Search Results
Deforestation and conversion
Definitions
- TREE COVER LOSS: a stand-replacement disturbance which is considered to be clearing of at least half of tree cover within a 30-meter pixel. The exact threshold is variable both through space and time, and is biome-dependent (updated from Hansen et al., 2013). Such a change may or may not be permanent. Non-permanent tree cover loss routinely occurs in the context of logging, fire, or shifting agriculture. Tree cover loss data is often analyzed as a first step to measure deforestation.
- DEFORESTATION: a tree cover loss event that is permanent in nature, e.g., when forest is converted to cropland or cleared for development; or when it occurs within humid tropical primary forest boundaries (Forest Declaration Assessment, 2022).
Historical data and current trend methodology
- Different global data and methods can be used to approximate deforestation, though none perfectly captures trends in permanent forest loss. Here, we provide an estimate of deforestation using a combination of datasets available on Global Forest Watch. These are annual tree cover loss (Hansen et al. 2013) updated to 2024, dominant drivers of tree cover loss (Sims et al. 2025) updated to 2024, and extent of primary humid tropical forest (Turubanova et al. 2018). Deforestation is calculated as the area of tree cover loss within areas where the dominant driver was classified as permanent agriculture, hard commodities or settlements and infrastructure, as well as humid tropical primary forest loss due to expansion of shifting cultivation.
- Tree cover loss from fire, however, can be temporary in nature or lead to permanent land-use change. To illustrate this, one methodology—described in the State of Climate Action report—excludes all tree cover loss due to fire (Tyukavina et al. 2022) for estimating permanent forest loss, regardless of the dominant driver causing the loss. The other methodology—described in the Forest Declaration Assessment—excludes wildfires (Curtis et al. 2018) from the drivers of permanent forest loss.
- Critically, these estimates of deforestation have several limitations. In Hansen et al. (2013), the tree cover baseline for the year 2000 does includes both natural forests and forestry plantations, making it impossible to differentiate whether the tree cover loss identified from the year 2001 onward represents the loss of natural or planted forests. The dataset may also underestimate smaller-scale forest clearings due to the limitations of detecting such losses with medium-resolution (30 m) satellite data. Finally, improvements in the detection of tree cover loss—achieved with the incorporation of new satellite data and methodology changes between 2011 and 2015—may result in higher estimates of loss in recent years as compared to earlier years. For details on the improved methodology, see the technical blogs by Potapov et al. (2015) and by Weisse & Potapov (2021). Due to these adjustments in the methodology, data are only displayed from 2015 onward.
- Estimates of deforestation derived from remotely sensed data differ from those reported by countries due to differences in definitions, methods, and data sources.
- The regional classification adopted for this indicator is based on the global distribution of forests by climatic domain (FAO, 2020). The geographic distribution of each climatic domain was overlaid with national borders, and each country was assigned to the climatic domain that overlapped with the largest percentage of its area. Regions are defined by a combination of continent and climatic domain. Country boundaries and continent assignments are based on the Database of Global Administrative Areas (GADM), version 3.6. The regional aggregation is detailed in Annex B of the 2024 Forest Declaration Assessment.
Historical data sources
Full description, licensing and other information available at the original data source.
Conserving the world’s forests is essential for addressing the interconnected challenges of nature loss and climate change. Forest habitats are home to 80% of terrestrial plant and animal species. Conserving forests also prevents the release of their large carbon stores into the atmosphere, safeguarding their ability to continue sequestering carbon, and for tropical forests, maintaining the biophysical mechanisms that help to cool the planet. To help secure these benefits, the Glasgow Leaders’ Declaration (GLD) has established the collective goal of halting forest loss – which includes reaching zero gross deforestation – by 2030.
The world permanently lost 8.07 million hectares (Mha) of forests in 2024. To be on track to achieve zero deforestation by 2030, no more than 5 million ha should have been deforested globally in 2024. However, with over 8 Mha of deforestation last year, that target was exceeded by 63%. Despite year-to-year fluctuations, global deforestation remains near levels at the beginning of the decade, when global forest commitments were agreed. Maintaining the current annual deforestation rate, the world will fail to meet, with devastating consequences for human well-being, biodiversity, health, and economic prosperity.
