Secondary forests on anthropogenic soils in Brazilian Amazonia conserve agrobiodiversity

André Junqueira

Throughout Brazilian Amazonia anthropogenic soils that are the product of pre-Columbian settlements are called Terra Preta de Índio (Indian Dark Earths, TPI). These soils are dramatically different from surrounding soils due to long-term human activity, but there is little information about how secondary forest succession is affected by these differences. We tested if community structure (density, richness and basal area), floristic composition and domesticated species’ richness and density were similar between TPI and non-anthropogenic soils (NAS) in 52 25 × 10 m secondary forest plots in different successional stages near three traditional communities along the middle Madeira River, Central Amazonia. We sampled 858 woody individuals on TPI (77 domesticated) and 1095 on NAS (27 domesticated); 550 understory palms on TPI (169 domesticated) and 778 on NAS (123 domesticated). We found 179 species on TPI (10 domesticated), 190 on NAS (8 domesticated), and 74 (25%) in both environments. Although community structure on TPI and NAS was fairly similar, they showed significantly distinctive floristic compositions, both for woody individuals and understory palms. The density and richness of domesticated species was significantly higher on TPI than on NAS for woody individuals, but not for palms. The intimate long-term association of TPI with human activity has lead to the formation of distinct secondary forests and has favored the concentration of domesticated populations of crop species. Hence, secondary forests on anthropogenic soils concentrate agrobiodiversity, offering advantages for in situ conservation of genetic resources, and are unique ecosystems that should be considered in conservation efforts.

Biodivers Conserv (2010) 19:1933–1961 DOI 10.1007/s10531-010-9813-1 ORIGINAL PAPER Secondary forests on anthropogenic soils in Brazilian Amazonia conserve agrobiodiversity André Braga Junqueira • Glenn Harvey Shepard Jr. Charles R. Clement • Received: 3 July 2009 / Accepted: 13 February 2010 / Published online: 6 March 2010 Ó Springer Science+Business Media B.V. 2010 Abstract Throughout Brazilian Amazonia anthropogenic soils that are the product of pre-Columbian settlements are called Terra Preta de Índio (Indian Dark Earths, TPI). These soils are dramatically different from surrounding soils due to long-term human activity, but there is little information about how secondary forest succession is affected by these differences. We tested if community structure (density, richness and basal area), floristic composition and domesticated species’ richness and density were similar between TPI and non-anthropogenic soils (NAS) in 52 25 9 10 m secondary forest plots in different successional stages near three traditional communities along the middle Madeira River, Central Amazonia. We sampled 858 woody individuals on TPI (77 domesticated) and 1095 on NAS (27 domesticated); 550 understory palms on TPI (169 domesticated) and 778 on NAS (123 domesticated). We found 179 species on TPI (10 domesticated), 190 on NAS (8 domesticated), and 74 (25%) in both environments. Although community structure on TPI and NAS was fairly similar, they showed significantly distinctive floristic compositions, both for woody individuals and understory palms. The density and richness of domesticated species was significantly higher on TPI than on NAS for woody individuals, but not for palms. The intimate long-term association of TPI with human activity has lead to the formation of distinct secondary forests and has favored the concentration of domesticated populations of crop species. Hence, secondary forests on anthropogenic soils concentrate agrobiodiversity, offering advantages for in situ conservation of genetic resources, and are unique ecosystems that should be considered in conservation efforts. A. B. Junqueira (&) C. R. Clement Coordenação de Pesquisas em Ciências Agronômicas, Instituto Nacional de Pesquisas da Amazônia, Avenida André Araújo, 2936, Manaus, Amazonas 69060-001, Brazil e-mail: abjunqueira@gmail.com C. R. Clement e-mail: cclement@inpa.gov.br G. H. Shepard Jr. Departamento de Antropologia, Museu Paraense Emı́lio Goeldi, Avenida Perimetral 1901, Belém, Pará 66077-830, Brazil e-mail: gshepardjr@gmail.com 123 1934 Biodivers Conserv (2010) 19:1933–1961 Keywords Community structure Amazonian dark earths Plant domestication Secondary succession Succession management Traditional resource management Abbreviations ANCOVA EMBRAPA (Empresa Brasileira de Pesquisa Agropecuária) IBGE (Instituto Brasileiro de Geografia e Estatı́stica) INPA (Instituto Nacional de Pesquisas da Amazônia) IUSS LTSP/INPA (Laboratório Temático de Solos e Plantas) MANOVA NAS NMDS PCA TPI (Terra Preta de Índio) Analysis of covariance The Portuguese acronym for the Brazilian Agricultural Research Corporation) The Portuguese acronym for the Brazilian Institute of geography and statistics The Portuguese acronym for the National Research Institute for Amazonia International Union of Soil Science. the Portuguese acronym for the Soil and Plant Thematic Laboratory of INPA Multivariate analysis of variance Non-anthropogenic soils. Non-metric multidimensional scaling Principal components analysis The Portuguese acronym for Indian black earths, also called Indian dark earths and Amazonian dark earths Introduction Throughout Brazilian Amazonia anthropogenic soils associated with pre-Columbian indigenous settlements are called Terra Preta de I´ndio (Indian Dark Earths, hereafter TPI; Woods and Denevan 2009). TPI is widely distributed in Amazonia, occurring in patches that may vary from a single hectare to hundreds of hectares (Smith 1980). TPI patches are especially common in Brazilian Amazonia (Sombroek et al. 2002), but they occur also in Colombia, Peru, Venezuela and the Guianas (Eden et al. 1984; Andrade 1986). Most TPI patches were formed between 500 and 2,500 years ago (Neves et al. 2003), but the specific cultural and ecological processes that created these soils are still poorly understood and are the focus of an intense, multidisciplinary scientific effort (Woods and Denevan 2009). Although edaphic conditions can be highly variable among different TPI patches (Falcão et al. 2009), a few properties are common to almost all of them when compared to adjacent soils: higher levels of phosphorous, calcium, organic matter, pH and cation exchange capacity (Lehmann et al. 2003a). These characteristics make these soils more suitable for agriculture than other upland (i.e., non-flooded) Amazonian soils (Glaser 2007), which generally have low fertility and soil organic matter contents (Chauvel et al. 1987; Lehmann et al. 2003a). For this reason, TPI is frequently associated with specific and more intensive forms of agriculture (German 2003a, b; Fraser and Clement 2008; Fraser et al. 2009), although fallows are still an important part of the agricultural systems. There are suggestions that the secondary forest succession on TPI occurs differently than on other upland soils (Clement et al. 2003). Early successional stages on TPI showed a higher percentage of soil coverage by new weeds, higher weed species richness, and a higher relative proportion of annual and leguminous plants when compared with adjacent non-anthropogenic soils (Major et al. 2005). When abandoned, swiddens on TPI are 123 Biodivers Conserv (2010) 19:1933–1961 1935 colonized by a characteristic group of aggressive weeds, including several species typically associated with human-disturbed environments (Major et al. 2005). Local farmers recognize diverse stages of forest regrowth on TPI through indicator plant species, even when vegetation is dense (Moran 1981; Sombroek et al. 2002; German 2003b). Farmers also recognize vegetation structural characteristics associated with TPI, such as lower canopies and denser understories (Woods and McCann 1999), smaller average diameter of adult trees, and a greater abundance of vines and plants with spines (German 2003b). In the Xingu River basin, the proximity to TPI patches was identified as one of the factors determining the occurrence of secondary forests dominated by lianas (Balée and Campbell 1990). Preliminary observations by Samuel Almeida and colleagues (Clement et al. 2009) indicate that old-growth forests [at least 300 years old (Kern 1996)] on TPI at Caxiuanã (Pará, Brazil) show differences in forest structure and species composition when compared to old-growth forests on non-anthropogenic soils. On the other hand, Paz-Rivera and Putz (2009) found few differences in the density of 17 useful tree species when comparing old secondary forests (at least 140 years old, probably much older) on anthropogenic and nonanthropogenic soils in a lowland forest in Bolivia. However, these authors suggested that on anthropogenic soils large individuals of long-lived species may be remnants of ancient cultivation (Paz-Rivera and Putz 2009). Still, apart from the experimental study of Major et al. (2005), the preliminary observations by Samuel Almeida and colleagues (Clement et al. 2009) and the quantitative ‘‘useful-species-focused’’ approach of Paz-Rivera and Putz (2009), all other observations are derived from qualitative and ethnographic data, and lack more detailed ecological investigation to be validated. Clement et al. (2003) raised the hypothesis that TPI could act as agrobiodiversity reservoirs, areas that concentrate considerable genetic diversity of native and exotic species with domesticated populations due to TPI’s long-term association with human activity. There is a growing need to locate areas of high diversity of crop wild relatives, which may or may not also be areas of high landrace diversity, in order to propose adequate conservation strategies for these areas (Maxted et al. 2008). We used an ecological approach to test the hypothesis that secondary forests on TPI concentrate agrobiodiversity, predicting that these environments would show a greater abundance and richness of species with domesticated populations when compared to adjacent secondary forests on non-anthropogenic soils. This is the first study that addressed the question of secondary succession on TPI. We compared secondary forests in several successional stages on TPI and nearby nonanthropogenic soils (hereafter NAS) with regards to forest structure (density, species richness and basal area) and species composition. We also identified TPI indicator species and tested the hypothesis that secondary forests on TPI concentrate agrobiodiversity. Methods Study site The study was