Greenhouse gas emissions from tropical forest degradation: an underestimated source
© The Author(s) 2017
Received: 29 October 2016
Accepted: 4 February 2017
Published: 14 February 2017
The degradation of forests in developing countries, particularly those within tropical and subtropical latitudes, is perceived to be an important contributor to global greenhouse gas emissions. However, the impacts of forest degradation are understudied and poorly understood, largely because international emission reduction programs have focused on deforestation, which is easier to detect and thus more readily monitored. To better understand and seize opportunities for addressing climate change it will be essential to improve knowledge of greenhouse gas emissions from forest degradation.
Here we provide a consistent estimation of forest degradation emissions between 2005 and 2010 across 74 developing countries covering 2.2 billion hectares of forests. We estimated annual emissions of 2.1 billion tons of carbon dioxide, of which 53% were derived from timber harvest, 30% from woodfuel harvest and 17% from forest fire. These percentages differed by region: timber harvest was as high as 69% in South and Central America and just 31% in Africa; woodfuel harvest was 35% in Asia, and just 10% in South and Central America; and fire ranged from 33% in Africa to only 5% in Asia. Of the total emissions from deforestation and forest degradation, forest degradation accounted for 25%. In 28 of the 74 countries, emissions from forest degradation exceeded those from deforestation.
The results of this study clearly demonstrate the importance of accounting greenhouse gases from forest degradation by human activities. The scale of emissions presented indicates that the exclusion of forest degradation from national and international GHG accounting is distorting. This work helps identify where emissions are likely significant, but policy developments are needed to guide when and how accounting should be undertaken. Furthermore, ongoing research is needed to create and enhance cost-effective accounting approaches.
KeywordsCarbon stock Deforestation Forest fire Woodfuel REDD+ Timber harvest
The degradation of forests in developing countries, particularly those within tropical and subtropical latitudes, is perceived to be an important contributor both to global greenhouse gas emissions and to development. Its impacts are understudied and poorly understood, and present a major challenge for national-level carbon inventories  and for addressing diminishing biodiversity . International emission reduction programs (especially reducing emissions from deforestation and degradation, conservation of forest carbon stocks, sustainable management of forests and enhancement of forest carbon stocks—REDD+) have focused mostly on deforestation, which is easier to detect and thus more readily measured and monitored than forest degradation . A key challenge for measuring and monitoring forest degradation is that it is difficult to detect using commonly-used remote sensing products, such as Landsat. Instead, much higher resolution imagery is needed to identify the more subtle changes in forest cover typical of forest degradation activity. The World Bank, a major REDD+ investor/donor, established a Carbon Fund  with a methodological framework that requires emissions from forest degradation to be accounted where ‘significant’, which is defined as more than 10% of ‘forest-related emissions’. Yet it is unclear how to quantify and meaningfully demonstrate “significance”, or how to account for emissions cost-effectively when significant.
Forest degradation occurs when there is a direct, human-induced decrease in carbon stocks in forests resulting from a loss of canopy cover that is insufficient to be classed as deforestation [11, 17]. Moreover, the decrease in carbon stocks should be persistent, although the duration of this persistence has not been defined. Common drivers of forest degradation include timber harvesting (legal and illegal), fuel wood collection, non-stand replacing fires, and animal grazing in the forest (preventing forest regeneration) .
A handful of studies have attempted to assess and quantify emissions from human-driven forest degradation, including an assessment of the importance of drivers of forest degradation made by Hosonuma et al. . This study was based on data only for the area of forest disturbed in 46 tropical and sub-tropical countries. Of the total area of disturbed forests in these countries, they found that 51% of the disturbed area was caused by timber harvesting, 31% by woodfuel harvest, 9% by fires, and 7% by grazing. While timber harvest was the most significant activity in South and Central America and Asia, woodfuel was the largest activity by proportion (48%) in Africa. These estimates only included a subset of tropical and subtropical countries; were not produced through an independent and consistent assessment; and offered no quantitative information on the magnitude of the greenhouse gas emissions and how they compare to those from deforestation.
Another assessment of the emissions from forest degradation in the tropics conducted by Houghton  was based on his bookkeeping model for the period 1990–2010. He estimated that the average annual net emissions from harvesting of timber and woodfuel (with the exclusion of the re-clearing of forest fallow within the shifting cultivation cycle) just 10% of the summed emissions from deforestation and degradation, with degradation emissions dominated by timber harvest with marginal emissions from woodfuel, and no emission from fires. Given the exclusion of other key causes of forest degradation, this study is incomplete and lacks consistency.
