forest area adjacent to communities surveyed around Nyungwe Forest.
The aim was to assess whether or not the canopy tree species that were most preferred
were capable of producing seedlings and saplings in the Forest.
Seedlings and
saplings were recorded for their numbers.
Quality of the most preferred timber and pole species
A quick visual assessment of the straightness of the tree bole in each plot was made.
Stem quality was evaluated on a six-point scale in terms of their suitability for
utilisation as timber or poles. The quality of the stem was matched with the quality of
the stem provided on Figure 2.4 below. Therefore, points 1 to 6 were used to mean
stems of excellent, very good, good, fair, poor, and very poor quality stems.
38
Stellenbosch University http://scholar.sun.ac.za
1 = excellent; 2 = very good; 3 = good; 4 = fair; 5 = poor; 6 = very poor.
Figure 2.4. Different levels ofstem quality used in the resource assessment,
Nyungwe Forest.
2.2.3.3 Data analysis
Tree species identification and composition
A
list showing the scientific and vernacular names of the tree species was made by
compiling all tree species found in all forest areas next to the Communes being
surveyed. Tree species composition was assessed for large trees (trees greater than 10
cm DBH), for medium trees (trees of 2.5 - 9.9 cm OBH) and for the seedlings and
saplings (individuals ofO.lO - 1.30 m height and < 2.5 cm DBH) at the different study
sites.
Density and diameter size-class distribution
Analysis of size-class distribution was limited only to the tree species most preferred
by local communities. The selection of these species was based on the results of the
household survey and on the relative frequency in the species composition data
obtained from forest inventory. Diameters of the most preferred tree species were
compiled into size-class histograms showing the distribution of the stem densities into
different OBH size classes. Species distribution curves were generated to reflect the
39 http://scholar.sun.ac.za
Stellenbosch University
tree species demographic pattern by plotting DBH size classes versus the stem
densities per hectare (ITFC-EMP,
1999).
The resulting histograms reflected the
current status of the Forest in relation to the most dominant canopy tree species.
A sample provides an estimate of the population parameter of interest, as it is
impossible to measure the entire population (Mead et al., 1993). Confidence limits
define a range of values around the sample mean, which is expected to include the
true population mean with a certain degree of confidence (Snedecor and Cochran,
1967).
In this study, stem densities were calculated as a measure of abundance of the most
preferred tree species for timber, poles and medicinal materials.
The calculated
confidence interval of 95% for the mean stem densities and the lower limit confidence
interval were used as conservative estimates of the density of the resource.
In
contrast, stem densities at the upper limit of the confidence interval were considered
abundant
enough for utilisation.
These confidence intervals were then used to
determine the level at which the preferred tree species were abundant enough for
utilisation.
In their research on stand density management
in the United States, Drew and
Flewelling (1979) quoted by Davis and Johnson (1987) found that the stem density of
0.55 stems per hectare is the level at which total growth per hectare level out and
competition-induced
mortality of trees begins. As the density of 1.0 is an empirically
fitted maximum in nature (Davis and Johnson, 1987), preferred tree species with a
stem density of 1.0 stems per hectare and above were considered to have a high
enough stem density to recommend them for utilisation.
The guiding concept for the
harvesting of timber, pole and medicinal materials was to reduce the density to a level
such that the density of the preferred tree species will grow back to or slightly above
the 0.55 stems per hectare at the next scheduled harvesting ofthese products.
40 http://scholar.sun.ac.za
Stellenbosch University
Species dominance
Species dominance was measured by the basal area (G), i.e. cross-sectional area of an
individual of a particular species measured at breast height (Philip, 1983). The
following formula was used for the calculation of the basal area:
G
= 7t
x ((DBH)2/4), where:
G= basal area (m/ha);
7t =
3.14;
DBH = diameter at breast height (cm).
The contribution of each individual species to the forest biomass was measured by the
relative dominance (D), which is defined as the cross-sectional area for each species
divided by the basal area for all species.
D = GsfGtob where;
D = relative dominance or relative basal area (m2/ha);
Gs = basal area for a given species (m2/ha);
Gtot = basal area for all species (m2/ha);
The calculated proportion was used to compare the contribution of each species to the
total standing biomass.
With a total index of 1.0 for all the species in the sample
forest area, a tree species with a relative basal area of 0.01 and more was subjectively
said to have a 'high contribution' to the forest standing biomass and that with less
than 0.01 to have a 'low contribution'.
Species of the latter category were
immediately not classified with harvestable tree species.
Regeneration characteristics
The most abundant and dominant preferred tree species were compared in terms of
corresponding
available seedlings and saplings.
Seedling and sapling densities of
these species were compared by means of histograms showing the relative abundance
of trees across the study sites.
abundance
of seedlings/saplings
The relationship between tree species and relative
was considered
to be a measure
of species
Stellenbosch University
41 http://scholar.sun.ac.za
regeneration and thus was used to predict the ability of the preferred tree species to
produce significant resources.
Quality of the boles of the most preferred timber and pole species
Data on the quality of the stems were presented as percentages.
The six levels of the
quality of the stems were aggregated into three categories namely good, fair and
acceptable.
For the most abundant and dominant preferred tree species, histograms
were constructed to provide a visual display of the percentage of individuals with
good, fair and acceptable stem quality.
Stellenbosch University
42 http://scholar.sun.ac.za
3
RESULTS
3.1
P ARTICIP ATORY RURAL APPRAISAL
This section presents the information collected during interviews with individual
households, small groups and key informants. The results of the participatory rural
appraisal
(PRA)
assessments
include
the
of the availability
socio-economic
of resources
status
of the
in Nyungwe
respondents,
the
Forest, their relative
abundance estimates and the prioritisation
of important resources to community
livelihoods.
a list of key institutions
The section
also presents
that were
acknowledged by the interviewees to be appropriate for better management of the
utilisation of Nyungwe Forest. Also presented here are the experiences of individual
key informants relating to Nyungwe Forest management problems and the strategies
adopted to reduce the conflicts between local people and forest managers.
In many
cases, the total percentages of responses exceeds 100% because the respondents
mentioned more than one item for examples the types of crops farmed, the resources
available in Nyungwe Forest, the uses of resources, the types of conflicts and the
strategies to avoid them, among others.
3.1.1
Socio-demographic characteristics
In the study area, interviews were conducted in 200 individual households of which
161 (80.5%) were represented by men and 39 (19.5%) by women. Thus, most of the
information
was collected from men who were the heads of the households.
Information on marital status indicated that 85.5% were married, 7.5% widows, 6.0%
single, and 1.0% divorced.
respondents.
Figure 3.1 illustrates the age distribution
of the
43
Stellenbosch University
http://scholar.sun.ac.za
Figure 3.1. Age classes of the respondents in the study area, n = 200
The majority (43.5%) of the households contained 5 - 7 members (Figure 3.2). This
coincides with the national statistics, which indicate an average family size of five
persons (MINECOFIN, 1998).
Large households (JO or more members) were
reported by 36% of the respondents. One of the reasons for large family size appears
to be the hosting of the children orphaned as a consequence of the 1994 genocide.
2 - 4 merTtlers
5 - 7 merri>ers
Figure 3.2. Family size of the respondents in the study area, n = 200
Table 3.1 shows the level of education of the respondents and indicates that the
majority of the respondents had little or no primary school education.
This was
particularly the case in the Kagano and Mudasomwa Communes, where people had
44
Stellenbosch University http://scholar.sun.ac.za
abandoned school education m order to engage in tea harvesting for income
generation.
Table 3.1. Education level of the respondents in the study area, n = 200
Educationlevel
No school
Some primaryschool
Primary schoolcompleted
Some secondaryschool
Secondaryschoolcornpleted
Total
%
34.5
39.5
19.0
6.0
1.0
100.0
Other reasons for school abandonment, such as the lack of school fees, were
expressed by 6% of the respondents. A very low percentage (1%) of the respondents
had completed their secondary education, and none of the respondents had reached
university level. This implies that people educated to secondary and university levels
were living outside their native area, probably due to better employment and income
generating opportunities in other parts of the country. As a result, less educated
people were encountered in all the households interviewed.
3.1.2
Farm size and land ownership
Land is a major form of capital for the production of crops, livestock and tree
resources. Figure 3.3 shows the farm sizes of sample households in the study area.
>2ha
11%
34%
1-2 ha
33%
0.5 - 1 ha
22%
Figure 3.3. Farm sizes ofthe respondents in the study area, n = 200
Stellenbosch University
45 http://scholar.sun.ac.za
Many respondents (35%) owned less than 0.5 ha of farmland.
The majority (41%)
obtained their land through state allocation (Figure 3.4).
50
40
<Il
C
30
...
20
0
""0
C
0
C<Il
0
.......
0
'cJ.
10
o
Allocation by
state
Inheritance
Purchase
Purchase and
inheritence
Rent
Illegal
occupation
Figure 3.4. Processes of land acquisition by the respondents in the study area,
n = 200.
Another 36% obtained their land through inheritance.
Some respondents (15%) had
purchased farms from neighbours, while a few respondents (6.0%) indicated that their
land was both purchased and inherited. Where land was acquired through donation by
the state, as in Musebeya Commune, the average farm size was 2 ha. Areas where
land was largely inherited, for instance in Nshili Commune, the farm size was less
than 1.0 ha. Previous research by MINAGRI (2000) found that the average farm size
in Gikongoro
Province, which includes three of the five Communes
surveyed
(Mudasomwa, Musebeya, Nshili), was 1.26 ha, comparatively higher than the 1.0 ha
owned by 79% of all households in Rwanda.
3.1.3
Livelihood activities and sources of income
Not a single respondent indicated only one livelihood activity. Most people combined
food production
and livestock farming with other activities such as beekeeping,
charcoal making, pitsawying, handcraft, employment, commerce and mining.
As a
Stellenbosch University
46 http://scholar.sun.ac.za
result, agriculture was a source of livelihoods to 99.5% of the respondents, and
livestock keeping to 74.0%.
Tree-related livelihood activities included pitsawying
(9%) and charcoal making (4.5%) (Figure 3.5). Only a few respondents ranked the
practices of beekeeping (one respondent) and mining (three respondents) in Nyungwe
Forest as important activities for people's livelihoods.
Beekeeping
Mining
Commerce
Charcoal making
Paid employment
Handcraft
Pitsawying
Labour •••
Livestock keeping ••••••••••••••
_
Food production
o
20
40
80
60
100
% of responses
Figure 3.5. Priority enterprises of the respondents in the study area, n = 200.
Agriculture was a source of income to 75% of the respondents, while 40% of the
respondents
obtained cash from livestock keeping.
Income earnings from tree
products were reported by 10% of the households. About 30% of the respondents had
off-farm income, and only 6% had paid employment.
respondents
A small percentage
(2.5%) were engaged as casual labourers on neighbouring
of
farms,
particularly in tea harvesting where tea factories were operational.
Other sources of cash income included the sale of craft works, the sale of fruit and
gold mining.
Respondents also indicated that credit systems in the study area were
not developed and, as a result, financial resources for many households (78%) were
self-generated, with little or no access to bank credits. A few respondents (8%) had
Stellenbosch University
47 http://scholar.sun.ac.za
established a mechanism of a 'rotating fund' among group members.
This system
enabled each group member to be awarded a lump sum of money from the monthly
contributions of all other members.
The respondents highlighted the importance of
this system in enabling members to cover high costs, including the expenditure for
house building, the purchase of livestock and the payment of school fees. Despite the
advantages of such schemes however, most (69%) of the respondents
registered with any group or association.
were not
Other respondents (20%) were members of
farmers' associations.
3.1.3.1
Cropfarming
All the respondents managed their farms predominantly for the production of food
crops.
They grew traditional food crops, including sweet potatoes, Irish potatoes,
bananas, maize, wheat, peas, and cassava, with sweet potatoes and maize being the
most prominent in the study area (Table 3.2). The principal objective of crop farming
was to generate enough food for domestic consumption and to meet other basic needs
through income accrued from the sales of excess production on local markets.
grown for food and cash income were predominantly
Crops
Irish potatoes, bananas and
wheat. A few respondents grew other cash crops such as tea (7.5%, n
=
15) and trees
(4.5%, n = 9).
Table 3.2. Important food and cash crops grown by the respondents in the study
area.
Crops
Irish potatoes
Maize
Wheat
Peas
Beans
Cassava
Sweet potatoes
Banana
Vegetables
Percentage
Food crops
17.0
34.7
20.5
23.5
37.0
19.0
45.5
6.0
11.5
of respondents
Food and cash crops
51.0
19.5
22.0
14.0
7.0
13.5
16.0
22.5
6.0
Generally, crop harvests were small, with the majority of the households harvesting
less than five bags (one bag is approx. 100 kg) even in areas where the crops were
most suited to soils and climate (Table 3.3). Irish potatoes, maize and wheat were
48 http://scholar.sun.ac.za
Stellenbosch University
common in Musebeya, Mudasomwa and Nshili. Roots and tubers (cassava and sweet
potatoes) were preferred crops in Busozo, and banana production was specific to the
agro-climatic conditions of Kagano Commune in the North-West of the study area.
Factors that were said to influence the quantity at harvest included the non-use of
improved inputs and the destruction of crops by wild animals.
Table 3.3. Estimates of crop harvests by the respondents in the study area.
Results are presented for important crops mentioned by more than 20% of the households.
No. bags
(1 bag =
100 kg)
<1
1-2
3-5
6-8
9-10
> 10
No
estimate
Total
3.1.3.2
Irish
potato
3.8
25.0
31.1
4.5
18.9
16.7
100
Maize
Percentage of respondents
Cassava
Wheat
Bean
Peas
26.2
43.0
14.0
2.8
11.2
1.9
0.9
33.3
37.0
23.5
1.2
2.5
62.5
29.5
6.8
65.8
25.0
7.9
-
-
-
-
1.1
1.3
100
100
2.5
100
100
11.6
40.6
24.6
5.8
5.8
1.4
10.1
Sweet
potato
3.9
16.4
18.8
11.7
3.1
4.7
41.4
100
100
Banana
10.7
89.3
100
Other plant species grown
Apart from crops, trees, shrubs and grasses were also farmed by many of the
respondents.
The presence of fruit, medicinal and fodder plant species is widespread
in the survey area. Exotic fruit plants on farms included avocado (Persea americana),
Japanese plum (Cyphomandra betacea), passion fruit (Pasiflora edulis), citrus (Citrus
lemon), guayava (Psidium guayava), mangoes (Mangifera
(Anona reticulata) and pawpaw (Carica papaya).
48.5% had planted avocado.
indica), custard apple
Of the households interviewed,
Only 21.5% and 19.5% of the respondents had plum
trees and passion fruit on their farms, respectively.
The remaining fruit plant species
were planted by fewer than five households.
The respondents mentioned the presence of 20 medicinal plants traditionally used to
treat humans and livestock on their farms (Appendix 2). These plant species were
indigenous, and they were planted or conserved on farms. The most common were
Iboza riparia (33.0%), Plecthranthus
(13%).
barbatus (20.5%) and Vernonia amygdalina
49 http://scholar.sun.ac.za
Stellenbosch University
Different fodder species were included as components of the farming system with the
objective of controlling erosion. At the same time, they were used to feed domestic
livestock during periods of fodder shortage.
Many respondents (59%) established
Setaria spp. on erosion control ridges or on terraces. Other fodder species mentioned
were Panicum hochstetteri, Cynodon aethipicus, Tripsacum laxum, Digitaria hackelii
and Triumfetta cordifolia. The elephant grass Pennisetum purpureum was present on
18.5% of farms and its leafy biomass and young plants were cut to feed livestock
during the dry season of July to August.
Horticultural crops grown in the study area were largely cabbages and amaranths.
the 200 households
interviewed,
39.5% produced cabbages, particularly
Of
in the
swamps. Amaranths were present in the home-gardens of 24.5% of the respondents.
The prevalence of these vegetables was attributed to an existing local market as well
as to the need for food supplements within the households.
Other vegetables and
fruits of limited interest included carrots, pineapples, onions and courgettes.
3.1.3.3 Livestock farming
Within the study area, livestock farming was complementary
to food production.
Different livestock types were present, but they were unevenly distributed among the
respondents (Table 3.4).
Table 3.4. Numbers and distribution of livestock types in the study area.
No. of livestock
None
1
2-4
5-7
8 - 10
> 10
Total
Cattle
63.5
9.5
21.0
4.0
2.0
100.0
Goats
44.5
5.0
24.0
18.5
4.0
4.0
100.0
% of res20ndents
Pigs
Shee~
59.5
78.0
6.5
12.5
18.0
7.5
11.0
2.0
5.0
Chicken
89.0
1.0
4.0
3.5
1.5
100.0
100.0
100.0
Rabbits
93.5
2.0
1.5
1.5
1.5
100.0
Goats (56.5%) and sheep (40.5%) were the most commonly kept livestock among the
respondents, while only 36.5% kept cattle. Pigs were important in the study area and
were kept by 22% of the people interviewed.
Chickens and rabbits were also kept.
Some households kept all the types of livestock, while others had none. The purpose
of livestock farming was predominantly for manure production, except in the case of
Stellenbosch University
50 http://scholar.sun.ac.za
rabbits and chickens, which provided about 30% of the respondents with protein.
Income was also an important aspect of livestock keeping, especially in the case of
sheep and goats. Many households obtained an income from sales of goats (70.8%),
sheep (56.3%) and pigs (44.4%). About 22% of the households obtained cash from
cattle through the sale of unproductive cows, calves or milk.
Of the households interviewed, 60 to 80% used manure from cattle, goats, sheep and
pigs on their farms owing to the low soil fertility caused by erosion and continued
cultivation.
For better productivity, when affordable, manure or compost combined
with inorganic fertilisers including NPK and lime, were used.
Grazing practices by the respondents are shown on Figure 3.6.
consisted predominantly
Grazing practices
of free range grazing (67.7%) and zero grazing (32.3%).
During the dry season, fodder plants were collected everywhere, from wetlands and
swamps, as well as from farms (terraces and erosion control ridges).
Croplands
Household residues
Woodlots and forests
Along roads
Fallow farms ••••••
Erosion control ridges ••••••••••••••
Wetlands
•••••••••••••••
o
5
10
15
20
% of responses
Figure 3.6. Sources of fodder during the dry season in the study area.
25
51
Stellenbosch University http://scholar.sun.ac.za
3.1.3.4 Tree crops
Apart from fruits and medicinal plants present on farms, 66.3% of the households
owned plantations of different sizes (Figure 3.7), usually as small woodlots of less
than 0.5 ha. The farm sizes of the respondents showed a strong positive relationship
with tree crop area (pearson's correlation r = 0.35, p<O.OOl, n
= 200). The planting of
Eucalyptus spp. was common, alone or in combination with other tree species, namely
Pinus patuia and Grevillea robusta (Figure 3.8).
1.0·2.0 ha
4.0%
0.5 • 1.0 ha
13.1%
None
33.7%
< 0.5 ha
49.2%
Figure 3.7. Plantation areas owned by the respondents in the study area.
60
.....
Cl)
c
Il)
"0
C
0
0Cl)
50
40
30
....
Il)
4-<
0
~
20
10
0
2
3
4
5
6
7
Planted trees
1 = Eucalyptus spp.; 2 = None; 3 = Eucalyptus spp. and Pinus patula; 4 = Cupressus lusitanica;
5 = P. patula and C. lusitanica, 6 = Arundinaria alpina, Cypress and Hagenia abyssinica 7 = Eucalyptus spp. and Grevillea
robusta.
Figure 3.8. Tree species planted by the respondents in the study area.
Stellenbosch University
52 http://scholar.sun.ac.za
The majority of people (54.4%) indicated that they planted trees as their main source
of building poles.
Those who planted Pinus patula, Cupressus lusitanica
and
Grevillea robusta (10.3%) aimed to produce timber by pitsawying, as they recognised
an increasing market demand for this product.
Trees were converted to fuelwood by 28.7% of the respondents and to charcoal by
9%. Charcoal was a common product for sale. Others said that money was obtained
from the sales of poles. On average, income from tree products for many households
did not exceed 10000 Rwandan Francsl (RWF) per year (Figures 3.9 through 3.11).
40
<Zl
.....
s::
11)
'0
s::
30
0
0...
<Zl
11)
I-<
~
0
20
'cf-
Annual income, RWF
Figure 3.9. Annual income from the sale of charcoal by the respondents in the
study area
1
At the time of writing, 1US$ was equivalent to 473 RWF
Stellenbosch University
53 http://scholar.sun.ac.za
30
Vl
.....
s::
<l)
"'0
s::
0
0..
Vl
20
....
<l)
4...;
0
~
10
Annual income, RWP
Figure 3.10. Annual income from the sale of lumber by the respondents in the
study area.
40
Vl
1::
<l)
"'0
s::
30
0
0..
CIl
<l)
....
~
0
20
~
Annual income, R WF
Figure 3.11. Annual income from the sale of building poles by the respondents in
the study area.
Stellenbosch University
54 http://scholar.sun.ac.za
3.1.4
Soil fertility status and management
Soil fertility in the study area was a key concern to many households.
When asked to
rate the fertility status of their farms, 55.5% of the respondents said it was moderate,
while others considered the fertility of their farms as very poor (18%) and poor
(16.5%).
Many people (85%) reported that soil fertility decline in the study area was
a major impediment to the productivity of crops. Many causes of soil fertility decline
were cited, but inadequate use of manure, fertiliser and compost was a crosscutting
problem among all the Communes surveyed (Table 3.5).
Table 3.5. Causes of soil fertility decline listed by the respondents
area.
Causes of soil fertility decline
Lack of use of manure, fertiliser and compost
'Senility' of land
Continued cultivation, no fallow
Erosion
Soil acidity
Lack of use of terraces
Lack of use of improved crop varieties
Small farm size
Inadequate agricultural practices
Overgrazing
No.
responses
131
64
33
20
9
7
6
4
3
1
in the study
Percent of
responses (%)
65.5
32.0
16.5
10.0
4.5
3.5
3.0
2.0
1.5
0.5
Farmers' strategies to improve the fertility status of their farms were varied, but the
majority (53.5%) were using manure, fertiliser, compost or a combination of these.
The respondents with land shortage problems said that they were renting land from
the neighbours (22%) or that they served as casual labourers on neighbours' fields
(15%).
The remaining respondents had adopted various solutions to cope with low
fertility, including the practice of natural fallow, the use of terraces to prevent soil
erosion, the registration as farmer's association members and the use of improved
seeds.
3.1.5
Respondents'
perception
of Nyungwe Forest resources
The 200 respondents listed 18 categories of resources present in Nyungwe Forest. Of
the 18 resource categories, only 14 were said to be important for the respondents'
Stellenbosch University
55 http://scholar.sun.ac.za
livelihoods.
Table 3.6 shows these resources, their abundance estimates and their
uses by the respondents.
The most important products from the Forest, according to
more than 50% of the interviewees, were land, timber and minable gold. Nyungwe
Forest was seen as a source of fodder and pastures but by few respondents (16%).
Appendix 3 shows a list of palatable tree and grass species mentioned by 5 or more
respondents.
The most mentioned were Panicum hochstetteri, Triumfetta cordifolia,
Mariscus thomaiophyllus and Digitaria hackelii.
Table 3.6. List of important resources for community's livelihoods in the study
area, n = 200.
Resources
% respondents
Abundance estimate
Land
68.5
D (49)
Timber
58.5
D (45)
Gold mine
56.5
F (72)
Fodder and pastures
16.0
F (8)
Wooden and woven item
13.5
D (21)
Fuelwood
11.0
D (18)
Poles
10.5
D (20)
Employment
9.0
A (5)
Honey (beekeeping)
7.5
A (6)
Charcoal
3.5
NM
Medicinal plants
3.5
A (5)
Rain and humidity
3.5
NM
Ecotourism
2.5
NM
Game-meat
1.0
F (5)
A = Abundant; D = Dominant; F = frequent; NM = not mentioned
S = subsistence; C = commercial; SIC = subsistence and commercial
( ) indicates the percentage of responses for each resource.
The perception
of resources
Uses
SIC (75.4)
C (62.9)
C (89.7)
SIC (59.4)
SIC (48.3)
S (78.3)
S(74.1)
S (90)
C (71.4)
C (87.5)
C (62.5)
NM
C (100)
S (77.8)
in the Forest by individual
interviews
compared
favourably with that of the group participants in all the study sites. On the whole, 21
resources were listed and their abundances estimated (Table 3.7).
Timber, poles,
fuelwood and beekeeping were the major abundant resources common to all study
survey sites.
However, land availability was a controversial resource on which the
participants had a long discussion before a consensus was reached. Land availability
was perceived differently in terms of its size and soil fertility. For some participants,
the land area was large but not productive enough as the soil fertility declined after
one to two years of continued use. Others classified land as abundant because there
existed fertile wetlands in the Forest.
A similar debate occurred again for the
participants' perceptions of indigenous timber trees. Some claimed that they did not
know what was in the Forest, as they were not allowed to go there. Others said that
timber trees were abundant because the Forest was protected, was regenerating and no
pitsawying was carried out, as it was in the past.
Stellenbosch University
56 http://scholar.sun.ac.za
Table 3.7. Assessment of the availability of resources in Nyungwe Forest by
groups in the study area.
Resources
Gold
Timber
Land
Medicinal plants
Charcoal
Ecotourism
Fuelwood
Wildlife
Rain and humidity
Goods
Poles
Employment
Beekeeping
Pastures
Bean stakes
Water
Vegetables
Mushrooms
Wild honey
Fruit
Bushrneat
Musebeya
F
0
D
0
D
F
0
0
0
0
0
D
D
0
D
0
R
R
R
R
F
Nshili
F
D
F
0
NM
NM
F
D
D
0
0
F
D
0
NM
D
R
R
NM
NM
NM
Kagano
A
D
D
A
F
NM
0
A
D
F
D
0
0
NM
NM
D
R
R
NM
R
F
Mudasomwa
F
D
0
F
D
NM
D
A
0
F
D
A
A
A
NM
D
R
R
R
R
0
Busozo
F
F
A
0
NM
NM
D
A
D
NM
A
A
D
F
NM
D
0
0
0
R
F
D = Dommant; A = Abundant; F = Frequent; 0 = Occasional; R = Rare; NM = Not mentioned
From household interviews, only a few respondents
indicated the availability of
building poles (10.5%, n = 21), fuelwood (11%, n = 22), medicinal plants (3.5%, n=7)
and charcoal (11%, n = 22) in the Forest.
Ecotourism potential was a resource
mentioned by a very few people, only 2.5% of the respondents,
and only 9%
identified the potential of the Forest to provide local employment.
The Forest's
ecological functions, such as its influence on local climate, were cited as resources by
3.5% of the respondents.
3.1.6
Resource preferences
and ranking
This exercise enabled the participants to list the resources that they thought would be
of value to their livelihoods if they were given access to use them.
In general,
however, as some resources were mentioned and listed, the participants from some
groups reacted by commenting that their preferences could not be met, as the Forest
was not theirs but for the State, the Rwanda Office of Tourism and National Parks
(ORTPN) and the foreign tourists.
Stellenbosch University
57 http://scholar.sun.ac.za
Preferred resources were ranked using both discussion and the pairwise ranking of
importance of resources to communities.