carried out in three riverside communities located in the municipality of Manicoré, on the middle Madeira River, Amazonas state, Brazil: Água Azul (5°490 5700 S; 61°330 5600 W); Barreira do Capanã (5°500 5100 S; 61°400 1100 W); and Terra Preta do Atininga (5°380 1900 S; 61°30 600 W; Fig. 1). The local climate is characterized as Af in the Köppen system, with mean annual temperatures between 27 and 28°C, and a main annual rainfall of about 2,500 mm, with a marked dry season from June to September. The natural vegetation 123 1936 Biodivers Conserv (2010) 19:1933–1961 Fig. 1 Location of study sites. White triangles represent the three communities where plots for secondary forest sampling were established, and the white circle indicates the town of Manicoré, located along the middle Madeira River, Amazonas, Brazil of the region is composed mainly of closed canopy evergreen terra firme forest, with savannas further from the rivers and seasonally flooded forests along the rivers (IBGE 2004; Silva and Pereira 2005; Rapp Py-Daniel 2007). The predominant soils are ferralsols, but patches of acrisols, lixisols and podzols occur, and a ‘‘strip’’ of gleysols extends along the Madeira River floodplain (IBGE 2001; soil nomenclature sensu IUSS 2007). The occurrence of anthropogenic soils is common on the Madeira River (Simões and Lopes 1987; Sombroek et al. 2002; Kern et al. 2003; Fraser et al. 2009). These communities are inhabited by caboclos, descendants of immigrants that arrived in the region from northeastern Brazil during the rubber boom in the beginning of the Twentieth century and intermarried to different extents with local indigenous people. About 42, 38 and 36 families live in Água Azul, Barreira do Capanã and Terra Preta do Atininga, respectively. Major subsistence and economic activities include shifting cultivation (predominantly manioc, Manihot esculenta, Euphorbiaceae), hunting, fishing and the extraction of Brazil nut (Bertholletia excelsa, Lecythidaceae) and rubber (Hevea brasiliensis, Euphorbiaceae). All three communities are located on or close to TPI patches that vary in area from eight to about 35 ha. A mosaic of agricultural areas and secondary forests in various stages of regrowth occupies the whole area. Sample design In each community we used the ‘‘snowball’’ technique (Bernard 2002) to select key informants, recognized by local residents as having considerable knowledge about the types of vegetation and soils in the region. With the key informants we walked along trails through agricultural areas, secondary forests, primary forests and other areas used by the community. During this process the informants were stimulated to identify the different vegetation and soils types with local nomenclature. Based on this information, we selected 123 Biodivers Conserv (2010) 19:1933–1961 1937 areas of secondary vegetation in various successional stages in areas recognized by local residents as anthropogenic soils (TPI) and non-anthropogenic soils (NAS), in which we established 25 3 10 m plots. Local informants provided us with information on each plot’s land-use history and time of regeneration since the last agricultural use (hereafter ‘‘plot age’’). Given the detailed local knowledge about the extension of anthropogenic soils, its well-defined position in the landscape and the relative scarcity of secondary forests on TPI, we consider that our sampling effort was adequate and site selection was not significantly affected by local farmer’s biases (such as choosing TPI areas with obviously different secondary forests to please the researchers). All plots were established in secondary forests following shifting cultivation, which is practiced by local residents in a radius of 2–3 km from the community centers. The intensity of previous land-use was highly variable among plots, both on NAS and on TPI. Independently of the soil, some areas are more intensively used than others, which may be due to many different factors, e.g., because they are closer to habitation areas or to sources of water. Nevertheless, we assumed that this heterogeneity in previous land-use is more or less evenly distributed among TPI and NAS plots. Also, even during the fallow period, secondary forests both on TPI and on NAS are subject to some kind of management, such as the extraction of firewood, fruits and medicinals, as well as the elimination of undesired species (spiny lianas, for example) and encouragement of others (useful and/or domesticated species), all of which may alter the course of secondary succession (Denevan and Padoch 1987; Irvine 1989). However, some characteristics of previous and present land use that might affect successional patterns are related to the type of soil [e.g., the more intensive land use that originated TPI patches in pre-Columbian times (Denevan 2001), or the generally shorter fallow period associated with TPI in present shifting cultivation (Fraser et al. 2009)]. We sampled plots in secondary forests with ages varying from five to 30 years, and we established about the same number of plots in areas recognized locally as TPI and NAS. At Água Azul and Terra Preta do Atininga we established 16 plots on each (eight on TPI and eight on NAS), and at Barreira do Capanã we established 20 plots (12 on TPI and eight on NAS), totaling 52 plots sampled (1.3 ha). Although we pre-classified each plot as TPI or NAS, the delimitation of the two treatments used in all statistical analysis was done a posteriori, using results from physical and chemical soil analysis and observation of field characteristics. At Água Azul, plots were located from 30 to 1,390 m from each other; at Barreira do Capanã, from 50 to 2,330 m; and at Terra Preta do Atininga, from 50 to 2,000 m. Due to the occurrence of TPI in relatively small patches (compared to surrounding NAS) and to its intensive use for living and for shifting cultivation, secondary forests on TPI were not abundant on the landscape, which imposed restrictions on plot distribution and lead to some plots quite close to one another. However, we consider these plots as independent samples since they are located in secondary forests with different ages, which makes them easily distinguishable. Also, we did spatial autocorrelation analyses to account for possible effects of distance on floristic composition. Data collection In each plot we sampled (1) all woody individuals with diameter at breast height (DBH) greater than 5 cm (including lianas), and (2) all palms more than 1 m high and with DBH smaller than 5 cm (hereafter understory palms). Many of the understory palms are seedlings or juveniles of species sampled as woody individuals when their DBH was 123 1938 Biodivers Conserv (2010) 19:1933–1961 appropriate. All woody individuals had their DBH measured. All individuals sampled (both woody and understory palms) were collected for botanical identification, which was done with the aid of specialists, taxonomic keys, identification guides, and by comparing the vouchers collected to specimens deposited at the INPA herbarium (Manaus, Brazil). Fertile specimens were deposited at the INPA herbarium, and non-fertile material was identified where possible and stored in an adjacent working collection, given the herbarium policy of not accepting non-fertile material. In each plot we collected five soil samples (the top 20 cm after removing the litter layer), obtained every 5 m along the long axis of the plot, which were bulked to produce a composite soil sample of each plot. Samples were air-dried, cleaned of roots and sieved through a 2 mm sieve. Analysis included soil texture (percentage of sand, clay and silt), pH (H2O and KCl), macronutrients (P, K, Ca, Mg) and micronutrients (Mn, Zn and Fe). All analyses were done at the LTSP/INPA, according to EMBRAPA (1999) methodology. Data analysis The results of the soil analyses were analyzed with a Principal Component Analysis (PCA) biplot. Data expressed in percentages (amount of sand, silt and clay) were transformed by the arcsine of the square root divided by 100 (Sokal and Rohlf 1981). PCAs were done separately for each community, since TPI originates from the same type of soil as the surrounding soils, which therefore should be used as a reference for TPI identification. Plots established in areas recognized locally as non-anthropogenic soil were included in the treatment NAS. Plots established in areas recognized as anthropogenic soils were included in the treatment TPI when they met the following criteria: presence of field characteristics usually associated with areas of anthropogenic soils (e.g., thick modified A0 horizon, potsherds, color darker than surrounding and subsurface soils); and levels of phosphorous and/or calcium higher than the mean plus two times the standard deviation of the NAS plots’ levels of these nutrients. Two of the 28 plots (7.1%) recognized locally as anthropogenic soils did not meet these criteria and were classified as NAS. Consequently, 26 plots (0.65 ha) were classified as NAS and 26 plots (0.65 ha) were classified as TPI. Vegetation structure was compared between the treatments through analysis of covariance (ANCOVA), using density of individuals, richness and basal area as dependent variables, and the time of regeneration (estimated secondary forest age) as a covariate. To control the effect of the density of individuals on species richness, we used a rarefaction method: the dataset of each plot was re-sampled 10,000 times, generating the expected number of species in a small collection of ten individuals (for understory palms) or 22 individuals (for woody individuals) drawn at random from the total pool of individuals of the plot (Gotelli and Colwell 2001). Floristic composition data were ordered through the multivariate analysis Non-metric Multidimensional Scaling (NMDS). Data were transformed using Wisconsin double standardization, where species are first standardized by the maximum and then sites are standardized by site totals. This standardization is generally found to improve the results of NMDS applied to community data (Oksanen et al. 2009). To calculate the biological dissimilarity matrix we used the Chao Jaccard estimator (Chao et al. 2005). The floristic differences between the two treatments were tested with non-parametric Multivariate Analysis of Variance (non-parametric MANOVA). To investigate the effects of spatial distance on floristic composition, we did an analysis of spatial autocorrelation. We performed Mantel tests with 10,000 randomizations to test if the biological dissimilarity matrix was correlated to the Euclidean distance between plots. 