Search

The influx of nearly a million refugees from Myanmar’s Rakhine state to Cox’s Bazar, Bangladesh, in August 2017 put significant pressure on the regional landscape leading to land degradation due to biomass removal to provide shelter and fuel energy and posed critical challenges for both host and displaced population. This article emphasizes geospatial applications at different stages of addressing land degradation in Cox’s Bazar. A wide range of data and methods were used to delineate land tenure, estimate wood fuel demand and supply, assess land degradation, evaluate land restoration suitability, and monitor restoration activities. The quantitative and spatially explicit information from these geospatial assessments integrated with the technical guidelines for sustainable land management and an adaptive management strategy was critical in enabling a collaborative, multi-disciplinary and evidence-based approach to successfully restoring degraded landscapes in a displacement setting.

Information on land cover is vital to numerous United Nations (UN) missions, including achieving the Sustainable Development Goals (SDGs). Because land cover data are developed by a variety of organizations for a range of objectives, they are based on different classification schemes and have discrepancies. In addition, the sustainability for land cover is hampered by limited access to information and documentation. Accordingly, international standards for land cover are developed to improve interoperability between different land cover datasets. However, the use and development of land cover datasets are limited by various factors including availability of properly documented land cover legends in support of different applications including change assessment, comparison, and international reporting. The purpose of this article is to highlight the importance of land cover in achieving several goals and to introduce the first international platform for land cover legend, named Land Cover Legend Registry (LCLR). This registry is a contribution to the international land cover community and the UN in effort to promote and support data harmonization processes and interoperability from local to global level, and vice versa. Users can not only use the registry for preparing consistent datasets, but also contribute to it by providing the latest data to ensure the long-term availability of both updated and existing datasets around the world. Moreover, building on the experience developing land cover legends with different nations, a brief explanation on the preparation of legends is also provided. Additionally, it is more important than ever to develop land cover registers to support the use, expansion, integration, and use uptake of land cover data, particularly for innovative remote sensing, machine learning, and information and communication technologies and techniques that build on existing and national contexts.