Recent work by Pearson et al.  focused on the perceived key cause of forest degradation: timber harvest and associated infrastructure (skid trails and logging roads). They showed that for nine major tropical timber producing countries, emissions from logging were on average equivalent to about 12% of those from deforestation. For those nine countries with relatively low emissions from deforestation, emissions from logging were found to be equivalent to half or more of those from deforestation, whereas for countries with the highest emissions from deforestation, emissions from logging were equivalent to <10% of those from deforestation. These estimates are supported by the work of Asner and others in the Brazilian Amazon. Asner et al.  estimated logged areas ranged from 60 to 123% of previously reported deforestation areas. Huang and Asner  estimated that the inclusion of timber harvest elevated emissions by 15–19% over the emissions from deforestation alone.
Collection of traditional woodfuel (firewood and charcoal) for cooking and heating is common throughout the tropics, and can lead to forest degradation where removals exceed regrowth. Where annual harvest of woodfuel exceeds the forest’s incremental growth in biomass, it is considered to be unsustainable, and leads to a decline of woody biomass and to net carbon emissions . Bailis et al.  estimated that 27–34% of woodfuel harvest was unsustainable, particularly in East Africa and South Asia, and thus leads to significant forest degradation.
Fire is an important cause of forest disturbance and is commonly used to manage forest lands in the tropics and subtropics . Fire is often used to transform forest, e.g. into croplands, but this is a land-use change and so is considered to be deforestation rather than forest degradation. When fires in forests are not associated with an intentional conversion for a land-use change, this is considered to be forest degradation. The work by van der Werf et al.  included an analysis for tropical latitudes that partitioned the forest fires into two classes: non-deforestation fires (i.e. forest degradation), and deforestation fires.
It is clear that no estimates of CO2 emissions currently exist that incorporate all major forms of forest degradation. Thus, a systematic, consistent calculation approach is needed to allow for an estimation of all significant emissions across all tropical and subtropical developing countries. Such improved knowledge on emissions from forest degradation would allow decision makers to understand the extent of forest degradation and what opportunities there are to reduce associated emissions. As such, the goals of our work were to: (1) provide a consistent estimate of CO2 emissions from the major causes of degradation in the tropical and subtropical forests of developing countries, and (2) compare the magnitude of the emissions caused by forest degradation and its sub-activities with those from deforestation in both absolute and relative terms. Results from such an analysis would provide guidance to national and international policy makers as to which forest lands to allocate resources so that national GHG emissions are reduced.
Selective timber harvest in native forests.
Woodfuel harvest—where removals exceed regrowth of forest C stocks.
Fire—wildfires that do not cause a change in land-use.
Summary of activities, spatial scale, pools and time frame included in the analysis
Above and belowground live biomass, harvested wood products
Wood fuel harvest
GADM Level 1
Above and belowground live biomass
Global Fire Emissions Database 
Summed to GADM Level 1
Above and belowground live biomass, dead wood, and litter
Summed to GADM Level 1
Above and belowground live biomass, dead wood, litter, and soil carbon
Selective timber harvest
Source of field data for development of timber harvesting emission factors (ELE extracted log emission, LDF logging damage factor, LIF logging infrastructure factor)
Republic of Congo
Rest of Africa
Central America and Caribbean
Andean countriesb (Bolivia, Colombia, Ecuador, Paraguay, Peru, Venezuela)
Guyana, Suriname, French Guyana
Average annual industrial roundwood production (IRP), a measure of the extracted volumes, for the period of 2005–2010 was obtained from the FAO Global Forest Resources Assessment database (FAOSTAT) , as well as the country reports submitted to the FAO as part of the Forest Resource Assessment (FRA) program. Because the reported IRP include volumes produced from native forests and forest plantations, the reported IRP was adjusted to ensure that only timber production from native forests was considered (to capture only emissions from selective logging). For the majority of timber-producing countries included in the analysis (representing 96% of the total IRP), country-specific harvest volumes from plantations for the 2005–2010 timeframe reported in Jürgensen et al.  were subtracted from the average total industrial roundwood production volume, as reported by FAOSTAT for the same time period. For countries not included in Jürgensen et al. , no adjustments were made, as we assumed that IRP from plantations (if they exist) were insignificant.