123 Biodivers Conserv (2010) 19:1933–1961 1939 The identification of TPI and NAS indicator species was done with the Dufrene and Legendre (1997) Indicator Species Analysis. This analysis combines information on species abundance in a given group and the ‘‘loyalty’’ of occurrence of a species in a given group. The analysis generates Indicator Values (IV) of each species in each group (TPI and NAS); its statistical significance was tested with the Monte Carlo technique, with 10,000 randomizations. To investigate if secondary forests on anthropogenic soils concentrate agrobiodiversity, we classified the species with domesticated populations in the plots following Clement (1999). The treatments were compared for density of individuals and richness of species with domesticated populations in the plots using Mann–Whitney tests. PCA were done with XLSTAT (Fahmy and Aubry 2003), ANCOVA and Mann– Whitney tests with SYSTAT 10.2 (Wilkinson 2002), rarefactions with Ecosim 7.0 (Gotelli and Entsminger 2001), Indicator Species Analysis with PC-ORD (McCune and Mefford 1999), and NMDS, Mantel tests and the non-parametric MANOVA with R Statistical Software (R Development Core Team 2008), using the vegan package (Oksanen et al. 2009). Results Species diversity We sampled 1953 woody individuals (858 on TPI, 1095 on NAS) belonging to 52 families and 279 species (Appendix, Table 3). 171 species were found on TPI and 174 on NAS. 66 species (23.6%) were found in both soils. Families with the highest number of species include Fabaceae (Faboideae: 21 spp; Mimosoideae: 21 spp.), Moraceae (20 spp.), Malvaceae (15 spp.) and Annonaceae (13 spp.). Apeiba tibourbou (Malvaceae), Cochlospermum orinocense (Cochlospermaceae) and Virola theiodora (Myristicaceae) were the most frequent species found on TPI, occurring in 50, 50 and 38.5% of the plots, respectively. Cochlospermum orinocense, Myrcia aff. aliena (Myrtaceae) and Casearia grandiflora (Salicaceae) were the most frequent species found on NAS, occurring in 50, 34.6 and 34.6% of the plots, respectively. 149 species (53.4% of the total) were found in only one plot. Among understory palms, we sampled 1328 individuals (550 on TPI, 778 on NAS) belonging to 25 species (Appendix, Table 4). 16 species were sampled on TPI and 21 on NAS. 12 species (46.1%) were sampled in both soils. Astrocaryum murumuru, Attalea cf. attaleoides and Euterpe precatoria were the most frequent species on TPI, occurring in 57.7, 57.7 and 46.2% of the plots, respectively. Astrocaryum aculeatum, Astrocaryum gynacanthum and Attalea cf. attaleoides were the most frequent species found on NAS, occurring in 92.3, 84.6 and 61.5% of the plots, respectively. Six species (24% of the total) were found in only one plot. Forest structure—woody individuals with DBH C5 cm The density of woody individuals was significantly related to the treatments TPI and NAS (ANCOVA; F = 14.031; P \ 0.001), to plot age (ANCOVA; F = 7.357; P = 0.009), and there was a significant interaction between these two independent variables (ANCOVA; F = 5.111; P = 0.028). The interaction shows that the effect of plot age on the density of wood individuals is different on each soil, which becomes evident in the linear regressions 123 1940 Biodivers Conserv (2010) 19:1933–1961 Fig. 2 Regressions between density (a, b), species richness (c, d) and basal area (e, f) of woody individuals with DBH C5 cm with secondary forest age in 26 plots on anthropogenic soils (TPI) and 26 plots on nonanthropogenic soils (NAS). 1Estimated species richness for 22 individuals through rarefaction made separately for each soil: on TPI, there was no significant relationship between the density of woody individuals and plot age (F = 0.112; P = 0.741; Fig. 2a), while on NAS the density of woody individuals was negatively correlated with plot age (r2 = 0.293; F = 11.345; P = 0.003; Fig. 2b). This means that secondary forests on TPI in early succession have a lower density of trees than NAS, and this value remains relatively unchanged as the forests age. On the other hand, NAS shows density values higher than TPI in early successional stages, but older secondary forests on NAS tend to show density values equivalent to older secondary forests on TPI. Richness of woody individuals was significantly related to plot age (ANCOVA; F = 28.084; P \ 0.001), but not to the soils (ANCOVA; F = 0.04; P = 0.843), nor was 123 Biodivers Conserv (2010) 19:1933–1961 1941 there interaction between the two independent variables (ANCOVA; F = 0.083; P = 0.775). On both soils, the richness of woody individuals was positively correlated to plot age (TPI: r2 = 0.256; F = 9.278; P = 0.006; Fig. 2c; NAS: r2 = 0.495; F = 25.484; P \ 0.001; Fig. 2d). A single individual of Brazil nut (Bertholletia excelsa, Lecythidaceae) occurred in plot 1 on TPI at Terra Preta do Atininga and a single individual of Guazuma ulmifolia (Ulmaceae) in plot 7 on TPI at Água Azul, representing 79.4 and 68.7% of the plot basal area, respectively. Local informants reported that these individuals were present before the last time the area was cleared for agricultural use and were much older than the plots’ age. These individuals were excluded from basal area analysis. There was a significant relationship between basal area and plot age (ANCOVA; F = 13.003; P = 0.001), but not with soils (ANCOVA; F = 0.037; P = 0.848), nor was there interaction between the two independent variables (ANCOVA; F = 0.432; P = 0.514). This shows that the effect of plot age on the increase in basal area is similar on the two soils, and that the basal area does not differ between TPI and NAS when plot age is controlled, although basal area on TPI is more variable. On both soils, the basal area was positively correlated with plot age, but the effect of plot age was weak, as indicated by the low r2 values (TPI: r2 = 0,183; F = 6,586; P = 0,017; Fig. 2e; NAS: r2 = 0,213; F = 7,769; P = 0,010; Fig. 2f). Forest structure—understory palms We found no significant relationship between the density of individuals of understory palms and the soils (ANCOVA; F = 2.241; P = 0.141), nor with plot age (ANCOVA; F = 1.877; P = 0.177), nor was there an interaction between the soils and plot age (ANCOVA; F = 0.170; P = 0.682). The same pattern was found for understory palms’ richness: we found no significant relationship between species richness and the soils (ANCOVA; F = 0.001; P = 0.970), nor with plot age (ANCOVA; F = 1.922; P = 0.173), nor was there an interaction between the soils and plot age (ANCOVA; F = 0.0; P = 1). These results show that there is no difference between TPI and NAS with respect to understory palms’ density of individuals and species richness, and that there is no effect of plot age on these variables. Floristic composition For woody individuals the NMDS ordination accounted for 30.2% of the variation in the data set. There is a statistically significant difference between the soils with respect to floristic composition (Fig. 3a; NPMANOVA: F = 3.13; P \ 0.001). A much clearer pattern was found for the understory palms (Fig. 3b): the variation explained by the model was 56.4% and the soils differed significantly in floristic composition (Fig. 3b; NPMANOVA: F = 13.43; P \ 0.001). Floristic composition was significantly correlated to the distance between plots at Água Azul for woody individuals (Mantel statistic r = 0.202, P = 0.028), but not for understory palms; at Barreira do Capanã, it was correlated both for woody individuals (Mantel statistic r = 0.353, P \ 0.001) and for understory palms (Mantel statistic r = 0.411, P = 0.002); and at Terra Preta do Atininga, it was correlated both for woody individuals (Mantel statistic r = 0.431, P \ 0.001) and for understory palms (Mantel statistic r = 0.328, P = 0.003). Although distance might be an important predictor of similarity in species composition in some cases, our analyses show a weak relationship between these two variables here (indicated by the low r values). This weak relationship may be due to the 123 1942 Biodivers Conserv (2010) 19:1933–1961 Fig. 3 Non-metric multidimensional scaling (NMDS) with two axes shown for a the community of woody individuals with DBH C5 cm and b the community of understory palms (B1 m high) sampled in secondary forest plots on anthropogenic soils (TPI) and non-anthropogenic soils (NAS) (percentage explained by the bidimensional model: a—30.2%; b—56.4%). Each point represents a plot, and its position in the graph is a bidimensional representation of the biological similarity matrix between the plots, obtained with the Chao Jaccard estimator patchy distribution of TPI, which is an intrinsic characteristic of TPI that obliges plots to be close together within a given community and thus subject to similar seed influx from surrounding vegetation, as well as similar genetic resources introduced by members of the communities. Nonetheless, the autocorrelations are quite small, which suggests that differences in other parameters, e.g., soils, are more important. TPI and NAS indicator species Among the woody species sampled, seven were TPI indicator species, and nine were NAS indicator species (Table 1). For understory palms, six species were TPI indicator species and three were NAS indicator species (Table 2). Three palm species [murumuru (Astrocaryum murumuru), urucuri (Attalea cf. phalerata) and caiaue´ (Elaeis oleifera)] were TPI indicator species both for woody individuals and for understory palms. Species with domesticated populations on TPI and NAS According to the classification of Clement (1999), 14 of the 280 woody species and four of the 26 understory palm species are considered to contain populations with some degree of domestication at the time of the European conquest of Amazonia (Appendix, Tables 3 and 4). For woody individuals, TPI showed higher density of individuals (Mann–Whitney U = 169.0; P = 0.002) and richness (U = 202.5; P = 0.009) of species with domesticated populations when compared to NAS. For understory palms, we found no difference between TPI and NAS with respect to density of individuals (U = 339.0; P = 0.985) and richness (U = 319.5; P = 0.691) of species with domesticated populations. Among the seven woody species that are TPI indicator species, three contain populations domesticated in some degree, as well as two of the six understory palms (Tables 1 and 2). Caiaue´, a species strictly associated with anthropogenic soils throughout Amazonia (Andrade 1983; Barcelos 1986; Balée and Gély 1989; McCann 2003), and murumuru, a palm with incipiently domesticated populations associated with archaeological sites (Morcote-Rı́os and Bernal 