Table 3.8 shows the ranked order of the
resources in order of their importance in relation to each other. The most important
resources listed by the participants were land, timber and gold, but the ranking
appeared to be dependent on the agro-ecological
characteristics of the population.
zone and the socio-economic
Land preference ranked high in areas where farm
sizes were small, as in N shili and Kagano, but not in Musebeya where people had an
average farm size of 2.0 ha. Whereas timber was the first priority in Mudasomwa,
gold mining was important for the community in Busozo.
These preferences partly explain the problems encountered by the local authorities
and the conservation
services
with regard to the illegal use of the Forest.
Deforestation for arable land was a severe problem in Busozo, particularly in the
Bweyeye Sector.
Table 3.8. Ranking of resource categories in the study area
Rank
order
1
2
3
4
5
6
7
8
9
10
11
12
l3
14
Musebeya
Nshili
Kagano
Mudasomwa
Busozo
Land
Timber
Fuelwood
Pastures
Employment
Poles
Ecotourism
Charcoal
Beekeeping
Rain
Gold
Medicinal plants
Goods
Bushmeat
Land
Timber
Gold
Bamboos
Pastures
Poles
Fuelwood
Employment
Beekeeping
Charcoal
Bushmeat
Land
Gold
Timber
Poles
Employment
Fuelwood
Goods
Beekeeping
Pastures
Medicinal plants
Bushmeat
Timber
Gold
Land
Charcoal
Fuelwood
Beekeeping
Medicinal plants
Bushmeat
Goods
Gold
Land
Timber
Pastures
Employment
Rain
Poles
Bushmeat
In Nshili and Kagano, available information suggested that both the buffer zone and
Nyungwe Forest were used to produce food crops. Food production and hunting in
the Forest were mostly practised by about 1 000 indigenous people, Batwa, who lived
inside the Forest. Unlawful users of the Forest were given fines or were jailed.
At
Kagano, the illegal use of the Forest and the buffer zone was condemned by a fine
ranging from 10 000 RWF to 20 000 RWF.
Strategies to reduce the rate of
deforestation were many and included the management of the largest wetlands in the
environs of the Forest, namely Kamiranzovu and Uwasenkoko.
Other suggested ways
of reducing the conflict from land problems were the delimitation and management of
Stellenbosch University
58 http://scholar.sun.ac.za
pasture lands, the intensification of agriculture involving the use of erosion control
methods and fertilisers, and the establishment of projects that could provide local
employment.
Timber was the first priority needed in Mudasomwa, second in Musebeya and Nshili
and third in Kagano. Although the participants denied the harvesting of timber from
the Forest, The Nyungwe
Forest Conservation
Project
(PCFN) and the local
authorities recognised illegal logging of commercially important timber tree species in
all the Communes being surveyed.
They indicated that illegal logging was taking
place throughout the year, with intense activity during the dry season. Timber was
sold locally, outside the area or in neighbouring countries such as in the Republic of
Burundi that borders with Nshili and Busozo Communes in the South of the Forest.
The local authorities and the conservation services were much concerned about
charcoal making and fuelwood gathering in Nyungwe Forest and in the buffer zone.
Proposed solutions for reducing the conflict over illegal timber exploitation, charcoal
making and fuelwood gathering included among others:
•
Licensed harvesting and pitsawying of mature or dead trees in specified areas of
the Forest;
•
Harvesting timber trees in the buffer zone;
•
Privatisation of the management of Nyungwe Forest;
•
Distribution
of seedlings
communities
lil
of fast growing
preferred
tree species to local
order to enable them to establish their own sources of tree
products;
•
Revival of forestry projects.
Gold mining
was the highest priority
lil
Busozo,
followed
by Kagano
and
Mudasomwa. At Busozo, mining was an alternative form of livelihood, and the group
participants indicated that it was the only source of income for the majority of
households, whereas people in Kagano and Mudasomwa were getting income from
tea factories. For instance, the Tea Factory at Gisakura in Kagano Commune
employed annually approximately
averaged 40 million RWF.
5 000 casual labourers for which the salaries
59 http://scholar.sun.ac.za
Stellenbosch University
The Rwanda Office of Tourism and National Parks suspended gold mining because
the activity led to the deviation and pollution of rivers and loss of biodiversity of the
Forest. Mining in the valleys in the vicinity of the Forest resulted in the cutting of
trees, erosion, hunting, agriculture and the sale of beer in the Forest. According to the
conservation services, mining rights in bottomlands and valleys could only be granted
to organised groups and the activity guided by a memorandum of agreement.
Except for the high rankings given to land, timber and gold resources, preferences for
other resources ranked variably, with exception of bushmeat which appeared to be of
less interest to all five communities.
Bamboo, Arundinaria alpina, bamboo ranked
among the most important resources at Nshili compared to other study sites.
This
indicated the extensive use of bamboo collected without permit from the Forest for
house construction and for basketry material.
Groups in Musebeya, Kagano and Mudasomwa
plants as important resources.
Communes
identified medicinal
Current use of the resource was permitted by the
Nyungwe Conservation Project (PCFN) and had shown little impact to the Forest.
The practice
of beekeeping
was also considered
important,
particularly
where
beekeepers' associations were formed. PCFN restricted the activity to the buffer zone
and in open areas inside Nyungwe Forest.
Beekeeping, however, was reported to
result in runaway fires in all the study sites.
Conservationists
estimated that the
possibility of runaway fires could be addressed through the formation of beekeeper's
associations.
Beekeeping was practised during the period of April to August when
tree species favoured by bees were in flowers. On the local market, honey was sold at
500 RWF to 800 RWF per kilogram. Honey was consumed as food, as an ingredient
of beer, and wax was used for the local production of candles.
3.1.7
Tree species preferences and uses by the respondents in the study area
3.1.7.1 Timber tree species
Hardwood tree species from Nyungwe Forest were preferred for their sizes, durability
and wood colour for different uses. Table 3.9 shows the most commonly used tree
species. Other preferred timber species are presented in Appendix 4.
60 http://scholar.sun.ac.za
Stellenbosch University
Table 3.9. Tree species most preferred for timber by the survey respondents
around Nyungwe Forest, n = 200
Only tree species mentioned by 10 or more respondents are presented.
Tree species
Percentage of
responses
64.0
56.0
38.0
15.5
13.5
12.0
11.5
10.0
10.0
8.0
7.5
6.0
5.0
5.0
Podocarpus latifolius
Entandrophragma excelsum
Podocarpus falcatus
Symphonia globulifera
Parinari excelsa
Faurea saligna
Carapa grandiflora
Dasylepis racemosa
Syzygium parvifolium
Hagenia abyssinica
Ocotea usambarensis
Newtonia buchananii
Polyscias fulva
Strombosia scheffleri
A range of timber species was mentioned by the respondents as being used for the
production of wooden items. Generally, goods were obtained from the following key
tree species: Polyscias fulva (a musical instrument, locally known as 'inanga'),
Ocotea usambarensis (household utensils, beer boat), Markhamia lutea (household
utensils), Carapa grandiflora (household utensils), Dasylepsis racemosa (beer boats)
and Asparagus falcatus (beer boats). Other wooden and woven goods were obtained
from non-timber forest species and included bamboo Arundinaria alpina, different
climbers (Smilax anceps, Urera cameroonensis,
U hypselodendron)
and shrubs
(Triumfetta cordifolia).
3.1. 7.2
Building poles
The preference of tree species for poles in the study area was based on the quality of
the stems. Major characteristics of suitable trees for poles included stem straightness,
durability and a suitable diameter of at least 10 em. Tree species preferred for quality
building poles are given in Table 3.10.
Poles were used for house building, for
fencing of livestock and the home compound.
Bamboo was also used for building
particularly in Nshili Commune next to the forest area where it occurs.
61 http://scholar.sun.ac.za
Stellenbosch University
Table 3.10. Tree species whose stems are preferred
study area, n = 200.
for building poles in the
Only tree species mentioned by 10 or more respondents are presented.
% of responses
Tree species
36.0
Macaranga neomildbraediana
28.5
Cassipourea ruwenzoriensis
26.0
Strombosia scheffleri
13.5
Syzygium parvifolium
12.5
Carapa grandiflora
12.5
Casearia runssorica
6.5
Olea hochstetteri
6.5
Arundinaria alpina
6.0
Parinari excelsa
5.5
Alchornea hirtella
5.5
Podocarpus latifolius
5.5
Oxyanthus speciosus
5.0
Faurea saligna
5.0
Cassipourea gummiflua
5.0
Mil/etia spp.
5.0
Symphonia globulifera
5.0
All tree species
Only 9.5% of the respondents indicated that Nyungwe Forest was a source of building
poles in the past, where quality-building poles were abundant.
Even now, of the 25
respondents who mentioned building poles as one of the most important resources for
their livelihoods, 80% (n = 20) categorised the availability of building poles in the
Forest as very abundant.
The majority of the respondents (52.1 %) said that poles
occurred dominantly everywhere in the Forest. Most people (74.1 %) estimated that
access to Nyungwe Forest would satisfy their subsistence needs for building poles.
3.1. 7.3 Medicinal plants
Respondents listed 45 medicinal tree species found in Nyungwe Forest (Appendix 5;
Table 3.11). Products from these plants particularly the bark, were traditionally used
by medical practitioners to treat people or livestock.
Bark material was used from
56.5% (26 out of 45 species) of the medicinal plants mentioned by the respondents.
For some species, both bark and leaves could be used as sources of medicine (Table
3.11).
Trees are the major life form category of the medicinal plants, with the
exception
of Asclepiadaceae
scandens which are climbers.
spp., Begonia
meyeri-johannis
and Sericostachys
Stellenbosch University
62 http://scholar.sun.ac.za
Most people reported that medicinal plants were used locally to treat intestinal
parasites in people and livestock.
were mentioned
There is no commercial trade in medicines.
Men
as the major harvesters of medicinal plants by 67.4% of the
respondents, while 30.2% said that both men and women harvested medicinal plants.
Table 3.11. Plant species used for medicinal purposes in the survey area, n =
200.
Onl~ Elant sEecies mentioned b~ 3 or more resEondents are Qresented.
Part used
Local name
Plant sEecies
Bark
Umushwati
Carapa grandiflora
Bark, roots
Umwumba
Prunus africana
Bark, leaves
Syzygium parvifolium
Umugote
Bark
Zanthoxyllum gilletii
Umuturirwa
Bark
Croton megalocarpus
Umunege
Dichrostachys cinerea
Umunkamba
Leaves
Fagaropsis angolensis
Umugomera
Bark
Ocotea usambarensis
Bark
Umutake
Rapanea melanophloeios
Bark, leaves
Uruneke
Ocotea michelsonii
Umuganza
Bark
Mitragyna rubrostipulosa
Leaves, bark
Umuzibaziba
Vernonia amygdalina
Umuravumba
Leaves
Asclepiadaceae div. spp.
Indarama
Tubers
Canthium oligocarpum
Bark, leaves
Umushabarara
Cyathula polycephala
Igifashi
Leaves
Maesa lanceolata
Leaves, roots
Umuhanga
Begonia meyeri-johannis
Leaves
Irebe
Xanthoxyllum usambarensis
Intareyirungu
Bark
Symphonia globulifera
Bark, leaves
Umushishi
Pentas decora
Leaves, bark
Isagara
Brillantaisia cicatricosa
Icyunga
Bark
Faurea saligna
Umutiti
Bark
Myrica kandtiana
Isubyo
Wood
Tabernamontana johnstonii
Umuronzi
Leaves, bark
Canthium g/abriflorum
Imvuvu
Bark
No. resEondents
51
27
22
16
16
10
9
8
8
8
7
7
6
6
6
5
5
5
3
3
3
3
3
3
3
The respondents varied widely in their estimates of the amount of medicinal plants
that could be collected from the Forest. From a total of 129 respondents who knew
about medicinal plants, 44.2% (n
while 36.4% (n
=
=
57) estimated the amount collected to be small
47) said moderate quantities were harvested.
Forty-four percent of
the respondents said that medicinal plants were found everywhere in the Forest, while
26% (n = 35) located many medicinal plants in the bottomlands.
The remaining
respondents found medicinal plants in several different locations, including valleys,
disturbed areas, and hilltops.
Medicinal plants such as Vernonia amygdalina, Iboza
riparia, Mitragyna rubrostipulosa,
and Lantana trifolia were also found outside
Nyungwe Forest in disturbed areas and on fallow farms.
Stellenbosch University
63 http://scholar.sun.ac.za
3.1.8
Uses of wild foods
Wild foods collected from the Forest (Table 3.12) are used either for subsistence or as
a source of modest income from their sales on the local market. They are seasonally
obtained from the Forest, most frequently during the wet seasons.
Table 3.12. Wild food resources found in Nyungwe Forest, as mentioned by the
respondents in the study area, n = 200.
Food
Percentage of
responses
88.5
14.5
32.0
66.5
11.5
Fruit
Game
Mushrooms
Vegetables
Wild honey
The most important edible fruits to the respondents are Myrianthus holstii (88.5%,
n=l72) and Rubus rigidus (17%, n = 34).
Other wild fruits mentioned by the
respondents are Cyphomandra betacea (5.5%, n = 11), Physalis peruviana (5.0%, n
=
10), Passiflora edulis (7.5%, n = 15) and Citrus lemon (0.5%, n = 1). These were all
reported to be available in Nyungwe Forest from seeds dispersed from plants
originally established on farms.
Edible wild mushrooms identified by their local names included ibizinu, ibihumyo,
imegeri, utwoba, intyabire and ubushikiri.
According to the respondents,
these
mushrooms are commonly growing in association with termite mounds or with dead
trees. Mushrooms provided seasonal income from sale in local markets.
Prices for
one mushroom ranged from 10 RWF to 50 RWF, depending on the location of the
market. Mushroom gathering was practised mostly by women and children.
Wild vegetables from Nyungwe Forest are another seasonally available resource that
was used as a dietary supplement or as a source of small income.
whose leaves are used as food include different Amaranthus
nigrum.
Annual herbs
spp. and Solanum
Local names of some Amaranthus spp. were bwizabwishyamba,
umuhuruza and umushokorankoko.
inderama,
The vegetables were found in various locations in
the Forest. Most respondents mentioned their occurrence in swamps (39.8%, n = 53),
and in disturbed areas, including forest areas that had been cleared for agriculture and
64 http://scholar.sun.ac.za
Stellenbosch University
burned areas (28.6%, n = 38).
The wild vegetables were mostly gathered for
subsistence use, but a local market for Solanum nigrum did exist. For this vegetable,
out of 75 respondents, 41.5% suggested a market price of 5 to 10 RWF per handful.
As for other NWFP from Nyungwe Forest, the major collectors of wild vegetables
were children and women who collected small amounts at a time.
Honey is a wild food that can be found in the Forest. A few respondents (11.5%, n =
23) mentioned its availability from ground or tree-nesting bees known locally as
inkura.
Wild honey is collected from hollow tree trunks or branches as well as
underground.
Wild honey is used solely for household
consumption.
Honey
production was listed as an important activity in the Forest, and associated with the
availability of plant species favoured by bees and hills inside the Forest that provided
the best sites for placing hives. Forest plants that were important sources of nectar
and pollen to bees were listed as Sericostachys scandens, Brillantaisia cicatricosa, B.
nitens, Mimulopsis excellens, Syzygium parvifolium and Vireetaria major. The areas
preferred for hive placement were those with high densities of flowering trees,
climbers and herbs.
Bushmeat was of value to adjacent communities.
Animal species were hunted for
subsistence. Other wild foods, consisting of two tuberous species, were mentioned by
only a few respondents.
Discorea spp. (2%, n
=
These were Afromomum angustifolium
(2%, n
=
4) and
4), reportedly gathered by poor people and the indigenous
people Batwa, who were living inside the Forest. The low numbers of users suggests
that the two wild foods were not commonly of value in all the survey sites.
3.1.9
Respondents' knowledge of wildlife species
The respondents mentioned 32 wildlife species and estimated their abundance in the
Forest.
Their presence was determined by the noise of wildlife from the Forest or
observations inside or outside the Forest.
Commonly cited wildlife included crop
raiding animals such as bushpigs (82%, n = 164), chimpanzees (48%, n = 97), grivet
monkeys (25.5%, n = 51), black and white colobus monkeys (44.5%, n = 86) and
olive baboons (73%, n = 146).
Stellenbosch University
65 http://scholar.sun.ac.za
Wild animals
caught
for domestic
meat
were listed
by many
respondents.
Respondents perceived most of the crop raiding animals to be abundant because they
were easily seen in groups on farms or heard from the Forest.
According to many
respondents (25% to 50%), dominant wild animals consisted of monkeys, baboons
and colobus.
A few wild animals were estimated to be rare in the Forest: elephant
(Loxodonta africana) (92%, n
=
23), buffalo (Syncerus caffer) (80%, n
=
4), civet
(Viverra civetta) (54.5%, n = 6) and bushbuck (Tragelaphus scriptus) (55.6%, n =
15). Generally, edible animals were said to be few.
Animals hunted for subsistence
included the black-fronted
duiker Cephalophus
nigrifrons (98.2%, n = 167), the yellow-backed duiker C sylvicultor (99%, n = 102),
the bushpig Potamochoerus poreus (100%, n = 127), the African elephant Loxodonta
africana (36.5%, n = 35), the African buffalo Syncerus caffer (98.7%, n = 75), and the
bushbuck Tragelaphus scriptus (98.9%, n
=
86). Many other small animals, including
the Gambian rats, rabbits, a bird species identified by its local name igikoyi, and a
small mammal known locally as inkezi, were valuable sources of animal proteins. Of
the 200 respondents, only 17 said that monkeys were hunted from the Forest either for
food (76.5%) or for reducing the population of crop raiding animals (23.5%).
Hunting was practised in all seasons.
Of 183 people who answered, 71% estimated
that hunting was done throughout the year, although a few argued that hunting is
frequent during the rainy season because animal footprints are visible on the ground.
Commonly cited hunting methods included dogs (22.3%, n
(29.3%, n
=
55), and snares (23.4%, n
=
=
42), dogs and spears
44).
According to the people interviewed, two factors, namely the reproduction rate and
the 'no hunting rule', contributed to the change in animal population size of wild
animals in the Forest. Compared to the past, an increase in the population of monkeys
was observed by 40 % of the people interviewed, baboons by 24.5%, bushpigs by 8%,
colobuses
by 6.5% and chimpanzees
by 4.5%.
Chimpanzee's
populations
particular were thought to be increasing because they are not eaten.
in
Where the
number of wild animals was perceived to have decreased, this decrease was attributed
to hunting. Respondents perceived a reduction in numbers of black-fronted duiker (C
nigrifrons), yellow-backed duiker (C sylvicultor), and porcupine Atherurus africanus.
Stellenbosch University
66 http://scholar.sun.ac.za
The hunting for meat, skins, teeth, horns and medicines caused the local extinction of
some wild animals as recognised by the respondents (Table 3.13).
Table 3.13. Respondents'
Forest, n = 200.
knowledge of locally extinct wild animals in Nyungwe
English names
Elephant
Buffalo
Leopard
African lion
3.1.10 Current
Latin names
Loxodonta africana
Syncerus caffer
Panthera pardus
Panthera leon
% of respondents
60.5
30.5
8.5
6.0
resource access and use
The majority of the respondents (97.5%) said they had no access to resources in
Nyungwe
Forest.
A few individuals
(3%) admitted
access to cut bamboo
(Arundinaria alpina) from the Forest. In Nshili Commune in the south-eastern zone
of the Forest, more than five persons said they were aware of the regulations but that,
due to poverty and lack of alternatives, they kept on harvesting bamboo from the
Forest.
Others perceived that they benefited from the ecological function of the
Nyungwe Forest in their areas, even though they were not permitted to access land
and other resources in the Forest.
In the buffer zone, however, the respondents
collected dead branches for use as fuelwood (31.5%) and grasses (23%) for fodder.
Agriculture in the bottomlands in the environs of the Forest was practised by 11.5 %
of the interviewees.
Others used the buffer zone as a settlement site because they
argued that they had not been compensated for leaving the forest plantations.
3.1.11 Respondents'
perception of conflicts and conflict resolution
In the study area, the respondents listed four types of conflicts arising from their
cohabitation with the Forest (Table 3.14), the major conflicts being between local
people and either the conservation
services or the government.
These conflicts
occurred when the Nyungwe Forest Conservation Project (PCFN) and the extension
services evicted the local people who were encroaching
respondents
also accused the Rwanda
on the Forest.
Office of Tourism
and National
Some
Parks
(ORTPN) of employing outsiders as forest guides and rangers. Conflicts between the
local people and the government were attributed to lack of government decisions
Stellenbosch University
67 http://scholar.sun.ac.za
permitting the use of resources in the Forest by neighbouring communities.
The
respondents adjacent to the Forest were mostly concerned about their crops that were
raided by wild animals.
While illegal users of the Forest were severely sanctioned,
respondents claimed that in contrast they were not compensated for crops damaged by
wild animals.
Table 3.14. Nature of conflicts listed by the respondents
200.
in the study area, n
=
% of respondents
Conflict
Local people vs. government
Local people vs. wildlife
Local people vs. conservation
Local people vs. militia
Local people vs. extension service
Within local people
45.5
24.5
48.0
20.5
2.5
0.5
Twenty percent of the survey respondents reported that, during the last two weeks,
armed bandits coming from the Forest had threatened their lives and stolen their
livestock.
This was an uncommon problem, and respondents proposed the use of
force for its resolution.
Apparently, tensions or competition among local villagers
over Nyungwe resources did not exist within the community surveyed (Table 3.15).
Table 3.15. Importance of conflicts over resource use and access, as rated by tbe
respondents in the study area.
Conflict
Not important
n
107
50
97
41
5
Local people vs. government
6.5
Local people vs. wildlife
Local people vs. ORTPN*
21.6
Local people vs. militia
2.4
Local people vs. agricultural
60.0
extension services
Within local people
1
100
* Rwanda Office of Tounsm and National Parks.
Importance of conflict
Important
Very important
9.7
6.0
23.7
12.2
-
12.7
8.0
23.7
7.3
40.0
-
-
Most
important
71.0
86.0
30.9
78.0
-
Only 89 respondents suggested strategies for the resolution of local - government
conflict.
Of these, approximately
24% suggested consultation between the local
people, the government authorities and the conservation services in order to establish
resources that could be used. Other commonly mentioned strategies included creation
of job opportunities in the area, the demarcation of a use-zone where resource use
Stellenbosch University
68 http://scholar.sun.ac.za
would be permitted, the use of fines, fencing of the Forest to avoid crop destruction
by wild animals and compensation for damaged crops by wildlife.
3.1.12 Respondents' perception of past and present forest management
Of the households interviewed during the survey, 85.7% said that the management of
Nyungwe Forest was under the responsibility of the state. When asked what role the
state played in forest management, 62.5% of the respondents said that the state was
involved in forest protection activities. Twenty percent thought that the management
of the Forest by the state had encompassed the allocation ofland to local people in the
Forest adjacent areas. Only 11.7% believed that the Forest belonged to everyone, as
access was open to all. Harvesting rights were said to be exclusive to the state and the
colonial authority, and the provision of harvesting licenses was recognised by 16% of
the respondents.
At the local level, agricultural extension services and the local
administration were reported to enforce regulations and to sensitise local people on
the importance of conserving Nyungwe Forest.
3.1.13 Respondents' views of appropriate institutions for forest management
Community organisations
(Wild and Mutebi, 1996).
are important for collaborative
management
of forests
Survey respondents were asked to name institutions in
which they had trust and confidence (Table 3.16) and to give reasons for their
answers.
Table 3.16. Proposed institutions for management by the respondents, n
Institutions
District
Cell
ORTPN
Local protection committee
Sector
MINAGRI*
Local administrative committee
Developmental Project
Government
Local farmers' associations and co-operatives
Total
* Ministry of Agriculture, Livestock and Forestry.
% of respondents
24.0
16.5
12.5
10.5
10.0
7.5
6.5
6.0
3.5
3.0
100.0
=
200
Stellenbosch University
69 http://scholar.sun.ac.za
Twenty-four percent were of the view that the District Administration was the most
appropriate institution that could enable community use of resources and manage the
Forest sustainably because of the power and authority that the District holds over all
development issues in the region. Many argued that local concerns were best known
at District level.
Other reasons for the choice of District as the appropriate organisation included: •
The capability of the District to provide employment to local people;
•
The ability to mobilise enough resources for management and protection;
•
The part of the Forest near to the respondents lies within the District geographical
zone;
•
The leadership at District level was elected by local people;
•
The District had a vested interest in the conservation of Nyungwe Forest, more
than any other organisation.
Similar institutions
Communes surveyed.
were identified
during interviews
with groups in the five
Generally, institutions in which local people had confidence
were local and concerned with local administration and development (Box 3.1).
Box 3.1. Proposed institutions
for management
of Nyungwe Forest utilisation
District Administration.
The District based administration has been introduced in Rwanda very
recently within the framework of good governance and decentralisation of administrative
authority.
A District is geographically based and comprises one or more adjacent exadministrative Communes. The leaders of the Districts were elected locally in order to manage
the affairs of the Districts and to coordinate activities in the administrative Sectors.
Administrative
Sector. This is a lower level of the District Administration. It is based on the
administration of at least 250 households within the same geographical area. The leaders at this
level consist of an executive committee of nine people elected on their merits by local people.
Each member has specific responsibilities
in the sector, including health, education,
communication, and security, among others. The sector Coordinator is the head of the Sector
and he is assisted by a secretary.
Administrative Cell. This is a further subdivision of the Sector and refers to an administrative
area occupied by at least 50 households. The structure of the Cell is similar to that of the Sector,
with the community leadership elected locally and its representation found in women, youth and
men.
Nyungwe Forest Conservation Project (PCFN). A conservation project managed by ORTPN
and funded by USAID and WCS. It was established in 1988. It deals with conservation,
environmental education and research on fauna and flora in Nyungwe Forest.
Stellenbosch University
70 http://scholar.sun.ac.za
The interviewees perceived key roles that these institutions should play with regard to
resource use and access in Nyungwe Forest. Below is a compilation of the roles that
came out from household and group interviews:
•
To participate in protection and conservation ofNyungwe Forest;
•
To employ local people in pitsawying mature trees and in enrichment activities on
degraded and burned areas in Nyungwe Forest;
•
To negotiate access to resources from the highest authorities and to participate in
the organisation of associations for resource utilisation (for example, associations
of beekeepers, gold-miners, herbalists, and basketry makers);
•
To identify and delimit use areas in the part of the Forest adjacent to communities,
including land for agricultural development;
•
To raise awareness on the importance of conserving Nyungwe Forest;
•
To provide technical and financial assistance to organised resource users;
•
To provide incentives for protection;
•
To resolve conflicts over resource use and access.
Individual
responsibility
for forest management
was assessed
by asking
the
respondents what they could do to sustain Nyungwe Forest resources. Answers were
many and diverse.
The responses in Table 3.17 show that the respondents have an
interest in having access to resources in the Forest.
Table 3.17. Individual responsibility of the respondents for sustainable forest
management in the study area, n = 200.