Tree diversity in Asia's tropical and subtropical forests is central to nature-based solutions. Species vulnerability to multiple threats, which affect provision of ecosystem services, is poorly understood. We conducted a region-wide, spatially explicit assessment of the vulnerability of 63 socioeconomically important tree species to overexploitation, fire, overgrazing, habitat conversion, and climate change. Trees were selected for assessment from national priority lists, and selections were validated by an expert network representing 20 countries. We used Maxent suitability modeling to predict species distribution ranges, freely accessible spatial data sets to map threat exposures, and functional traits to estimate threat sensitivities. Species-specific vulnerability maps were created as the product of exposure maps and sensitivity estimates. Based on vulnerability to current threats and climate change, we identified priority areas for conservation and restoration. Overall, 74% of the most important areas for conservation of these trees fell outside protected areas, and all species were severely threatened across an average of 47% of their native ranges. The most imminent threats were overexploitation and habitat conversion; populations were severely threatened by these factors in an average of 24% and 16% of their ranges, respectively. Our model predicted limited overall climate change impacts, although some study species were likely to lose over 15% of their habitat by 2050 due to climate change. We pinpointed specific natural areas in Borneo rain forests as hotspots for in situ conservation of forest genetic resources, more than 82% of which fell outside designated protected areas. We also identified degraded areas in Western Ghats, Indochina dry forests, and Sumatran rain forests as hotspots for restoration, where planting or assisted natural regeneration will help conserve these species, and croplands in southern India and Thailand as potentially important agroforestry options. Our results highlight the need for regionally coordinated action for effective conservation and restoration.Especies de Árboles Valoradas y Amenazadas de Asia Tropical y Subtropical Resumen La diversidad de árboles en los bosques tropicales y subtropicales de Asia es un eje central para las soluciones basadas en la naturaleza. La vulnerabilidad de las especies ante las múltiples amenazas, las cuales afectan el suministro de servicios ambientales, es un tema poco comprendido. Realizamos una evaluación regional espacialmente explícita de la vulnerabilidad de 63 especies de árboles de importancia socioeconómica ante la sobreexplotación, incendios, sobrepastoreo, conversión del hábitat y cambio climático. Los árboles se seleccionaron para su evaluación a partir de listas nacionales de prioridades, y las selecciones fueron validadas por una red de expertos de 20 países. Usamos el modelado de idoneidad Maxent para predecir el rango de distribución de las especies, conjuntos de datos espaciales de libre acceso para mapear la exposición a las amenazas y rasgos funcionales para estimar la susceptibilidad a las amenazas. Con base en la vulnerabilidad a las amenazas actuales y al cambio climático, identificamos las áreas prioritarias para su conservación y restauración. En general, el 74% de las áreas más importantes para la conservación de estos árboles quedó fuera de las áreas protegidas y todas las especies estaban seriamente amenazadas en promedio en el 47% de su distribución nativa. Las amenazas más inminentes fueron la sobreexplotación y la conversión del hábitat; las poblaciones estuvieron seriamente amenazadas por estos factores en promedio en el 24% y 16% de su distribución, respectivamente. Nuestro modelo predijo un impacto general limitado del cambio climático, aunque algunas especies estudiadas tuvieron la probabilidad de perder más del 15% de su hábitat para el 2050 debido a este factor. Identificamos áreas naturales específicas en las selvas de Borneo como puntos calientes para la conservación in situ de los recursos genéticos forestales, más del 82% de los cuales estaban fuera de las áreas protegidas designadas. También identificamos áreas degradadas en los Ghats Occidentales, los bosques secos de Indochina y las selvas de Sumatra como puntos calientes para la restauración, en donde la siembra o la regeneración natural asistida ayudarán a conservar estas especies. Además, identificamos campos de cultivo al sur de India y Tailandia como potenciales opciones importantes de agrosilvicultura. Nuestros resultados resaltan la necesidad de acciones regionales coordinadas para la conservación y restauración efectivas.【摘要】保护亚洲热带和亚热带森林的树木多样性是基于自然的解决方案的关注重点。物种面对多种威胁的脆弱性会影响生态系统服务的供给, 但目前人们对此仍知之甚少。我们对63种具有社会经济意义的树种面对过度开发、火灾、过度放牧、栖息地转换和气候变化的脆弱性进行了全区域空间显式评估。我们进行评估的树种来自国家优先保护名录, 树种的选择得到代表 20个国家的专家网络的认可。研究使用Maxent适宜性模型预测了物种分布范围, 并用可获取的空间数据集绘制了威胁暴露情况, 用功能特征估计了威胁敏感性。我们利用暴露地图和敏感性估计结果得到了物种特异的脆弱性地图, 并根据物种对当前威胁和气候变化的脆弱性, 确定了保护和恢复的优先区域。总的来说, 对这些树种的保护最重要的区域有 74%在保护区之外, 所有树种平均在其原生分布区 47%的范围内受到严重威胁。最紧迫的威胁是过度开发和栖息地转换;树木种群平均在其 24%和16%的范围内受到这些因素的严重威胁。我们的模型预测得到的气候变化整体影响有限, 尽管一些物种到2050年可能会因气候变化而失去超过15%的栖息地。我们发现婆罗洲雨林中特定的自然区域应作为森林遗传资源原地保护的热点地区, 该区域82%以上的面积位于已有保护区之外。我们还确定了西高止山、中南半岛旱林和苏门答腊雨林的退化地区为恢复热点地区, 在这些地区种植树木或辅助自然再生将有助于保护树种, 而印度南部和泰国的耕地则是潜在的重要农林地。我们的结果强调了为有效保护和恢复而采取区域性协调行动的必要性。【翻译: 胡怡思;审校: 聂永刚】.

Allometric models are commonly used to estimate biomass, nutrients and carbon stocks in trees, and contribute to an understanding of forest status and resource dynamics. The selection of appropriate and robust models, therefore, have considerable influence on the accuracy of estimates obtained. Allometric models can be developed for individual species or to represent a community or bioregion. In Bangladesh, the nation forest inventory classifies tree and forest resources into five zones (Sal, Hill, Coastal, Sundarbans and Village), based on their floristic composition and soil type. This study has developed allometric biomass models for multi-species of the Sal zone. The forest of Sal zone is dominated by Shorea robusta Roth. The study also investigates the concentrations of Nitrogen, Phosphorus, Potassium and Carbon in different tree components. A total of 161 individual trees from 20 different species were harvested across a range of tree size classes. Diameter at breast height (DBH), total height (H) and wood density (WD) were considered as predictor variables, while total above-ground biomass (TAGB), stem, bark, branch and leaf biomass were the output variables of the allometric models. The best fit allometric biomass model for TAGB, stem, bark, branch and leaf were: ln (TAGB) = -2.460 + 2.171 ln (DBH) + 0.367 ln (H) + 0.161 ln (WD); ln (Stem) = -3.373 + 1.934 ln (DBH) + 0.833 ln (H) + 0.452 ln (WD); ln (Bark) = -5.87 + 2.103 ln (DBH) + 0.926 ln (H) + 0.587 ln (WD); ln (Branch) = -3.154 + 2.798 ln (DBH) - 0.729 ln (H) - 0.355 ln (WD); and ln (Leaf) = -4.713 + 2.066 ln (DBH), respectively. Nutrients and carbon concentration in tree components varied according to tree species and component. A comparison to frequently used regional and pan-tropical biomass models showed a wide range of model prediction error (35.48 to 85.51%) when the observed TAGB of sampled trees were compared with the estimated TAGB of the models developed in this study. The improved accuracy of the best fit model obtained in this study can therefore be used for more accurate estimation of TAGB and carbon and nutrients in TAGB for the Sal zone of Bangladesh.