2001), were TPI indicator species both for woody individuals and for understory palms. One species with semi-domesticated populations, tucumã (Astrocaryum aculeatum), was a NAS indicator species among the understory palms, and 123 Biodivers Conserv (2010) 19:1933–1961 1943 Table 1 Indicator species of woody individuals with diameter at breast height (DBH) C5 cm sampled in secondary forests plots on anthropogenic (TPI) and non-anthropogenic soils (NAS) at three communities along the middle Madeira River, Amazonas, Brazil Soil Species Family Common name IV observed IV from randomized groups p DD Average SD TPI Apeiba tibourbou Malvaceae Oicima 40.5 22.2 5.5 0.005 – Astrocaryum murumuru Arecaceae Murumuru 30.8 12.8 4.3 0.004 id Spondias mombin Anacardiaceae Taperebá 26.9 11.6 4.1 0.010 sd Eugenia muricata Myrtaceae Araçá-branco 26.9 11.5 4.1 0.010 – Elaeis oleifera Arecaceae Caiaué 24.2 12.8 4.2 0.024 d Attalea cf. phalerata Arecaceae Urucuri 23.1 10.6 3.9 0.024 – NAS Myrcia aff. aliena Myrtaceae Murtinha 34.6 14.1 4.6 0.001 – Casearia grandiflora Celastraceae Cachimbeira 34.6 13.9 4.4 0.002 – Vismia cayennensis Clusiaceae Lacre-branco 30.8 12.8 4.3 0.004 – Miconia cf. poeppigii Melastomataceae Sapateiro, Cuandú 30.8 13.9 4.9 0.005 – 26.1 13.7 4.7 0.030 – Lacmellea gracilis Apocynaceae Tucujá, Jacataca, Sorvinha Bellucia grossularioides Melastomataceae Goiaba-de-anta, Papa-terra 23.1 11.8 4.2 0.022 – Maximiliana maripa Arecaceae Inajá 19.2 9.1 3.3 0.048 id Cecropia aff. ulei Urticaceae Embaúba, Embaúcabranca 19.2 9.6 3.8 0.050 – IV indicator value, SD Standard deviation, p Proportion of 10,000 randomized trials with IV equal to or exceeding the observed indicator value, DD Degree of domestication sensu Clement (1999), id incipiently domesticated, sd semi-domesticated one species with incipiently domesticated populations, inajá (Maximiliana maripa), was a NAS indicator species among woody individuals. Discussion Our study confirms previous observations that secondary succession on anthropogenic soils is different from that on non-anthropogenic soils (Major et al. 2003; Clement et al. 2003; Major et al. 2005), leading to the formation of secondary forests that are structurally and floristically different from secondary forests on non-anthropogenic soils. Forest structure differences and similarities Density of individuals on NAS declined with plot age while TPI maintained constant density over time; this was the only structural difference we found between these two 123 1944 Biodivers Conserv (2010) 19:1933–1961 Table 2 Indicator species of understory palms B1 m high sampled in secondary forests plots on anthropogenic (TPI) and non-anthropogenic soils (NAS) at three communities along the middle Madeira River, Amazonas, Brazil Soil Species Common name IV observed IV from randomized groups Average TPI p DD SD Astrocaryum murumuru Murumuru 57 21.9 5.4 0.000 id Elaeis oleifera Caiaué 36.9 16.6 5.1 0.002 id Oenocarpus minor Bacabinha 34.6 14.2 4.6 0.001 – Attalea cf. phalerata Urucuri 28.7 17.1 5.3 0.045 – Bactris concinna Marajá, Marajá-preto 24.5 14.2 4.6 0.048 – Marajazinho 21.2 11.6 3.9 0.037 – Tucumã 81.9 35.5 5.4 0.000 sd Astrocaryum gynacanthum Mumbaca 80.6 31.8 5.5 0.000 – Tucumã-ı́ 26.9 12.4 4.4 0.010 – Geonoma deversa NAS Astrocaryum aculeatum Astrocaryum acaule IV indicator value, SD Standard deviation, p Proportion of 10,000 randomized trials with IV equal to or exceeding the observed indicator value, DD Degree of domestication sensu Clement (1999), id incipiently domesticated, sd semi-domesticated environments. Density is a rather unpredictable parameter in secondary succession, since it is affected by factors that operate at different spatial and temporal scales, and that may have different effects on different size classes (Chazdon et al. 2007). Density of individuals tends to be higher in secondary than in old-growth forests (Guariguata and Ostertag 2001; Chazdon et al. 2007), but during the first years or decades after abandonment it may increase to an asymptote (Aide et al. 1995), show an intermediate peak (Aide et al. 2000; Feldpausch et al. 2005), or may not be correlated with forest age (Peña-Claros 2003). Our findings suggest that anthropogenic soils may show specific patterns of density change with time, which may be due to their higher fertility. Soil fertility may influence the assemblage of colonizing species (Chazdon 2003), and different species may respond differently to nutrient availability (Gehring et al. 1999). Species with high growth rates tend to be disproportionately favored by ample resource levels, leading to their overdominance during early succession (Chapin et al. 1986). On high fertility soils the abundance of shortlived species tends to be higher than on low-fertility soils (Tilman 1987; Dieleman et al. 2000). Major et al. (2005) showed that the abundance of annual herbaceous species was higher on anthropogenic soils than on adjacent soils. These intrinsic species life-history differences may drive successional changes in density of individuals (Chazdon et al. 2007). Unlike density, trends in species richness and basal area were similar on anthropogenic and non-anthropogenic soils, showing a linear increase with time. Basal area (which is strongly correlated to aboveground biomass; Brown 1997) and species richness follow more predictable patterns in secondary succession (Chazdon et al. 2007). In general, species richness increases with time, and may reach levels similar to old-growth forests in a relatively short time (Finegan 1996; DeWalt et al. 2003; Capers et al. 2005). Increment of basal area (or biomass) with time follows a similar pattern in tropical secondary succession (see reviews by Guariguata and Ostertag 2001; Chazdon et al. 2007). However, both species accumulation and biomass increment with time are strongly influenced by soil fertility and previous land-use (Moran et al. 2000; Pascarella et al. 2000; Gehring et al. 2005). Several studies have shown that richness diminishes with an increase in resource 123 Biodivers Conserv (2010) 19:1933–1961 1945 availability and productivity (e.g., Waide et al. 1999; Mittelbach et al. 2001; Stevens and Carson 2002), while others show an unimodal distribution across a productivity gradient, where richness is higher in intermediate levels of productivity (Grime 1979; Stevens and Carson 1999). Soil fertility can also positively affect biomass accumulation (Gehring et al. 1999; Laurance et al. 1999; Castilho et al. 2006), and basal area (Chinea 2002). Some of the characteristics of anthropogenic soils (e.g., higher cation exchange capacity, higher pH, lower Al saturation; Lehmann et al. 2003a) permit its intensive agricultural use (Major et al. 2005). In general, swidden duration is longer and fallow time is shorter than on non-anthropogenic soils (German 2003b; Fraser and Clement 2008; Fraser et al. 2009). Sites that have been under more intensive agricultural use tend to accumulate species and biomass slower than ‘‘light-use’’ sites (Uhl et al. 1988; Aide et al. 1995; Fearnside and Guimarães 1996), even when land use differences are subtle (Gehring et al. 2005). We would, therefore, expect anthropogenic soils to show different successional patterns regarding richness and basal area increment with time, since these soils are more fertile than adjacent soils and are usually associated with more intensive land use. However, despite their generally higher fertility, anthropogenic soils do not necessarily have all the nutrients required for plant growth. Some nutrients in anthropogenic soils show similar or even lower levels than in non-anthropogenic soils, especially potassium, and some nutrients present in higher levels may not be available to plants due to interactions between nutrients (Lehmann et al. 2003b). Also, species in secondary forests that are well adapted to the nutrient-poor soils that predominate in the region may not respond significantly to higher nutrient availability. In addition, the simultaneous and sometimes antagonistic effects of soil fertility and land use may have obscured possible structural differences between secondary forests on anthropogenic and non-anthropogenic soils. Further studies aiming to compare these environments’ vegetation structures should examine the interaction between these two variables. Differences in species composition The main difference we found in secondary succession between anthropogenic and nonanthropogenic soils was in species composition. We found significant floristic differences between TPI and NAS, both for woody individuals and for understory palms. Our results give quantitative support to other studies that have reported, through ethnographic observation, that local residents recognize anthropogenic soils through species composition (Moran 1981; Woods and McCann 1999; Sombroek et al. 2002; German 2003b). The floristic composition of anthropogenic soils should be strongly influenced by longterm human activity in these environments, since the floristic composition of secondary forests may be influenced by the intensity, duration and type of land-use before ‘‘abandonment’’ (see reviews by Guariguata and Ostertag 2001; Chazdon 2003; Chazdon et al. 2007). In Mesoamerica, the long-term human activity and the intensive use and management of the landscape by the pre-Columbian Maya resulted in secondary forests with different floristic composition and high economic and ecological value (Campbell et al. 2006; Ford and Nigh 2009). Different types of land use may result in different assemblages of species that colonize the site (Mesquita et al. 2001), and during succession the vegetation may undergo several types of human intervention that add or eliminate some species according to a desired objective (Denevan and Padoch 1987; Irvine 1989), leading to many possible successional trajectories. 