Activity
Participation in protection and management
Respect of laws, regulations and guidelines
Reporting offenders
Non-destructive use of the Forest
Harvesting mature trees and enrichment planting
Participation in awareness raising
Use of demarcated use zone
Selective harvesting of mature trees
Rotational harvesting and enrichment planting
Enforcement of patrol
% of
respondents
29.0
21.0
21.0
19.5
11.5
9.5
8.5
4.0
2.5
1.0
The most common suggestion for community involvement (29%, n = 200) in forest
protection and management was through participation in related activities, followed
Stellenbosch University
71 http://scholar.sun.ac.za
by the respect of laws, regulations and guidelines (21%, n
=
200) and the report of
offenders to authorities (21%, n = 200). Only 19.5% of the respondents said that they
would commit themselves to what they pretended to be non-destructive activities in
the Forest, including agriculture and mining in the swamps, gathering of fuelwood
and other non-wood products.
3.2
FOREST RESOURCE ASSESSMENT
Species composition and tree abundance were analysed for distribution among and
within the forest areas adjacent to the communities surveyed in Musebeya, Nshili,
Kagano, Mudasomwa
and Busozo Communes.
Abundances
of tree species in
different diameter sizes are presented for the three categories as indicated in Table
3.18.
Table 3.18. Categories of trees used during the analysis.
Categories
Small trees
Medium trees
Large trees
Height (m)
0.30 - 1.30
> 1.30
DBH (cm)
< 2.5
2.5 - 9.9
2::10.0
The subdivision into small, medium and large trees corresponds to the definition of
seedlings and saplings, poles and trees by Mwima (2000) and Owiunji (1997),
respectively. The small trees were recorded for the assessment of the regeneration of
tree species represented
by large and medium trees.
Data on tree and species
abundance are presented on the basis of diameter size-class distributions among and
within the study sites. Further, the assessment of abundance, dominance, size-class
distribution and regeneration is presented for the most preferred tree species identified
during participatory rural appraisal (PRA).
3.2.1
Identification of tree species
The local and scientific names of all tree species are given in Appendix 6. Some local
names were applied at "generic level," and a single local name could correspond to
two or more species, as reported by Cunningham (1996).
local name "ingongo"
corresponded
For example, the single
to three Cassipourea
species, Cassipourea
ruwenzoriensis, C. ndando, and C. congoensis; "umunkamba" was a local collective
72 http://scholar.sun.ac.za
Stellenbosch University
name applied to various species including Ormocarpum tricocarpum, Dichrostachyus
cinerea, and an on-farm medicinal plant, Clematis simensis.
The matching of local
names to scientific names also required attention because a local name given to a
single plant species sometimes differed from one area to another. For instance, local
people in Nshili and Busozo Communes used the local name "rwamba" to describe
Prunus africana whereas others used "umwumba"
to describe the same species.
Despite this shortfall, local knowledge of tree species represented a practical, less
time consuming and cost-effective method for identifying resource of interest to local
communities in the absence of an operational Herbarium and qualified ethnobotanists
in Rwanda.
3.2.2
Tree species abundance and density analysis for the study area
For the whole sample area, there was an average of 439 large trees per hectare
(standard deviation = 52 stemslha), represented by 54 species. The mean density of
medium trees per hectare was 1 407 (standard deviation = 542 sterns/ha), also
represented by 54 tree species.
Small trees averaged 5 146 individuals per hectare
(standard deviation = 3776 sterns/ha) but they were represented by only 34 tree
species. The large standard deviations indicate that the stem densities differed greatly
among the study sites. The results of the analysis of variance indicated significant
differences in stem densities among the three tree categories (Table 3.19).
Table 3.19. Analysis of variance of the densities of trees among tree categories
Tree category
Small trees
Medium trees
Large trees
F-value
p-value
*
Stem density/ha
5146A
1407AB
439 B
6.97*
0.02
indicates significance at p < 0.05.
The Tukey's test of the comparison of the means showed that the densities of small
trees did not differ significantly from the densities of medium trees but differed
significantly from the densities of large trees.
The analysis of the whole sample
indicated that there were no significant differences in the densities of individuals
between the study sites (d.f = 4, F=0.59, p>0.5) with a minimum
difference of 8380 stems per hectare.
significant
Stellenbosch University
73 http://scholar.sun.ac.za
A separate Chi-square analysis of the whole sample showed that the distribution of
trees among the five study sites was not even (X2 = 1880.31, d.f. = 8, p<O.OO1).
2
However, tree species were evenly distributed among these sites (X = 11.55, d.f. = 8,
p = 0.17).
The difference in the distributions may reflect the heterogeneity in the
study area, as the plots were located in areas with different altitudes, soils and
different degrees of disturbance.
Some species had many representative individuals, contributing significantly to the
total density of trees recorded on each study site. The contribution of each tree species
to the total density of individuals recorded in each study site and at different tree
categories is presented in Appendices 7 through 9.
Table 3.20 summarises the
occurrence of the 10 most dominant species by site and tree categories.
Table 3.20. Occurrence
category.
of the 10 most dominant tree species, by site and tree
Tree species
Strombosia schejjleri
Carapa grandiflora
Cleistanthus polystachyus
Myrianthus holstii
Casearia runssorica*
Symphonia globulifera
Cassipourea ruwenzoriens is *
Parinari excelsa*
Macaranga neomildbraediana*
Chrysophyllum gorungosanum
..
1 = Musebeya; 2 = Nshili, 3 = Kagano, 4
* Species that were not among the 10
dominant as large and medium trees.
Tree categories
Large trees
Medium trees
Small trees
1,2,3,4,5
1,3,4,5
2,3,4,5
1,2,3,4,5
1,3,4,5
1,2,3,4,5
2,3,5
2,3
2,3,5
1,2,3,4,5
1,2,3,4,5
1,2,5
1,2,3,4,5
1,2,3,5
1,2,3,4,5
1,2,3,4,5
1,2,3,4,5
1,2,3,4,5
3,5
1,2,3,4,5
1,2,3,4,5
1,3,4
1,2,3,4,5
3
3,5
1,2,3,4,5
1,2,3,5
1,3,4,5
1,2,3,4,5
1,2,4,5
= Mudasomwa; 5 = Busozo .
most dominant species in the small tree category but were
Seven tree species were represented in the large trees and the small trees in all the five
sites. Only seedlings and saplings of Symphonia globulifera were recorded in all the
five study sites. The marked differences in the presence of seedlings and saplings of
tree species in all the study sites could be attributed to environmental conditions, to
shade intolerance of some species, and to disturbances in the periphery of Nyungwe
Forest. These differences are also reflected in the stem densities of the species (Table
3.20 above). Figure 3.12a to 3.12c provide a comparison of the stem densities of the
10 most dominant tree species recorded in the three tree categories.
Figure 3.13
indicates that the densities of small and medium trees were generally higher than
those of large trees in all the study sites.
74
Stellenbosch University http://scholar.sun.ac.za
2500
2000
os
.é
....
1500
8..
.g
'"I:
1000
E
500
-8
b3
0
2
5
4
3
6
8
7
10
9
Species
Figure 3.l2a. Stem densities per hectare of the 10 most dominant tree species represented by
small trees
350 l
...
300
0.
;.-,
250
<1)
.~ 200
~'" 150
~ 100
ei)
50
o
2
3
4
5
6
7
9
8
10
Species
Figure 3.l2b. Stem densities per hectare of the 10 most dominant tree species represented by
medium trees.
150
ii 120
Co
.~
'""
90
"0
ë
~
en
60
30
0
2
3
4
5
6
7
8
9
10
Species
• Musebeya
• Nshili D Kagano
• Mudasomwa
• Busozo
1 = Carapa grandiflora; 2 = Casearia runssorica; 3 = Cassipourea ruwenzoriensis; 4 = Chrysophyllum gorungosanum; 5 =
Cleistanthus polystachyus, 6 = Macaranga neomildbraediana; 7 = Myrianthus holstii; 8 = Parinari excelsa; 9 = Strombosia
scheflleri; 10 = Symphonia globulifera.
Figure 3.12c. Stem densities per heetare of tbe 10 most dominant tree species
represented by large trees.
75
Stellenbosch University
http://scholar.sun.ac.za
Ste 2=N;hili
~1=MEeh:ya
12345
6
7
8
9 ID
5 6
7
8 9 ID
I 2 3 4 5 6 7 8 9 10
~
~3=Kag;m
I 234
12345
6
7
8 9 ID
~
ill)
1<Ul
.!!Il)
~(fi)
~<U)
jm
O~~~~~~~~~-.~~
123
456
7
8
9 ID
1 = Carapa grandiflora; 2 = Casearia runssorica; 3 = Cassipourea ruwenzoriensis; 4 =
Chrysophyllum gorungosanum; 5 = Cleistamhus polystachyus, 6 = Macaranga neomildbraediana; 7 =
Myrianthus holstii; 8 = Parinari e:xcelsa;9 = Strombosia scheffleri; 10 = Symphonia globulifera
Figure 3.13. Comparison of stem densities per hectare ofthe 10 most dominant
species between tree categories, by study site.
76 http://scholar.sun.ac.za
Stellenbosch University
3.2.3
Diameter size-class distribution
of trees among and within the study sites
Table 3.21 shows the densities of trees by size-class categories in all the five study
sites.
Table 3.21. Diameter size-class distribution
of large trees among the study sites.
1 . included
.
The analysis
only species WhiiChifi
registered a frequency greater or equa to rve
DBH size classes (cm)
10.0- 19.9 20.0 - 29.9 30.0- 39.9 40.0 -49.9 > 50
Sites
Stem densities per ha
s
46
38
84
Musebeya
19
184
88
64
30
Nshili
10
60
62
60
56
32
62
Kagano
10
142
90
32
20
Mudasomwa
154
58
13
88
34
42
36
Busozo
12
194
80
s = number of species With a frequency greater than five.
Kolmogorov - Smimov tests were applied to the data to test the differences in the
distributions
of stem densities between pairs of sites as these tests are more
appropriate than the chi-square test when the number of cases is small and when the
expected frequencies at each category are also small (Bryman and Cramer, 1997; Zar,
1999). The results of the analysis are given in Table 3. 22 below.
Table 3.22. Kolmogorov - Smirnov tests comparing distributions
different sizes in pairs of sites
Sites
Musebeya vs. N shili
Musebeya vs. Kagano
Musebeya vs. Mudasomwa
Musebeya vs. Busozo
Nshili vs. Kagano
Nshili vs. Mudasomwa
Nshili vs. Busozo
Kagano vs. Mudasomwa
Kagano vs. Busozo
Mudasomwa vs. Busozo
Most extremes differences
Absolute
Positive
Negative
-0.60
0.60
0.20
-0.20
0.20
0.20
-0.40
0.40
0.00
0.40
-0.40
0.20
-0.20
0.40
0.40
-0.40
0.40
0.40
-0.40
0.40
0.40
0.20
-0.20
0.20
-0.40
0.40
0.20
-0.20
0.40
0.40
of trees of
K-SZ
0.95*
0.32
0.63*
0.63*
0.63*
0.63*
0.63*
0.32
0.63*
0.63*
2-tailed p
0.33
1.00
0.82
0.82
0.82
0.82
0.82
1.00
0.82
0.82
K-S Z: Kolmogorov - Smrrnov Z
* indicates
significance at p-value = 0.05.
The tests indicated that there was no difference in the distribution of stem densities
between Musebeya
and Kagano, and between Kagano and Mudasomwa.
The
observed differences in the distributions of trees of many pairs of sites illustrated
differences in the two distributions.
This could result from some species being
77 http://scholar.sun.ac.za
Stellenbosch University
represented by more trees than expected at some diameter size classes, to some
species that do not grow up to larger sizes or to the effect of selective logging of
suitable species and sizes for timber or poles.
The analysis of distribution indicated that some tree species of different sizes were
evenly distributed among two or more sites. Table 3.23 compares the distribution of
the species among the study sites. Carapa grandiflora, Casearia runssorica, Maesa
lanceolata, and Cleistanthus polystachyus were found in all the five study sites, but
they were unevenly represented among the diameter size classes.
Among the tree
species that occurred in all the study sites, the size-class distribution was even for
Myrianthus holstii and Cassipourea ruwenzoriensis.
Table 3.23. Chi-square analysis to investigate the distribution of species among
the study sites and within diameter size classes.
The differences in degrees of freedom were due to species that were not common to all the study sites
and
sneci Wit
. h no representative
. III
. dilVIiduals in some diameter size-c
.
Iasses.
n to species
?
Occurrence
Tree species
DBH size
n
d.f.
Xclasses
1,2,3,4,5
A,B,C,D,E
Carapa grandiflora
126
16
32.53**
1,2,3,4,5
Casearia runssorica
A,B,C,D,E
62
16
29.43*
1,2,3,4,5
A,B,C,D,E
Cassipourea ruwenzoriensis
45
12
8.67
A,B,C,D,E
1,2,4,5
Chrysophyllum gorungosanum
28
12
9.99
Cleistanthus polystachyus
20.78**
A,B,C,D,E
1,3,5
68
8
Cremaspora trifolia
1,3,4
21
1.40
A,B
2
Diospyros gabonensis
17
0.62
A,B,C,D
3,4
3
Grewia mildbraedii
A,C,D,E
1,4
Il
3
0.24
A,B,C,D
1,5
Maesa lanceolata
20
20.00***
3
Myrianthus holstii
A,B,C,D
1,2,3,4,5
68
12
11.32
Newtonia buchananii
18
1
0.07
A,E
3,4
Oxyanthus troupinii
A,B,C
1,4
19
2
0.28
Parinari excelsa
28
14.16
A,B,C,D,E
2,3,4
8
Strombosia scheffleri
1,2,3,4,5
217
16
52.12***
A,B,C,D,E
Symphonia globulifera
A,B,C,D,E
1,2,4,5
44
12
13.04
* Significant at p<0.05, ** significant at p<O.OI, *** significant at p<O.OOI
1 = Musebeya, 2 = Nshili; 3 = Kagano; 4 = Mudasomwa; 5 = Busozo.
A = 10.0 - 19.9 cm; B = 20.0 - 29.9 cm; C = 30 - 39.9 cm; 0 = 40.0 - 49.9 cm; E = ~ 50 cm
3.2.4
Presence of preferred tree species in the study area
During the PRA, the respondents identified different tree species for different uses. A
minimum frequency of 10responses
was used as a guiding criterion for narrowing
down the list of preferred tree species. The forest resource assessment established that
not all woody resources mentioned as being used by the respondents were found in
the Forest. Only 12 (i.e. 85.7%) of the 14 most preferred timber species were found
as large trees and 12 (i.e. 75%) of the 16 preferred for pole species were found as
78
Stellenbosch University http://scholar.sun.ac.za
medium trees. Some of the most preferred timber tree species that were not found
included Faurea saligna and Ocotea usambarensis. Among the tree species preferred
as building poles, Oxyanthus speciosus, Faurea saligna and Milletia sp. were not
encountered.
Of the 45 most preferred medicinal trees, only 17 species were
recorded. Some medicinal plants mentioned during PRA surveys were not found at
all during the forest resource assessment, including Prunus africana and Ocotea
usambarensis.
Of the six tree species preferred for wooden goods, only three were
found in the sample area. These were Po/yscias fulva, Dasylepsis racemosa and
Carapa grandiflora. Tables 3.24a and 3.24b show the most preferred tree species
identified in the sample area.
For the whole sample area, 54 species including the most preferred tree species for
timber, poles, wooden items and medicinal materials, were recorded as large and
medium trees. Based on the total number of individuals recorded in the sample area
(1097 large trees and 995 medium trees), tree species occurring 10 or fewer times
were defmed as 'rare species', those occurring 10 - 40 times as 'average species' and
41 times or more (nearly twice the average number of trees) as 'common species'.
Tables 3.24a and 3.24b and Figure 3.14 show the occurrence categories of the tree
species recorded in all the study sites. For timber tree species, five species were
classified as rare. Tree species with average occurrence were four. Timber tree
species of common occurrence were Strombosia scheffleri, Carapa grandiflora and
Symphonia globulifera. Of the 16 medicinal tree species, rare tree species in the large
tree category were six (37.5%) whereas rare ones recorded as medium trees were
nine. Five medicinal tree species had 'average occurrence' .
...
.~
10
.,c,
8
'"
6
..0
4
""'...0
.,
El
='
Z
2
0
Average
Rare
Abundance
[] Timber
• Medicinal. large trees
Common
classes
• Poles
• Medicinal, medium trees
Figure 3.14. Occurrence categories of preferred tree species in the sample area
Stellenbosch University
79 http://scholar.sun.ac.za
Table 3.24a. List of preferred
timber and pole species and their occurrence
Tree species
Timber
Carapa grandiflora
Dasy/epsis racemosa
Entandrophragma exce/sum
Hagenia abyssinica
Newtonia buchananii
Podocarpus /atifolius
Parinari exce/sa
Podocarpus fa/catus
Po/yscias ju/ya
Strombosia scheffleri
Symphonia g/obulifera
Syzygium parvifolium
categories
in the study area
Occurrence value
Occurrence category
126
4
18
1
12
2
35
3
8
217
47
21
Common
Rare
Average
Rare
Average
Rare
Average
Rare
Rare
Common
Common
Average
1
82
47
2
76
17
5
13
3
41
Rare
Common
Common
Rare
Common
Average
Rare
Average
Rare
Common
Common
Average
Poles
Alchornea hirtella
Carapa grandiflora
Casearia runssorica
Cassipourea gummiflua
Cassipourea ruwenzoriensis
Macaranga neomildbraediana
Olea hochstetteri
Parinari excelsa
Podocarpus latifolius
Strombosia scheffleri
Symphonia globulifera
Syzygium parvifolium
Table 3.24b. List of preferred
99
39
medicinal species and their occurrence
Tree species
Medicinal sl!ecies, lar~e tree cate~orl:
Bridelia bride/iifolia
Canthium oligocarpum
Carapa grandiflora
Entandrophragma exce/sum
Maesa lanceo/ata
Neoboutonia macroca/yx
Newtonia buchananii
Olea hochstetteri
Oxyanthus troupinii
Parinari exce/sa
Pauridiantha paucinervis
Rapanea melanoph/oes
Symphonia globulifera
Syzygium parvifolium
Tabernamontana johnstonii
Zanthoxyjum gil/etii
Medicinal sl!ecies, medium tree cate~on:
Bridelia brideliifolia
Canthium oligocarpum
Carapa grandiflora
Entandrophragma exce/sum
Maesa /anceolata
Neoboutonia macroca/yx
Newtonia buchananii
Olea hochstetteri
Oxyanthus troupinii
Parinari exce/sa
Pauridiantha paucinervis
Rapanea me/anoph/oes
Symphonia g/obulifera
Syzygium parvifolium
Tabernamontana johns toni i
Zanthoxylum usambarense
Zanthoxy/um gil/etii
categories
in the study area.
Occurrence value
Occurrence category
1
4
126
18
21
12
12
1
20
35
13
9
47
21
19
2
Rare
Rare
Common
Average
Average
Average
Average
Rare
Average
Average
Average
Rare
Common
Average
Average
Rare
1
21
82
5
Rare
Average
Common
Rare
Rare
Rare
Average
Rare
Common
Average
Average
Rare
Common
Average
Rare
Rare
Rare
I
1
32
5
79
13
II
3
99
39
5
3
3
Stellenbosch University
80 http://scholar.sun.ac.za
Preferred tree species for building poles were identified among the species in the
medium tree category and consisted of trees of2.5 - 9.9 cm OBH. Their distribution
into occurrence categories revealed that Alchornea hirtella, Podocarpus latifolius and
Olea hochstetteri were rare.
Syzygium parvifolium, Macaranga neomildbraediana
and Parinari excelsa had average occurrence while at least five of the tree species
preferred for building poles occurred frequently in the sample area and were defined
as 'common
species'.
They included Carapa grandiflora,
Casearia runssorica,
Symphonia globulifera and Cassipourea ruwenzoriensis.
3.2.5
Abundance of the most preferred tree species in the sample area
The stem densities of the preferred tree species were calculated relative to the stem
densities of all species per hectare. The stem densities ranged from 0.05 to 7.59 stems
per hectare for timber tree species and 1.35 to 5.77 stems per hectare for poles. For
medicinal trees, the stem densities ranged from 0.59 to 3.52 and 0.72 to 4.05 stems
per hectare for the preferred tree species recorded as large and medium trees,
respectively.
Tables 3.25a to 3.25c show the relative densities of the preferred tree
species by tree categories.
Table 3.25a. Relative densities of the preferred timber tree species.
Tree species
Strombosia schejJleri
Carapa grandiflora
Symphonia globulifera
Parinari excelsa
Newtonia buchananii
Syzygium parvifolium
Entandrophragma excelsum
Polyscias fulva
Dasylepsis racemosa
Podocarpus falcatus
Podocarpus latifolius
Hagenia abyssinica
Mean stem density
Standard error
Confidence level (95%)
Stem density
Relative stem
per ha
density
86.8
19.77
11.48
50.4
4.28
18.8
14
3.19
8.4
1.91
8.4
1.91
7.2
1.64
3.2
0.73
0.36
1.6
0.27
1.2
0.8
0.18
0.09
0.4
3.82
1.71
3.77
81 http://scholar.sun.ac.za
Stellenbosch University
Table 3.25b. Relative densities of the preferred pole tree species.
Tree species
Symphonia globulifera
Carapa grandiflora
Cassipourea ruwenzoriensis
Casearia runssorica
Strombosia scheffleri
Syzygium parvifolium
Macaranga neomildbraediana
Parinari excelsa
Olea hochstetteri
Podocarpus latifolius
Alchornea hirtellaf glabrata
Cassipourea gummiflua
Mean stem density
Standard error
Confidence level (95%)
Stem density Relative stem
per ha
density
140.03
9.95
115.98
8.24
107.50
7.64
66.48
4.72
57.99
4.12
55.16
3.92
24.05
1.71
18.39
1.31
7.07
0.50
4.24
0.30
1.41
0.10
2.83
0.20
3.56
1.00
2.21
Table 3.25c. Relative densities of the preferred medicinal tree species.
Medium trees
Stem
density/ha
Symphonia globulifera
140.03
Carapa grandiflora
115.98
Oxyanthus troupinii
111.74
Syzygium parvifolium
55.16
Newtonia buchananii
45.26
Canthium oligocarpum
29.70
Parinari excelsa
18.39
Pauridiantha paucinervis
15.56
Entandrophragma excelsum 7.07
Olea hochstetteri
7.07
Tabernamontana johnstonii 7.07
Rapanea melanophloeios
4.24
Zanthoxyllum gilletii
4.24
Zanthoxyllum usambarense 4.24
Bridelia brideliifolia
1.41
Maesa lanceolata
1.41
Neoboutonia macrocalyx
1.41
Mean stem density
Standard error
Confidence level (95%)
Tree species
Relative
stem density
9.95
8.24
7.94
3.92
3.22
2.11
1.31
1.11
0.50
0.50
0.50
0.30
0.30
0.30
0.10
0.10
0.10
2.38
0.78
1.66
Large trees
Stem
densit}:/ha
18.80
50.40
8.00
8.40
8.40
1.60
14.00
5.20
7.20
0.40
7.60
3.60
0.80
Relative
stem density
4.28
11.49
1.82
1.91
1.91
0.36
3.19
1.19
1.64
0.09
1.73
0.82
0.18
0.40
8.40
4.80
0.09
1.91
1.09
2.11
0.69
1.46
Using a density of equal or greater than 1.0 stems per hectare (at the lower limit of the
95% confidence interval) as a cut off limit for harvestable
timber tree species
Stellenbosch University
82 http://scholar.sun.ac.za
following Davis and Johnson (1987), there were seven tree species which could be
considered abundant enough for utilisation.
Of these species, commonly available
tree species were Strombosia scheffleri and Carapa grandiflora.
Among medicinal tree species, there were 11 species that were considered abundant
enough to allow collection
of medicinal
materials
from large trees.
Carapa
grandiflora, Symphonia globulifera and Parinari excelsa were the most abundant
medicinal tree species in the large tree category.
Within the medium tree category, only eight medicinal tree species were abundant
enough to be listed as harvestable species.
Carapa grandiflora
and Symphonia
globulifera were predominant in the Forest and were classified among the abundant
medicinal tree species.
Finally, seven tree species of the medium tree category were abundant for potential
supply of poles.
The five most abundant were, in decreasing
order of their
abundance, Symphonia globulifera, Carapa grandiflora, Cassipourea ruwenzoriensis,
Casearia runssorica and Strombosia scheffleri.
3.2.6
Dominance of the most preferred tree species
All abundant preferred timber species had a relative dominance greater or equal to
0.01 and contributed more to the forest standing biomass. Of the most abundant and
preferred pole species, the only tree species which showed a relative dominance of
less than 0.01 was Macaranga neomildbraediana.
However, many of the medicinal
tree species represented by large trees showed a low contribution to the total biomass
of the Forest.
Tree species with sufficient contribution owing to their relatively
higher densities of their stems included Carapa grandiflora, Syzygium parvifolium,
Parinari excelsa, Symphonia globulifera and Newtonia buchananii.
The medicinal
tree species of the medium tree category contributed sufficiently to the standing
biomass, with the exception of Canthium oligocarpum.
The relative dominance of the
species in the large tree category is presented in Appendix 10.
83 http://scholar.sun.ac.za
Stellenbosch University
3.2.7
Size-class distribution and regeneration of most abundant preferred tree
species.
The distribution of stem densities into different diameter size classes differed among
the preferred tree species (Figure 3.15). Of the 11 species, only Carapa grandiflora
and Symphonia globulifera approximated the 'inverse J' curve that is characteristic of
a mature and stable system or a population that is self-perpetuating (Bazzaz, 1998;
Luken, 1990).
regeneration.
This suggests that there were shade-tolerant species with abundant
For some species including Cassipourea ruwenzoriensis,
Parinari
excelsa, Casearia runssorica, Oxyanthus troupinii and Syzygium parvifolium,
the
infrequency of seedling establishment caused a reduction in the smaller size class,
followed directly with a greater number of saplings or individuals of 2.5 - 9.9 cm
DBH.
Similarly, Newtonia buchananii and Pauridiantha paucinervis
showed no
overall pattern, with intermediate classes that contain no individuals at all.
This
suggests that the existing level of saplings were insufficient to re-stock these classes.
The shape of the curve of Entandrophragma
excelsum was also irregular, with the
lack of regeneration among the smaller size class and the presence of saplings and few
large trees in other diameter size classes. Generally, E. excelsum was found at lower
density compared to other species. It can be seen from the curves of E. excelsum and
Parinari excelsa that their size-class distributions exhibit relatively more large trees at
the largest size class (i.e. z50 cm DBH) than those found at the intermediate sizes.
This could indicate temporarily variable recruitment in the stages oftheir life history.
The observed pattern in the size-class distribution of Oxyanthus troupinii represents a
population that has very few large trees and relatively abundant seedlings and
saplings.
This pattern illustrates a population
of a long-lived tree species that
experienced a period of very poor recruitment or past intense harvesting of large trees.
Figure 3.16 shows that the regeneration of the species is very variable across the study
sites.
The regeneration
appears to be least at Nshili and Mudasomwa
for some
preferred tree species. Only Symphonia globulifera produced seedlings and saplings
in all five study sites while seedlings of Newtonia buchananii were only recorded in
the part of the Forest next to Kagano Commune.