Emissions from woodfuel were derived using the WISDOM model  that estimates the fraction of non-renewable biomass (NRB) in relation to supply and demand potential . In the WISDOM model, woodfuel derived as a byproduct of deforestation activities was not included in order to avoid double-counting deforestation emissions. As the WISDOM model estimates only include the aboveground biomass pool, an expansion factor of 1.32 was applied to conservatively estimate the total biomass, based on the American Carbon Registry’s Energy efficiency measures in thermal applications of nonrenewable biomass methodology , based on the CDM-approved methodology AMS‐II.G, Version 05.0. This factor assumes that for every unit of biomass extracted from the forest, an additional 10% is left in the field from uncollected aboveground biomass. A further 20% is conservatively estimated to remain from root biomass.
The Global Fire Emissions Database (GFED; ) was used for estimates of emissions from forest fire. The GFED provides a global monthly layer with a cell size of 0.5 decimal degrees (approx. 50 × 50 km) of dry matter emissions that are classified into different sources and land cover types. Within the humid tropical forest biome, fire emissions from deforestation are decoupled from other emissions based on fire persistence (the length of time for which a fire burns in the same location). To avoid double-counting with deforestation emissions, only emissions from GFED-classified forest fires within latitudes 23° North and South (and not deforestation fires) were used in this degradation category. The GFED3 monthly layers from 2005–2010 were used for this study, and emissions estimates for only CO2 are reported here in order to be consistent with other degradation activities.
Although there are several estimates of CO2 emissions from tropical deforestation published fairly recently (e.g. [1, 6, 13, 15, 26, 30]), these estimates were not used because they were not consistent with respect to carbon pools included, area of study, definition of forest, inclusion of other land-use changes, gross versus net emissions, and years covered. As one of our goals was to compare estimates of degradation emissions with those of deforestation, we believed it was important to estimate the emissions from deforestation in a manner consistent with our analysis of forest degradation (Table 1).
Emissions were obtained by multiplying the average forest carbon stocks for each administrative unit by the area of forest loss. We used the Hansen et al.  dataset, derived from Landsat 7 ETM+ satellite images, to determine the area of deforestation. Deforestation data was based on a canopy closure of 20% to ensure that deforestation in countries with more open, drier forests were well captured. Areas shown as loss (between 2005 and 2010) were considered to be deforested, and were summed across level-one subnational administrative units as defined by the GADM (Database of Global Administrative Areas; political boundaries reflecting states or districts).
Source of data for calculating emissions from deforestation
Saatchi et al. biomass map (; and unpublished update to 2011 increasing resolution from 500 to 250 m and adding additional ground data)
Forest mask for year 2005 from Hansen et al.  to exclude non-forest biomass pixels
Equations from Mokany et al. 
Dead organic matter
Fraction of aboveground biomass 
Soil organic matter
Peat soil emissions—annual emission factor for drained organic soil applied for 10 years (5.3 t CO2 ha−1 year−1; )
Non-peat soil emissions: C stock in top 30 cm from HWSD database
Land use change soil factors from IPCC 
Carbon stocks of the non-soil pools were derived as detailed in Table 3. Biomass was averaged across the subnational administrative units and carbon stocks from all pools were assumed to be committed to the atmosphere immediately at the time of deforestation. Emissions were obtained by multiplying the average forest carbon stocks for each administrative unit by the area of forest loss.
Estimated annual emissions from deforestation and forest degradation and relative proportions
Annual emission (Gt CO2e year−1)
Comparison of emissions from forest degradation
This study offers the first complete and consistent analysis of gross emissions from activities associated with the degradation of forest lands in developing countries in the tropical and subtropical latitudes. We estimated total forest degradation emissions of 2.1 Gt CO2e year−1, of which 53% was derived from timber harvest, 30% from woodfuel harvest, and 17% from forest fires.
Proportion of total forest degradation emissions by degrading activity by region
For another comparison we can specifically compare emissions from timber harvesting in the Brazilian Amazon. Huang and Asner  estimated annual gross emissions as 0.15–0.18 Gt CO2e year−1. Comparing just the nine Brazilian states that comprise the Amazon region, our study estimates emissions to be 0.28 Gt CO2e year−1, or more than 1.5 times higher than those reported by Huang and Asner. However, the Huang and Asner study explicitly stated that their estimate of gross annual emissions was likely to have substantially underestimated emissions due to the exclusion of areas that were deforested in subsequent years.