123 1946 Biodivers Conserv (2010) 19:1933–1961 The shorter fallow periods in shifting agriculture on anthropogenic soils represent an increase in the frequency of burning, which may reduce the density of heat-intolerant seeds in the soil seedbank (Ewel et al. 1981; Uhl et al. 1981), reduce the contribution of sprouting to regeneration (Uhl et al. 1982; de Rouw 1993), but also favor the resprouting of some plants with stems and/or roots that are heat-tolerant (Mesquita et al. 2001). Also, the reduction of fallow duration can favor the maintenance of short-lived species in the soil seedbank, as shown by Major et al. (2005) for herbaceous species on anthropogenic soils in Central Amazonia. These factors have most likely influenced the species composition of the later successional stages we sampled. Finally, due to the close association between anthropogenic soils and human settlements, they must have been (and still are) areas of more intensive agroforestry practices, resulting in secondary forests with a high concentration of useful species, many with domesticated populations. The role of secondary forests on anthropogenic soils in concentrating agrobiodiversity Owing to the fact that anthropogenic soils are strictly associated with past agricultural and settlement areas, they were probably environments where pre-Columbian people and later residents concentrated plant resources, as still happens today in homegardens and in fallows of shifting agriculture (Denevan and Padoch 1987; Irvine 1989; Denevan 2001; Kumar and Nair 2004). In these environments, the introduction of species with distinct degrees of domestication and the use of various forms of vegetation management (selective weeding, transplanting, etc.) may cause modifications in the course of natural succession, and these effects may persist for a long time. Long-lived species, such as Brazil nut (Bertholletia excelsa) that contains incipiently domesticated populations, or understory trees, such as cocoa (Theobroma cacao) that contains semi-domesticated populations, may persist indefinitely in the secondary forest, and in some cases species may persist for a considerable time as sprouts or seeds (Clement et al. 2003). Balée and Campbell (1990) determined that some species with incipiently domesticated populations (e.g., Bertholletia excelsa, Maximiliana maripa, Theobroma speciosum) are among the most important in liana-dominated anthropogenic forests close to TPI patches in the Xingu River basin. We found that secondary forests on anthropogenic soils concentrate species with incipiently or semi-domesticated populations, many of which are commonly associated with archaeological sites and anthropogenic soils in Amazonia (Balée and Gély 1989; Clement et al. 2003). All of the nine indicator species of anthropogenic soils are useful species (see Tables 1 and 2), three of which contain semi-domesticated populations [Astrocaryum murumuru (reported in archaeological sites as early as 9,000 years before present—yBP; Morcote-Rı́os and Bernal 2001), Elaeis oleifera (reported as early as 6,000 yBP (idem) and believed to have been distributed in Amazonia by humans; Balée and Gély 1989) and Spondias mombin], and two of which are widely used for food, fiber and other purposes by traditional populations now and certainly in the past (Oenocarpus minor, Attalea cf. phalerata). These observations suggest that the current species composition of secondary forests on anthropogenic soils represent a continuum between the introduction of useful species (many with domesticated populations) in the past and their maintenance by modern management and encouragement practices. In fact, several of our informants recall high concentrations of useful species, including with domesticated populations, on the TPI sites even before the communities were established at their present sites. It is worth noting, however, that two NAS indicator species, Astrocaryum aculeatum and Maximiliana 123 Biodivers Conserv (2010) 19:1933–1961 1947 maripa, contain semi-domesticated and incipiently domesticated populations, respectively. This reinforces the idea that secondary succession on NAS is also subject to considerable human agency [as previously noted by Denevan and Padoch (1987) and Irvine (1989) on secondary forests on non-anthropogenic soils]. Nevertheless, our results support a much more intensive human influence in secondary succession on anthropogenic soils than on adjacent non-anthropogenic soils. Our findings contrast with those of Paz-Rivera and Putz (2009), who found no difference in the density of 17 useful tree species between anthropogenic soils and nonanthropogenic soils in a much older (at least 140 years old) lowland forest in Bolivia. Although Paz-Rivera and Putz (2009) selected their 17 species based on their being edible or otherwise useful, it is possible that their subsample of the total species pool is insufficient to identify previous human intervention, since it may include species that, despite being useful, were not useful enough to justify a type of intensive management that favored their occurrence. Instead, we chose to identify possible human intervention in the vegetation by looking at species with domesticated populations that where cultivated in Amazonia at the time of European conquest (Clement 1999). These are species that are much more intensively manipulated and therefore they bring more relevant information to the question of how secondary forests on TPI might have been affected by human agency [only six of the 17 species selected by Paz-Rivera and Putz (2009) were species with domesticated populations according to Clement (1999)]. Also, adjacent nonanthropogenic areas are subjected to management and introduction of useful species (Paz-Rivera and Putz 2009), but the management focused on species with domesticated populations probably occurred more intensively on TPI than on NAS, since the former originated from pre-Columbian homegardens and swiddens, where indigenous experimentation is more evident (Clement et al. 2003; Hiraoka et al. 2003). Although we did not sample secondary forests on TPI older than 30 years, it is probable that the differences between TPI and NAS in terms of species composition will persist in later successional stages. Samuel Almeida and colleagues (Clement et al. 2009) sampled trees with DBH C10 cm in a one-hectare plot in an old secondary forest [at least 300 years old (Kern 1996)] on TPI at Caxiuanã (Pará, Brazil) and compared the species richness and abundance to another one-hectare plot in a mature forest on an adjacent nonanthropogenic soil. They found that, although the density of individuals was similar between TPI and NAS, there were clear differences regarding species richness (39.7% higher on NAS than on TPI) and composition (the similarity between the sites was only 6.9%). These results support the idea that human-induced modifications in secondary succession may persist for a long time, and are likely to be detected in old-growth secondary forests on TPI, even though natural processes could mask these differences as vegetation ages (Paz-Rivera and Putz 2009). We suggest further studies to compare TPI and NAS regarding richness and density of domesticated species in older secondary forests to examine the durability and dynamics of the human-induced modifications in the secondary succession of these forests. Although Posey’s (1984) widely cited example of actively managed ‘‘forest resource islands’’ among the Kayapó Indians has drawn sharp criticism (Parker 1992; see also Posey 1992), our results suggest similar processes shaping the introduction (whether active or passive) of useful species, many with domesticated populations, in TPI secondary forest sites in a different part of Amazonia. This result likewise supports Clement et al. (2003), who hypothesized that anthropogenic soils could act as agrobiodiversity reservoirs through the concentration and maintenance of genetic resources of native and exotic plant species with domesticated populations. 123 1948 Biodivers Conserv (2010) 19:1933–1961 Conclusions Although anthropogenic soils are widely distributed in Amazonia and emerged in different environmental, cultural and socioeconomic contexts, they share an intense association with human activity in the past and, frequently, with specific forms of current use and management. This is most noticeable in terms of the floristic composition of the secondary forests on anthropogenic soils, which is clearly distinct from secondary forests on adjacent nonanthropogenic soils. The total extent of anthropogenic soils in Amazonia is still unknown (Erickson 2003), with estimates varying from 0.1 to 0.3% (Sombroek et al. 2003) of the total area of the Amazon basin. The relevance of these areas for understanding settlement and agricultural patterns in pre-Columbian Amazonia has been increasingly recognized (Neves et al. 2003; Woods and Denevan 2009). Our study shows that anthropogenic soils also maintain unique secondary forests and concentrate species with domesticated populations, thereby concentrating agrobiodiversity. Hence, they offer advantages for in situ conservation of genetic resources, and should to be considered in conservation efforts since they aggregate heterogeneity and biodiversity in Amazonian landscapes. Acknowledgments Our sincere thanks for the invaluable help and kindness of the residents of Água Azul, Barreira do Capanã and Terra Preta do Atininga, especially to the field assistants Raimundo Nonato Soares Barros, Silvestre Arcanjo de Souza, José Rodrigues de Souza and Raimundo Furtado Neto. Thanks also to a number of people at INPA who helped with plant identifications, especially José Ferreira Ramos and Paulo Apóstolo. André Braga Junqueira received a graduate scholarship from the Brazilian National Research Council (CNPq), and this work is part of his master’s dissertation at INPA. The Brazilian International Institute of Education (IEB)/The Moore Foundation provided financial support and IdeaWild provided equipment for fieldwork. Tânia Pimentel and the staff from the LTSP/INPA provided support with the laboratorial soil analyses. We thank Ana Catarina Jakovac, James A. Fraser, Bruce Walker Nelson, Laura German, Lin Chau Ming and William Balée for suggestions to improve earlier versions of this manuscript. Charles R. Clement is a fellow of CNPq. Appendix: Secondary forest species abundance and frequency See Tables 3 and 4. Table 3 List of the woody species found on 52 secondary forest plots at three communities along the middle Madeira River, Amazonas, Brazil Family Species Number of individuals DD3 Frequency TPI1 NAS2 Total TPI1 NAS2 Total Achariaceae Lindackeria paludosa (Benth.) Gilg 12 8 20 4 7 Anacardiaceae Anacardium occidentale L. 1 0 1 1 0 1 Astronium lecointei Ducke 0 1 1 0 1 1 Mangifera indica L. 1 0 1 1 0 1 Spondias mombin L. 11 0 11 7 0 7 Tapirira guianensis Aubl. 8 23 31 5 7 12 Thyrsodium spruceanum Benth. 0 3 3 0 3 3 123 11 sd sd Biodivers Conserv (2010) 19:1933–1961 1949 Table 3 continued Family Species Number of individuals DD3 Frequency TPI1 NAS2 Total TPI1 NAS2 Total Annonaceae Apocynaceae Annona foetida Mart. 1 0 1 1 0 Diclinanona calycina (Diels) R. E. Fr. 0 1 1 0 1 1 1 Ephedranthus amazonicus R.E. Fr. 1 0 1 1 0 1 1 Guatteria cf. scytophylla Diels 0 1 1 0 1 Guatteria foliosa Benth. 