Stellenbosch University
84 http://scholar.sun.ac.za
Carapa
1000
800
600
400
200
oE
q
il,i
1 I
J,
o
ABC
DBH
Cassipourea
oE
500
.~
400
c
300
~ 200
~
I O~
C. runssorica
grandiflora
I
T
D
E
size classes,
T
T
F
G
T
T
T
D
E
F
G
size classes,
cm
Newtonia
ABC
E
F
size classes,
,.I,L
C
o
DBH
J:...__"r'.L
.... ,'~J:......_T'~.L"'_r, ..J:..._,
E
F
size classes,
cm
ABC
D
troupinii
500
:g 300
~ 200
E 100
~
(/) 0
I
I
D
E
F
size classes,
cm
ABC
DBH
Pauridiantha
.I.
I
F
size classes,
cm
I
G
paucinervis
, 1. , .1
..
G
A
B
DBH
Symphonia
scheffleri
G
400
cm
E
bosia
excelsum
j
G
excelsa
B
Strom
I
, .I.
G
Oxyanthus
D
Parinari
F
cm
D
DBH
T
DBH
A
40
buchananii
•
E
DB H size classes,
f : l-+l----r, ...1--., ...
oE
q
,
.I
Entandrophragma
.I , .I
DBH
T
ABC
cm
oE
ABC
240
:g 180
~
120
E
60
s
(/)
0
r u w e n z o r ie n s is
i, ,
300
C
D
I
E
F
size classes,
cm
G
g lo b u life r a
1000
,
.-
.- , ~ ,
.L
, .L
E
F
G
size classes,
cm
,
ABC
D
DBH
llHI,I"
T
D
T
T
E
F
G
size classes,
cm
ABC
DBH
T
Sy zy g iu m p a rv ifo liu m
T
A
B
DBH
T
I
eDE
size classes,
A = <2.5 em; B = 2.5-9.9cm; C
F = 40.0 - 49.9cm; G = ;;::50cm
=
T
T
F
G
cm
10.0-19.9cm; 0 = 20.0 - 29.9cm; E
=
30.0-39.9cm;
Figure 3.15. Size-class distribution of the most abundant and dominant trees
preferred for timber, poles and medicines.
Stellenbosch University
85 http://scholar.sun.ac.za
Carapa
:!?
-:c
~
Casearia
grandiflora
;J<
;: 30
c
" 25
"0':> 20
" 15
<!:
10
::" 5
1ii
0;
0
40
30
g- 20
:>
<!::
.e"
'"
~
runssorica
10
0
A
B
D
C
c.::
E
A
B
Sites
Cassipourea
:!?
ruwenzoriensis
Newtonia
30
25
20
ec
u
:>
:!?
~
g-
g- IS
u
10
.e 5
Ii 0
~
<!::
u
.~
r-l.-,
A
B
Oxyanthus
u
:>
0-
u
<!::
u
.~
1ii
"ii
et:
D
C
~
et:
E
D
E
D
E
30
20
IS
10
5
0
A
B
Parinari
troupinii
;t.
~
c
s
0-
u
<!::
u
.~
ee
l,
B
~
30
25
20
15
10
5
0
E
D
C
excelsa
A
B
C
Sites
Sites
Pauridiantha
C
Sites
30
25
20
15
10
5
0
A
Strombosia
paucinervis
;t.
;t.
~
~
e
30
25
sg. 20
u
15
<!:
10
u
.e
5
1ii
"ii
0
et:
scheffleri
JOL
25
20
u
15
<!::
u
10
.<:
5
u
:>
0-
~
o
et:
A
B
A
E
D
C
Symphonia
B
C
D
E
Sites
Syzygium
globulifera
;t.
;t.
30
25
~
t:
s 20
u
15
<!::
10
.e"
5
1ii
~ 0
parvifolium
30
-: 25
"
~
gt:
20
IS
10
"<:
5
'" 0
0;
et:
0-
<!::
A
B
D
C
A
E
B
= Nshili;
B
C
D
E
Sites
Sites
= Musebeya;
II
...
Sites
A
E
buchananii
Sites
~
c
D
-:o 25
c
<!::
;t.
C
Sites
C
= Kagano;
D
=
Mudasomwa; E
=
Busozo
Figure 3.16. Relative abundance of seedlings and saplings of the abundant and
dominant tree species across the five sites.
Stellenbosch University
86 http://scholar.sun.ac.za
3.2.8
Quality of harvestable tree species for timber and poles
Figures 3.17 and 3.18 show the quality of harvestable trees for timber and poles.
Figure 3.17 shows that the proportion of good quality stems suitable for timber was
higher for Symphonia globulifera, Entandrophragma excelsum, Parinari excelsa, and
Strombosia scheffleri.
Boles of these tree species were cylindrical in form, from the
ground to the first branch, and therefore were good enough for use as sawn timber.
Carapa grandiflora and Newtonia buchananii had irregular taper and protuberances
and could only be used with a considerable
proportion
of waste.
Syzygium
parvifolium, for which many stems ranged from fair to acceptable, would be better
used for a product where stem straightness is not a major requirement.
preferred tree species for building poles, Symphonia globulifera
quality stems, followed by Cassipourea ruwenzoriensis
Of the most
produced many
(Figure 3.18). However, less
than 20% of the individuals of Strombosia scheffleri were of good quality stems.
87
Stellenbosch University http://scholar.sun.ac.za
100%
<Il
80%
Il)
~
.......
0
c:
0
Ot
0
0-
60%
40%
e
0-
20%
0%
Species
Figure 3.17. Proportion of trees with stem suitable for timber.
100%
<Il
Il)
80%
~
.......
0
60%
s::
0
Ot
0
00
0-
...
40%
20%
0%
Species
• Good
0 Fair • Acceptable
Figure 3.18. Proportion of trees with stem suitable for building poles.
88 http://scholar.sun.ac.za
Stellenbosch University
4
DISCUSSION
In this chapter, both the results obtained from the household survey and from the
forest resource assessment are discussed.
The participatory assessment was carried
out as a preliminary study aimed at collecting information on resource preferences
and uses and the perceptions of local people regarding the appropriate institutions for
the management of utilisation in Nyungwe Forest.
Participatory
techniques
are
commonly used to strengthen any management decisions. For example, Scott (1992)
used PRA techniques to collect information on which to base recommendations
multiple use activities at Bwindi Impenetrable National Park in Uganda.
for
In the
resource assessment, the areas of the Forest next to the Communes surveyed were
assessed for their potential to provide the preferred tree products to the communities.
Cunningham (1996) carried out a similar forest survey prior to recommending use of
resources at Mgahinga Gorilla National Park in Uganda.
Preferred tree products from Nyungwe Forest are discussed in relation to their
availability, abundance,
size-class distribution and regeneration.
The discussion
focuses on the tree species most preferred for timber, poles and medicinal materials as
identified during the forest assessment.
Other uses of the Forest are also discussed,
including the production of wooden goods, fuelwood, charcoal, beekeeping, fodder,
honey, bean stakes and land. A focus on these products is justified by their economic
potential and the role they play in the livelihoods of forest-dependent
communities
(Neumann and Hirsch, 2000; Prasaad, 1999). Further, having discussed the resources
available in Nyungwe Forest, the long-term potential of the Forest to provide these
resources is presented together with recommendations
on the management of the
Forest for utilisation.
4.1
4.1.1
RESOURCE
PREFERENCES,
USES AND AVAILABILITY IN NYUNGWE FOREST
Wood products
4.1.1.1 Timber
The survey established that 29 tree species were preferred for use as timber, of which
the most sought after were Podocarpus latifolius, P. falcatus and Entandrophragma
89 http://scholar.sun.ac.za
Stellenbosch University
excelsum.
These timber trees were found in low densities in the Forest, suggesting
that they were targeted for pitsawying.
selective and unmanaged,
Logging of valuable trees has been very
and it has resulted in some tree species becoming
endangered (Gapusi and Mugunga, 1997; Habiyambere, 1999; Sorg, 1978).
As an alternative to the loss of access to indigenous timber, the respondents were
found to manage fast growing exotic timber species to cope with the short-term
demand for furniture and construction materials or else to generate income needed by
the households.
The lack of indigenous timber tree species on the respondents' farms
may reflect past social forestry programmes that emphasised the planting of exotic
species, without a focus on the indigenous tree species preferred by the local people.
Illegal logging for indigenous timber by neighbouring communities has been reported
to be a major threat to the conservation of the Forest (MINITERRE, 2000). Since the
local communities
favour certain species for timber, harvesting is likely to be
selective. Selective logging affects mature forests and species composition, as a result
of canopy gap formation (Howard, 1991).
distributions
In addition, logging results in patchy
of both the species and size-class categories
harvested (Cunningham,
of trees and shrubs
1996). Timber species in the Forest showed this pattern,
with a limited abundance of marketable sizes of timber tree species.
These species
were found at low frequencies, with fewer harvestable sizes and an increasing number
of young trees. The small number of large sized hardwoods was probably a result of
overexploitation.
Sustainable harvesting of timber is based on the methods of regeneration, silvicultural
systems and yield regulation (Geldenhuys,
1998; ITTO, 1990).
The ecological
management of natural forest emphasises the extraction of resources that show a low
impact of removal. Selective logging is thought to be a viable option if it is practised
outside the conservation
area of a forest (Struhsaker,
1997).
Therefore,
a
management system could be developed for sustainable logging of timber over 10%
of the total area of Nyungwe Forest as was established by a management
developed
in 1988 (Fimbel and Kristensen,
effectively implemented.
1994; MINAGRI,
plan
1984) but not
Stellenbosch University
90 http://scholar.sun.ac.za
4.1.1.2
Building poles
Nyungwe Forest is surrounded by human settlements.
collected building poles from the Forest.
In the past, local people had
In the study area, managed woodlots are
source of building materials. This is in common with other parts of the country where
communities have managed trees for various uses, including small to medium sized
poles which can be sold (den Biggelaar, 1996).
The loss of access to poles and other wood products from the Forest is compensated
by tree planting on farms (Olson et al., 1995). Ample quantities of poles appeared to
be cultivated, but indigenous poles from Nyungwe Forest were apparently more
preferred for their straightness and durability.
From field observation, many houses
were found to be built from cultivated Eucalyptus poles but in the south-eastern zone
bamboo from the Forest was also used, with the use of the species decreasing with
distance away from the edge of the Forest. The common use of cultivated poles in the
study area was a result of the development of community nurseries and the supply of
seedlings to local people by past donor-funded
forest projects in order to reduce
dependence on the Forest (MINAGRI, 1990).
There are no available data on the consumption of indigenous poles. Due to the high
population density (350 people/km') and the number of households around the Forest,
the demand for poles is expected to be very high, leading to increased illegal logging
for poles.
Programmes of tree planting on farms with suitable indigenous species
could partly alleviate illegal cutting of poles from the Forest in the long-term.
4.1.1.3
Fuelwood and charcoal
In common with other rural areas in Rwanda, fuelwood provides the major source of
household
energy (MINECOFIN,
1998).
Due to its high conservation
status,
fuelwood is the only commodity that a few people access from the buffer zone of
Nyungwe Forest.
Similar to poles, there have been attempts by foresters to provide
fuelwood to adjacent communities to the Forest by encouraging tree planting on farms
(Olson et al., 1995). The production of fuelwood for household consumption was the
second priority next to pole production in the survey area.
91 http://scholar.sun.ac.za
Stellenbosch University
Charcoal production was considered as an income generating activity in the study area
around the Forest.
woodlots.
The respondents produced charcoal from their own Eucalyptus
Despite the availability of trees in farmers' fields, charcoal making has
been reported as a threat to the conservation of the Forest (MINITERRE, 2000). This
implies that there is a demand for charcoal from indigenous trees. Other reasons for
exploitation of wood for charcoal from the Forest may be related to the shortage of
land, which prevents some households from establishing woodlots.
charcoal in cities also provides
an economic
The demand for
incentive that encourages
some
individuals to encroach on the Forest.
The annual increase in the rate of fuelwood consumption in Rwanda is estimated at
2%, translating to an annual increase of three million m3 per year, exceeding the limits
of sustainable exploitation (Habiyambere, 1999). With an estimated forest cover of
7% of the country's area and the increasing demand for fuelwood and charcoal, the
survival of the remnant indigenous forests is uncertain (Mitchell, 1997). It will likely
depend on the development of suitable alternatives such as the use of biogas, the use
of efficient cooking stoves and an active programme of community
forestry in
Rwanda (GEC, 2000; Percival and Dixon, 1995)
4.1.1.4
Wooden items
Households adjacent to Nyungwe Forest rely on locally produced materials.
The
local producers are highly selective of species since certain attributes are required of
the wood for specific purposes.
The most prominent characteristics are flexibility,
durability, strength and resistance to splitting.
The selective harvesting of these
species can result in genetic erosion of the species as vigorous and quality individuals
are removed from the population (Cunningham, 1996). The current rarity of these
species could be partly explained by the effect of illegal harvesting.
Bamboo is used by local people in the south-eastern zone of Nyungwe Forest where it
is predominant.
Considerable quantities of materials made of bamboo are sold both
locally and outside the area, or used for building purposes.
The major incentive for
the unlawful cutting of bamboo from the Forest is the economic advantages for
individuals to benefit from the sale of goods. The use of bamboo for commercial and
Stellenbosch University
92 http://scholar.sun.ac.za
subsistence needs might result in overexploitation if not regulated or cultivated on
farms for alternative sources of basketry and granary materials (Butynski, 1984).
Bamboo can easily grow in people's gardens as shown by Cunningham (1996) as part
of the management of Bwindi Impenetrable National Park in Uganda.
Other tree
species including Markhamia lutea and Polyscias fulva could receive attention in the
development of agroforestry in Rwanda since they are indigenous species traditionally
planted or conserved on farms (Niang and Styger, 1990) and valued for the making of
various wooden items.
4.1.1.5 Bean stakes
Beans, both bush and climbing varieties, are among the staple foods produced in the
zone surrounding the Forest. In the survey area, climbing beans were preferred for
their high yield and resistance to heavy rains during the growing seasons. Bean stakes
are essential for climbing bean production, and huge quantities of bean stakes are
required. In the past, stakes were a common product obtained from Nyungwe Forest
by neighbouring farmers.
The most favoured tree species was Alchornea hirtella,
which is presumably the reason for its low frequencies in the study area.
Many farmers in the area use branches of Pinus patula, Eucalyptus spp. and Grevillea
robusta from their own farms or from the buffer zone for Nyungwe Forest as bean
stakes. Far from the edge of the Forest, the shortage of adequate staking materials is
reflected by the use of Pennisetum purpureum and stalks of sorghum and maize. As
the population depends on agriculture for its subsistence, the demand for bean stakes
is expected to increase. The demand cannot be met from the supplies available within
the agricultural systems. It is recommended that in addition to assessing ways for the
community to use bean stakes from Nyungwe Forest, development
programmes
should encourage the local people to cultivate their own supplies of bean stakes by
establishing fast growing shrubs including Mimosa scabrella and Chamaecytisus
palmensis identified by the Agroforestry Research in Rwanda to be suited on local
soils and climate. Other species with high coppicing abilities could be introduced for
the production of large quantities of stakes. For instance, communities surrounding
Bwindi Forest in Uganda use bamboo for stakes (Wild and Mutebi, 1996).
Stellenbosch University
93 http://scholar.sun.ac.za
4.1.2
4.1.2.1
Non-wood products
Land
Nyungwe Forest forms a physical barrier between two agro-ecological regions, with
different farming systems found on either side (Renner, 1991). The sample area falls
within the two agro-ecological zones, and so the farming conditions of the different
groups of respondents vary.
These differences were reflected in farm sizes, crop
types and soil fertility.
The decreasing availability of new farmland means the population around the Forest
is unable to support itself from agriculture.
Problems of soil infertility result from
acidic soils. These soils are severely degraded, which leads to low crop production
and increased poverty (Olson, 1994). The communities adjacent to Nyungwe Forest
lack access to production
resources including
livestock for manure production,
fertilisers, and other means to manage soil fertility.
This result supports the finding
by Blaikie and Brookfield (1987) that access to production resources is felt differently
between households within the same region depending on their ability to afford
inputs.
In the past, adjacent communities used land outside and inside Nyungwe Forest in an
extensive manner (Olson et al., 1995) until they were relocated without compensation
onto land outside the Forest (Renner, 1992).
With the current process of land
acquisition through inheritance, the farm size is unable to meet the subsistence needs
of the local population.
There is also a lack of alternatives to agriculture, including
diversified sources of income.
opportunities
Newbury (1992) attributed the lack of off-farm
in the southwest of Rwanda to the past government
regimes that
concentrated government funds on the development of their home region".
Local perception of Nyungwe Forest as a potential land resource could therefore be
attributed to the historical background, including the influx of farmers looking for
2 Newbury (1992) argued that the southwest of Rwanda had fewer opportunities than other areas as the
former President Habya/imana Juvenal and the people he selected for important government posts
concentratedfunds in the North of the country where they originated.
Stellenbosch University
94 http://scholar.sun.ac.za
fertile lands to the eastern edge of the Forest (Olson, 1990; Cambrezy, 1984). The
preference for farmland from within the Forest stems from the past practice of
clearing the Forest for food production (Barabwiliza,
1980).
Overuse of forest
products would result in loss of biodiversity in the long-term, but clearance for
agriculture is the greatest threat (MINAGRI, 2000; Mitchell, 1997). This pressure is
driven by poverty and a growing population in adjacent areas.
Subsistence
agriculture
was the major livelihood
strategy amongst the survey
population.
Revenues from crops are relatively low, due to low prices given to
agricultural produce (Ben Chabanne and Cyiza, 1992; World Bank, 1993). Income
accrued by people around Nyungwe Forest can not meet their economic and financial
requirements, resulting in an increased dependence on natural resources.
Unless
adequate strategies are adopted, increased land for agriculture and livestock farming
will continue to be a major concern for local communities.
4.1.2.2 Gold mining
Local villagers have high expectations of earning cash income (Barakabuye, 2001)
and therefore illegal mining is considered to be a major law enforcement
and
conservation problem (MINITERRE, 2000). Mining in defined zones was listed by
MINAGRI (1993) with the priority activities that were considered for authorisation to
organised groups. Recognising access of local groups to mineral reserves can help in
improving incomes to local households.
Habimana (1982) found that 28% and 10%
of the sample population in Mudasomwa and Kivu Communes next to the Forest
obtained cash income
from gold mining, respectively.
Although
the sample
population was too small to make inference about the large adjacent population as a
whole, it is assumed that access to minerals in the Forest could be an alternative
source of income to the neighbouring communities.
Due to the negative impact of
mining on the conservation of biodiversity, it is necessary to define and demarcate
mining areas and to monitor activities by miners' groups.
95 http://scholar.sun.ac.za
Stellenbosch University
4.1.2.3 Wildlife and ecotourism
In the past, neighbouring communities extracted bushmeat from Nyungwe Forest
through hunting and the use of traps (Storz, 1983).
The survey established that
wildlife is not managed for utilisation as part of the livelihoods of the people adjacent
to the Forest.
Hunting for household consumption is regarded as a threat to the
conservation of wildlife, including rare and protected animals such as chimpanzees
and buffaloes (MINITERRE, 2000). The value of the resource in money terms cannot
be estimated without an in-depth survey, as off-take from the Forest is illegal and
bushmeat trade said to be non-existent.
In Equatorial Africa, hunting and bushmeat trade occupy a central place in the
economy, and the volumes traded are substantial.
For instance, the annual bushmeat
trade in Gabon is valued at US $ 22 million on informal markets and at US $ 3 million
on formal markets (Bailon 1996, quoted by Klein and van der Wall, 1998). In the
Republic of Congo, 26 000 animals are said to be sold in Pointe Noire monthly,
implying that half a million animals are killed for this market each year (Wilson and
Wilson, quoted in Colchester, 1994). In Cameroon and in the Democratic Republic of
Congo, the bushmeat consumption is estimated at 21 and 41 kg per person per year
(Wilkie and Carpenter, 1999).
Wildlife can be managed as a source of subsistence or income.
For example, the
annual income from hunting and trapping in villages around the Korup National Park
in Cameroon amounted to approximately US $ 1050 per hunter and accounted for
50% of the total village income (Infield, 1988).
In Nyungwe Forest, the non-
consumptive use of wildlife through ecotourism is favoured by the government
managing
institution,
ORTPN
(Barakabuye,
2001).
The communities
around
Nyungwe Forest do not enjoy the benefits of ecotourism, apart from the employment
of a few local guides, because all the economic benefits accrue to Government.
In the
year 2000, Nyungwe Forest registered 777 visitors from different foreign countries,
from which US $ 6 722 were obtained by ORTPN (Barakabuye, 2001). This revenue
is low compared with the revenues obtained before 1994.
In the early 1990s,
ecotourism in the Forest raised US $ 15 000 per year, but this amount was only
enough to pay the staff and the upkeep of the Forest (Offutt, 1992). Since all of these
96 http://scholar.sun.ac.za
Stellenbosch University
revenues go to government funds, they have little impact on communities around the
Forest.
When carefully planned and managed, an ecotourism development in a
tropical forest can provide a sustainable return, much of which can remain in the local
economy (Horwich, 1988).
In many countries,
attempts have been made to involve local people in the
management of forest reserves and to create opportunities for local communities to
benefit from the forests.
This has been the case for Budongo Forest in Uganda
(Langoya and Long, 1998), Serengeti National Park in Tanzania (Emerton and
Mfunda, 1999), woodlands in Zimbabwe (Hasler, 1999) and wildlife management in
South Asia (Kothari et al., 2000).
The potential for community involvement
in
wildlife management in Nyungwe Forest exists and it is justified for the following
reasons: •
Local people around the Forest have knowledge of the resources, including rare
and endangered wildlife;
•
They are aware of the value of wildlife as bushmeat and as an ecotourism
attraction;
•
They have knowledge of the reproductive cycles and the habitats of wildlife.
Nevertheless,
with the present condition of poverty and absence of alternative
livelihoods strategies, the sustainability of the wildlife resource is threatened.
It is
essential for the Government to consider approaches to natural resource management
taken by many other governments which offer an immediate value to themselves and
benefits to local communities.
4.1.2.4
Medicinal plants
Medicinal plants are an important element of traditional medicine on which 80% of
the world population currently depends (WHO et al., 1993). Of the 250 000 higher
plants worldwide, one fifth are used as medicinal plants (WWF, 1994). The use of
medicinal plants is one of the most important functions of non-timber forest products
(Fedlmeir,
1998).
Due to its illegal status, medicinal plants used by the local
communities are supposed to be taken from farms.
97 http://scholar.sun.ac.za
Stellenbosch University
Although traditional healers do exist in the region, it appears that most people in the
region treat themselves or seek advice from friends, in particular from elder persons.
No data are availed on the number of traditional healers in Rwanda. With an estimate
of more than 500 medical practitioners who were registered to treat livestock in 1992
(Mbarubukeye, 1992), there is no doubt that traditional healers could number more
than medical doctors. Medical clinics are found in the Communes next to the Forest.
However, they are difficult for some people to reach, lack adequate physicians, and
the drugs prescribed are often too expensive. This explains the dependence on locally
collected medicinal plants for the treatment of various illnesses including headaches,
diarrhoea, and dental cavities.
The leaf material is the major component of medicinal plants commonly collected on
farms. Due to the low diversity of the species available on farms, additional material
could be collected from the Forest.
The majority of medicinal trees found in
Nyungwe forest were preferred for their bark material. The debarking of trees for the
gathering of medicinal material can result in death of the tree resource or its
susceptibility to fungal attack and other diseases.
In Madagascar, for instance, the
collection of the bark of Prunus africana has resulted in its death (Kollert, 1998).
Cunningham and Mbekum (1993) reported that stripping the bark off two opposing
quarters of the trunk of Prunus africana in Cameroon enabled the tree to resist total
injury. Thus the use of improved extraction methods of the bark of medicinal trees is
a means of achieving sustainable utilisation of medicinal trees. Some tree species are
unable to re-grow the bark removed and thus more vulnerable than those that can.
Other factors that affect the sustainable harvesting of medicinal plants include the
amount harvested, the level of commercial trade and the part used (Cunningham,
1996). Since medicinal trees in the Nyungwe Forest could be exploited largely by
debarking, these trees are at high risk of death if the collection methods are not
improved.
Medicinal plants have been shown to be economically important for sustaining local
livelihoods.
At present, in Nyungwe Forest, illegally collected medicinal plants are
used for subsistence needs instead of commercial use. In some countries, however,
medicinal plants are produced in large quantities, sold at local and international
markets and contribute to the national economy.
In Nepal, for instance, official
Stellenbosch University
98 http://scholar.sun.ac.za
records indicated an average of 50 tons per year of aromatic and medicinal plants
commercially exported from the country (Aumeeruddy,
1998).
In Ethiopia, seven
medicinal plants are sold by vendors on local markets (Deffar, 1998). In South Africa
in 1993, the annual trade of 20 000 tonnes of medicinal plants worth US $ 60 million
were annually traded at the national level (Cunningham,
higher today.
1993), and this figure is
In the year 2000, the economic value of medicinal plants in South
Africa approximated a value of US $ 41 million (approx. R 270 million) per annum
(Mander, 1998, quoted by Lawes et al., 2000).
The survey of medicinal plant use showed that local communities have considerable
knowledge about many aspects of medicinal plants, including their abundance, uses
and methods of harvesting.
There is no doubt that medicinal plants cannot be
managed effectively without the co-operation of the local communities, who are in the
best position to protect and manage them.
Although PCFN is seeking ways of
managing the local collection of medicinal plants, agreements are needed between
traditional healers' associations and the Nyungwe Forest authorities to confer certain
rights and responsibilities.
The agreements would define which medicinal plants
should be used, the amounts to be collected, and by whom.
This could result in an
increased interest among local collectors in the protection and conservation of the
resources for their continuing benefits.
4.1.2.5 Fodder plants and pastures
The forest legislation in Rwanda states that livestock must not be grazed in state
owned forests (JORR,
1989).
An exception
has been made for the Acacia
melanoxylon and Pinus patuIa plantations in the buffer zone for Nyungwe Forest,
where cattle, goat and sheep were grazed on the herb layer (Barbier, 1992). In the
survey, evidence was found that livestock is still being grazed in the buffer zone with
the permission of the agricultural extension agents in the area. Available fodder grass
and shrub species mentioned by the respondents are similar to those reported by
Gasana (1988), the most common belonging to the families of Poaceae, Asteraceae,
Cyperaceae and Tiliaceae.
99 http://scholar.sun.ac.za
Stellenbosch University
Livestock keepers lack diversified means of coping with shortages of fodder during
the dry season, the most common solution being free range grazing.
Fodder is not
stored, and currently there is no market available for fodder. The study did not focus
on fodder productivity,
abundance and composition
in Nyungwe Forest.
Field
observation and household interviews revealed that, apart from patches of fodder
along roads and in some gaps inside the Forest, N yungwe Forest was poor in fodder
quality and quantity. This was so because the herb layer is densely occupied by ferns
and lianas, most of which are not eaten by livestock.