Emissions from deforestation versus forest degradation
The estimate of gross deforestation emissions presented in this study (average annual for 2005–2010 is 6.22 Gt CO2) is included primarily to serve as a basis for consistent comparison with the estimates of degradation emissions. Recent published estimates of deforestation emissions [1, 6, 13, 15, 26, 30] have been smaller than our estimate, ranging from 2.3 to 4.2 Gt CO2 year−1. There are several reasons for the discrepancy between these estimates, including a focus on net rather than gross emissions, different time periods which will capture lower historical rates of deforestation—e.g. 2000–2005  to 2001–2013 —and different study areas. All of the estimates generally include only aboveground biomass carbon stocks in trees (except , which also included belowground biomass), yet our estimate includes all five IPCC carbon pools, including aboveground, belowground, dead wood, litter, soil, and peat. Belowground biomass of forests is about 20% or more of aboveground biomass and dead wood and litter will account for at least another 5% of aboveground biomass. Emissions from mineral soil due to cultivation generally account for another 20–25% of aboveground stocks. Taking all these factors into account, the emissions from the other studies could increase by as much as a factor of 1.5, or to a range of 3.5–6.3 Gt CO2 year−1, while still not including significant peat soil emissions in Indonesia and Malaysia. In light of all this, we conclude that our estimate of deforestation emissions is in line with other recently published estimates mentioned above.
Emissions from forest degradation are not an insignificant source of CO2 and account for 25% of the summed emissions from deforestation and forest degradation of 8.28 Gt CO2 year−1. In other words, degradation emissions are equivalent to about a third of those from deforestation. According to the World Bank’s Carbon Fund, if emissions from forest degradation are more than 10% of all forest-related emissions, they must be included and accounted for. As we have shown, emissions from all sources of forest degradation were less than 10% in only 11 out of the 74 countries, and thus all the remaining countries would need to include forest degradation in their accounting system. The guidelines, however, only give instructions on summed forest degradation but not on individual activities. For example, in Colombia summed degradation emissions were equal to 9% of total emissions, but all the emissions are from timber harvest and thus could be excluded under FCPF rules. In contrast, the summed degradation emissions in Peru were 11% but the timber harvest emissions comprised 8% of total degradation. While Peru’s emissions from timber degradation are less significant than in Colombia, since total degradation emissions make up more than 10%, Peru would be required to also account for fire and woodfuel even though they sum to just 3% of emissions. Thus, there is a need for policies that better articulate the inclusion and exclusion of activities rather than the summed forest degradation level.
Significance of degradation emissions
Uncertainties and omitted sources
The purpose of this analysis was to demonstrate the scale of emissions from forest degradation in a manner that is to the best of our knowledge consistent and accurate. This requires accurate information on extent of the type of forest degradation and the associated emissions. For selective logging, there was concern about the data used to estimate emissions, as it may have included timber volumes derived from plantations. However, steps were taken to ensure that our estimates of IRP capture extraction rates only from native forests. The logging emission factors were developed using data only from a limited number of countries yet have very small error bounds, and the emission sources considered have significant relationships with forest characteristics . The fire analysis is spatially-specific and globally-consistent, and was designed to avoid double counting fire degradation emissions with fire emissions resulting from or associated with deforestation. The most uncertain emission source is woodfuel, given that the data are derived from a single year.
Estimates of emissions from timber harvest are likely to be underestimated due to the omission of illegal logging, assuming illegal logging is not included in national official statistics of IRP. It is important to acknowledge that research indicates that as much as 72% of logging is illegal in the Brazilian Amazon, 61% in Indonesia and 65% in Ghana .
Another omission is degradation through overgrazing. This source was included in Hosonuma et al. , who reported that this activity is responsible for 7% of the pantropical area of forest degradation (the least important form of degradation in the study). In addition, the impact of grazing is predominantly on regeneration, with damage to seedlings and saplings. The impact on forest carbon stocks is therefore small in the short term, though may be greater in later years as future generations of emergent trees are removed.
Our estimates show annual forest degradation emissions of 2.1 billion tons of carbon dioxide across 74 developing countries. To further illustrate the significance of this number: it exceeds both the total emission from highway vehicles (1.7 billion tons of carbon dioxide equivalents per year; fueleconomy.gov accessed 1/27/17), and the total emissions from power generation in the USA (1.9 billion tons of carbon dioxide equivalents per year; eia.gov accessed 1/27/17).