0 2 2 0 2 2 Guatteria megalophylla Diels 0 1 1 0 1 1 Guatteria olivacea R. E. Fr. 0 7 7 0 2 2 Guatteria pteropus Benth. 5 3 8 4 3 7 Guatteriella sp. 1 0 6 6 0 4 4 Guatteriopsis sp. 1 1 0 1 1 0 1 Rollinia cuspidata Mart. 2 1 3 2 1 3 Rollinia exsucca (DC. Ex Dunal) A. DC. 6 3 9 2 2 4 Xylopia aromatica (Lam.) Mart. 2 19 21 2 5 7 Apocynaceae sp. 1 1 0 1 1 0 1 Himatanthus cf. sucuuba (Spruce ex Müll. Arg.) Woodson 3 4 7 2 3 5 Himatanthus aff. stenophyllus Plumel 2 2 4 2 2 4 Himatanthus sp. 1 1 0 1 1 0 1 Lacmellea gracilis (Müll. Arg.) Markgr. 1 30 31 1 7 8 Rauvolfia sprucei Müll. Arg. 1 2 3 1 2 3 Tabernaemontana cymosa Jacq. 0 2 2 0 2 2 Araliaceae Schefflera morototoni (Aubl.) Maguire, Steyerm. & Frodin 6 8 14 3 8 11 Arecaceae Astrocaryum aculeatum Meyer 7 13 20 3 8 11 sd Astrocaryum murumuru Mart. 15 0 15 8 0 8 id Attalea cf. attaleoides (Barb. Rodr.) Wess. Boer 2 3 5 2 3 5 Attalea cf. speciosa Mart. ex Spreng. 3 4 7 2 2 4 Attalea cf. phalerata Mart. ex Spreng. 10 0 10 6 0 6 Elaeis oleifera (Kunth.) Cortés 18 2 20 7 1 8 Euterpe precatoria Mart. 13 0 13 2 0 2 Iriartella setigera (Mart.) H. Wendl. 2 0 2 2 0 2 Maximiliana maripa (Aubl.) Drude 0 5 5 0 5 5 Oenocarpus minor Mart. 2 0 2 2 0 2 Asteraceae Vernonia scabra Pers. 2 0 2 2 0 2 Bignoniaceae Adenocalymma sp. 1 0 1 1 0 1 Arrabidaea brachypoda (DC.) Bureau 2 0 2 2 0 2 39 14 53 4 7 11 Lundia densiflora DC. 2 0 2 1 0 1 Scobinaria sp. 1 2 0 2 1 0 1 Tabebuia sp. 1 0 1 1 0 1 1 Jacaranda copaia (Aubl.) D. Don id id 123 1950 Biodivers Conserv (2010) 19:1933–1961 Table 3 continued Family Species Number of individuals DD3 Frequency TPI1 NAS2 Total TPI1 NAS2 Total Boraginaceae Burseraceae Cordia scabrifolia A. DC. 4 1 5 2 1 Cordia sellowiana Cham. 1 9 10 1 6 3 7 Cordia sprucei Mez 0 1 1 0 1 1 Crepidospermum goudotianum (Tul.) Triana & Planch. 2 0 2 1 0 1 Crepidospermum rhoifolium (Benth.) Triana & Planch. 0 4 4 0 3 3 Protium nitidifolium (Cuatrec.) D. C. Daly 0 1 1 0 1 1 Protium robustum (Swart) D.M. Porter 1 0 1 1 0 1 Tetragastris panamensis (Engl.) Kuntze 1 0 1 1 0 1 Trattinnickia burserifolia Mart. 0 2 2 0 2 2 Trattinnickia glaziovii Swart 0 5 5 0 2 2 4 Trattinnickia peruviana Loes. 1 5 6 1 3 Trattinnickia rhoifolia Willd. 1 1 2 1 1 2 Cannabaceae Trema micrantha Blume 1 0 1 1 0 1 Caricaceae Jacaratia aff. digitata (Poepp. & Endl.) Solms 1 0 1 1 0 1 Chrysobalanaceae Hirtella elongata Mart. & Zucc. 2 Clusiaceae 4 0 4 2 0 Licania gracilipes Taub. 0 1 1 0 1 1 Licania micrantha Miq. 0 1 1 0 1 1 Licania sp. 1 1 0 1 1 0 1 Rheedia macrophylla (Mart.) Planch. & Triana 1 0 1 1 0 1 Symphonia globulifera L. f. 1 0 1 1 0 1 Vismia aff. gracilis Hieron. 2 0 2 1 0 1 Vismia cayennensis (Jacq.) Pers. 0 27 27 0 8 8 Vismia gracilis Hieron 0 7 7 0 3 3 Vismia japurensis Reichardt 0 1 1 0 1 1 27 9 36 6 2 8 Vismia sandwithii Ewan Vismia sp. 1 Cochlospermaceae Cochlospermum orinocense (Kunth) Steud. 0 2 2 0 2 2 81 175 256 13 13 26 Dilleniaceae Doliocarpus sp. 1 0 2 2 0 1 1 Euphorbiaceae Acalypha brasiliensis Müll. Arg. 1 0 1 1 0 1 Acalypha macrostachya Jacq. 0 1 1 0 1 1 Alchorneopsis floribunda (Benth.) Müll. Arg. 0 1 1 0 1 1 123 Aparisthmium cordatum (Juss.) Baill. 0 32 32 0 5 5 Chaetocarpus cf. schomburgkianus (Kuntze) Pax & K. Hoffm. 0 2 2 0 1 1 Conceveiba guianensis Aubl. 0 1 1 0 1 1 id Biodivers Conserv (2010) 19:1933–1961 1951 Table 3 continued Family Species Number of individuals DD3 Frequency TPI1 NAS2 Total TPI1 NAS2 Total Croton aff. sampatik Müll. Arg. 1 0 1 1 0 1 Croton palanostigma Klotzsch 0 15 15 0 3 3 Croton sp. 1 1 0 1 1 0 1 Mabea angustifolia Spruce ex Benth. 0 29 29 0 2 2 1 0 1 1 0 1 1 4 5 1 3 4 Machaerium hoehneanum Ducke 0 1 1 0 1 1 Macrolobium limbatum Spruce ex Benth. 0 1 1 0 1 1 13 5 18 6 2 8 Fabaceae– Bauhinia guianensis Aubl. Caesalpinoideae Bauhinia longicuspis Spruce ex Benth. Schizolobium amazonicum Huber ex Ducke Sclerolobium sp. 1 0 1 1 0 1 1 13 33 46 5 7 12 Tachigali sp. 1 1 0 1 1 0 1 Tachigali sp. 2 0 1 1 0 1 1 Tachigali sp. 3 0 11 11 0 3 3 Alexa grandiflora Ducke 1 0 1 1 0 1 Dalbergia spruceana (Benth.) Benth. 0 1 1 0 1 1 Dialium guianense (Aubl.) Sandwith 1 1 2 1 1 2 Diplotropis cf. triloba Gleason 0 3 3 0 1 1 Senna silvestris (Vell.) H.S. Irwin & Barneby Fabaceae– Faboideae Fabaceae– Mimosoideae Diplotropis sp. 1 1 0 1 1 0 1 Dipteryx odorata (Aubl.) Willd. 1 2 3 1 2 3 Machaerium floribundum Benth. 2 0 2 2 0 2 Mucuna sp. 0 1 1 0 1 1 Ormosia aff. grossa Rudd 2 2 4 1 1 2 Ormosia discolor Spruce ex Benth. 1 0 1 1 0 1 Poecilanthe effusa (Huber) Ducke 1 13 14 1 5 6 Pterocarpus rohrii Vahl 0 1 1 0 1 1 Swartzia apetala Raddi 1 4 5 1 2 3 Swartzia cuspidata Spruce ex Benth. 0 1 1 0 1 1 Swartzia laurifolia Benth. 0 2 2 0 2 2 Swartzia laxiflora Bong. ex Benth. 1 5 6 1 5 6 Swartzia polyphylla DC. 0 1 1 0 1 1 Swartzia sp. 1 0 1 1 0 1 1 Swartzia sp. 2 0 2 2 0 1 1 Swartzia tessmannii Harms 0 1 1 0 1 1 Enterolobium schomburgkii (Benth.) Benth. 1 4 5 1 4 5 Inga alba (Sw.) Willd. 3 2 5 2 2 4 Inga cf. disticha Benth. 21 29 50 5 6 11 Inga cf. edulis Mart. 12 0 12 4 0 4 d 123 1952 Biodivers Conserv (2010) 19:1933–1961 Table 3 continued Family Species Number of individuals DD3 Frequency TPI1 NAS2 Total TPI1 NAS2 Total Inga cf. grandiflora Wall. 6 0 6 2 0 2 Inga cf. ingoides (Rich.) Willd. 1 0 1 1 0 1 Inga cf. lomatophylla (Benth.) Pittier 1 0 1 1 0 1 Inga cf. marginata Willd. 1 0 1 1 0 1 Inga heterophylla Willd. 2 11 13 2 5 7 7 Inga lateriflora Miq. 1 8 9 1 6 24 28 52 6 3 9 Inga macrophylla Humb. & Bonpl. ex Willd. 7 4 11 4 2 6 Inga obidensis Ducke 2 4 6 1 3 4 Inga umbellifera (Vahl) Steud. 2 0 2 2 0 2 Parkia nitida Miq. 0 1 1 0 1 1 Parkia pendula (Willd.) Benth. ex Walp. 0 1 1 0 1 1 Inga longiflora Spruce ex Benth. Samanea saman (Jacq.) Merr. 15 0 15 4 0 4 Stryphnodendron guyanense (Aubl.) Benth. 0 1 1 0 1 1 Stryphnodendron pulcherrimum (Willd.) Hochr. 0 2 2 0 2 2 Vatairea sp. 1 1 0 1 1 0 1 Zygia racemosa (Ducke) Barneby & J. W. Grimes 0 1 1 0 1 1 Goupiaceae Goupia glabra Aubl. 1 10 11 1 5 6 Hippocrateaceae Cheiloclinium cognatum (Miers) A.C. Sm. 1 0 1 1 0 1 Indeterminate Indet Liana 1 1 0 1 1 0 1 Indet Liana 2 1 0 1 1 0 1 Lacistema aggregatum (P. J. Bergius) Rusby 0 1 1 0 1 1 Lacistemaceae Lauraceae Lecythidaceae 123 Endlicheria cf. formosa A. C. Sm. 2 1 3 1 1 2 Mezilaurus itauba (Meisn.) Taub. ex Mez 0 3 3 0 2 2 Ocotea guianensis Aubl. 0 1 1 0 1 1 Ocotea longifolia Kunth 15 7 22 7 4 11 Ocotea oblonga (Meisn.) Mez 0 1 1 0 1 1 Ocotea splendens (Meisn.) Baill. 0 1 1 0 1 1 Persea americana Mill. 1 0 1 1 0 1 Rhodostemonodaphne sp. 0 1 1 0 1 1 Bertholletia excelsa Bonpl. 10 4 14 8 4 12 Couratari stellata A.C. Sm. 0 1 1 0 1 1 Eschweilera atropetiolata S. A. Mori 0 1 1 0 1 1 Eschweilera chartaceifolia S. A. Mori 1 0 1 1 0 1 d id Biodivers Conserv (2010) 19:1933–1961 1953 Table 3 continued Family Species Number of individuals DD3 Frequency TPI1 NAS2 Total TPI1 NAS2 Total Eschweilera gigantea (R. Knuth) J. F. Macbr. 1 0 1 1 0 1 Eschweilera ovalifolia (DC.) Nied. 0 1 1 0 1 1 Eschweilera sp. 1 1 0 1 1 0 1 Eschweilera sp. 2 0 1 1 0 1 1 Eschweilera truncata A. C. Sm. 0 1 1 0 1 1 Gustavia augusta L. 1 0 1 1 0 1 Malpighiaceae Byrsonima crispa A. Juss. 0 1 1 0 1 1 Malvaceae Apeiba membranacea Spruce ex Benth. 2 0 2 2 0 2 Apeiba tibourbou Aubl. 30 7 37 13 3 16 Ceiba pentandra (L.) Gaertn. 1 0 1 1 0 1 Ceiba sp. 1 3 0 3 1 0 1 Eriotheca aff. globosa (Aubl.) A. Robyns 0 1 1 0 1 1 Eriotheca longitubulosa A. Robyns 0 2 2 0 2 2 Guazuma ulmifolia Lam. 1 0 1 1 0 1 Heliocarpus americanus L. 26 0 26 3 0 3 Luehea sp. 1 0 1 1 0 1 Pachira sp. 1 2 0 2 2 0 2 Sterculia aff. apetala (Jacq.) H. Karst. 1 0 1 1 0 1 Sterculia aff. frondosa Rich. 3 0 3 2 0 2 Theobroma cacao L. 1 0 1 1 0 1 Theobroma obovatum Klotzch ex Bernoulli 1 0 1 1 0 1 Theobroma speciosum Willd. Ex Spreng. 0 1 1 0 1 1 Melastomataceae Bellucia acutata Pilg. 0 4 4 0 2 2 Meliaceae Bellucia grossularioides (L.) Triana 0 24 24 0 6 6 Miconia affinis DC. 5 13 18 4 4 8 Miconia cf. pilgeriana Ule 0 1 1 0 1 1 Miconia cf. poeppigii Triana 0 44 44 0 8 8 Miconia cf. tillettii Wurdack 0 3 3 0 3 3 Miconia cuspidata Mart. ex Naudin 0 10 10 0 2 2 25 4 29 6 3 9 Miconia minutiflora (Bonpl.) DC. Menispermaceae Miconia prasina (Sw.) DC. 0 1 1 0 1 1 Miconia symplectocaulos Pilg. 0 1 1 0 1 1 Guarea humaitensis T.D. Penn. 1 0 1 1 0 1 Guarea kunthiana A. Juss. 4 0 4 3 0 3 Trichilia guianensis Klotzsch ex C. DC. 0 1 1 0 1 1 Trichilia micrantha Benth. 1 0 1 1 0 1 Abuta grandifolia (Mart.) Sandwith 1 0 1 1 0 1 sd id 123 1954 Biodivers Conserv (2010) 19:1933–1961 Table 3 continued Family Species Number of individuals DD3 Frequency TPI1 NAS2 Total TPI1 NAS2 Total Miristicaceae Virola theiodora (Spruce ex Benth.) Warb. Monimiaceae Moraceae Myrtaceae 21 8 29 10 6 16 Bracteanthus glycycarpus Ducke 1 0 1 1 0 1 Siparuna cf. depressa Jangoux 0 1 1 0 1 1 Siparuna cf. guianensis Aubl. 0 1 1 0 1 1 Siparuna sarmentosa Perkins 1 0 1 1 0 1 Brosimum acutifolium Huber 0 1 1 0 1 1 Brosimum guianense (Aubl.) Huber 2 1 3 1 1 2 Brosimum lactescens (S. Moore) C. C. Berg 1 0 1 1 0 1 Brosimum sp. 1 0 1 1 0 1 1 Castilla ulei Warb. 1 0 1 1 0 1 Clarisia aff. biflora Ruiz & Pav. 2 0 2 1 0 1 Clarisia ilicifolia (Spreng.) Lanj. & Rossberg 2 0 2 2 0 2 Ficus citrifolia Mill. 2 0 2 2 0 2 Ficus maxima Mill. 1 1 2 1 1 2 Ficus obtusiuscula (Miq.) Miq. 2 1 3 1 1 2 Helianthostylis sprucei Baill. 1 0 1 1 0 1 Helicostylis tomentosa (Poepp. & Endl.) Rusby 1 3 4 1 2 3 Maclura tinctoria (L.) D. Don ex Steud 2 0 2 2 0 2 Maquira calophylla (Poepp. & Endl.) C. C. Berg 1 1 2 1 1 2 Maquira sclerophylla (Ducke) C. C. Berg 1 3 4 1 3 4 Perebea mollis (Poepp. & Endl.) Huber 1 8 9 1 1 2 Pseudolmedia laevis (Ruiz & Pav.) J. F. Macbr. 2 0 2 1 0 1 Sorocea guayanensis W. C. Burger 0 1 1 0 1 1 Sorocea hirtella Mildbr. 1 0 1 1 0 1 Trymatococcus amazonicus Poepp. & Endl. 0 4 4 0 2 2 Eugenia aff. citrifolia Poir. Eugenia muricata DC. 123 0 2 2 0 1 1 14 0 14 7 0 7 Eugenia omissa McVaugh 2 0 2 2 0 2 Myrcia aff. aliena McVaugh 0 87 87 0 9 9 Myrcia aff. huallagae McVaugh 1 6 7 1 2 3 Myrcia aff. paivae O. Berg 1 0 1 1 0 1 Myrcia aliena McVaugh 0 8 8 0 4 4 Myrcia bracteata (Rich.) DC. 7 3 10 2 2 4 Myrcia cf. gigas McVaugh 1 0 1 1 0 1 Biodivers Conserv (2010) 19:1933–1961 1955 Table 3 continued Family Species Number of individuals DD3 Frequency TPI1 NAS2 Total TPI1 NAS2 Total Nyctaginaceae Myrcia servata McVaugh 0 1 1 0 1 Psidium apiculatum Mattos 6 0 6 3 0 1 3 Guapira opposita (Vell.) Reitz 1 0 1 1 0 1 Neea cf. filipes Huber 1 0 1 1 0 1 Neea sp. 1 0 1 1 0 1 1 Neea sp. 2 1 0 1 1 0 1 Ochnaceae Ouratea castaneifolia (DC.) Engl. 1 0 1 1 0 1 Olacaceae Aptandra tubicina (Poepp.). Benth. ex Miers 5 0 5 3 0 3 Heisteria cf. densifrons Engl. 1 0 1 1 0 1 Passifloraceae Passiflora nitida Kunth 0 1 1 0 1 1 Piperaceae Piper aduncum L. 87 0 87 5 0 5 Rubiaceae Duroia sp. 1 4 0 4 1 0 1 Faramea sp. 1 0 2 2 0 1 1 Ferdinandusa cf. rudgeoides (Benth.) Wedd. 0 1 1 0 1 1 Palicourea grandifolia (Willd. ex Roem. & Schult.) Standl. 