The insufficient grazing resources outside the Forest could explain the perception of
the respondents of the Forest as a source of fodder and pasture.
reported cases of uncontrolled
Gapusi (1999)
and excessive grazing in the Forest.
Livestock
population around N yungwe Forest is on the increase owmg to the demand for
manure (from cattle) and income (from goats and sheep). There is a need for a welldefined national policy on grazing to sustain fodder supply by non-forest sources and
by instituting change in livestock composition and grazing practices.
4.1.2.6
Beekeeping
Beekeeping has a low impact on the forest, as has been the case at Bwindi
Impenetrable and Mgahinga Gorilla National Parks of Uganda (Wild and Mutebi,
1996).
Beekeeping
in Nyungwe Forest was controlled by forest projects, and
activities of beekeeper's groups were guided by contracts and regulations (Barbier,
1992). The only concern was possible fires that could be set by the honey collectors.
Due to low production, honey is used for subsistence needs of the household in the
study area, with only small quantities sold on the local market.
In some countries,
large quantities of honey are produced by a considerable number of beekeepers.
In
Malawi, some 8 000 beekeepers produce 1 000 tonnes and 150 tonnes of honey and
beeswax per annum respectively (Chanyenga, 1999). The average export of honey
and beeswax in Ethiopia is estimated at 3.05 tonnes and 270.34 tonnes per year,
respectively (Deffar, 1998).
Stellenbosch University
100 http://scholar.sun.ac.za
The low productivity of honey in Rwanda could be due to the use of traditional
production technologies.
Production can be increased if improved methods, including
the use of modem beehives, are used.
Honey production also depends on the
availability of certain plant species that are associated with bees. Known bee forage
species listed by the respondents were less diverse compared to those found in the
Forest and reported by Habiyambere (1999). Forest disturbances could be one of the
reasons for the loss of tree species favoured by bees. Increasing bee forage species
would certainly boost honey production.
between beekeeper's
Past experiences based on agreement
associations and forest projects could guide decisions over
beekeeping in the Forest for the benefit of the local communities.
4.1.2.7 Wildfoods
Although an assessment of the availability of wild foods has not been conducted in
the Forest, the households adjacent to Nyungwe Forest mentioned a number of
seasonal wild foods that supplement the diet or are used as a source of income. The
ranking of these non-wood products in relation to the other resources showed that
these products were not of great importance to the population.
Practical examples
demonstrate how fruit and mushroom harvesting could be integrated into local forest
management plans and can constitute sources of income for the population.
harvesting of wild foods appears to be sustainable or to have low impact.
The
This is
because:
•
Harvesting of wild vegetables, mushrooms and fruit is a seasonal activity, and
practised by the poorest people. For wild vegetables, edible leaves of Amaranthus
spp. and Solanum nigrum are collected during the rainy season;
•
The use of the fruit of Myrianthus holstii encourages the conservation
of the
female tree (Cunningham, 1996) and the population around Nyungwe Forest has
planted this species on their farms;
•
Farmers in the area produce fruit and vegetables in their gardens (for example:
Amaranthus spp., cabbages, carrots, onions), and thus do not rely entirely on the
foods from the Forest.
Stellenbosch University
101 http://scholar.sun.ac.za
As the gathering of wild foods has relatively little impact on the forest dynamics, this
is one category of resources that the management of Nyungwe Forest could allow
local people to access, as a way of their gaining benefit from the conservation of the
Forest. Mutual trust between parties would be essential here, however, to ensure that
other, restricted, resources are not harvested at the same time.
4.2
CHARACTERISTICS
OF THE PREFERRED TREE SPECIES
Due to time and financial constraints, the study concentrated on the assessment of the
standing stock of the most preferred tree species for timber, poles, and medicinal
plants.
These species were assessed for their density, size-class distribution and
regeneration using plots in areas of the Forest adjacent to local people.
were available
for the population
biology of the preferred
As no data
tree species, their
distribution and responses to exploitation, the maximum sustainable yield could not
be determined (see Lund, 1998; Ruiz Pérez and Arnold, 1996).
The information
provided by this study can only be used to understand the importance of Nyungwe
Forest in the livelihoods of local communities as well as to justify the management
policies of the Forest.
To be able to determine sustainable levels of resource use requires information about
the density, the distribution of the resources within the forest, the population structure,
the productivity of these resources and the ecological impact of different harvesting
levels (Cunningham,
1999).
In this study, plots were localised, and therefore
particular tree species that are restricted to certain habitats could have been missed.
However, as the plots were located in areas of the Forest near the local communities,
the study provides baseline data on the availability, abundance, size-class distribution
and regeneration of the tree species of relevance to local people. This study can also
be used to inform managers about the impact of illegal use of the species and as an aid
to developing potential management strategies for future use.
The impact of illegal logging in Nyungwe
Forest could be partly reflected
in
differences in densities of seedlings and medium trees between the study sites. These
differences
could also be attributed
to environmental
conditions
and types of
disturbances including fire and grazing. The periphery of the Forest has been altered;
102 http://scholar.sun.ac.za
Stellenbosch University
this has been substantiated by the chi-square analysis of the densities of trees of
different sizes among and within the study sites.
The presence of many small
individuals of the preferred tree species at many sites suggests that the periphery of
Nyungwe Forest is in the process of secondary succession.
species
including
neomildbraediana
Polyscias
fulva,
Neoboutonia
It is likely that some
macrocalyx
and Macaranga
which are characteristic of secondary forests (ORINFOR, 1978)
will become more common where human use is heavy.
The total stem density of all species and the density of the trees of the preferred
species determined the relative abundance of each species within each category of
product.
The density of preferred trees can only convey an impression of available
resources because the majority of the species were preferred for various products.
The densities provided could be used as conservative criteria to set harvesting quotas
which do not rely on an estimate of maximum sustained yield. A resource plant may
be abundant but for some reason not sufficiently appropriate for use for different
products. For instance, preferred tree species for timber and poles do exist in the
Forest but the densities of quality stems were sometimes in low proportion relative to
the stem densities of some species classified as potentially utilisable for timber and
poles. Other characteristics including size-class distribution and regeneration should
be examined in order to determine sustainable utilisation of the species.
Size-class distributions of trees may be used to predict the age structure and state of a
forest (Lorimer, 1980), to determine the regeneration status of dominant tree species
(Geldenhuys and Muray, 1992), and as an indicator of natural or human disturbance
history (Lorimer, 1980). Size-class distributions of trees prior to harvesting provide
important baseline information for future monitoring work (Cunningham, 1999). A
comparison of the size-class distribution prior to and after harvesting
indicates
changes in the rate of seedling and sapling establishment in the harvested population
(ITFC-EMP, 1999).
The analysis of size-class distributions of the preferred tree species indicated that
many populations examined were unstable, with variable transitions from one size
class to the next.
between successive
A stable population is characterised
size classes (Taylor and Walker,
by low ratios of change
1984), corresponding
to
Stellenbosch University
103 http://scholar.sun.ac.za
marginally convex curves without clear inflections (Knowles and Grant, 1983).
Usually size-class
information
is analysed in terms of a negative exponential
distribution (e.g. Midgley et al., 1990). With the exception of Carapa grandiflora
and Symphonia globulifera, most of the species did not conform to this ideal. Since
they showed
a negative
exponential
distribution,
the populations
of Carapa
grandiflora and Symphonia globulifera had more juvenile than adult trees, indicating
they were more stable in densities and were self-replacing (Geldenhuys, 1992; Hall
and Bawa, 1993). The depicted size classes for the remaining species represented
unstable populations.
Some species exhibited either the loss of reproductives or had temporarily variable
recruitment.
Peters
(1994)
attributed
the resulting
size-class
distribution
in
undisturbed stands to sporadic or irregular seedling establishment where the level of
regeneration maintains the population with the 'infrequency' of occurrence causing
'peaks' and 'valleys' in the size class distribution.
The populations which exhibit this
pattern are commonly species that depend on canopy gaps for regeneration or whose
regeneration has been interrupted by seed mortality, physical damage to seedlings or
lack of pollinators or dispersal agents (Armesto and Fuentes, 1988; van Wyk et al.,
1996).
Entandrophragma excelsum has shown exception in that it lacks seedlings.
Midgley
et al. (1995) expressed concern at the apparent absence of regeneration of some trees
in Southern African forests. They speculated that there might be some recent changes
in disturbance regimes in these areas.
In retrospect, the lack of regeneration of E.
excelsum could be due to the lack of major disturbances like those caused by elephant
and buffalo when their herds were present in Nyungwe Forest.
The low density of
large trees of E. excelsum could also be due to intensive pitsawying of the species for
valuable timber in the past (Habiyambere,
1999).
Being a primary species, the
population of this species could permanently disappear from the Forest if the current
conditions do not change.
Size-class distributions of the preferred tree species indicated predominantly that a
few number of species were in natural or unharvested condition.
Many species
exhibited a size-class distribution frequently encountered among light demanding,
104 http://scholar.sun.ac.za
Stellenbosch University
early pioneer species, which require large canopy gaps for regeneration.
absence of such a disturbance, these species whose regeneration
Parinari excelsa, Pauridiantha paucinervis,
In the
is limited (e.g.
Casearia runssorica and Cassipourea
ruwenzoriensis) may temporarily disappear from the Forest, their populations being
represented only by seeds lying dormant in the soils. The results of the study concur
with the findings of Zimmermann et al. (1994) that regeneration in tropical wet
forests is an important life history attribute of the species.
Significant differences in the abundance of seedlings of different species across sites
indicated poor regeneration and recruitment of some species in some sites. This may
be caused, in addition to differences in environmental conditions (e.g. soils, rainfall)
and the ability of the species to reseed, by differences in canopy cover or light gaps in
the Forest (Armesto and Fuentes, 1988).
Fire is reported to be the most common
disturbance in Nyungwe Forest (Gapusi, 1999) and could have affected differently the
part ofthe Forest being surveyed.
4.3
POTENTIAL
SUPPLY AND VULNERABILITY
OF THE PREFERRED
TREE SPECIES
TO EXPLOIT ATION
The discussion in section 4.2. has attempted to show that Nyungwe Forest has
different species with a variety of ecological characteristics that make sustainable
harvesting a very difficult objective. The major problems identified by the study are
the high diversity and low population density, the unstable size-class distribution and
the low density of regeneration of many species. Ignoring these characteristics could
cause an unsustainable exploitation of timber, poles and medicinal materials.
According to Hall and Bawa (1993), extraction is considered sustainable if the impact
of harvesting has no significant effect on the reproduction and regeneration of the
population being harvested.
A combination of criteria including density, size-class
distribution and regeneration could convey an impression about the vulnerability of
tree species to exploitation (Tables 4.1a and 4.1b).
105 http://scholar.sun.ac.za
Stellenbosch University
Table 4.1a. Classification of the preferred timber and pole species into
'vulnerable' and 'non-vulnerable' categories.
Vulnerable
Tree resources
Timber
Carapa grandiflora
Entandrophragma excelsum
Newtonia buchananii
Parinari excelsa
Polyscias fulva
Strombosia scheffleri
Symphonia globulifera
Syzygium parvifolium
Non-vulnerable
x
x
x
x
x
x
x
x
Poles
Carapa grandiflora
Casearia runssorica
Cassipourea ruwenzoriensis
Macaranga neomildbraediana
Strombosia scheffleri
Symphonia globulifera
Syzygium parvifolium
x
x
x
x
x
x
x
Table 4.1b. Classification of the preferred medicinal tree species into
'vulnerable' and 'non-vulnerable' categories
Tree resources
Medicinal tree s[!ecies2 laq:;e tree catelj:;ory
Carapa grandiflora
Entandrophragma excelsum
Maesa lanceolata
Newtonia buchananii
Oxyanthus troupinii
Parinari excelsa
Pauridiantha paucinervis
Symphonia globulifera
Syzygium parvifolium
Tabernamontana johnstonii
Medicinal s[!ecies2 medium tree catelj:;o[Y
Canthium oligocarpum
Carapa grandiflora
Newtonia buchananii
Oxyanthus troupinii
Parinari excelsa
Pauridiantha paucinervis
Symphonia globulifera
Syzygium parvifolium
Vulnerable
Non-vulnerable
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
As the information provided is a baseline data, the grouping of these species into
vulnerable and non-vulnerable categories could serve as a benchmark for describing
the initial ecological conditions that have to be monitored in the process of resource
exploitation.
Obviously,
abundant
species with many seedlings
(e.g. Carapa
grandiflora, Symphonia globulifera and Strombosia scheffleri) are less vulnerable to
106 http://scholar.sun.ac.za
Stellenbosch University
exploitation. More important than the overall abundance of the preferred tree species
is the size-class distribution of individuals of each species.
Stem densities of some species including Cassipourea ruwenzoriensis and Newtonia
buchananii were dominated by the preponderance of small trees, large trees being
very few. The exploitation of such species could not be maintained on a sustained
yield basis and therefore species which approximate an inverse J-shaped curve are
recommended
for exploitation.
Entandrophragma
Felling
of large trees
of species
including
excelsum, Newtonia buchananii and Syzygium parvifolium
for
timber could eliminate these species from the Forest in a much shorter period of time
because of low densities of their stems and instability of their populations.
For many
species including Carapa grandiflora, Symphonia globulifera, Strombosia scheffleri
and Cassipourea ruwenzoriensis, poles were more abundant compared to the densities
of harvest able individuals for timber.
Another criteria, part of plant used, could be added to the three criteria in order to
determine the vulnerability
of harvesting medicinal materials.
As no data was
available on the impact of harvesting medicinal materials from the species, the
information about vulnerability of medicinal species to exploitation is incomplete.
In
addition, it is clear that many of the tree species are favoured for more than one
products, which might lead to overexploitation as observed by Fedlmeier (1998).
Harvesting methods also have an impact on the sustainability
of the resources.
Geldenhuys (1998) proposed methods of timber harvesting based on single tree
selection in the management of natural forest in South Africa. The methods include
harvesting trees selected from a stand on the basis of mortality rates and harvesting of
dead and dying trees of commercially valuable species.
The harvesting of over-
mature and dying trees in Nyungwe Forest can be applied with the aim of maintaining
a stable population of the preferred tree species.
As a way of achieving this, a
silvicultural system that could be employed would exploit products from overstocked
size-classes as it has been applied in the South African forests (Geldenhyus, 1992).
However, as for all natural forests, the system should aim to increase growing space
for faster growth of useful stems and to establish the regeneration of the desirable
species to replace the stem removed in logging (Baur, 1964; Roach, 1974).
If the
Stellenbosch University
107 http://scholar.sun.ac.za
removal of trees across the different size classes is proportional to the relative
densities of the stems in each size class, then there will be no change in the size-class
distribution
of the species at a time when absolute
densities
are declining.
Information on the life history of the species can guide decision over exploitability of
species, as the removal of some trees might not enhance the regeneration of shadetolerant species.
Other factors such as growth rate and coppice response to exploitation of products
need to be considered in order to determine species that are either vulnerable or not
It can be hypothesised that vulnerable species are those
vulnerable to exploitation.
highly preferred for timber and poles, with slow growth rate and weak coppicing
ability. Species that are likely to be less vulnerable would be those with relatively fast
growth rate. No data are available on growth rate and coppicing ability of the species
studied, and this need to be investigated.
Monitoring
operations
and harvesting
adjustments
enable
to
determine
the
sustainability of harvesting levels (Peters, 1994). The most important characteristic
that must be monitored is the density of regeneration.
densities,
Peters (1994) recommends
remedial
In the case of low seedling
treatments
including
enrichment
planting, selective weeding, cutting and removal of woody vines from the crown of
adult trees. These treatments could result in reducing competition and increasing of
light levels into the understorey. The periphery ofNyungwe Forest surveyed requires
such treatments in order to improve the availability and recruitment of seedlings of
the majority of the species including for instance Casearia runssorica, Cassipourea
ruwenzoriensis,
parvifolium.
Parinari
excelsa,
Pauridiantha
paucinervis
and
Syzygium
Species such as Entandrophragma excelsum require enrichment planting
or the re-introduction of locally extinct herbivores (e.g. elephant, buffalo) that will
induce disturbances needed for the regeneration of the species.
4.4
POTENTIAL
FOR JOINT MANAGEMENT
OF NVUNGWE
FOREST
After recounting the needs, problems and preferences of local communities as well as
the potential supply of preferred tree species from Nyungwe Forest, the potential for
community participation in the protection and management of the Forest is assessed.
Stellenbosch University
108 http://scholar.sun.ac.za
The discussion is guided by the experiences of joint forest management (JFM) or
collaborative management in countries where these techniques have been applied.
In Rwanda, JFM or collaborative management of protected forests has not previously
existed.
In the case of Nyungwe Forest, however, there have been programmes of
social forestry that dealt with development activities in the rural areas adjacent to the
Forest (Barbier, 1992; Labrousse, 1992).
These were provided as incentives for
conservation and to reduce dependence of local communities to the forest resources.
Deforestation has been reported to continue at an alarming rate (e.g. Gapusi, 1999;
MINITERRE, 2000) to provide in the daily food needs of the adjacent household to
the Forest.
Cutting of trees for various uses is an important cause of deforestation,
arising from poverty and insufficient land to establish sources of tree products.
This
degradation causes a decline of species suitable for wooden goods, furniture, and
basketry materials.
The consultancy work by Gapusi (1999) indicated a number of
wrong perceptions which relate to the meaning of sustainability and the dependence
of adjacent households to the Forest. This accentuates the need for implementation of
collaborative resource management strategies in the adjacent areas to the Forest.
The effective and meaningful involvement of local communities in natural resource
management has been attempted under JFM in India by linking socio-economic
incentives and forest management (Badola, 1999; Tewari, 1992). Many countries in
the world have undertaken a series of initiatives aimed at devolving management,
control and ownership of natural forest and woodland resources to local communities.
In India, resource degradation has led the government to enter into partnership
agreements with communities whereby the responsibility for forest management is
placed with communities.
In return, these communities are allowed to harvest certain
species in certain areas and to receive a fixed percentage of revenues from harvesting
by the state forest department (Mukerji, 1992).
Other countries
community-based
in Africa have changed their national
forest management.
forest policies
In Ghana, for instance,
toward
a collaborative
management scheme has been adopted in order for the community to benefit from the
forest resources (Prah, 1997). In Zimbabwe, the CAMPFIRE programme insures that
109 http://scholar.sun.ac.za
Stellenbosch University
revenues derived from wildlife reach councils and communities (Scoones and Matose,
1993).
Elsewhere in Africa, community based approaches to forest management have been
adopted: in Tanzania in the Duru-Haitemba and Mgori forests (Sjoholm and Wily,
1995; Wily and Haule, 1995), in the Bwindi and Mount Elgon forests of Uganda
(Wily, 1993), the Kilum and Ijim montane forests of Cameroon (Nurse et al., 1994),
the forests of the Senegal Valley (Diouf, 1994), the Toumousseni forest in Burkina
Faso (Hagberg, 1992), the Gshaka Gumti Park in Nigeria (Dunn, 1994), woodlands
near Mopti in Niger (Kerkhot, 1990), the El Ain woodlands of Sudan (El Din and
Shanks, 1992) and the woodlands in Mali (Skinner, 1988). JFM is also emerging as a
strategy for conservation and sustainable use of forests as in Sri Lanka (Carter et al.,
1994), Thailand and the Philippines (Poffenberger and McGean, 1998).
Joint forest management approaches that work elsewhere may not be applicable to
Rwanda in the context of the current legal and policy provisions
particularly for protected areas, including Nyungwe Forest.
of forests,
Currently, the major
focus is on the conservation of biological diversity and integrity.
The presence of a
unique wealth of biodiversity makes Nyungwe Forest an extremely sensitive area.
The high conservation
status, however, does not imply the exclusion
of local
communities to access certain resources and to participate in the management of the
Forest. The collaborative management at Bwindi Impenetrable and Mgahinga Gorilla
National Parks in Uganda is an example of community use of wood and non-wood
products (Wild and Mutebi, 1996; Cunningham,
1996).
The conservation
and
management of Nyungwe Forest could follow the examples of the Uganda National
Parks and allow access to some resources, including those identified by this study.
Many factors
favour the development
of JFM in Nyungwe
Forest.
Firstly,
communities adjacent to Nyungwe Forest have used products from the Forest in the
past.
Secondly, the Action Plan for the conservation and management of natural
forests in Rwanda adopted in 1984, conceived a three-zone system as follows (Fimbel
and Kristensen, 1994):
•
Forest fringe zone where some timber harvesting was permitted (10% of all the
forest area);
110 http://scholar.sun.ac.za
Stellenbosch University
•
Natural reserve zone where minimal use was allowed (40% of all the forest area);
•
High protection zone, where resource use was restricted (50% of all the forest
area).
The subdivision of Nyungwe Forest into different zones approximated the system of
buffer zones adopted in Uganda (see Wild and Mutebi, 1996).
The restriction of
resource access to the buffer zone can imply the loss of access to some resources, for
example bamboo, which is found in the south-eastern part of Nyungwe Forest.
However, since bamboo can be grown in gardens, it is possible to produce this
resource outside the Forest.
Thirdly, the decentralisation of power and authority by the central government aims to
involve local people and their leadership in the management of all affairs related to
socio-economic development in their regions.
The role of community leadership is
one of the important factors affecting the success of collaborative management (Wild
and Mutebi,
1996).
The survey of the respondents'
perceptions
of potential
institutions for the management of utilisation found different organisations that could
operate at local level for the management
including administrative
of the Forest.
These organisations,
District, Sectors and Cells, work with or have a direct
relationship with the government structures and interact already with the forest
authorities and conservation services.
Due to its power, authority, organisational
structure and mandate, a District could be one of the actors involved in management
and conservation of the Forest.
respondents
The various categories of institutions listed by the
as being more appropriate
show the importance
of having vanous
stakeholders including both users and non-users in forest management.
Fourthly, the conservation
services do not exercise adequate
harvesting of resources, which leads to overexploitation
control over the
and loss of biodiversity.
Since 1994, there has not been effective management of the Forest. The conservation
project staff is too small, without resources and therefore ineffective (WCS and
ORTPN, 1995).
There is patrolling of the Forest, but still there is intense illegal
activities going on.
Without effective management, the Forest will continue to be
exploited, albeit illegally, by adjacent communities.
The lack of an approach toward
Stellenbosch University
III http://scholar.sun.ac.za
developing partnership between the conservation services and local organisations
could exacerbate current conflicts over resource use and access.
The local use of resources from the Forest can be based on the subdivision of the
Forest into different management zones and on the vulnerability of some resources to
exploitation
(Peters,
1996).
Because collaborative
forest management
or JFM
involves agreement between resources users and the owner of the forest (in this case
the government), further clarification of who can use the resources and who will be
responsible for the management of the Forest may be required.
Obviously, as has
been the case in other countries, the foremost users are those people who live next to
the Forest, but they need to be organised into user groups.
People living in the environs of the Forest already work in associations or groups in
various sectors such as crop production, livestock farming and beekeeping, among
others. These associations provide opportunities for the formation of users groups for
collaborative management.
However, because the users are many and the Forest is
large, the use of the resources
could be demarcated
following
administrative
boundaries, instead of natural boundaries as has been the case in Tukucha and
Banskharka panchayats of Nepal (Hobley, 1987). This could enable the involvement
of local administration in decision-making regarding utilisation and management of
the part of the Forest falling within its boundaries.
In order to make use of the Forest
sustainable, it is desirable to link the forest produce with local enterprises, for both
wood and non-wood products (Encarnacion, 1999). The development of alternatives
to resource use could be an important complementary
strategy for improving the
livelihoods of the adjacent communities.
Enabling access to some resources in the Forest alone cannot solve the livelihood
problems of the population.
Substitution activities can provide for some of the
preferred and requested resources as a remedial strategy to low resource availability.
Alternative activities could include local cultivation of indigenous tree and shrub
species for the production of medicinal plants, basketry materials, poles, and wooden
and woven products, following the example of the Guira Banso project in Ghana
(Prah, 1997).
112 http://scholar.sun.ac.za
Stellenbosch University
Incentives for community involvement in forest protection and management can also
come in the form of income, goods, and power to control the forest by local
communities.
Ecotourism in Nyungwe Forest could potentially provide sustainable
financial activity, thus encouraging local people in the sustainable management of
forest resources as has been the case for Budongo Forest Reserve in Uganda (Langoya
and Long, 1998). In the long-term, the communities that live in the area surrounding
the Forest can benefit directly from wildlife. This may be either through protein offtake or from financial benefit from revenue taken from tourists visiting to view
wildlife. Such strategies are akin to the CAMPFIRE programme set up in Zimbabwe
or the community conservation programmes successfully running in Kenya. In return
for the benefit that local communities gain from wildlife, they can act to conserve the
wildlife.
A revenue sharing scheme has to be installed as a one of the means for resolving
conflicts between forest authorities and adjacent communities.
Money that is accrued
from the Forest, once it has moved towards self-sustainability, can be used to answer
to the immediate development needs of the local communities.
Until the conservation
service of the Forest has sufficient income from ecotourism revenues, the Rwanda
Government in partnership with the donor community can have to work closely with
the local human populations.
In response to wood requirements, local communities could be encouraged, trained
and provided with appropriate materials to establish their own sources of wood
products.
Forest projects in the region would, for example, aim at reducing the
fuelwood and pole requirements.
An alternative to fuelwood can be kerosene.
The
provision of efficient kerosene cookers and a campaign to familiarise people with its
use would reduce the fuelwood requirements.
If charcoal is used, then people could
be trained how to produce efficient cookers.
The most difficult, but most important aspect for the adjacent communities can be for
the projects to induce a change from agriculture and livestock based economy to a
monetary economy.
The reasons for this change are as follows: (a) many aspects of
households will increasingly need to acquire things that require money (e.g. clothing,
school fee, house construction, medical fee); (b) the long-term impacts of the present
113 http://scholar.sun.ac.za
Stellenbosch University
economic system will prove to lead to reductions in farm sizes and heavy use of
pastures, resulting in the households to become increasingly poor. Micro-economic
projects that facilitate this move could be encouraged and subsidised.
Such projects can also provide incentives for protection and conservation of the Forest
through the development of local enterprises such as the production of bricks and
roofing tiles, and the production of efficient ceramic cookers that could reduce the
amount of fuel and building wood that is required by the households adjacent to the
Forest. Incentives can be in the form of credits provided by projects to support the
management of the Forest, as was done by the SAFIRE (Southern Alliances for
Indigenous Resources) project in Zimbabwe (Grundy and Le Breton, 1998).
The
management of Nyungwe Forest should seek to diversify livelihood opportunities as
much as possible among the communities living adjacent to the Forest.
In the process of collaborative management, communities could be employed in the
rehabilitation and enrichment planting of the degraded areas and in other silvicultural
treatments inside the Forest. This tactic has shown success in some projects in India
where local people were used as a source of labour (Anderson
et al., 1998;
Poffenberger et al., 1990). In return, communities could benefit from employment
and rights to harvest certain abundant species sustainably.
In order to avoid overexploitation,
agreements between communities
and forest
managers are needed and could establish resource users, species and quantities to be
harvested by each user and monitoring activities for managing resource use.
It is
possible to develop a joint venture between some resource user groups and the forest
management authority to harvest some resources on sustainable basis, integrated with
other forest management activities. The Forest authority can control the harvesting of
resources through a permit system, but ensuring that permit rates do not exceed what
the people can afford to pay, which could reduce illegal activities.