Against a backdrop of growing concerns over REDD+ implications for tenure of forest-dependent communities, very few scientific studies have attempted to assess the linkages between the two. This paper offers empirical evidences for impacts of REDD+ intervention based on a case study of Thailand, where de jure state property and de facto tenure apparently co-exist. In contrast to the existing literature proposing positive tenure impacts, the findings suggest that REDD+ would not be sufficient to incentivize the government to embark on tenure reforms or any improvement of customary tenure and challenge REDD+ about its incentives for countries to accelerate tenure reforms. The findings also reveal that risks associated with tenure insecurity discouraged REDD+ investment leading to the withdrawal of the fund altogether. The findings enrich the on-going debates on the tenure impacts, tenure reform and competing agendas and provide insights to support the development of future national strategies and frameworks. The paper moreover proposes measures to create enabling conditions for REDD+.

The implementation of forest-based projects to mitigate greenhouse gas emissions requires the estimation of emission factors (here the difference in biomass stocks between two forest types). The estimation of these quantities using forest inventory data and allometric models implies different sources of errors that need to be prioritized to improve the precision of estimation. Using data from permanent sample plots in a tropical moist forest in central Africa and considering four allometric models with equal likelihood, the largest source of error in the estimate of the difference of biomass between intact and logged-over forest was that due to the model choice (40 % of the sum of squares). The error due to the model choice did not cancel out in the difference due to an interaction between the model’s prediction and the diameter structure of the forest. The variability in biomass between plots was the second largest source of error, but was underestimated because of post-stratification. The error due to the model choice could be reduced by weighting the models’ predictions.

1 IntroductionGiven the pressing need to quantify carbon fluxes associatedwith terrestrial vegetation dynamics, an increasing number ofresearchers have sought to improve estimates of tree volume,biomass, and carbon stocks. Tree allometric equations arecritical tools for such purpose and have the potential toimprove our understanding about carbon sequestration inwoody vegetation, to support the implementation of policiesand mechanisms designed to mitigate climate change (e.g.CDM and REDD+; Agrawal et al. 2011), to calculate costsand benefits associated with forest carbon projects, and toimprove bioenergy systems and sustainable forest manage-ment (Henry et al. 2013).

GlobAllomeTree is an international platform for tree allometric equations. It is the first worldwide web platform designed to facilitate the access of the tree allometric equation and to facilitate the assessment of the tree biometric characteristics for commercial volume, bio-energy or carbon cycling. The webplatform presents a database containing tree allometric equations, a software called Fantallomatrik, to facilitate the comparison and selection of the equations, and documentation to facilitate the development of new tree allometric models, improve the evaluation of tree and forest resources and improve knowledge on tree allometric equations. In the Fantallometrik software, equations can be selected by country, ecological zones, input parameters, tree species, statistic parameters and outputs. The continuously updated database currently contains over 5000 tree allometric equations classified according to 73 fields. The software Fantallometrik can be also used to compare equations, insert new data and estimate the selected output variables using field inventory. The GlobAllomeTree products are freely available at the URL: http://globallometree.org for a range of users including foresters, project developers, scientist, student and government staff.