0 2 2 0 1 1 Palicourea guianensis Aubl. 1 1 2 1 1 2 Uncaria guianensis (Aubl.) J.F. Gmel. 0 1 1 0 1 1 Warszewiczia coccinea (Vahl) Klotzsch 6 2 8 5 2 7 Rutaceae Zanthoxylum rhoifolium Lam. 6 0 6 2 0 2 Salicaceae Casearia aff. grandiflora Cambess. 0 3 3 0 2 2 Casearia cf. pitumba Sleumer 0 2 2 0 1 1 Casearia duckeana Sleumer 0 1 1 0 1 1 Casearia grandiflora Cambess. 0 21 21 0 9 9 Casearia javitensis Kunth 1 9 10 1 4 5 Casearia ulmifolia Vahl ex Vent. 3 2 5 2 1 3 Laetia procera (Poepp.) Eichler 1 0 1 1 0 1 Cupania cf. racemosa (Vell.) Radlk. 4 0 4 3 0 3 Cupania rubiginosa (Poir.) Radlk. 2 1 3 2 1 3 Cupania scrobiculata Rich. 0 2 2 0 1 1 Cupania sp. 1 2 0 2 2 0 2 Pseudima frutescens (Aubl.) Radlk. 2 0 2 2 0 2 Talisia cf. guianensis Aubl. 0 1 1 0 1 1 Toulicia guianensis Aubl. 0 3 3 0 3 3 Ecclinusa lanceolata (Mart. & Eichler) Pierre 0 1 1 0 1 1 Micropholis guyanensis (A. DC.) Pierre 0 1 1 0 1 1 Sapindaceae Sapotaceae Pouteria caimito (Ruiz & Pav.) Radlk. 0 1 1 0 1 1 d Pouteria cf. macrophylla (Lam.) Eyma 0 1 1 0 1 1 sd 123 1956 Biodivers Conserv (2010) 19:1933–1961 Table 3 continued Family Species Number of individuals DD3 Frequency TPI1 NAS2 Total TPI1 NAS2 Total Simaroubaceae Simaba cedron Planch. 1 0 1 1 0 1 Simaba polyphylla (Cavalcante) W. W. Thomas 0 1 1 0 1 1 1 Simarouba cf. amara Aubl. 0 1 1 0 1 Solanaceae Solanum quaesitum C. V. Morton 1 0 1 1 0 1 Strelitziaceae Phenakospermum guyannense (Rich.) Endl. 5 2 7 1 1 2 Ulmaceae Ampelocera edentula Kuhlm. 1 0 1 1 0 1 Celtis iguanaea (Jacq.) Sarg. 2 0 2 1 0 1 Cecropia aff. ficifolia Warb. ex Snethl. 5 13 18 3 8 11 Cecropia aff. purpurascens C.C. Berg 2 0 2 1 0 1 Cecropia aff. ulei Snethl. 0 12 12 0 5 5 Urticaceae Cecropia sciadophylla Mart. 2 7 9 2 4 6 Coussapoa orthoneura Standl. 0 1 1 0 1 1 Coussapoa sp. 1 0 1 1 0 1 1 Pouroma villosa Trécul 0 1 1 0 1 1 Pourouma guianensis Aubl. 16 1 17 5 1 6 Pourouma minor Benoist 1 1 2 1 1 2 Violaceae Leonia crassa L.B. Sm. & A. Fernández 2 0 2 2 0 2 Paypayrola grandiflora Tul. 1 0 1 1 0 1 Vochysiaceae Vochysia aff. maxima Ducke 0 1 1 0 1 1 Vochysia sp. 1 7 0 7 3 0 3 1 Anthropogenic soils 2 Non-anthropogenic soils DD Degree of domestication sensu Clement (1999), id incipiently domesticated, sd semi-domesticated, d domesticated Table 4 List of the understory palm species found on 52 secondary forest plots at three communities along the middle Madeira River, Amazonas, Brazil Species Astrocaryum acaule Mart. Number of individuals Frequency TPI1 TPI1 NAS2 Total NAS2 DD Total 0 55 55 0 7 7 Astrocaryum aculeatum Meyer 14 110 124 6 24 30 Astrocaryum gynacanthum Mart. 18 363 381 4 22 26 Astrocaryum murumuru Mart. Attalea cf. attaleoides (Barb. Rodr.) Wess. Boer Attalea cf. phalerata Mart. ex Spreng. Attalea sp. 1 123 84 1 85 15 1 16 127 118 245 15 16 31 42 4 46 8 4 12 0 38 38 0 4 4 sd id Biodivers Conserv (2010) 19:1933–1961 1957 Table 4 continued Species Number of individuals 1 TPI 2 NAS Total Frequency 1 TPI DD 2 NAS Total Bactris arundinacea Trail 0 1 1 0 1 1 Bactris cf. bifida Mart. 0 1 1 0 1 1 Bactris concinna Mart. 9 30 3 33 7 2 Bactris hirta Mart. 0 2 2 0 1 1 Bactris maraja Mart. 7 9 16 4 4 8 Bactris simplicifrons Mart. 1 6 7 1 5 6 32 0 32 4 0 4 Desmoncus cf. leptospadix Mart. 2 4 6 1 4 5 Desmoncus macroacanthos Mart. 0 2 2 0 2 2 Elaeis oleifera (Kunth) Cortés 71 3 74 10 1 11 Euterpe precatoria Mart. 80 14 94 12 10 22 Geonoma deversa (Poit.) Kunth 7 Bactris sp. 1 11 1 12 6 1 Iriartella setigera (Mart.) H. Wendl. 1 0 1 1 0 1 Maximiliana maripa (Aubl.) Drude 0 9 9 0 4 4 Oenocarpus bacaba Mart. 0 1 1 0 1 1 Oenocarpus minor Mart. 29 0 29 9 0 9 Syagrus sp. 1 0 33 33 0 4 4 Wendlandiella sp. 1 0 1 1 0 1 1 Anthropogenic soils 2 Non-anthropogenic soils id id DD Degree of domestication sensu Clement (1999), id incipiently domesticated, sd semi-domesticated References Aide TM, Zimmermann JK, Herrera L, Rosario M, Serrano M (1995) Forest recovery in abandoned tropical pastures in Puerto Rico. For Ecol Manage 77:77–86 Aide TM, Zimmerman JK, Pascarella JB, Rivera L, Marcano-Vega H (2000) Forest regeneration in a chronosequence of tropical abandoned pastures: implications for Restoration Ecology. Rest Ecol 8:328–338 Andrade EB (1983) Relatório de expedição para coleta de germoplasma de caiaué [Elaeis oleifera (H.B.K.) Cortés] na Amazônia brasileira. EMBRAPA/CNPSD, Manaus Andrade A (1986) Investigación arqueologica de los antrosoles de Araracuara. Fundación de Investigaciones Nacionales 31:1–103 Balée W, Campbell DG (1990) Evidence for the successional status of liana forest (Xingu River basin, Amazonian Brazil). Biotr 22:36–47 Balée W, Gély A (1989) Managed forest succession in Amazonia: the Ka’apor case. In: Posey DA, Balée W (eds) Resource management in Amazonia: indigenous and folk strategies. Advances in economic botany, vol 7. The New York Botanical Garden, New York, pp 129–158 Barcelos E (1986) Caracterı́sticas genético-ecológicas de populações naturais de caiaué [Elaeis oleifera (H.B.K.) Cortés) na Amazônia Brasileira. Dissertation, Instituto Nacional de Pesquisas da Amazônia/ Fundação Universidade do Amazonas Bernard HR (2002) Research methods in anthropology: qualitative and quantitative approaches. Alta Mira Press, Walnut Creek 123 1958 Biodivers Conserv (2010) 19:1933–1961 Brown S (1997) Estimating biomass and biomass change of tropical forests: a primer. FAO forestry paper 134. http://www.fao.org/documents. Cited 2 Jun 2009 Campbell DG, Ford A, Lowell K, Walker J, Lake JK, Ocampo-Raeder C, Townesmith A, Balick M (2006) The Feral forests of the Eastern Petén. In: Balée W, Erickson L (eds) Time and complexity in historical ecology: studies in the neotropical lowlands. Columbia University Press, New York Capers RS, Chazdon RL, Brenes AR, Alvarado BV (2005) Successional dynamics of woody seedling communities in wet tropical secondary forests. J Ecol 93:1071–1084 Castilho CV, Magnusson WE, Araújo RNO, Luizão RCC, Luizão FJ, Lima AP, Higuchi N (2006) Variation in aboveground tree live biomass in a central Amazonian forest: effects of soil and topography. For Ecol Manage 234:85–96 Chao A, Chazdon RL, Colwell RK, Shen TJ (2005) A new statistical approach for assessing similarity of species composition with incidence and abundance data. Ecol Let 8:148–159 Chapin FS, Vitousek PM, Van Cleve K (1986) The nature of nutrient limitation in plant communities. Am Nat 127:48–58 Chauvel A, Lucas Y, Boulet R (1987) On the genesis of the soil mantle of the region of Manaus, central Amazonia, Brazil. Experientia 43:234–241 Chazdon RL (2003) Tropical forest recovery: legacies of human impact and natural disturbances. Perspect Plant Ecol Evol Syst 6:51–71 Chazdon RL, Letcher SG, van Breugel M, Martı́nez-Ramos M, Bongers F, Finegan B (2007) Rates of change in tree communities of secondary Neotropical forests following major disturbances. Phil Trans R Soc B 362:273–289 Chinea JD (2002) Tropical forest succession on abandoned farms in the Humacao Municipality of eastern Puerto Rico. For Ecol Manage 167:195–207 Clement CR (1999) 1492 and the loss of Amazonian crop genetic resources: I. The relation between domestication and human population decline. Econ Bot 53:188–202 Clement CR, McCann JM, Smith NJH (2003) Agrobiodiversity in Amazonia and its relationships with Dark Earths. In: Lehmann D, Kern D, Glaser B, Woods W (eds) Amazonian dark earths: origin, properties, and management. Kluwer Academic Publ, Dordrecht Clement CR, Klüppel MP, German LA, Almeida SS, Major J, Aragão LEOC, Guix JC, Lleras E, WinklerPrins AMGA, Hecht SB, McCann JM (2009) Diversidade vegetal em solos antrópicos da Amazônia. In: Teixeira WG, Kern DC, Mádari B, Lima HN, Woods W (eds) As Terras Pretas de Índio na Amazônia: sua caracterização e uso deste conhecimento na criação de novas áreas. Embrapa Amazônia Ocidental, Manaus de Rouw A (1993) Regeneration by sprouting in slash and burn rice cultivation, Tai rain forest, Cote d’Ivoire. J Trop Ecol 9:387–408 Denevan WM (2001) Cultivated landscapes of native Amazonia and the Andes. Oxford University Press, Oxford Denevan WM, Padoch C (1987) Introduction: the Bora agroforestry project. In: Denevan WM, Padoch C (eds) Swidden fallow agroforestry in the Peruvian Amazon. Advances in economic botany, vol 5. The New York Botanical Garden, New York, pp 1–7 DeWalt SJ, Maliakal SK, Denslow JS (2003) Changes in vegetation structure and composition along a tropical forest chronosequence: implications for wildlife. For Ecol Manage 182:139–151 Dieleman JA, Mortensen DA, Buhler DD, Ferguson RB (2000) Identifying associations among site properties and weed species abundance. II: Hypothesis generation. Weed Sci 48:576–587 Dufrene M, Legendre P (1997) Species assemblages and indicator species: the need for a flexible assymetrical approach. Ecol Monogr 67:345–366 Eden MJ, Bray W, Herrera LF, McEwan C (1984) Terra preta soils and their archaeological context in the Caquetá Basin of southeast Colombia. Am Antiquity 49:125–140 Empresa Brasileira de Pesquisa Agropecuária (1999) Manual de análises quı́micas de solos, plantas e fertilizantes. Embrapa Solos, Embrapa Informática Agropecuária, Embrapa Comunicação para Transferência de Tecnologia, Brası́lia Erickson C (2003) Historical ecology and future explorations. In: Lehmann J, Kern D, Glaser B, Woods W (eds) Amazonian dark earths: origin, properties and management. Kluwer Academics Publisher, Dordrecht Ewel JJ, Berish C, Brown B, Price N, Raich J (1981) Slash and burn impacts on a Costa Rican wet forest site. Ecology 62:816–829 Fahmy T, Aubry P (2003) XLSTAT, version 7.0. Society Addinsoft, New York Falcão NPS, Clement CR, Tsai SM, Cornerford NB (2009) Pedology, fertility and biology of Central Amazonian dark earths. In: Woods W, Teixeira WG, Lehmann J, Steiner C, Winklerprins A (eds) Terra Preta Nova: a tribute to Wim Sombroek. Springer, Berlin Heidelberg 123 Biodivers Conserv (2010) 19:1933–1961 1959 Fearnside PM, Guimarães WM (1996) Carbon uptake by secondary forests in Brazilian Amazonia. For Ecol Manage 80:35–46 Feldpausch TR, Riha SJ, Fernandes ECM, Wandelli E (2005) Development of forest structure and leaf area in secondary forests regenerating on abandoned pastures in Central Amazonia. Earth Interact 9:1–22 Finegan B (1996) Pattern and process in Neotropical secondary rain forests: the first 100 years of succession. Trends Ecol Evol 11:119–124 Ford A, Nigh R (2009) Origins of the Maya forest garden: Maya resource management. J Ethnobio 29:213– 236 Fraser JA, Clement CR (2008) Dark earths and manioc cultivation in Central Amazonia: a window