Stellenbosch University
114 http://scholar.sun.ac.za
5
CONCLUSION AND RECOMMENDATIONS
This research has shown that Nyungwe Forest is important to the livelihoods of many
people living on its periphery, through the provision of wood and non-wood forest
products. The preference and demand for forest products result from declining levels
of soil fertility and the lack of livelihood alternatives in the study area. As a result of
the demand for more livelihood alternatives, the managers of Nyungwe Forest are
faced with a difficult situation arising from the effect of population pressure and rural
poverty (Mitchell, 1997). This has led to illegal cutting of trees for various uses, and
poaching of animals for food, gathering of wild vegetables and fruit, the harvesting of
medicinal plants and mining in the valleys of the Forest.
Although communities adjacent to the Forest have farmed trees, the products are less
diverse and not enough to cater for their needs, implying increased reliance on the
Forest for their livelihoods.
However, not all resources preferred by adjacent
communities are found in the Forest, and many available resources are present in
insufficient abundance to be recommended for sustainable utilisation.
As more than
50% of the preferred tree species for various uses were vulnerable to exploitation, the
periphery of Nyungwe Forest is now considered to be an area of low resource
potential for timber, pole and medicinal materials.
Only a few tree species can be considered utilisable due to the fact that populations of
many species were unstable.
The uneven distribution of the range of size classes
found for many species could be a cause for concern
implications.
and has management
The removal of harvesting pressure for poles and timber may result in
more stable populations of the preferred tree species (Shackleton, 1993). The effect
of this is the development and the establishment of new seedlings at regular intervals
leading to a standard size-class distribution of the species.
The management of the
preferred species for the sustainable supply of timber and poles can be achieved
through appropriate silvicultural systems that maintain the characteristics of a stable
population.
As the density of harvestable trees for timber was very low, a silvicultural system
based on single tree selection could allow the production of timber from overstocked
Stellenbosch University
115 http://scholar.sun.ac.za
diameter size classes in a well-defined
zone of the Forest.
However, timber
harvesting should be controlled in order to avoid overexploitation and adverse impact
on the growth dynamics of the harvested population.
As the densities of poles were
more than those of timber trees, the area of the Forest surveyed had more potential for
the supply of poles than timber.
Many tree species are favoured by local people for multiple uses. It is important to
determine the use of each tree species if harvesting is to be sustainable.
medicinal tree species have had their bark harvested
for medicinal
Many
materials.
Debarking of trees for the supply of medicinal material and using the same species for
timber harvesting can adversely affect the regeneration and the genetic composition of
the future population.
Therefore tree species indicated for medicinal purposes should
not be used for timber exploitation except if medicinal material is gathered only from
the small trees or if the medicinal material is gathered before timber harvesting.
Other resources of importance to the livelihoods of local communities
gathered from Nyungwe Forest and are used for household consumption.
are also
Sustainable
utilisation of these products by local communities can be one of the possible means of
establishing positive linkages between local households and conservation of Nyungwe
Forest.
The gathering of these resources should also be organised, regulated and
monitored.
Substitutions
have
always
management of forests.
been
considered
an important
component
of the
Some of the preferred tree species can be propagated on
farms (example Entandrophragma
excelsum for timber, Arundinaria
alpina for
basketry material and building purposes, Alchornea hirtella for bean stakes, Casearia
runssorica for poles), which would reduce the pressure on these resources inside the
Forest. Identification of alternatives to resources, which are less abundant and more
vulnerable to exploitation, could be done in co-operation with the local communities.
This study suggests a way of achieving a balance between community utilisation and
the conservation ofNyungwe Forest. Collaborative management of the Forest can be
a viable arrangement given well-structured
local institutions in which local people
have confidence, the government policy of decentralising power and authority to the
Stellenbosch University
116 http://scholar.sun.ac.za
lowest administrative level, and past experiences in the management of Nyungwe
Forest. This management option could draw on approaches adopted in other countries
such as India, where JFM emerged, and Uganda's Bwindi Impenetrable National Park
where resource use by local people is permitted on specified zones of the Park. The
use of resources in the Forest requires an enabling policy environment with a view to
addressing
the conflicts
over resource
use and access, silvicultural
problems,
productivity, and benefit sharing, through participation of local communities.
Below is a summary of some suggestions regarding the use of Nyungwe Forest and
areas for future research:
•
The management of Nyungwe Forest could permit the gathering of edible wild
foods in areas outside the strict conservation zone of the Forest. Studies aimed at
providing information on the availability, identification, abundance, seasonality,
distribution and harvesting impact for these products have to be carried out in
order to guide the frequency and quantities of removals.
•
Beekeeping can be a source of income to local communities.
The current
initiative of PCFN to stimulate the formation of beekeepers' associations which
should use the resource should be encouraged.
the responsibility
In this way, beekeepers can take
for preventing fires and avoiding the damaging use of the
Forest. The traditional methods of honey production should be improved upon.
•
Bamboo is mostly used by people in the southwest ridge of the Forest, where it is
dominant.
A limited number of harvesters could be licensed to use the resource
until sufficient supplies can be grown on farms.
Research is needed to provide
information on the effect of continued harvesting on the production of biomass in
order to ensure long-term sustainability of bamboo extraction.
•
Mining can be carried out, but in accordance with the regulations of the Ministry
of Commerce, Industry and Tourism and in areas defined by the conservation
services.
Local miners' groups can have access to the resource upon obtaining a
mining title.
Stellenbosch University
117 http://scholar.sun.ac.za
•
The illegal use of wood for the production of wooden items for household use has
resulted in the scarcity of some tree species.
This is a consequence of highly
selective harvesting that removes the most fit individuals. Felling of these trees in
the Forest should be stopped.
Attention
should be focussed on providing
alternative sources of wood for wooden items, including, for instance, the planting
of Polyscias fulva and Markhamia lutea outside the Forest or on farms.
•
Timber harvesting can be done for those tree species which have more individuals
at the larger diameter size class than expected (e.g. Parinari excelsa, Strombosia
scheffleri). The removals should be regulated and based on selective harvesting.
•
Indigenous pole species are preferred to exotic pole species. The indigenous poles
are likely to be much sought after by the adjacent communities,
overharvesting.
leading to
The management of Nyungwe Forest can determine harvesting
cycles and the quantity of poles to be harvested for a particular species.
The
removal of poles as well as the harvesting of timber should ensure that the
population of the species will perpetuate over time.
•
Traditional healers' groups can be permitted to collect medicinal plants.
Due to
the effect of debarking, the gathering of leaf materials should be encouraged along
with the use of growth form categories other than trees (e.g. climbers, shrubs).
Research
is needed to evaluate the response
of preferred
tree species to
disturbance, including harvesting of fruits and bark.
•
The management
and conservation
of Nyungwe
Forest should promote
the
establishment of substitute resources, including the cultivation of preferred tree
species, and assist the local communities
to practice sustainable
agriculture
through the maintenance of productive cropping systems and the development of
small, micro and medium enterprises based on cultivated resources.
•
The periphery of Nyungwe Forest is a secondary forest as a result of disturbances.
Highly disturbed sites need to be identified and studied in terms of species
Stellenbosch University
118 http://scholar.sun.ac.za
composition and size classes.
This can provide useful information for forest
rehabilitation and restoration.
•
Studies on the ecology and the status of the preferred tree species should be
carried out. This may involve the study of the dynamics, the constraints of life
stages,
productivity
distributions.
and
growth
rates,
habitat
requirements
and
spatial
This can provide information on the management of the species for
a sustainable supply of resources.
The methods to be used should combine both
plots and transects and should allow the evaluation of the spatial distribution of
the preferred tree species along gradients and in different habitats.
Stellenbosch University
119 http://scholar.sun.ac.za
REFERENCES
ADB (African Development Bank), 2001. African Development Report 2001.
Fostering Good Governance in Africa. Oxford University Press, USA.
AGRIFOR CONSULT, 1997. Réhabilitation du sous secteur forestier au Rwanda.
Rapport de démarrage 27 Janvier - 30 Avril 1997. AGRIFOR CONSULT,
Kigali.
Anderson, J., Clément, 1., and Crowder, L.V., 1998. Accommodating conflict
interests in forestry: concepts from pluralism. Unasylva 194: 1-56.
ARDI, 1998. Stage de formation sur la culture des champignons. Agenee Rwandaise
de Développement Intégré, Kigali.
Armesto, J.J., Fuentes, E.R., 1988. Tree species regeneration in a mid-elevation,
temperate rain forest in Isla de Chiloé, Chile. Vegetatio 74: 151-159. Kluger
Academic Publishers, Dordrecht.
Arnold, J.E.M., 1990. Social forestry and communal management in India, Social
Forestry Network, Paper 11b Overseas Development Network, London.
Aumeeruddy, Y.T., 1998. People and Plants applied ethnobotany project at Ayubi
National Park, Pakistan. WWF, UNESCO and Royal Botanical Garden, Kew.
Babbie, E. and Halley, F., 1998. Adventure in Social Research: Data analysis using
SPSS for Windows 95. Pine Forge Press, California.
Badola, R., 1999. People and protected areas in India. Unasylva 199: 12-14.
Bahigiki, E. and Vedder, A., 1987. Etude socio-économique et proposition écologique
sur la Forêt de Nyungwe (UGZ3). Min. Agric. Elev. et des Forêts, Direction des
Forêts, Kigali.
Barabwiliza, R. 1992. Problématique des réserves écologiques dans un pays ft
économie essentiellement agricole - les forêts de montagnes et des volcans du
Rwanda. Unpublished MSc. thesis. Fondation Universitaire Luxembourgeoise,
Centre Interuniversitaire des Sciences de l'Environement, Luxembourg.
Barabwiliza, R., 1980. Impact de l'homme sur les formations forestiêres des zones
marginales au Rwanda. Bulletin Agricole du Rwanda 13: 14 -23.
Barakabuye, N., 2001. Integration of local communities in the conservation of
Nyungwe Natural Forest. ORTPN, Kigali.
Barbier, C., 1992. Evaluation du Plan d'Action pour la conservation et l'aménagement
des forêts naturelles de la Crête Zaïre-Nil.
Rapport du consultant en
sylviculture et aménagement. Min. Agric. Elev. et des Forêts, Direction des
Forêts, Kigali.
Baur, G.N., 1964. The ecological basis of rain forest management.
Forestry
Commission of New South Wales, Sydney.
Bazzaz, F.A., 1998.
Plant in changing environments: linking physiological,
population and community ecology. Cambridge University Press, Cambridge.
Ben Chaabane, H. and Cyiza, P., 1992. Méthodologie d'élaboration des données sur
le revenu rural des communes Rwandaises.
Direction de la Planification,
Division des Stratégies de Développement Communal et Régional, Ministêre du
Plan, Kigali.
Bitariho, R., 1999. The abundance, distribution and use of montane bamboo in
Bwindi Impenetrable and Mgahinga Gorilla National Parks, S.W. Uganda.
Unpublished MSc. thesis. Makerere University, Institute of Environment and
Natural Resources, Kampala.
120 http://scholar.sun.ac.za
Stellenbosch University
Blaikie , P.M. and Broofield, H.P., 1987. Land degradation and society, Methuen,
New York.
Bosshard, W. 1971. Rapport sur la situation forestiere du Rwanda et proposition pour
la suite du projet. Birmensdorf.
Bouxin, G., 1974. Etudes phytogéographique des plantes vasculaires du marais
Kamiranzovu (Forêt de Rugege, Rwanda). Bulletin Jardin Botanique Belge 44:
141-159
Bryman, A., and Cramer, D., 1997. Quantitative Data Analysis with SPSS for
Windows: a Guide for Social Scientists. Rowtledge, New York.
Budowski, G., 1976. Proposition pour un programme de sauvegarde de la Forêt de
Nyungwe. Ronéo, Min, Agric., Elevage, Kigali.
Butynski, T.M., 1984. Ecological survey of the Impenetrable (Bwindi) Forest,
Uganda, and recommendations for its conservation and management. Wildlife
Conservation International, New York Zoological Society, USA.
Cambrezy, L., 1984. Le surpeuplement en question: organisation spatiale et écologie
des migrations au Rwanda. ORSTOM, Paris.
Carter, J. Connelly, S. and wilson N., 1994. Participatory forestry in Sri Lanka: Why
so limited. Change on the Horizon. Rural development Forestry Network,
Document 17b, London.
CGF, 1993. Rapport Annuel 1992. Département de Foresterie, ISAR-RUHANDE,
Butare.
Chanyenga, T.F., 1999. Non-wood forest products in Malawi. From:
http://www.fao.org/forestry.
Colchester, M., 1994. Slave and enclave: The political economy of Equatorial Africa.
World Rainforest Movement, Penang.
Cunningham, A.B., 1993. African medicinal plants: Setting priorities at the interface
between conservation and primary healthcare. People and plant Initiative,
Division of Ecological Sciences, Paris.
Cunningham, A.B., 1996. People, park and plant use. Recommendation for multipleuse zones and development alternatives around Bwindi Impenetrable National
Park, Uganda. People and plants Initiatives. Division of Ecological Sciences,
UNESCO, Paris.
Cunningham, A.B., 1999. Woodcarvings, drums and monitoring tree resources. A
test-case of local-level monitoring in Mpanga Forest, Uganda. Field workshop,
8 - 12 March 1999. Unpublished report. WWFIUNESCO/Kew People and
Plants Initiative. Paris.
Cunningham, A.B., 2000. Methods for selection and monitoring of plants species in
multiple-use zones. A training course at the Institute of Tropical Forest
Conservation, Bwindi Impenetrable National Park, Uganda.
WWFIUNESCO/Kew people and Plants Initiative. Unpublished Report.
Cunningham, A.B., and Mbenkum, F.T., 1993. Sustainability of harvesting Prunus
africana bark in Cameroon.
People and Plants Working Paper No.2,
UNESCO, Paris.
Davis, L. S., Johnson, K. N., 1987. Forest management. Third Edition. McGrawHill, New York.
Deffar, G., 1998. Non-wood forest products in Ethiopia. From:
http://www.fao.org/forestry
den Biggelaar, C., 1996. Farmer experimentation and innovation: a case study of
knowledge generation processes in agroforestry systems in Rwanda. FAO,
Rome.
121 http://scholar.sun.ac.za
Stellenbosch University
Dessouter, S., 1991. Pharmacopée humaine et véterinaire au Rwanda. Institut
National de Recherche Scientifique, Butare.
DGF, 1985. Projet de la loi forestiere. Direction Générale des Forêts, Kigali, Rwanda.
Diouf, B., 1994. The responsibility of local people for the management of forest
resources in North Senegal. Rural Development Forestry Network, Network
Paper l8d, , ODI, London.
Dubois, 0., 1998. Rights and wrongs of rights to land and forest resources in SubSaharan Africa: Bridging the gap between customary and formal rules. Forest
Participation Series No.lO. Forestry and Land use Programme, lIED, London.
Dunn, A., 1994. Gashala Gumti, Nigeria - from game reserve to national park., Rural
Development Forestry Network, Network Paper 25d ODI, London.
El Din, A. and Shanks, E., 1992. Natural forest management project, Kozdofan,
Sudan. Mid-term review. ODI, London.
Emerton, L., Mfunda, 1., 1999. Making wildlife economically viable for communities
living around the Western Serengeti, Tanzania. lIED Biodiversity Group,
London.
Encarnacion, C.S., 1999. Community based enterprises and conservation: the
Kalahan Forest Farms Development Projects. Unasylva 199: 15-17.
FAO, 1989. Food and Agricultural Legislation 38: pp. 122 -125, FAO, Rome
FAO, 1992. Produits forestiers non-ligneux : Quel avenir? FAO, Rome.
FAO, 1996. Projet de Stratégie pour le développement agricole national - Horizon
2000 Rwanda. Suivi du Sommet Mondial de I' Alimentation. FAO, Kigali.
FAO, 1999. Etudes des resources forestiêres naturelles et les plantations forestiêres du
Rwanda. FAO, Rome.
Fedlmeier, e., 1998. The development of secondary forest on abandoned grazing
areas, using Costa Rica as an example. Plant Research and Development,
Volume 47/48. Institute for Scientific Co-operation, Tubingen.
Fimbel, R.A. and Kristensen, K.A., 1994. Gold mining activities within the UGZ4
management Zone, Nyungwe Forest Reserve, Rwanda. Report to WCS. Natural
Resources Management Project, Kigali.
Fischer, E., et Hinkel, H., 1992. La nature du Rwanda: Apercu sur la flore et la faune.
Mayence.
Fisher, R.J., 1999. Devolution and decentralisation of forest management in Asia and
the Pacific. Unasylva 199: 3-5.
Gapusi, R.J. and Mugunga, P.e., 1997. Espêces rares et en extinction dans la flore
Rwandaise. Revue bibliographique. ISAR-RUHANDE, Butare.
Gapusi, R.J., 1999. Identification et Analyse des Options pour la Conservation de la
Biodiversité
dans la Forêt Naturelle
de Nyungwe,
Direction
de
l'Environnement, Kigali.
Gasana K.J., 1988. La foresterie sociale au projet Crête Zaire-Nil. CZN, Gikongoro
GEC, 2000. Conservation programs. Green Environmental Conservation, Kigali.
Geldenhuys, C.J. and Murray, B., 1992. Floristic and structural composition of
Hanglip forest, nothern Transvaal. Report No. FOR-DEA 432, Division of
Forest Science and Technology, CSIR, Pretoria.
Geldenhuys, C.J., 1992.
The use of diameter distributions in sustained-use
management of forests: examples from South Africa.
The Ecology and
Management of Indigenous Forests in South Africa. Oxford University Press,
Cape Town.
Geldenhuys, C.J., 1998. Development of a forest management system for sustainable
resource use: basic principles with an example from South African evergreen
122 http://scholar.sun.ac.za
Stellenbosch University
forest. Paper presented at International Workshop on 'Sustainable management
of African Rainforests today', Libreville.
Grison, F., et Braesco, A, 1984. Forêt de Nyungwe: Factibilité du Projet Forestier.
Rapport Scientifique aux actions forestiêres, Caisse Centrale de Coopération
Economique, CTFT.
Grundy, I. and Le Breton, G., 1998. Le programme MIT! de la SAFlRE - une
nouvelle approche en matiere de gestion des resources naturelles dans les zones
communales au Zimbabwe. Rural Development Forestry Network 22e, ODI,
Londres.
Habimana, P., 1982. Etude géographique de l'impact de la Forêt de Nyungwe.
Influence socio-économique sur les Districts de Mudasomwa et Nyamagabe.
Mémoire présenté en vue de I' obtention du grade de licencié es sciences.
Université Nationale du Rwanda, Fac. de Lettres, Butare.
Habiyambere T., 1999. Etude pour l'élaboration d'un plan d'action stratégique pour la
conservation et la gestion de la biodiversité des ecosystémes. Cas du Rwanda.
Direction des Forêts, Kigali.
Habiyambere, T., 1997. Mot d'ouverture il l'occasion du séminaire-atelier sur « La
Nécessité d'une agriculture durable: Contribution il l'agroforesterie ». in: La
Nécessité d'une agriculture Durable: Contribution il l'agroforesterie. Rapport
du Séminaire- Atelier. ISAR-RUHANDE, Butare.
Hagberg, S., 1992. Management of natural forests. The gazetted forest reserve of
Toumousseni, Burkina Faso. Working Paper 215, Swedish University of
Agricultural Sciences, Uppsala.
Hall, P. and Bawa, K., 1993. Methods to assess the impact of extraction of nontimber tropical forest products on plant population. Economic Botany 47: 234247. New York Botanical Garden, New York.
Hasler, R. 1999. Overview of the social, ecological and economic achievements and
challenges of Zimbabwe's CAMPFIRE Programme.
Evaluating Eden EE
DP03, Biodiversity Group, lIED, London.
Hayes, W.AF., 1991. La culture des champignons au Rwanda. TECHNO SERVE
Inc., Kigali.
Hecq, J. and Froment, D., 1961. Contribution il l'étude des problêmes de reboisement
et de· la conservation des sols - Structures agraires traditionnelles,
rationalisation de l'agriculture, conservation du sol et amélioration de la
productivité - le Rwanda et le Burundi. Institut National pour l'Etude
Agronomique du Congo. Brussels.
Hobley, M., 1987. Involving the poor in forest management: Can it be done? The
Nepal - Australia Project experience. ODI, Agricultural Education Unit,
London.
Horward, P.C., 1991. Nature conservation in Uganda's tropical forest reserves.
Forest Conservation Programme, IUCN, Gland.
Horwich, R.H., 1988. The Community Baboon Sanctuary. An approach to the
conservation of private land in Belize. In: Bradwold and Greenberg, eds.
Saving the tropical forest. Earthscan.
Infield, M., 1988. Hunting, trapping and fishing in villages within and on the
periphery of the Korup National Park. WWF, Yaoundé.
Inglis, AS., 1999. Implication of devolutiion for participatory forestry in Scotland.
Unasylva 199: 45-51.
ITFC-EMP, 1999. The potential supply of weaving and medicinal plant resources in
the proposed Remera multiple-ues zone of Bwindi Impenetrable National Park,
Stellenbosch University
123 http://scholar.sun.ac.za
S.W. Uganda. Institute of tropical Forest Conservation, Ecological Monitoring
Programme, Kabale.
ITFC- EMP, 2001. Protocol for monitoring harvesting impacts of key utilised plant
species in Bwindi Impenetrable National Park, S.W. Uganda. Institute for
Tropical Forest Conservation, Ecological Monitoring Programme, Kabale.
ITTO, 1990. Guidelines for the sustainable management of natural forest. Technical
series No.5, International Tropical Timber Organisation, Yokohama.
IUCN, 1979. Action Strategy for Protected Areas the Afrotropical Realm. IUCN,
Gland, Switzerland and Cambridge.
IUCN, 1987. Action strategy for protected areas in the Afromontane Realm. IUCN,
Gland, Switzeland and Cambridge.
IUCN/WWF, 1985. Rapport d'une mission au Zaïre et Rwanda. IUCN, Gland and
Cambridge.
JORR (Journal Officiel de la République Rwandaise), 1989. Loi No.4711988 du 5
Décembre 1988 portant organisation du régime forestier. 23eme Année.
Journal Officiel de la République Rwandaise 3; 192 - 220.
Kamatenesi, M., 1997. The conservatiion and utilisation status of Nyakibazi
(Rytigynia spp.) in the multiple-use areas of Bwindi Impenetrable National
Park, Uganda. Unpublished MSc. thesis. Makerere University, Institute of
Environment and Natural Resources, Kampala.
Kayinamura T., 1986. La Médecine TraditionnelIe Rwandaise: Centre de Médecine
TraditionnelIe de Bare. Centre de Médecine TraditionnelIe de Bare, Kibungo.
Kent, M. and Coker, P., 1992. Vegetation Description and Analysis: A Practical
Approach. CRC Press, Boca Raton, Florida.
Kerkhof, P., 1990. Agroforestry in Africa: A Survey of Project Experiences. Panos,
London.
Klein, M. and Van der Wal, M., 1998. About tropical hardwood, hunters and gorillas:
Conservation of forest fauna in South Cameroon. In: The Congo Bassin. IUCN,
The Netherlands.
Knowles, P., and Grant, M.C., 1983. Age and size class structure analyses of
Engelmann spruce, Ponderosa pine, Lodgepole pine, and Limber pine in
Corolado. Ecology 64: 1-9.
Kollert, W., 1998. Sustainable resources management in Sabah, Malaysia: Licensing
policies and economics. Plant Research and Development, Volume 47/48.
Institute for Scientific Co-operation, Tiibingen.
Kothari, A., Patha, K.., and Vania, F., 2000. Where communities care. Community
based wildlife and ecosystem management in South Asia. Evaluating Eden
EE03, Biodiversity Group, lIED, London.
Kruger, L.M., Midgley JJ. and Cowling, R.M., 1997. Resprouters vs reseeders in
South African forest trees: a model based on forest canopy height. Functional
Ecology 11: 101 - 105. British Ecological Society, Great Britain.
Labrousse, R. 1992. Evaluation conjointe du Plan d'Action pour la conservation et
l'aménagement des forêts de la Crête Zaire-Nil. Section 2. Rapport sectoriel:
Economie. Caisse Francaise de Development, Kigali.
Langoya, C.D. and Long, C., 1998. Local communities and ecotourism development
in Budongo Forest Reserve, Uganda. Rural Development Forestry Network,
Network paper 222. ODI, Portland House, Stag Place, London.
Lawes, M.l., Mander, M., and Cawe, S., 2000. The value and uses of natural forests.
South African Forestry Handbook, Vol.2, pp. 613-624. South African Institute
of Forestry, Pretoria.
124 http://scholar.sun.ac.za
Stellenbosch University
Lonely Planet, 2001. Rwanda Map. Available at: www.lonelyplanet.com.
Lorimer, CJ., 1980. Age structure and disturbance history of a southern Appalachian
virgin forest. Ecology 61: 1169 - 1184.
Luken, J.O., 1990. Directing Ecological Succession. Chapman and Hall, Great
Britain.
Lund, H.G., 1998. IUFRO guidelines for designing multiple resources inventories.
IUFRO World Series Vol. 8. IUFRO Secretariat, Vienna.
MacKinnon, J. and MacKinnon, K. 1986. Review of the protected areas system in the
Afrotropical Realm. IUCN, Gland, Switzerland and Cambridge, UKlUNEP,
Nairobi.
Mbarubukeye S., 1992. La médecine vétérinaire traditionnelle : Rapport du 1er atelier
des guérisseurs traditionnels du bétail au Rwanda. LVNR, Kigali.
Mbarubukeye, S. et Niang, A.I., 1996. Les usages particuliers des plantes ligneuses
autochtones du Rwanda: Utilisation en médecine traditionnelle humaine et
vétérinaire, en alimentation humaine, en apiculture et en artisannat. LVNR,
Kigali.
Mbonyintwari, A., 1989. L'évaluation écologique des terres dans le cadre de
l'aménagement des forêts naturelles de la Crête Zaire-Nil du Rwanda. Mémoire
présenté en vue de I'obtention du grade de maitre es sciences (MSc.), Ecole des
Gradués, Université Laval
Mead, R., Curnow, R.N. and Hasted, A.M., 1993. Statistical Methods in Agriculture
and Experimental Biology. Second Edition. Chapman & Hall, Great Britain.
MICOMART, 1998. Proposition d'un projet d'étude de réalisation d'une unité pilote
de production d'huiles essentielles au Rwanda. Min. Commerce, Artisan et
Industrie, Kigali.
Midgley, J., Seydack, A., ReyneIl, D., and McKelly, D., 1990. Fine-grain pattern in
Southern Cape plateaux forests. Journal of Vegetation Science 1: 539-546.
Opulus Press, Uppsala.
Midgley, J.J., Everard, D.A., and van Wyk, G., 1995. Relative lack of regeneration of
shade-intolerant canopy species in some South African forests. South African
Journal of Science 91: 7-8.