With the growing interest in estimating carbon stocks in forests, available allometric equations have been compiled. These compilations often reveal contrasting models for the same species and site. Rather than choosing a model with the risk of not selecting the best available one, Bayesian model averaging (BMA) offers a way to combine different allometric equations into a single predictive model. In the deterministic version of BMA, existing models with known coefficients are combined. In the statistical version, competing models are at the same time fitted and combined. Using the BMA of deterministic models, we combined three existing multispecies pan-tropical biomass equations for tropical moist forests. The resulting model brought a relatively minor although consistent improvement of the predictions of the aboveground dry biomass of trees. These three models were particular cases of a family of models that were subsequently combined using the BMA of statistical models. Again, the resulting model was able to capture features in the biomass response to diameter that no single model was able to fit. BMA thus is an alternative to model selection that allows integrating the biomass response from different models. (Resume d'auteur)

Summary 1. Regional above-ground biomass estimates for tropical moist forests remain highly inaccurate mostly because they are based on extrapolations from a few plots scattered across a limited range of soils and other environmental conditions. When such conditions impact biomass, the estimation is biased. The effect of soil types on biomass has especially yielded controversial results. 2. We investigated the relationship between above-ground biomass and soil type in undisturbed moist forests in the Central African Republic. We tested the effects of soil texture, as a surrogate for soil resources availability and physical constraints (soil depth and hydromorphy) on biomass. Forest inventory data were collected for trees ‡20 cm stem diameter in 2754 0.5 ha plots scattered over 4888 km 2 . The plots contained 224 taxons, of which 209 were identified to species. Soil types were characterized from a 1:1 000 000 scale soil map. Species-specific values for wood density were extracted from the CIRAD’s data base of wood technological properties. 3. We found that basal area and biomass differ in their responses to soil type, ranging from 17.8 m 2 ha )1 (217.5 t ha )1 )t o 22.3 m 2 ha )1 (273.3 t ha )1 ). While shallow and hydromorphic soils support forests with both low stem basal area and low biomass, forests on deep resource-poor soils are typically low in basal area but as high in biomass as forests on deep resource-rich soils. We demonstrated that the environmental filtering of slow growing dense-wooded species on resource-poor soils compensates for the low basal area, and we discuss whether this filtering effect is due to low fertility or to low water reserve. 4. Synthesis. We showed that soil physical conditions constrained the amount of biomass stored in tropical moist forests. Contrary to previous reports, our results suggest that biomass is similar on resource-poor and resource-rich soils. This finding highlights both the importance of taking into account soil characteristics and species wood density when trying to predict regional patterns of biomass. Our findings have implications for the evaluation of biomass stocks in tropical forests, in the context of the international negotiations on climate change.

This study gives an outlook on the carbon balance of Sub-Saharan Africa (SSA) by presenting a summary of currently available results from the project CarboAfrica (namely net ecosystem productivity and emissions from fires, deforestation and forest degradation, by field and model estimates) supplemented by bibliographic data and compared with a new synthesis of the data from national communications to UNFCCC. According to these preliminary estimates the biogenic carbon balance of SSA varies from 0.16 Pg C y−1 to a much higher sink of 1.00 Pg C y−1 (depending on the source data). Models estimates would give an unrealistic sink of 3.23 Pg C y−1, confirming their current inadequacy when applied to Africa. The carbon uptake by forests and savannas (0.34 and 1.89 Pg C y−1, respectively,) are the main contributors to the resulting sink. Fires (0.72 Pg C y−1) and deforestation (0.25 Pg C y−1) are the main contributors to the SSA carbon emissions, while the agricultural sector and forest degradation contributes only with 0.12 and 0.08 Pg C y−1, respectively. Savannas play a major role in shaping the SSA carbon balance, due to their large extension, their fire regime, and their strong interannual NEP variability, but they are also a major uncertainty in the overall budget. Even if fossil fuel emissions from SSA are relative low, they can be crucial in defining the sign of the overall SSA carbon balance by reducing the natural sink potential, especially in the future. This paper shows that Africa plays a key role in the global carbon cycle system and probably could have a potential for carbon sequestration higher than expected, even if still highly uncertain. Further investigations are needed, particularly to better address the role of savannas and tropical forests and to improve biogeochemical models. The CarboAfrica network of carbon measurements could provide future unique data sets for better estimating the African carbon balance.