Our study demonstrates that, almost without exception, forest degradation emissions are significant. Indeed, by our estimates 85% of the countries studied surpass the defined minimum threshold and would be required to estimate forest degradation emissions under World Bank requirements for participation in the Carbon Fund REDD+ program.
Yet emissions from forest degradation are overlooked and not accounted in any complete or systematic way. It is imperative that this source of greenhouse gas emissions be better understood so that strategies that tap into the mitigation potential of addressing them may be developed. These strategies might in turn also offer significant economic and development opportunities.
This paper serves as a starting point to demonstrate the importance of forest degradation as a source of greenhouse gases, and to show where emissions are most significant—and thus where interventions may have the greatest impact.
- CO2 :
extracted log emissions
Food and Agriculture Organization
harvested wood products
Intergovernmental Panel on Climate Change
industrial roundwood production
logging damage factor
logging infrastructure factor
reducing emissions from deforestation and degradation, conservation of forest carbon stocks, sustainable management of forests and enhancement of forest carbon stocks
The study and manuscript preparation was led and conceptualized by TRHP, SB provided scientific guidance on analysis and was the co-lead author, GS led the spatial analysis and LM conducted timber harvest analyses and data visualization. All authors read and approved the final manuscript.
We acknowledge helpful comments from anonymous reviewers. We thank Jeff Murray for copy editing of an advanced version of the manuscript.
The authors declare that they have no competing interests.
Availability of data and materials
Data will be available on Winrock International’s website.
The database that forms the basis of this analysis was initially prepared under funding from the World Bank (contract 7167342). Subsequent support for analysis was derived from the Interamerican Development Bank (contract INE/CCS-RG-T2036-SN2/14).
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.
- Achard F, Ebeuchle R, Mayaux P, et al. Determination of tropical deforestation rates and related carbon losses from 1990 to 2010. Glob Change Biol. 2014;20:1–15.View ArticleGoogle Scholar
- American Carbon Registry. Switch from non-renewable biomass for thermal applications. Approved carbon accounting methodology under American Carbon Registry. http://americancarbonregistry.org/carbon-accounting/standards-methodologies/switch-from-non-renewable-biomass-for-thermal-applications. Accessed 25 Oct 2016.
- Asner GP, Knapp DE, Broadbent EN, Oliveira PJC, Keller M, Silva JN. Selective logging in the Brazilian Amazon. Science. 2005;310:480–2.View ArticleGoogle Scholar
- Bailis R, Drigo R, Ghilardi A, Masera O. The carbon footprint of traditional woodfuels. Nat Clim Chang. 2015;5:266–72.View ArticleGoogle Scholar
- Barlow J, Lennox GD, Ferreira J, et al. Anthropogenic disturbance in tropical forests can double biodiversity loss from deforestation. Nature. 2016. doi:10.1038/nature18326.Google Scholar
- Baccini A, Goetz SJ, Walker WS, et al. Estimated carbon dioxide emissions from tropical deforestation improved by carbon-density maps. Nat Clim Chang. 2012;2:182–5.View ArticleGoogle Scholar
- Bustamante MMC, Roitman I, Aide TM, Alencar A, et al. Toward an integrated monitoring framework to assess the effects of tropical forest degradation and recovery on carbon stocks and biodiversity. Glob Chang Biol. 2016;22:92–109.View ArticleGoogle Scholar
- FAO/IIASA/ISRIC/ISS-CAS/JRC. Harmonized world soil database (version 1.2). FAO, Rome, Italy and IIASA, Laxenburg, Austria. 2012.
- Feldpausch TR, Jirka S, Passos CAM, Jasper F, Riha SJ. When big trees fall: Damage and carbon export by reduced impact logging in southern Amazonia. For Ecol Manag. 2005;219(2–3):199–215View ArticleGoogle Scholar
- Food and Agriculture Organization of the United Nations. FAOSTAT database on agriculture. http://faostat3.fao.org/browse/F/*/E. Accessed 25 Oct 2016.
- GOFC-GOLD. A sourcebook of methods and procedures for monitoring and reporting anthropogenic greenhouse gas emissions and removals associated with deforestation, gains and losses of carbon stocks in forests remaining forests, and forestation. GOFC-GOLD Report version COP19-2, (GOFC-GOLD Land Cover Project Office, Wageningen University, The Netherlands); 2013.