Mihigo, A., 1999. Situation du secteur forestier et des statistiques forestiéres. In
Workshop on data collection and analysis for sustainable forest managementLinking national and intrnational efforts EC - FAO partnership programme
(1998-2000). Projet GCP/INT/679/EC, Lambarene, Gabon, 27 September - 1
October 1999. Ministêre de l"agriculture, de l'Elevage et des Forêts, Kigali.
MINAGRI, 1984. Plan d'Action pour la conservation et l'aménagement des forêts
naturelles de la Crête Congo-Nil. Direction Générale des Forêts, Kigali.
MINAGRI, 1990. Rapport annuel 1990. Projet Crête Zaïre-Nil.
Min. Agric.
Elevage, Kigali.
MINAGRI, 1991. Synthése des réalisations des UGZ. Min. Agric. Elevage, Kigali.
MINAGRI, 1992. Evaluation du Plan d' Action pour la conservation et l'aménagement
des forêts naturelles de la Crête Zaïre-Nil. Section 1. Ministêre de l'agriculture,
de l'Elevage et des Forêts, Direction des Forêts, Kigali.
MINAGRI, 1993. Orientation prioritaires ft prendre en compte pour la poursuite de la
mise en oeuvre du Plan d' Action pour la conservation et l'aménagement des
forêts naturelles de la Crête Zaïre Nil. Document de Travail. Direction des
Forêts, Kigali.
MINAGRI, 1997. Agenda 21 National. Min. Agric. Elevage, Environ. Dev. Rural,
Kigali.
Stellenbosch University
125 http://scholar.sun.ac.za
MINAGRI, 1998. Strategy and action plan for food security. Ministry of Agriculture,
Livestock, Environment and Rural Development, Kigali.
MINAGRI, 2000. Food Security Survey: Phase I. Agricultural production and land
use. FSRP and DSA, Kigali
MINECOFIN, 1998. Socio-Demographic Survey of Rwanda 1998. Office National de
la Population, Kigali.
MINITERRE, 2000. Stratégies et Plan d' Actions pour la Conservation de la
Biodiversité
dans les Aires Protégées.
MINITERRE,
Direction
de
l'Environnement, Kigali.
Mitchell, T., 1997. Rwanda and Conflict. ICE Case Studies, No. 23.
Muderevu, F., 1974. Estimation de la régression de la forêt naturelle du Rwanda par
photo-interprétation. Projet Pilote Forestier, Kibuye.
Muhwezi, 0., 1997. The biology, use and impact of harvesting on the Afromontane
liana (Loeseneriella apocynoides) in and around Bwindi Impenetrable National
Park, Uganda. Unpublished MSc. thesis. Makerere University, Institute of
Environment and Natural Resources, Kampala.
Mukerji, I.F.S., 1992. Greening of Himalayas through people's participation in the
State of Himachal Pradesh (Western Himalayas).
Paper prepared for the
conference on wise management of tropical forests. Oxford.
Murekezi, A., 1999. Les données statistiques sur les produits forestiers non-ligneux au
Rwanda. FAO, Kigali.
Murererehe, S., 1999. Etudes sur les ressources naturelles et les plantations forestiêres
du Rwanda, FAO, Kigali.
Mvukiyumwami, J., 1987. Les páturages du Mutara et leurs problématiques. In:
Bulletin Agricole du Rwanda 12: 214 - 232, Min. Agric. Elevage, Kigali.
Mwima, P.M., 2000. Forest regeneration and ecological recovery within the Mbwa
River Tract in Bwindi Impenetrable National Park, Uganda.
Institute of
Tropical Forest Conservation, Kabale.
Ndayambaje, J.D., 1999. Identification et analyse des options pour la conservation de
la biodiversité dans le Pare National de l' Akagera. Min. Agric. Elevage,
Environ. Dev. Rural, Direction de l'Environnement, Kigali.
Neuman, W.L., 2000.
Social Research Methods: Qualitative and Quantitative
Approaches. Allyn and Bacon, Massachusetts.
Neumann, R.P., and Hirsch, E., 2000. Commercialisation of Non-Timber Forest
Products: Review and Analysis of Research. Center for International Forestry
Research, Bogor.
Newafrica, 2000. Rwanda National Parks Maps. Available at: www.newafrica.com.
Newbury, C., 1992. Rwanda: recent debate over governance and rural development.
In: Hyden, G. and Bratton, M, eds. Governance and Politics in Africa.
Nhira, c., and Matose, F., 1996. Joint forest management and resource sharing:
Lesson from India for Zimbabwe. Forest Participation Series No.5, Forestry
and Land use Programme, ODI, London.
Niang, A.I. et Styger, E., 1990. Les systêmes d'utilisation des terres et leur potentiel
agroforestier au Rwanda. ICRAF/ISAR, Butare.
Nurse, M., McKay, c., Young, J., and Asanaga, C., 1994. Biodiversity conservation
through community forestry in the montane forests of Cameroon. Rural
Development Forestry Network, Network Paper 18d, ODI, London.
Nyilimana, S., 1997. Politique de développement apicole au Rwanda. ARDI, Kigali.
Offut, M., 1992. Tourism as a conservation strategy in Nyungwe Forest Reserve,
Rwanda. IUCN/WCMC, Cambridge.
126 http://scholar.sun.ac.za
Stellenbosch University
Ogwal, S.F., 1998. A study of abundance, utilisation and management of smilax
anceps in Bwindi Impenetrable National Park, S.W Uganda. Unpublished MSc.
thesis. Makerere University, Institute of Environment and Natural Resources,
Kampala.
Olson, J.M., 1990. The impact of socio-economic factors on migration patterns in
Rwanda. M.A. thesis. Department of Geography, Michigan State University.
Olson, J.M., 1994. Farmer responses to land degradation in Gikongoro, Rwanda.
Ph.D. dissertation, Department of Geography, Michigan State University.
Olson, J.M., Manyara, G., Campbell, D.J., Lusch, D.P., and Hu, J., 1995. Rwanda
Society Environment Project.
Department of Geography, Michigan State
University.
ORINFOR, 1978. La Forêt de montagne du Rwanda. Office de l'lnformation, Kigali.
ORTPN, 1989. Rapport Annue11989. Min. Agric. Elevage, Kigali.
ORTPN, 1991. Plan de conservation de l'éléphant au Rwanda. Unpublished report.
Office Rwandais du Tourisme et des Pares Nationaux, Kigali.
Owiunji, I., 1997. The biodiversity survey of Bwindi Impenetrable National Park.
Makerere University, Institute of Environment and Natural Resources,
Kampala.
Palm, C., Izac, A.M. and Vosti, S., 1993. Procedural Guidelines for Characterisation
and Diagnosis. ICRAF, Nairobi.
PCFN, 1989. Proceedings from the first international workshop for the conservation
and management of afromontane forests, 19 - 23 June 1989 Cyangugu, Rwanda.
Project de conservation de la Forêt de Nyungwe, Cyangugu.
Percival, V. and Dixon, T.H., 1995. Environmental scarcity and violent conflict: The
case of Rwanda. American Association for the Advancement of Science and
the University of Toronto, Washington D.C.
Perlez, J., 1988. Compromise keeps Rwanda's dwindling families of Gorillas alive.
The Globe and Mail. P.A. 15.
Peters, e.M., 1994. Sustainable harvest of non-timber plant resources in tropical
moist forest: An ecological primer. USAID Biodiversity Support Programme,
Washington D.C.
Peters, C.M., 1996. Beyond nomenclature and use: A review of ecological methods
for ethnobotanists.
In: Alexiades M.N. eds. Selected guidelines for
ethnobotanists research: Afield manual, pp. 241 - 276.
Philip, M.S., 1983. Measuring trees and forests: A textbook written for students in
Africa. University of Dar-es-Salaam, Dar-es-Salaam.
Poffenberger, M. and McGean, B., 1998. Village voices, forest choices. Joint forest
management in India. Oxford University Press, New Delhi.
Poffenberger, M., Bhatia, K., and McGean, B., 1990. Forest management
partnerships: regenerating India's Forests. Ford Foundation, New Delhi.
Prah, E.A., 1997.
Joint Forest Management - the Gwira- Banso experience.
Commonwealth Forestry Review 76: 171 - 174.
Prasaad, R., 1999. Joint forest management in India and the impact of state control
over non-wood forest products. Unasylva 198: 58-62.
Renner, D., 1991. Recherche Action: Initiation d'un processus pour la mise en oeuvre
du programme de sensibilsation - vulgarisation concernant la conservation et
l'utilisation durable de la Forêt Naturelle de Nyungwe. Premiere étape.
Décembre 1989 - Décembre 1990. UGZl, Kibuye.
Stellenbosch University
127 http://scholar.sun.ac.za
Renner, D., 1992. Discussion sur la phase initiale 1991/92 du programme de
sensibilisation-vulgarisation
de l'UGZl
concernant la conservation
et
l'utilisation durable du massifforestier de Nyungwe. AUGZ1, Kibuye.
Roach, B.A., 1974. Selection cutting and group selection. Applied Forestry Research
Institute Miscellaneous Report No.5, State University of New York, New York.
Ruiz Pérez, M. and Arnold, J.E.M., 1996. Current issues in non-timber forest
products research. Proceedings of workshop 'Research on NTFP'. Hot Springs,
Zimbabwe 28 August - 2 September 1995. Center for Internatioanl Forestry
Research, Bogor.
Sarin, M., and SARTHl, 1996. The view from the ground: community perspectives
on joint forest management in Gujarat, India. Forest Participation Series No.4,
lIED, London.
Scoones, 1. And Matose, F., 1993. Local woodland management: constraints and
opportunities for sustainable resource use. In: Bradley, P.N. and McNamara,
K., eds. Living with trees: policies for forestry management in Zimbabwe, pp.
157 - 198. World Bank Technical Paper 210, World Bank, Washington DC.
Scott, P.J., 1992. Fringe benefits: Minor forest product collection within buffer zones
as a potential tool for conflict resolution in Bwindi Impenetrable National Park.
MSc. Thesis. Agricultural University of Norway.
Shackleton, e.M., 1993. Demography and dynamics of the dominant woody species
in a communal and protected area of the eastern Transvaal Lowveld. S. Afric. J
Bot. 59: 569-574.
Sjoholm, H. and Wily, L., 1995. Finding a way forward in natural forest management
in Tanzania. The emergence of village forest reserves. IRDC Currents 10,
International Rural Development Centre, Swedish University of Agricultural
Sciences, Uppsala.
Skinner, J., 1988. Toward better woodland management in the sahelian Mali.
Pastoral Development Network, Network Paper 25d" ODI, London.
Smith, LR., Wells, D.A., and Welsh, P., 1985. Botanical survey and monitoring
methods for grasslands.
Focus on Nature Conservation 10.
Natural
Conservation, Council.
Snedecor, G.W., and Cochran, W.G., 1964. Statistical Methods. Sixth Edition. The
Iowa State University Press, Iowa.
Sood, M.P., 1994. New forest initiatives in Himachal Pradesh. Forest Participation
Series No.3, lIED, England.
Sorg, J.P., 1978. La Forêt de Nyungwe: Exploitation ou protection? In:
Schweizerischen, zeitschrift fur forstwesen 129: 445-452.
Spinage, C., 1972. The ecology and problems of the Volcano National Park, Rwanda.
Biological Conservation 4: 194-204.
Stanley, A.C., and de Olieviera Castro,G.M., 1959. Manual of Vegetation Analysis.
Harper & Brothers, Publishers, New York.
Stebler, J.D., Darricau, Y., Kassianoff, P., Combe, J., Sentenac, A. et D'Huart, J.P.,
1984. Mission d'évaluation sur la conservation et l'aménagement des forêts
naturelles de la Crête Zaire-Nil. Rapport de Synthêse, Direction Générale des
Forêts, Kigali.
Storz M., 1983. La Forêt de Nyungwe et sa Faune. Projet Pilote Forestier, Direction
des Eaux et Forêts, Kigali.
Struhsaker, T.T., 1997. Ecology of an African rain forest: logging in Kibale and the
conflict between conservation and exploitation. Gainesville, University Press of
Florida.
Stellenbosch University
128 http://scholar.sun.ac.za
Sun, C., Kaplin, B., Kristensen, K.A., Munyaligoga, V., Mvukiyumwami, J.,
Kajondo, E., and Moermond, T.e., 1996. Tree phenology of a tropical montane
forest in Rwanda. Biotropica 28: 668-681.
Taylor, S.E., and Walker, B.H., 1984. Autecology of an invading population of the
cactus Cerus peruvianus (queen of the night) in the central Transvaal. South
African Journal of Botany 52: 397-402.
Tewari, D.N., 1992. Joint (participatory) forest management. The Indian Forester
119: 687-698.
Troupin, G., 1978. Flore du Rwanda: Spermatophytes. Volume I. Institut National de
Recherche Scientifique, Publication No. 18, Butare.
Troupin, G., 1983. Flore du Rwanda: Spermatophytes. Volume II. Agenee de
Cooperation Culturelle et Technique, Musée Royal de l"Afrique Centrale,
Tervuren.
Troupin, G., 1985. Flore du Rwanda: Spermatophytes. Volume III. Agenee de
Cooperation Culturelle et Technique, Institut National de Recherche
Scientifique, Butare.
Troupin, G., and Bridson, D.M., 1982. Flore des plantes ligneuses du Rwanda.
Institut National de Recherche Scientifique, Publication No. 21, Butare ..
Troupin, G., and Donneaux, A., 1982. Plantes ligneuses du Rwanda: Lexique des
Noms Kinyarwanda. Institut National de Recherche Scientifique, Butare.
Twarabamenye, E., 1999. Identification et Analyse des Options pour la Conservation
de la Biodiversité dans le Pare National des Volcans. Min. Agric. Elevage,
Environ. Dev. Rural, Direction de l'Environnement, Kigali.
UFRIC, 2000. Joint forest management. A community training manual for forest
communities.
Uganda Forestry Resources and Institution Centre, Makerere
University, Kampala.
UGZ1, 1987. Directives pour les interventions dans la zone tampon. AUGZ1, Kibuye.
van Wyk, G.F., Everard, D.A., Midgley, J.J., and Gordon, l.G., 1996. Classification
and dynamics of a southern African subtropical coastal lowland forest. S. Afric.
J Bot.62: 133-142.
Vedder, A. and Weber, B. 1984. Forest Conservation in Rwanda and Burundi. Swara
7: 32 - 35.
Wambugu, E.W., 1999. Women's role in agroforestry: a case study of Embu District,
Kenya. Unpublished MSc.thesis. University of Stellenbosch, South Africa.
WCS and ORTPN, 1995. Nyungwe Forest Conservation Project. Progress Report
April- September 1995. Office Rwandais du Tourisme et des Pares Nationaux,
Kigali.
Weber, A.W, 1992. Evaluation du Plan d' Action pour la Conservation et
l' Aménagement des Forêts Naturelles de la Crête Zaïre-Nil. Rapport Sectoriel
sur les Interactions Forêt - Population. Section 2. Ministêre de l' Agriculture, de
l'Elevage et des Forêts, Direction des Forêts, Kigali.
Weber, A.W., 1989. Conservation and development on the Zaire-Nil divide: An
analysis of value conflicts and convergence in the management of afromontane
forests in Rwanda. Ph.D. Thesis. University of Wisconsin, Madison.
WHO, IUCN, and WWF, 1993. Principes directeurs pour la conservation des plantes
médicinales. World Health Organisation, Gland.
Wild, R.G., and Mutebi, 1., 1996. Conservation through community use of plant
resources: Establishing collaborative management at Bwindi Impenetrable and
Mgahinga Gorrilla National Parks, Uganda. People and plants Initiatives.
Division of Ecological Sciences, UNESCO, Paris.
Stellenbosch University
129 http://scholar.sun.ac.za
Wilkie, D.S. and Carpenter, IF., 1999. Bushmeat hunting in the Congo Bassin: An
assessment of impacts and options for mitigation.
Biodiversity and
Conservation.
Williams, S.D., and Ntayombya, P., 1999. Akagera Biodiversity Project. Min. Agric.
Elevage, Environ. Dev. Rural, Direction de l'Environnement, Kigali.
Wilson, R., 1991. The conflict in Rwanda and its impart on the mountain gorilla. Oryx
25: 119 - 120.
Wily, L. and Haule, H., 1995. Good news from Tanzania. Village forest reserves in
the making: the story of Duru-Haitemba. Forest trees and People Programme,
Newletter No. 29, Uppsala.
Wily, L., 1993. Establishing the basis of joint forest management. Integrating
information collection with participatory planning process in regard to Mount
Elgon, Uganda. IUCN, Gland.
Wollenberg, E., 1999. A conceptual framework and typology for explaining the
outcome oflocal forest management. Journal of Forest Management 9:. 1-35.
World Bank, 1993. Inventoring and monitoring biodiversity in the Nyungwe Forest.
Mimeo. World Bank, Washington D.C.
World Bank, 1998. World Development Indicators. World Bank, Washington, USA.
WWF, 1994. La diversité végétale: une richesse vitale. WWF, Gland.
Zar, J.H., 1999. Biostatistical Analysis. Fourth edition. Prentice - Hall, Inc. New
Jersey.
Zimmermann, J.K., Everham, E.M., Waide, R.B., Lodge, D.J., Taylor, C.M., and
Brokaw, N.V.L., 1994. Responses of trees to hurricane winds in tropical wet
forests in Puerto Rico: Implications for tropical life histories. Journal of
Ecology 82: 911-922.
Stellenbosch University
130 http://scholar.sun.ac.za
APPENDICES
Appendix 1. Questionnaire for household interviews
I. Background information of the respondent
1. Respondent Ref. No.
_
3. Commune:
Sector:
2. Date of interview: -----Cell:
_
4. Gender: --------5. Marital status: -----6. Age:
years
II. Socio-economics
7. What is the family size of your household?
member(s)
8. What is your highest level of education?
_
9. Mention, in priority order, the enterprises undertaken by the household
1.
2.
3.
4.
5.
10. What is the most important source of your livelihood (survival strategy)?
11. Could you rank the sources of income for the household?
1.
2.
3.
4.
5.
12. How do you access to financial resources?
13. Do you belong to any group or association? If yes, which one?
_
III. Land tenure and land use
14. What is the size of your farm?
15. What is the process through which land has been acquired?
_
Stellenbosch University
131 http://scholar.sun.ac.za
16. How do you rate the fertility of your farm?
1. Very poor:
_
2. Poor:
_
3. Moderate: -----
4. High:
_
5. Veryhigh:
_
6. Don't know: ---
17. What is the status of soil fertility in the region during the last 5 years?
1. Static
3. Decreased --2. Increased
4. Don't know --18. What are the causes of soil fertility decline, increase or stability?
1. Stability: a
b.
c.
2. Increase: a
b.
c.
3. Decrease: a
b.
c. ------4. Don't know: ---
_
19. How does the household solve land availability problems?
IV. Crops
20. What are the types of crops grown by the household?
1. Cash crops:
2.Foodcrops:
3. Crops grown for food and cash income:
_
_
21. For the most important, whether food or cash crops, could you tell us the number
of bags yielded at harvest?
Most important Crops
1.
------------2.
_
3.
_
4.
------------5.
_
Number of bags at harvest
22. Do you also farm fruit trees, medicinal plants and fodder plants? If yes, what are
they?
1. Fruit trees: a.
b.
c.-------2. Medicinal plants: a.
b
c.
_
3. Fodder plants: a.
b
c.------4. Other horticultural materials: a.
b.
c.-----23. How do you compare revenues from crops with those from trees?
Stellenbosch University
132 http://scholar.sun.ac.za
V. Livestock farming
24. What are the types of livestock kept and how many are they?
Livestock types
Numbers
1.
_
2.
_
3.
__
25. What is the most important purpose oflivestock kept?
Livestock types
Purpose
1.
_
2.
_
3.
__
4.
_
5.
_
26. What is the average income from different animal products?
Animal products
Income (RWF)
1.
_
2.
_
3.
___
27. How is grazing practised? (grazing practices)
28. What are the sources of fodder during critical periods?
29. List, in order of preference, the most three palatable tree and grass species that
are found in Nyungwe Forest
Tree species
1.
2.
3.
Grass species
1.
2.
3.
Stellenbosch University
133 http://scholar.sun.ac.za
30. What are the particular locations where these plants can be obtained in Nyungwe
Forest? Why are they found there?
Tree species
1.
2.
3.
Grass species
1.
2.
3.
_
_
_
_
_
_
Location
1.
2.
3.
Location
1.
2.
3.
Reason for location
_
_
_
_
_
_
VI. Availability and uses of natural resources
31. Do you own any tree crop? If yes, what is its size and species planted?
Area:
speCIes:
_
32. What is the most important purpose of the planted trees?
_
33. What is the average income from tree products?
Tree products
1.
2.
3.
Average income (RWF)
34. What are the resources available in Nyungwe Forest?
1.
2.
3.
5.
6.
7.
4. -----8.
_
35. Before Nyungwe Forest becomes a forest reserve, mention in order of importance
the resources to which local people have had access.
2.
5.
2.
6.
3.
7.
4.
8. ------
_
36. To which resources do you have access to and use in Nyungwe Forest and in the
Buffer zone?
1. Nyungwe Forest: a.
b.
c.
_
2. Buffer zone: a.
b.
c. ------37. List in order of importance for your livelihood the resources present in Nyungwe
Forest and in the buffer zone.
1.
2.
3. --------
Stellenbosch University
134 http://scholar.sun.ac.za
38. For the above resources, could you tell how much is available using the scale of 1
to 5, where 1 means "very little and 5 "very abundant"?
Resources
Grade
1.
_
2.
3.
4.
_
_
_
39. What could be the different uses of these resources?
Resources
1.
_
2.
_
3.
4.
_
_
40. For timber, poles, medicinal plants, fodder and fruits, could you list in order of
importance the preferred species present in Nyungwe Forest?
Timber: --------------------------------------------------Poles:
___
Medicinal plants:
_
Fruits: ---------------------------------------------------Fodder:
__
41. How is each of the products above harvested?
Timber:
--------------------------------------------------Poles: ---------------------------------------------------Medicinal plants
___
Fruits:
-------------------------------------------------Fodder:
-------------------------------------------------42. If timber, poles, medicinal plants, fodder and wild fruits from Nyungwe Forest are
used for commercial purposes, where do you normally sell them?
Products
1.
_
2.
_
3.--------------4.
_
5.
_
Market location
43. For marketed product, how much is sold and at what cost?
Products
1.
Quantity
Cost (RWF)
_
2. -------3.
_
44. Where are these resources most abundant in Nyungwe Forest?
Stellenbosch University
135 http://scholar.sun.ac.za
Timber:
Poles: ----------------------------------------------------Medicinal plants:
Fruits: --------------------------------------------------Fodder:
__
_
__
45. Who do the collection of timber, poles, medicinal herbs and fruit from Nyungwe
Forest?
Timber:
__
Poles: ----------------------------------------------------Medicinal herbs:
_
Fruit: ----------------------------------------------------Fodder:
__
46. Using a scale of 1 to 5, where 1 means "very little" and 5 "very high" could you
tell us how much each key resource is harvested at a time?
Timber:
Poles:
Medicinal herbs:
Fruit:
Fodder:
47. For medicinal plants, which parts of plants are collected or harvested?
Medicinal plants
Parts of plant used
1. ------------2.
_
3.
_
4.
_
5. ------------6.
_
7.
_
8.
_
48. In which season, medicinal plants, fruit and fodder are most abundant?
Medicinal plants:
_
Fruits:
---------------------------------------Fodder:
---------------------------------------
49. Are there any changes that have occurred in Nyungwe Forest resources? If yes,
why?
Changes
Resources concerned
Reason for change
1.
_
2.
_
3.
_
4.
_
5. ------------6.
_
Stellenbosch University
136 http://scholar.sun.ac.za
VII. Wildlife
50. What wild animals occur in Nyungwe Forest and how abundant are they?
Animal types
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
Rare
-1
-1
-1
-1
-1
-1
-1
-1
-1
-1
Moderate
-2
-2
-2
-2
-2
-2
-2
-2
-2
-2
Abundant
-3
-3
-3
-3
-3
-3
-3
-3
-3
-3
Very
abundant
Most
abundant
-4
-4
-4
-4
-4
-4
-4
-4
-4
-4
-5
-5
-5
-5
-5
-5
-5
-5
-5
-5
Don't
know
-6
-6
-6
-6
-6
-6
-6
-6
-6
-6
51. What types and for what most important purposes are animal hunt? (Mention one
for each)
Game
Most important purpose
1.
___
2. ----------------------3.
___
4.
___
52. What kind of change has occurred in animal population size and why?
Change
Wildlife species
Reason for change
1. ---2.
_
3.
_
4.
_
53. For the major games, say when hunting is done?
Animal types
Time for hunting
1.
2.
3.
4.
5.
VIII. Institutions
54. In the past, where there any local institutions that were responsible
management ofNyungwe Forest? If yes, which ones?
1.
2.
3.
for the
_
Stellenbosch University
137 http://scholar.sun.ac.za
55. What was the role of local leadership in the management and conservation of
Nyungwe Forest resources during that time? (before Nyungwe became a forest
reserve)
1.
2.
3.
56. Give examples of some conflicts that arise in Nyungwe resource access and use.
1.
__
2.
_
3.
___
4.
_
5.
__
6.
_
57. Could you tell how each conflict is important using the scale of 1 to 4 where 1
means "not important" and 4 " most important"?
Nature of conflict
1.
_
2.
------------------------------3.
__
4.
_
5.
--------------------------------
Importance of conflict
58. For each conflict, what is the most appropriate strategy to avoid it?
Conflict
1.
2.
3.
4.
5.
Use
fine
-1
-1
-1
-1
- 1
of
Consultation!
meeting
-2
-2
-2
-2
-2
59. How are the civil structures / institutions
conservation of natural resources
a.
b.
c.
courts
Customary
approach
-4
-4
-4
-4
-4
-3
-3
-3
-3
-3
involved
In
sustainable
None
-6
-6
-6
-6
-6
use and
60. According to your opinion, which institutional arrangement could be responsible
for management of the Forest for the benefit of local people and why?
Stellenbosch University
138 http://scholar.sun.ac.za
61. What might be the role of local institutions in the management ofNyungwe Forest
resources?
1.
2.
3.
4.
5.
62. What would you do to improve and maintain availability of Nyungwe Forest
resources in the future?
1.
2.
3.
4.
5.
Stellenbosch University
139 http://scholar.sun.ac.za
Appendix 2. List of on-farm medicinal plants mentioned by the respondents in
the survey area.
Scientific name
Local name
Parts used
Asteracae spp.
Idoma
Leaves, whole plant
Barleria grandicalyx
Bugangabukari
Leaves
Cenecio cydoniifolius
Irarire
Leaves
Clematis simensis
Umunkamba
Leaves
Cynura scandens
Ikizimyamuriro
Leaves
Dychoriste tricocalyx
Akanyamapfundo
Leaves
Iboza riparia
Umuravumba
Leaves
Lysimachia ruhmeriana
Umuyobora
Leaves
Mitragyna rubrostipulosa
Umuzibaziba
Leaves, roots
Momordica foetida
Umwishya
Leaves
Oeimum swave
Umwenya
Leaves
Pentas decora
Isagara
Leaves
Physalis peruvian a
Iperi
Leaves
Plecthranthus barbatus
Igicunshu
Leaves
Scrabera alata
Umubanga
Leaves
Solanum aculeastrum
Umutobotobo
Leaves
Solanum nigrum
Urusenda
Leaves
Strychnos usambarensis
Umuhoko
Leaves
Tephrosia vogelii
Umuruku
Leaves, roots
Vernonia amygdalina
Umubirizi
Leaves
Stellenbosch University
140 http://scholar.sun.ac.za
Appendix 3. Palatable tree and grass species mentioned by 5 or more
respondents in the survey area, n = 200.