on preColumbian agricultural systems? Bol Mus Para Emı́lio Goeldi Ciências Humanas 3:175–194 Fraser JA, Cardoso T, Junqueira AB, Falcão NPS, Clement CR (2009) Historical ecology and dark earths in whitewater and blackwater landscapes: comparing the Middle Madeira and Lower Negro Rivers. In: Woods W, Teixeira WG, Lehmann J, Steiner C, Winklerprins A (eds) Terra Preta Nova: a tribute to Wim Sombroek. Springer, Berlin Heidelberg Gehring C, Denich M, Kanashiro M, Vlek PLG (1999) Response of secondary vegetation in Eastern Amazonia to relaxed nutrient availability constraints. Biogeochemistry 45:223–241 Gehring C, Denich M, Vlek PLG (2005) Resilience of secondary forest regrowth after slash-and-burn agriculture in central Amazonia. J Trop Ecol 21:519–527 German LA (2003a) Historical contingencies in the coevolution of environment and livelihood: contributions to the debate on Amazonian black earth. Geoderma 111:307–331 German LA (2003b) Ethnoscientific understandings of Amazonian dark earths. In: Lehmann J, Kern D, Glaser B, Woods W (eds) Amazonian dark earths–origin, properties, and management. Kluwer Academic Publisher, Dordrecht Glaser B (2007) Prehistorically modified soils of central Amazonia: a model for sustainable agriculture in the twenty-first century. Phil Trans R Soc B 362:187–196 Gotelli NJ, Colwell RK (2001) Quantifying biodiversity: procedures and pitfalls in the measurement and comparison of species richness. Ecol Lett 4:379–391 Gotelli NJ, Entsminger GL (2001) EcoSim: null models software for ecology, Version 7.0. Acquired Intelligence Inc. & Kesey-Bear, Jericho. http://garyentsminger.com/ecosim.htm Grime JP (1979) Plant strategies and vegetation processes. John Wiley and Sons, New York Guariguata MR, Ostertag R (2001) Neotropical secondary forest succession: changes in structural and functional characteristics. For Ecol Manage 148:185–206 Hiraoka M, Yamamoto S, Matsumoto E, Nakamura S, Falesi I, Baena ARC (2003) Contemporary use and management of Amazonian dark earths. In: Lehmann J, Kern D, Glaser B, Woods W (eds) Amazonian Dark Earths: origin, properties, and management. Kluwer Academic Publishers, Dordrecht Instituto Brasileiro de Geografia e Estatı́stica (2001) Mapa de solos do Brasil. http://www.ibge.gov.br. Cited 2 Jun 2009 Instituto Brasileiro de Geografia e Estatı́stica (2004) Mapa de vegetação do Brasil. http://www.ibge.gov.br. Cited 2 Jun 2009 Irvine D (1989) Succession management and resource distribution in an Amazonian rain forest. In: Posey DA, Balée W (eds) Resource management in Amazonia: indigenous and folk strategies. Advances in economic botany, vol 7. The New York Botanical Garden, New York, pp 223–237 IUSS Working Group WRB (2007) World reference base for soil resources 2006, first update 2007. World soil resources reports 103. FAO, Rome. http://www.fao.org/ag/AGL/agll/wrb/default.stm. Cited 2 Jun 2009 Kern DC (1996) Geoquı́mica e pedoquı́mica em sı́tios arqueológicos com terra preta na floresta de Caxiuanã (Portel), Pará. PhD Thesis, Universidade Federal do Pará Kern DC, D’Aquino G, Rodrigues TE, Frazão FJL, Sombroek W, Myers TP, Neves EG (2003) Distribution of Amazonian dark earths in the Brazilian Amazon. In: Lehmann J, Kern D, Glaser B, Woods W (eds) Amazonian dark earths–origin, properties, and management. Kluwer Academic Publisher, Dordrecht Kumar BM, Nair PKR (2004) The enigma of tropical homegardens. Agroforestry Syst 61:135–152 Laurance WF, Fearnside PM, Laurance SG, Delamonica P, Lovejoy TE, Rankin-de-Merona JM, Chambers J, Gascon C (1999) Relationship between soils and Amazon forest biomass: a landscape-scale study. For Ecol Manage 118:127–138 Lehmann J, Kern D, German L, McCann J, Martins GC, Moreira A (2003a) Soil fertility and production potential. In: Lehmann J, Kern D, Glaser B, Woods W (eds) Amazonian dark earths–origin, properties, and management. Kluwer Academic Publisher, Dordrecht Lehmann J, Silva JP Jr, Steiner C, Nehls T, Zech W, Glaser B (2003b) Nutrient availability and leaching in an archaeological Anthrosol and a Ferrasol of the Central Amazon basin: fertilizer, manure and charcoal amendments. Plant Soil 249:343–357 123 1960 Biodivers Conserv (2010) 19:1933–1961 Major J, DiTommaso A, German L, McCann J (2003) Weed population dynamics and management on Amazonian dark earth. In: Lehmann J, Kern D, Glaser B, Woods W (eds) Amazonian dark earths– origin, properties, and management. Kluwer Academic Publisher, Dordrecht Major J, DiTommaso A, Lehmann J, Falcão NPS (2005) Weed dynamics on Amazonian dark earth and adjacent soils of Brazil. Agr Ecosyst Environ 111:1–12 Maxted N, Kell SP, Ford-Lloyd BV (2008) Crop wild relative and use: establishing the context. In: Maxted N, Ford-Lloyd BV, Kell SP, Irinodo JM, Dulloo ME, Turok J (eds) Crop wild relative conservation and use. CABI International, Wallingford McCann JM (2003) Subsidy from culture: anthropogenic soils and vegetation in Tapajônia, Brazilian Amazonia. Dissertation, University of Wisconsin McCune B, Mefford MJ (1999) PC-ORD for Windows: multivariate analysis of ecological data, version 4.25. MjM Software, Oregon Mesquita RCG, Ickes K, Ganade G, Williamson GB (2001) Alternative successional pathways in the Amazon Basin. J Ecol 89:528–537 Mittelbach GG, Steiner CF, Scheiner SM, Gross KL, Reynolds HL, Waide RB, Willig MR, Dodson SI, Gough L (2001) What is the observed relationship between species richness and productivity? Ecology 82:2381–2396 Moran EF (1981) Developing the Amazon. Indiana University Press, Bloomington Moran EF, Brondizio ES, Tucker JM, Silva-Forsberg MCS, McCracken S, Falesi I (2000) Effects of soil fertility and land-use on forest succession in Amazonia. For Ecol Manage 139:93–108 Morcote-Rı́os G, Bernal R (2001) Remains of palms (Palmae) at archaeological sites in the new world: a review. Botanical Rev 67:309–350 Neves EG, Petersen JB, Bartone RN, da Silva CA (2003) Historical and socio-cultural origins of Amazonian dark earths. In: Lehmann J, Kern D, Glaser B, Woods W (eds) Amazonian dark earths–origin, properties, and management. Kluwer Academic Publisher, Dordrecht Oksanen J, Kindt R, Legendre P, O’Hara B, Simpson GL, Solymos P, Stevens HH, Wagner H (2009) Vegan: community ecology package. R package version 1.15-4. http://CRAN.R-project.org/package=vegan Parker E (1992) Forest islands and Kayapó resource management in Amazonia: a reappraisal of the Apêtê. Am Anthropol 94:406–428 Pascarella JB, Aide TM, Serrano MI, Zimmerman JK (2000) Land-use history and forest regeneration in the Cayey Mountains, Puerto Rico. Ecosystems 3:217–228 Paz-Rivera C, Putz FE (2009) Anthropogenic soils and tree distributions in a lowland forest in Bolivia. Biotropica 41:665–675 Peña-Claros M (2003) Changes in forest structure and species composition during secondary forest succession in the Bolivian Amazon. Biotropica 35:450–461 Posey DA (1984) A preliminary report on diversified management of tropical forest by the Kayapó Indians of the Brazilian Amazon. In: Prance GT, Kallunki JA (eds) Ethnobotany in the neotropics. Advances in economic botany, vol 1. The New York Botanical Garden, New York, pp 112–126 Posey DA (1992) Reply to Parker. Am Anthropol 94:441–443 Rapp Py-Daniel LH (2007) Caracterização da área amostrada. In: Rapp Py-Daniel LH, Deus CP, Henriques AL, Pimpão DM, Ribeiro OM (eds) Biodiversidade do médio Madeira: bases cientı́ficas para propostas de conservação. INPA, Manaus R Development Core Team (2008) R: a language and environment for statistical computing. R foundation for statistical computing, Vienna. http://www.r-project.org Silva KE, Pereira KP (2005) Alterações na cobertura vegetal dos municı́pios do Sul do Amazonas. In: Anais do XII Simpósio Brasileiro de Sensoriamento Remoto, Goiânia, 16–21 April 2005 Simões MF, Lopes DF (1987) Pesquisas arqueológicas no baixo/médio rio Madeira (Amazonas). Revista de Arqueologia 4:117–134 Smith NJH (1980) Anthrosols and human carrying capacity in Amazonia. Ann Assoc Am Geogr 70:553–566 Sokal RR, Rohlf FJ (1981) Biometry–the principles and practice of statistics in biological research, 2nd edn. W.H. Freeman & Co, New York Sombroek W, Kern D, Rodrigues TI, Cravo MS, Jarbas TC, Woods W, Glaser B (2002) Terra Preta and Terra Mulata: pre-Columbian Amazon kitchen middens and agricultural fields, their sustainability and their replication. Paper presented at the 17th World congress of soil science, Bangkok, 14–20 August 2002 Sombroek W, Ruivo MDL, Fearnside PM, Glaser B, Lehmann J (2003) Amazonian dark earths as carbon stores and sinks. In: Lehmann J, Kern D, Glaser B, Woods W (eds) Amazonian dark earths–origin, properties, and management. Kluwer Academic Publisher, Dordrecht Stevens MHH, Carson WP (1999) Plant density determines species richness along an experimental fertility gradient. Ecology 80:455–465 123 Biodivers Conserv (2010) 19:1933–1961 1961 Stevens MHH, Carson WP (2002) Resource quantity, not resource heterogeneity, maintains plant diversity. Ecol Lett 5:420–426 Tilman D (1987) Secondary succession and the pattern of plant dominance along experimental nitrogen gradients. Ecol Monogr 57:189–214 Uhl C, Clark K, Murphy P (1981) Early plant succession after cutting and burning in the upper Rio Negro region of the Amazon basin. J Ecol 69:631–649 Uhl C, Jordan C, Clark K, Clark H, Herrera R (1982) Ecosystem recovery in Amazon caatinga forest after cutting, cutting and burning, and bulldozer clearing treatments. Oikos 38:313–320 Uhl C, Buschbacher R, Serrao AS (1988) Abandoned pastures in eastern Amazonia: I. Patterns of plant succession. J Ecol 76:663–681 Waide RB, Willig MR, Steiner CF, Mittelbach G, Gough L, Dodson SI, Juday GP, Parmenter R (1999) The relationship between productivity and species richness. Annu Rev Ecol Syst 30:257–300 Wilkinson L (2002) Systat software, v. 10.2. Systat software, Richmond Woods WI, Denevan WM (2009) Amazonian dark earths: the first century of reports. In: Woods W, Teixeira WG, Lehmann J, Steiner C, Winklerprins A (eds) Terra Preta Nova: a tribute to Wim Sombroek. Springer, Berlin Heidelberg Woods WI, McCann JM (1999) The anthropogenic origin and persistence of Amazonian dark earths. In: The yearbook of the conference of Latin Americanist Geographers, vol 25. Southern Illinois University, Edwardsville, pp 7–14 123 All in-text references underlined in blue are linked to publications on ResearchGate, letting you access and rea

RELATED PAPERS

RELATED TOPICS

Log In

Secondary forests on anthropogenic soils in Brazilian Amazonia conserve agrobiodiversity

Secondary forests on anthropogenic soils in Brazilian Amazonia conserve agrobiodiversity

Related Papers

RELATED PAPERS

RELATED TOPICS