The main goal of national forest programs is to lead and steer forest policy development and implementation processes in an inter-sectoral way (FAO 2006). National forest monitoring systems contribute to forest programs through monitoring forest changes and forest services over time (FAO 2013). To do so, they generally collect and analyze forest-related data and provide knowledge and recommendations at regular intervals. The collection of forest-related data and their analyses have continually evolved with technological and computational advances (Kleinn 2002). For instance, ground measurements, such as diameter or height measurements, which were typically measured with measuring tape or forest compasses and relascopes, are now enhanced with new technologies, such as laser range finders. Furthermore, remote sensing is being increasingly used to improve ground sampling strategies (Maniatis and Mollicone 2010), to calculate forested land area and area changes (INPE 2006; INPE 2008; Hansen et al. 2013), and to detect many variables of interest such as forest fires, pest outbreaks, or trees outside forests (Barducci et al. 2002). The use of remotely sensed data together with ground-based observations has gained a lot of attention for estimating greenhouse gas emissions and removals associated with forests, particularly in the context of REDD+ (GOFC-GOLD 2010; GFOI 2014). During the last decades, the amount of information collected during forest inventories has thus grown rapidly and has, in turn, improved our ability to survey and manage many services such as biodiversity, carbon sequestration, or recreation. However, national forest monitoring approaches remain very heterogeneous from one country to another, and many national systems have still not taken the full advantage of newly operational technologies, despite an increasing availability of free, or at least less costly, data. This is probably because the use of these technologies to assess forest structural properties is, for the most part, used by only a few specialists and is largely confined to the research sector. The objective of this paper is to raise awareness by presenting, in a comprehensible way, some existing and promising technologies for supporting national forest monitoring.

Key message Three options are proposed to improve the accuracy of national forest biomass estimates and decrease the uncertainty related to tree model selection depending on available data and national contexts. Introduction Different tree volume and biomass equations result in different estimates. At national scale, differences of estimates can be important while they constitute the basis to guide policies and measures, particularly in the context of climate change mitigation. Method Few countries have developed national tree volume and biomass equation databases and have explored its potential to decrease uncertainty of volume and biomasttags estimates. With the launch of the GlobAllomeTree webplatform, most countries in the world could have access to country-specific databases. The aim of this article is to recommend approaches for assessing tree and forest volume and biomass at national level with the lowest uncertainty. The article highlights the crucial need to link allometric equation development with national forest inventory planning efforts. Results Models must represent the tree population considered. Data availability; technical, financial, and human capacities; and biophysical context, among other factors, will influence the calculation process. Conclusion Three options are proposed to improve accuracy of national forest assessment depending on identified contexts. Further improvements could be obtained through improved forest stratification and additional non-destructive field campaigns.

The increasing number of model types that are used to predict tree biomass from diameter, height and wood density has brought questioning about the biological relevance of complex allometries (i.e. non-power models). Statistical issues such as collinearity among predictors and unreliable coefficient estimates have also been associated with complex allometric models. Using a data set of 225 trees from central Africa, we assessed the relevance of simple allometry (i.e. power model) versus complex allometry to predict tree biomass. A complex allometric model of biomass was developed based on a model of resource partition between dbh and height growths. Although being a good model for biomass prediction, the power model was outperformed by the complex allometric model. A careful examination showed that the power model could be segmented into two pieces of power models. Using tree diameter and height as separated predictors improved the biomass prediction, irrespective of the collinearity between these two predictors. A critical value of 25% for the PRSE statistic used to assess the reliability of coefficient estimates corresponded to a significance level of 10 - 5 – 10 - 4 and was thus unreasonably low. We conclude that growth theories should be developed to explain allometric models, but that the arbitration between these models should ultimately rely on observed data, not on theories.

International audience; Advances in ecology face the complexity of ecosystems with dynamics longer than a single scientist’s career. In forestry and REDD+ practice, in particular, our ability to understand forest ecosystem dynamics and to manage them for mitigation and adaptation strongly relies on the combination of long-term research efforts and on data sharing. However, data collected by many measurement campaigns are regularly lost because of a lack of capacity to archive and maintain such information. Much progress would be achieved by encouraging researchers to provide access to primary data or publish “data-papers”