Scientific name
Local name
Occurrence
Panicum hochstetteri
Triumfetta cordifolia
Mariscus thomaiophyllus
Digitaria hackelii
Ipomea invo/ucrata
Exotheca abyssinica
Cynodon aethiopicus
Vireetaria major
Pennisetum purpureum
Botriocline /ongipes
Ferns
Arundinaria a/pina
Sericostachys scandens
Eragrostis olivacea
Cynodon n/emfuensis
Syzygium parvifolium
Rubus rigidus
Asystacia gangetica
Hyparrhenia div. spp.
Ikirumbi
Umunaba
Igikerakezi
Urwili
Everywhere
Bottomland
Swamps
Forest gaps
Bush
Forest gaps
Everywhere
Hilltops
Disturbed area
Everywhere
Hilltops
Hilltops
Everywhere
Forest gaps
Everywhere
Disturbed area
Everywhere
Everywhere
Forest gaps
s:
Inyovu
Umunigi
Urukiryi ""_,
Urubingo
Igihehe
Ib ish ihe
Umugano
Umukipfu
Ishinge
Umucaca
Umugote
Umukeri
Ijojwe
Umukenke
%of
res onses
73
25.5
17.5
16.5
14
13.5
11.5
11.0
10.5
8.5
6.5
5.0
4.5
4.0
4.0
3.5
3.0
3.0
3.0
Stellenbosch University
141 http://scholar.sun.ac.za
Appendix 4. List of preferred timber tree species by the respondents in the
study area, n = 200.
Scientific names
Vernacular names
Afrocrania volkensii
Umujegeshi
Bersama abyssinica
Umukaka
Carapa grandiflora
Umushwati
Cassipourea ruwenzoriensis
Ingongo
Chrysophyllum gorungosanum
Umutoyi
Croton megalocarpus
Umunege
Dasylepsis racemosa
Umugoma, imbayu
Dichaetanthera corymbosa
Umuhube
Entandrophragma excelsum
Umuyove
Faurea saligna
Umutiti
Hagenia abyssinica
Umugeti
Macaranga neomildbraediana
Umusekera
Maesopsis eminii
Umuhumuro
Markhamia lutea
Umusave
May tenus acuminata
Inembwe
Newtonia buchananii
Umukereko
Ocotea michelsonii
Umuganza
Ocotea usambarensis
Umutake
Oxyanthus troupini
Ingendajoro
Parinari excelsa
Umunazi
Podocarpus falcatus
Umufu
Podocarpus latifolius
Umuhulizi
Polyscias fulva
Umwungo
Prunus africana
Umwumba, rwamba
Rhamnus prinoides
Umusasa
Strombosia schefJleri
Umushyika
Symphonia globulifera
Umushishi
Syzygium parvifolium
Umugote
Zanthoxyllum gil/etii
Umuturirwa
Stellenbosch University
142 http://scholar.sun.ac.za
Appendix 5. List of preferred medicinal plants found in Nyungwe Forest, as
mentioned by the respondents.
Scientific name
Vernacular name
Asc/epiadacae div. spp.
Indondori
Begonia meyeri-johannis
Irebe
Bridelia brideliifolia
Umugimbu
Brillantaisia cicatricosa
Icyunga
Canthium glabriflorum
Imvuvu
Canthium oligocarpum
Umushabarara
Carapa grandiflora
Umushwati
Clerodendron div. spp.
Umukuzanyana
Crabbea velutina
Sinkangwinguma
Croton megalocarpus
Umunege
Cyathula polycephala
Igifashi
Dichaetanthera corymbosa
Umuhube
Dichrostachys cinerea
Umunkamba
Embelia schimperi
Umukaragata
Entada abyssinica
Umusange
Entandrophragma excelsum
Umuyove
Eucalyptus sp.
Inturusi
Fagaropsis angolensis
Umugomera
Fagodia obovata
Umutanoga
Faurea saligna
Umutiti
Iboza riparia
Umuravumba
Indigofera arrecta
Umutarabana
Lantana trifolia
Umuhengeri
Maesa lanceolata
Umuhanga
Mikaniopsis tedlei
Umuhokoro
Mitragyna rubrostipulosa
Umuzibaziba
Momordica foetida
Umwishywa
Myrica kandtiana
Isubyo
Neobutonia macrocalyx
Umwanya
Newtonia buchananii
Umukereko
Ocotea michelsonii
Umuganza
Ocotea usambarensis
Umutake
Olea hochstetteri
Intobo
Oxyanthus troupini
Umutovu
Parinari excelsa
Umunazi
Pauridiantha paucinervis
Umusibya
Pentas decora
Isagara
Prunus africana
Umwumba
Rapanea melanophloes
Uruneke
Sericostachys scandens
Umukipfu
Symphonia globulifera
Umushishi
Synadenum grantii
Umukoni
Syzygium parvifolium
Umugote
Tabernamontana johnstonii
Umuronzi
Vernonia amygdalina
Umuravumba
Zanthoxyllum usambarense
Intareyirungu
Zanthoxylum gilletii
Umubavu, umuturirwa
F- Forest; 0 = Outside Forest; ND = Not determined.
Part used
Occurrence
Roots
Whole plant
Roots
Roots
Bark
Bark
Bark
Leaves, roots
Leaves
Bark
Leaves
Bark
Leaves
Roots, wood
Leaves, roots
Bark
Leaves
Bark
Leaves, roots
Bark
Leaves
Roots
Leaves
Leaves, roots
Bark
Leaves, bark
Leaves
Wood
Leaves
Bark
Bark
Bark
Bark
Bark
Bark
Leaves
Leaves, bark
Bark, leaves
Bark, leaves
Leaves
Bark, leaves
Roots
Bark, leaves
Leaves, bark
Leaves
Bark
Bark
0
NO
F
NO
NO
F
F
0
0
NO
NO
NO
0
NO
NO
F
0
NO
NO
NO
0
NO
NO
F
NO
F,O
0
NO
F
F
NO
NO
F
F
F
F
0
NO
F
F
F
NO
F
F
0
F
F
Stellenbosch University
143 http://scholar.sun.ac.za
Appendix 6. List of scientific and vernacular names of tree species recorded in all study sites.
Scientific name
Agauria salicifolia HOOK
Alangium chinense REDHER
Albizia gummifera C.A. SMITH
Alchornea hirtella BENTH.
Allophylus kiwuensis GILG
Bersama abyssinicaFRESEN.
Bridelia brideliifolia FEDDE
Canthium oligocarpum HIERN
Carapa grandiflora SPRAGUE
Case aria runssorica GlLG
Cassipourea gummiflua 1. LEWIS
Cassipourea ruwenzoriensis ALSTON
Chassalia subcordata ROBYNS
Chrysophyllum gorungosanum ENGL
Chrysophyllum rwandense TROUPIN
Cleistanthus polystachyus HOOK
Combretum orophilum LIBEN
Cremaspora triflora SCHUMANN
Dasylepis racemosa OLIVER
Dichapetalum heudelotii BAILON
Diospyros gabonensis GUERKE
Ekebergia capensis SPARRMAN
Entandrophragma excelsum SPRAGUE
Galiniera coffeoides DELILE
Grewia mildbraedii BURRET
Hagenia abyssinica J.F. GMELIN
Harungana montana SPIRLET
Ikinyarubabi
Ilex mitis RADLK.
Lindackeria kivuensis BAMPS
Macaranga neomildbraediana LEBRUN
Maesa lanceolata FORSSKAL
Magnistipula butayei BRENAN
May tenus acuminata LOES
Memecylon walikalense A. and R. FERNANDES
Millettia psilopetala HARMS
Myrianthus holstii ENGL.
Neoboutonia macrocalyx PAX
Newtonia buchananii GILBERT and BOUTIQUE
Ochna afzelii R. BR. ex OLIV.
Olea hochstetteri BAKER
Oricia renieri G. GLIBERT
Oxyanthus troupinii BRIDSON
Parinari excelsa SABINE
Pauridiantha paucinervis BREMEK.
Pentadesma reyndersii SPIRLET
Pleiocarpa pycnantha STAPF
P odocarpus fa/catus PILGER
Podocarpus latifolius RENDLE
Polysciasfulva HARMS
Rapanea melanophloeios MEZ
Rhamnus prinoides L 'HERIT.
Rinorea gracilipes ENGL.
Ritchiea albersii GILG
Rytigynia kiwuensis ROBYNS
Strombosia scheffleri ENGL.
Symphonia g/obu/ifera L. f.
Syzygium parvifolium MILDBR.
Tabernamontanajohnstonii
PICHON
Xymalos monospora BAILLON
Zanthoxylum gilletii WATERMAN
Zanthoxylum usambarense KOKWARO
Local name
Umukarakara
Intogota, umurangara, umuvugangoma
Umusebeya
Bwizabwishyamba
Imbayu
Umukaka
Umugimbu
Umushabarara
Umushwati
Umuhanda
Intiritiri
Ingongo
Ikibonobono
Umutoyi
Urushehe
Umusamba
Urubabara
Kanyabwoya
Ingomwa, urubumburi
Umumenamabuye
Umunyakayumbo
Umufumba, umujuga
Umuyove
Umubonobono
Urushingati
Umugeti
Umushayishayi
Ikinyarubabi
Umunywande
Umunywamazi
Umusekera
Umuhanga
Intambasha
Inembwe, umunembwe
Umusuri
Umunaniranzovu
Umwufe
Umwanya
Umukereko
Isanzu, urushingo
Intobo
Umuzo
Umutovu, umuyebe
Inkungu, umunazi
Umusibya
Umwasa
Ikinesha
Umufu
Umuhulizi
Umwungo
Uruneke
Umusasa
Umwicaranyenzi
Umuhu
Urusarabatemyi
Umushyika
Umushishi
Umugote
Umuronzi
Umuhotora
Isoyo, umuturirwa
lntareyirungu
Stellenbosch University
144 http://scholar.sun.ac.za
Appendix 7. Relative frequencies of trees recorded within the large tree
category.
5 Total
4
Tree species
1
2
3
Agauria salicifolia
0.00
0.00
0.00
1.28
5.60
0.00
Alangium chinense
0.47
0.85
0.46
0.80
0.00
0.00
Albizia gummifera
2.14
0.40
2.21
0.00
0.47
1.00
Alchornea hirtella
0.00
1.10
0.90
0.95
0.00
0.55
Allophylus kiwuensis
0.00
1.20
0.00
0.00
0.95
0.46
Bersama abyssinica
1.71
0.40
0.00
0.00
0.00
0.46
Bridelia brideliifolia
0.00
0.43
0.09
0.00
0.00
0.00
Canthium oligocarpum
0.90
0.95
0.00
0.36
0.00
0.00
Carapa grandiflora
12.71
13.12
14.69
8.12
11.49
9.60
Casearia runssorica
4.80
7.18
3.17
8.53
5.13
5.65
Cassipourea ruwenzoriensis
2.80
3.87
4.07
5.21
4.70
4.10
Chassalia subcordata
0.95
0.00
1.37
3.20
0.55
1.81
Chrysophyllum gorungosanum
2.84
2.99
2.40
4.42
0.00
2.46
Chrysophyllum rwandense
0.40
0.00
0.00
0.00
0.00
0.09
Cleistanthus polystachyus
0.00
3.31
24.43
0.00
3.42
6.20
Combretum orophilum
0.00
1.10
0.45
0.47
0.85
0.55
Cremaspora triflora
2.40
0.00
2.71
4.27
0.00
1.91
Dasylepis racemosa
0.40
1.10
0.00
0.47
0.00
0.36
Diospyros gabonensis
0.40
2.21
3.62
4.27
0.43
2.10
Ekebergia capensis
0.00
0.55
0.00
0.00
0.00
0.09
Entandrophragma excelsum
2.80
1.66
0.00
1.90
1.71
1.64
Galiniera coffeoides
0.00
0.55
0.00
0.00
0.43
0.18
Grewia mildbraedii
2.40
0.00
0.00
2.37
0.43
1.09
Hagenia abyssinica
0.40
0.00
0.00
0.00
0.00
0.09
Harungana montana
0.00
0.00
0.00
0.00
0.43
0.09
Lindackeria kivuensis
0.40
0.00
0.00
0.00
0.00
0.09
Macaranga neomildbraediana
9.60
1.66
0.45
0.00
1.28
2.83
Maesa lanceolata
7.20
0.00
0.00
0.00
1.28
1.91
Magnistipula butayei
0.80
0.00
0.00
0.47
0.00
0.27
Memecylon walikalense
1.20
0.55
0.00
1.42
0.43
0.73
Myrianthus holstii
6.80
4.97
2.26
8.06
8.55
6.20
Neoboutonia buchananii
0.00
0.55
4.07
0.00
0.85
1.09
Newtonia macrocalyx
0.00
0.00
0.00
4.27
1.28
1.09
non identified
0.40
0.00
0.00
0.47
0.00
0.18
Ochna afzelii
0.00
1.10
0.00
0.00
2.56
0.73
Olea hochstetteri
0.40
0.00
0.00
0.00
0.00
0.09
Oricia renieri
0.40
1.66
0.90
1.90
1.28
1.19
Oxyanthus troupiniii
4.00
0.00
4.27
0.45
0.00
1.82
Parinari excelsa
1.20
4.97
4.07
4.74
1.71
3.19
P auridiantha pauc inerv is
2.80
0.00
1.36
1.42
0.00
1.19
Pentadesma reyndersii
1.60
0.00
0.45
0.00
0.00
0.46
Pleiocarpa pycnantha
0.00
0.55
0.45
0.47
0.00
0.27
Podocarpus falcatus
0.00
1.66
0.00
0.00
0.00
0.27
Podocarpus latifolius
0.80
0.00
0.00
0.00
0.00
0.18
Polyscias fulva
0.40
0.55
0.00
0.47
2.14
0.73
Rapanea melanophloeios
3.60
0.00
0.00
0.00
0.00
0.82
Rhamnus prinoides
2.80
0.00
0.00
1.90
0.00
1.00
Rinorea gracilipes
0.00
6.08
0.90
0.00
1.71
1.55
Ritchiea albersii
0.00
0.00
0.45
0.00
0.00
0.09
Strombosia scheffleri
6.00
25.41
24.89
17.06
27.78
19.78
Symphonia globulifera
3.60
3.31
1.36
2.84
9.83
4.28
Syzygium parvifolium
5.60
2.21
0.90
0.47
0.00
1.91
Tabernamontana johnstonii
0.00
1.66
1.81
0.00
5.13
1.73
Zanthoxylum gilleiii
0.00
0.55
0.00
0.00
0.43
0.18
Total
100.00 100.00 100.00 100.00 100.00
100.00
..
1 - Musebeya. 2 = Nshili; 3 = Kagano; 4 = Mudasomwa; 5 = Busozo
Stellenbosch University
145 http://scholar.sun.ac.za
Appendix 8. Relative frequencies of trees recorded within the medium tree
category
Tree species
Alangium chinense
Albizia gummifera
Alchornea hirtella
Allophylus kiwuensis
Bersama abyssinica
Bridelia brideliifolia
Canthium oligocarpum
Carapa grandiflora
Casearia runssorica
Cassipourea gummiflua
Cassipourea ruwenzoriensis
Chassalia subcordata
Chrysophyllum gorungosanum
Cleistanthus polystachyus
Combretum orophilum
Cremaspora triflora
Dasylepsis racemosa
Dichapetalum heudelotii
Diospyros gabonensis
Ekebergia capensis
Entandrophragma excelsum
Grewia mildbraedii
Unidentified 'Ikinyarubabi'
llex mitis
Lindackeria kivuensis
Macaranga neomildbraediana
Maesa lanceolata
Magnistipula butayei
May tenus acuminata
Memecylon walikalense
Myrianthus holstii
Neoboutonia macrocalyx
Newtonia buchananii
Ochna afzelii
Olea hochstetteri
Oricia renieri
Oxyanthus troupiniii
Parinari excelsa
Pauridiantha paucinervis
Pentadesma reyndersii
Pleiocarpa pycnantha
Podocarpus latifolius
Polyscias fulva
Rapanea melanophloeios
Rinorea gracilipes
Ritchiea albersii
Rytigynia kiwuensis
Strombosia scheffleri
Symphonia globulifera
Syzygium parvifolium
Tabernamontana johnstonii
Xymalos monospora
Zanthoxylum gilletii
Zanthoxylum usambarense
Total
I
= Musebeya;
2
..
= Nshili;
3
= Kagano;
2
1
0.00
2.34
0.00
0.00
0.00
0.00
7.42
8.59
2.34
0.39
0.39
9.38
3.52
0.00
0.00
3.13
0.00
0.00
0.00
0.39
1.17
1.17
0.78
0.78
0.39
2.34
0.39
0.39
4.69
0.39
0.78
0.00
3.13
0.00
1.17
3.52
11.33
1.17
0.78
0.00
0.00
1.17
0.00
1.17
0.00
2.34
6.25
1.17
2.73
5.86
0.39
6.64
0.00
0.00
100
4
= Mudasomwa;
5 = Busozo
Total
4
5
3
0.40
2.30 0.35 0.50 0.00
3.02
12.64 0.00 2.50 4.71
0.10
0.00 0.00 0.50 0.00
0.10
0.00 0.35 0.00 0.00
1.18
0.80
5.75 0.35 0.00
0.10
0.00 0.00 0.50 0.00
2.11
0.00 0.00 1.00 0.00
8.24
13.79 6.03 8.50 8.24
2.30 5.67 8.00 4.12
4.72
0.00 0.00 0.50 0.00
0.20
7.64
4.60 16.31 4.50 9.41
4.32
1.15 2.48 1.00 5.29
3.42
6.90
1.77 4.50 2.94
1.76
1.81
2.30 4.61 0.00
0.00 0.35 0.00 8.82
1.61
0.00
1.42 2.50 0.00
1.71
0.00 0.35 0.00 0.00
0.10
2.31
0.00 0.00 6.50 5.88
1.18
0.00
1.42 2.50
1.11
0.00 0.00 0.00 0.00
0.10
0.00 0.00
1.00 0.00
0.50
0.00 0.35
1.50 1.76
1.01
0.00 0.00 0.00 0.00
0.20
0.00 0.71 0.50 0.00
0.50
0.00 0.00 0.00 0.00
0.10
2.30
1.42 1.00 1.76
1.71
0.00 0.00 0.00 0.00
0.10
0.00 0.71 0.50 0.00
0.40
0.00 0.35 0.50 0.00
1.41
0.00 0.00 0.00 0.59
0.20
1.15 0.35 2.50 5.29
1.81
0.00 0.00 0.00 0.59
0.10
4.60 0.00 5.00 5.88
3.22
4.60 0.00 0.50 0.00
0.50
0.00 0.35 0.50 0.00
0.50
6.90 0.00 11.50 3.53
4.42
1.15 10.64 9.00 0.59
7.94
0.00
1.42 3.00 0.00
1.31
0.00
1.42 1.50 1.18
1.11
0.00 0.71 0.00 0.00
0.20
3.45 3.55 0.50
1.18
1.61
0.00 0.00 0.00 0.00
0.30
6.90 0.35 0.50 0.59
0.90
0.00 0.00 0.00 0.00
0.30
5.75 1.06 0.00 0.00
0.80
0.00 4.26 3.00 0.59
2.51
0.00 0.00 0.00 0.00
1.61
0.00 5.67 3.00 9.41
4.12
11.49 15.25 8.50 12.94
9.95
0.00 8.16 0.50 0.00
3.92
0.00 0.35
1.00 0.59
0.50
0.00 0.00 0.00 0.00
1.71
0.00
1.06 0.00 0.00
0.30
0.00 0.35 1.00 0.00
0.30
100
100
100
100
100
Stellenbosch University
146 http://scholar.sun.ac.za
Appendix 9. Relative frequencies of small trees recorded within the small tree
category.
Tree species
Albizia gummifera
Alchornea hirtella
Carapa grandiflora
Casearia runssorica
Cassipourea ruwenzoriensis
Chassalia subcordata
Chrysophyllum gorungosanum
Cleistanthus polystachyus
Combretum orophilum
Dichapetalum heudelotii
Diospyros gabonensis
lkinyarubabi
Macaranga neomildbraediana
Magnistipula butayei
Milletia psilopetala
Myrianthus holstii
Newtonia buchananii
Ochna afzelii
Olea hochstetteri
Oricia renieri
Oxyanthus troupiniii
Parinari excelsa
Pauridiantha paucinervis
Pentadesma reyndersii
Pleiocarpa pycnantha
Polyscias fulva
Rhamnus prinoides
Rinorea graci/ipes
Ritchiea albersii
Strombosia scheffleri
Symphonia globulifera
Syzygium parvifolium
Tabernamontana johnstonii
Xymalos monospora
Total
I
= Musebeya;
4
1
2
3
0.00 11.11 0.00 0.00
0.00 0.00 0.00 0.00
20.00 0.00 16.07 28.57
1.79 0.00
3.33 7.41
0.00 0.00 3.57 0.00
20.00 0.00 0.89 0.00
3.33 0.00 8.93 17.86
0.00 7.41 6.25 0.00
0.00 14.81 0.00 0.00
0.00 3.70 0.00 3.57
0.00 3.70 0.00 0.00
3.33 0.00 0.00 0.00
0.00 0.00 0.89 0.00
0.00 3.70 0.00 14.29
0.00 0.00 0.00 3.57
3.33 7.41 0.00 0.00
0.00 0.00 30.36 0.00
0.00 3.70 0.00 0.00
0.00 0.00 0.89 0.00
0.00 0.00 0.00 3.57
23.33
0.00 2.68 0.00
0.00 0.00 0.89 0.00
3.33 0.00 0.00 3.57
0.00 0.00
1.79 0.00
0.00 3.70 0.00 0.00
0.00 3.70
1.79 0.00
0.00 7.41 0.00 0.00
0.00 7.41 0.00 0.00
0.00 0.00 0.89 0.00
0.00 3.70 0.89 7.14
6.67 11.11 16.96 17.86
10.00 0.00 4.46 0.00
0.00 0.00 0.00 0.00
3.33 0.00 0.00 0.00
100
100
100
100
..
2 = Nshili; 3 = Kagano; 4 = Mudasornwa; 5 = Busozo
Total
5
0.00
1.21
0.40
2.00
12.00
15.38
2.00
2.43
4.00
2.43
4.00
3.64
2.00
6.88
0.00
3.64
18.00
5.26
4.00
1.62
2.00
0.81
0.00
0.40
2.00
0.81
0.00
2.02
0.00
0.40
10.00
3.24
0.00
13.77
0.00
0.40
0.00
0.40
6.00
1.62
2.00
4.45
0.00
0.40
2.00
1.21
0.00
0.81
0.00
0.40
0.00
1.21
2.00
1.21
0.00
0.81
0.00
0.40
8.00
3.24
16.00
14.98
0.00
3.24
2.00
0.40
0.00
0.40
100
100
Stellenbosch University
147 http://scholar.sun.ac.za
Appendix 10. Basal areas of tree species within the large tree category in the sample area.
Tree species
Frequency
basal area, m"
Relative basal area
mean basal area, nr'
Agauria salicifolia
Alangium chinense
Albizia gummifera
Alchornea hirtella
Allophylus kiwuensis
Bersama abyssinica
Bridelia brideliifolia
Canthium oligocarpum
Carapa grandiflora
Casearia runssorica
Cassipourea ruwenzoriensis
Chassalia subcordata
Chrysophyllum gorungosanum
Chrysophyllum rwandense
Cleistanthus polystachyus
Combretum orophilum
Cremaspora triflora
Dasylepis racemosa
Diospyros gabonensis
Ekebergia capensis
Entandrophragma excelsum
Galiniera coffeoides
Grewia mildbraedii
Hagenia abyssinica
Harungana montana
lkinyarubabi
Lindackeria kivuensis
Macaranga neomildbraediana
Maesa lanceolata
Magnistipula butayei
Memecylon walikalense
Myrianthus holstii
Neoboutonia macrocalyx
Newtonia buchananii
Ochna afzelii
Olea hochstetteri
Oricia renieri
Oxyanthus troupinii
Parinari excelsa
Pauridiantha paucinervis
Pentadesma reyndersii
Pleiocarpa pycnantha
Podocarpus falcatus
Podocarpus latifolius
Polyscias fulva
Rapanea melanophloeios
Rhamnus prinoides
Rinorea gracilipes
Ritchiea albersii
Strombosia scheffleri
Symphonia globulifera
Syzygium parvifolium
Tabernamontana johnstonii
Zanthoxylum gilletii
Total
14
5
Il
6
5
5
1
4
126
62
45
15
27
I
68
6
21
4
23
I
18
2
12
1
I
2
1
31
21
3
8
68
3
21
8
1
13
20
35
13
5
3
3
2
8
9
Il
17
1
217
47
21
19
2
1097
1.9761
0.1251
0.4490
0.0797
0.0605
0.1075
0.0839
0.0468
9.2044
4.4855
2.0773
0.6195
3.9447
0.0094
7.9985
0.0837
0.3294
0.1261
1.7815
0.1988
3.9164
0.0247
1.9279
0.0314
0.0326
0.0198
0.0082
1.5163
0.7440
0.0888
0.2945
2.5679
0.0386
15.4484
2.2266
0.0452
0.1820
0.7252
12.3693
0.8350
0.3354
0.0656
0.2635
0.1433
0.9756
1.0812
2.7289
0.5200
0.0143
17.8591
6.8172
2.6544
0.9918
0.9033
112.1841
0.0176
0.0011
0.0040
0.0007
0.0005
0.0010
0.0007
0.0004
0.0820
0.0400
0.0185
0.0055
0.0352
0.0001
0.0713
0.0007
0.0029
0.0011
0.0159
0.0018
0.0349
0.0002
0.0172
0.0003
0.0003
0.0002
0.0001
0.0135
0.0066
0.0008
0.0026
0.0229
0.0003
0.1377
0.0198
0.0004
0.0016
0.0065
0.1103
0.0074
0.0030
0.0006
0.0023
0.0013
0.0087
0.0096
0.0243
0.0046
0.0001
0.1592
0.0608
0.0237
0.0088
0.0081
1.0000
0.1412
0.0250
0.0408
0.0133
0.0121
0.0215
0.0839
0.0117
0.0731
0.0723
0.0462
0.0413
0.1461
0.0094
0.1176
0.0140
0.0157
0.0315
0.0775
0.1988
0.2176
0.0123
0.1607
0.0314
0.0326
0.0099
0.0082
0.0489
0.0354
0.0296
0.0368
0.0378
0.0129
0.7356
0.2783
0.0452
0.0140
0.0363
0.3534
0.0642
0.0671
0.0219
0.0878
0.0716
0.1219
0.1201
0.2481
0.0306
0.0143
0.0823
0.1450
0.1264
0.0522
0.4517
0.1023