- Hansen MC, Potapov PV, Moore R, et al. High-resolution global maps of 21st-century forest cover change. Science. 2013;342:850–3.View ArticleGoogle Scholar
- Harris NL, Brown S, Hagen SC, et al. Baseline map of carbon emissions from deforestation in tropical regions. Science. 2012;336:1573–6.View ArticleGoogle Scholar
- Hosonuma N, Herold M, De Sy V, De Fries RS, et al. An assessment of deforestation and forest degradation drivers in developing countries. Environ Res Lett. 2012;7:1–12.View ArticleGoogle Scholar
- Houghton RA. Carbon emissions and the drivers of deforestation and forest degradation in the tropics. Curr Opin Environ Sustain. 2012;4:597–603.View ArticleGoogle Scholar
- Huang M, Asner GP. Long-term carbon loss and recovery following selective logging in Amazon forests. Global Biogeochem Cycles. 2010. doi:10.1029/2009GB003727.Google Scholar
- IPCC. Definitions and methodological options to inventory emissions from direct human-induced degradation of forests and devegetation of other vegetation types. In: Penman J, Gytarsky M, Hiraishi T, Krug T, Kruger D, Pipatti R, Buendia L, Miwa K, Ngara T, Tanabe K, Wagner F, editors. Miura: Institute for Global Environmental Strategies (IGES); 2003.
- IPCC. 2006 IPCC guidelines for national greenhouse gas inventories. Prepared by the National Greenhouse Gas Inventories Programme. In: Eggleston HS, Buendia L, Miwa K, Ngara T, Tanabe K, editors. IGES, Japan. Volume 4 Agriculture, forestry and other land use. Paustian K, Ravindranath NH, Van Amstel A (coordinating lead authors); 2006.
- IPCC. 2013 Supplement to the 2006 guidelines: wetlands. Prepared by the task force on national greenhouse gas inventories of the IPCC. Chapter 2: Drained inland organic soils. Drosler M, Verchot LV, Freibauer A, Pan G (coordinating lead authors); 2013.
- Jürgensen C, Kollert W, Lebedys A. Assessment of industrial roundwood production from planted forests. Planted Forests and Trees Working Paper Series No. 48. Rome: FAO; 2014.
- Lawson S, MacFaul L. Illegal logging and related trade; indicators of the global response. London: Chatham House; 2010.Google Scholar
- Mokany K, Raison JR, Prokushkin AS. Critical analysis of root: shoot ratios in terrestrial biomes. Glob Chang Biol. 2006;12:84–96.View ArticleGoogle Scholar
- Page SE, Rieley JO, Banks CJ. Global and regional importance of the tropical peatland carbon pool. Glob Chang Biol. 2011;17:798–818.View ArticleGoogle Scholar
- Pearson TRH, Brown S, Casarim FM. Carbon emissions from tropical forest degradation caused by logging. Environ Res Lett. 2014. doi:10.1088/1748-9326/9/3/034017.Google Scholar
- Saatchi SS, Harris NL, Brown S, et al. Benchmark map of forest carbon stocks in tropical regions across three continents. Proc Natl Acad Sci. 2011;108:9899–904.View ArticleGoogle Scholar
- Tyukavina A, Baccini A, Hansen MC, et al. Aboveground carbon loss in natural and managed tropical forests from 2000 to 2012. Environ Res Lett. 2015;10:1–14.View ArticleGoogle Scholar
- UNFCCC. Estimation of carbon stocks and change in carbon stocks in dead wood and litter in A/R CDM project activities Version 2.0.0. EB 67 Report Annex 23. 2012.
- van der Werf GR, Randerson JT, Giglio L, et al. Global fire emissions and the contribution of deforestation, savannah, forest, agriculture, and peat fires (1997–2009). Atmos Chem Phys. 2010;10:11707–35.View ArticleGoogle Scholar
- World Bank. Carbon Fund Methodological Framework. 2013. https://www.forestcarbonpartnership.org/carbon-fund-methodological-framework. Accessed 25 Oct 2016.
- Zarin DJ, Harris NL, Baccini A, et al. Can carbon emissions from tropical deforestation drop by 50% in five years? Glob Chang Biol. 2015. doi:10.1111/gcb.13153.Google Scholar