International audience; Terrestrial carbon stock mapping is important for the successful implementation of climate change mitigation policies. Its accuracy depends on the availability of reliable allometric models to infer oven-dry aboveground biomass of trees from census data. The degree of uncertainty associated with previously published pantropical aboveground biomass allometries is large. We analyzed a global database of directly harvested trees at 58 sites, spanning a wide range of climatic conditions and vegetation types (4004 trees ≥ 5 cm trunk diameter). When trunk diameter, total tree height, and wood specific gravity were included in the aboveground biomass model as co-variates, a single model was found to hold across tropical vegetation types, with no detectable effect of region or environmental factors. The mean percent bias and variance of this model was only slightly higher than that of locally fitted models. Wood specific gravity was an important predictor of aboveground biomass, especially when including a much broader range of vegetation types than previous studies. The generic tree diameter-height relationship depended linearly on a bioclimatic stress variable E, which compounds indices of temperature variability, precipitation variability and drought intensity. For cases in which total tree height is unavailable for aboveground biomass estimation, a pantropical model incorporating wood density, trunk diameter and the variable E outperformed previously published models without height. However, to minimize bias, the development of locally derived diameter-height relationships is advised whenever possible. Both new allometric models should contribute to improve the accuracy of biomass assessment protocols in tropical vegetation types, and to advancing our understanding of architectural and evolutionary constraints on woody plant development.

International audience; The development of tree allometric equations is crucial to accurate forest carbon assessment. However, very few allometric equations exist for sub-Saharan Africa and as a result generalized allometric equations, often established for forests in other continents, are used by default. The objectives of this study were (1) to propose a sampling methodology and calculation procedures to assess biomass for tropical tree species of contrasted tree shapes in Africa, (2) to identify factors affecting within and between trees wood density, (3) to propose an allometric model that integrates these factors and (4) to evaluate the reliability of using generalized allometric equations in this type of forests. Models were developed to predict wood density and phytomass of the trees based on the harvesting of 42 trees from 16 species, representing three guild status in the wet evergreen forest of Boi Tano in Ghana. Results indicated that the wood density was highly influenced by the tree species, guild status, size of the tree and pith to bark distance. Dry mass of a tree was influenced by diameter at breast height, crown diameter and wood density. The wood density depends on the position of the wood within the tree and the guild status considered. The use of generalized allometric models in literature is limited by the specific climate zone, the consideration of tree height and species specific wood density. In considering those factors, using generalized allometric equations could result in an error of 3%. Further research should better consider the bigger trees and the influence of the topography and ecosystem history.

International audience; While Carbon (C) sequestration an farmlands may contribute to mitigate CO2 concentrations in the atmosphere, greater agro-biodiversity may ensure longer term stability of C storage in fluctuating environments. This study was conducted in the highlands of western Kenya, a region with high potential for agroforestry, with the objectives of assessing current biodiversity and aboveground C stocks in perennial vegetation growing on farmland, and estimating C sequestration potential in aboveground C pools. Allometric models were developed to estimate aboveground biomass of trees and hedgerows, and an inventory of perennial vegetation was conducted in 35 farms in Vihiga and Siaya districts. Values of the Shannon index (H), used to evaluate biodiversity, ranged from 0.01 in woodlots through 0.4-0.6 in food crop plots, to 1.3-1.6 in homegardens. Eucalyptus saligna was the most frequent tree species found as individual trees (20%), in windrows (47%), and in woodlots (99%) in Vihiga and the most frequent in woodlots (96%) in Siaya. Trees represented the most important C pool in aboveground biomass of perennial plants growing on-farm, contributing to 81 and 55% of total aboveground farm C in Vihiga and Siaya, respectively, followed by hedgerows (13 and 39%, respectively) and permanent crop stands (5 and 6%, respectively). Most of the tree C was located in woodlots in Vihiga (61%) and in individual trees growing in or around food crop plots in Siaya (57%). The homegardens represented the second C pool in importance, with 25 and 33% of C stocks in Vihiga and Siaya, respectively. Considering the mean total aboveground C stocks observed, and taking the average farm sizes of Vihiga (0.6 ha) and Siaya (1.4 ha), an average farm would store 6.5 +/- 0.1 Mg C farm(-1) in Vihiga and 12.4 +/- 0.1 Mg C farm(-1) in Siaya. At both sites, the C sequestration potential in perennial aboveground biomass was estimated at ca. 16 Mg C ha(-1). With the current market price for carbon, the implementation of Clean Development Mechanism Afforestation/Reforestation (CDM A/R) projects seems unfeasible, due to the large number of small farms (between 140 and 300) necessary to achieve a critical land area able to compensate the concomitant minimum transaction costs. Higher financial compensation for C sequestration projects that encourage biodiversity would allow clearer win-win scenarios for smallholder farmers. Thus, a better valuation of ecosystem, services should encourage C sequestration together with on-farm biodiversity when promoting CDM A/R projects. (C) 2008 Elsevier B.V. All rights reserved. [References: 40]