Note: This estimate of deforestation uses an updated global map of tree cover loss drivers (from Sims et al. 2025), which results in approximately 20% higher deforestation estimates than previous years’ Forest Declaration Assessment estimates, which relied on a previous estimate of tree cover loss drivers (Curtis et al. 2018). Key changes in the underlying data arise from higher resolution, more specific driver classifications, and new treatment of fire-related forest loss. The new classification was applied retroactively across the entire time series.
Definitions
- GROSS GHG EMISSIONS FROM DEFORESTATION: estimates of gross greenhouse gas (GHG) emissions resulting from deforestation. These include aboveground carbon, belowground carbon, deadwood and litter carbon, as well as soil organic carbon. Carbon dioxide (CO2), methane (CH4), and nitrous oxide (N2O) emissions from peat drainage resulting from deforestation are also included (Harris et al., 2021).
Historical data and current trend methodology
- GHG emissions from global deforestation (measured in metric tons of carbon dioxide equivalent per year) are estimated by combining maps of benchmark carbon stocks in 2000, tree cover loss, gain, and extent and additional contextual geospatial data (Gibbs et al. 2025, updated through 2024). Our estimates of gross GHG emissions include aboveground carbon, belowground carbon, deadwood and litter carbon, as well as soil organic carbon. CO2, CH4, and N2O emissions from peat drainage and forest fires are also included. Emissions are attributed to deforestation using Sims et al. (2025, updated through 2024) following the same categories used for the global deforestation indicator. Gross GHG emissions from humid tropical primary forest loss (tCO2e/yr) are estimated by overlaying gross emissions from Gibbs et al. 2025 with humid tropical primary forest extent in 2001 (Turubanova et al., 2018).
- Analysis of current trends in GHG emissions from deforestation follows methods developed by the Forest Declaration Assessment and relies on data available on Global Forest Watch.
- The regional classification adopted for this indicator is based on the global distribution of forests by climatic domain (FAO, 2020). The geographic distribution of each climatic domain was overlaid with national borders, and each country was assigned to the climatic domain that overlapped with the largest percentage of its area. Regions are defined by a combination of continent and climatic domain. Country boundaries and continent assignments are based on the Database of Global Administrative Areas (GADM), version 3.6. The regional aggregation is detailed in Annex B of the 2024 Forest Declaration Assessment.
Historical data sources
Full description, licensing and other information available at the original data source.
Deforestation remains the primary source of gross emissions from land use, land-use change, and forestry (LULUCF), accounting for almost 10% of global net anthropogenic greenhouse gas (GHG) emissions in 2022.
In 2024 alone, gross GHG emissions from deforestation reached 4.6 billion metric tons of carbon dioxide equivalent (GtCO2e) – an increase compared to 4.5 GtCO2e, which was the annual average during the baseline period from 2018 to 2020. To put the scale of these emissions in perspective, if deforestation was its own country, it would have been the third-highest emitter after China and the United States in 2024.
Definitions
- HUMID TROPICAL PRIMARY FOREST: mature natural humid tropical forest cover that has not been completely cleared and regrown in recent history (Turubanova et al., 2018).
Historical data and current trend methodology
- Humid tropical primary forest loss estimates tree cover loss occurring within humid tropical primary forests (Turubanova et al., 2018).
- This estimate is limited to the humid tropics, as no corresponding map of primary forest is available globally.
- Finally, estimates of humid tropical primary forest loss derived from remotely sensed data differ from those reported by countries, due to differences in definitions, methods, and data sources.
- Analysis of current trends in humid tropical primary forest loss follows methods developed by the Forest Declaration Assessment and relies on data available on Global Forest Watch.
- The regional classification adopted for this indicator is based on the global distribution of forests by climatic domain (FAO, 2020). The geographic distribution of each climatic domain was overlaid with national borders, and each country was assigned to the climatic domain that overlapped with the largest percentage of its area. Regions are defined by a combination of continent and climatic domain. Country boundaries and continent assignments are based on the Database of Global Administrative Areas (GADM), version 3.6. The regional aggregation is detailed in Annex B of the 2024 Forest Declaration Assessment.
Historical data sources
Full description, licensing and other information available at the original data source.
Humid, tropical primary forests – the bulk of which are found in major forest basins in the Amazon, the Congo Basin, and Southeast Asia – are among the world’s most important ecosystems for carbon storage and biodiversity. Urgent action to conserve these valuable forests is essential for meeting the goal of halting forest loss by 2030.
Yet global progress toward eliminating humid tropical primary forest loss remains woefully inadequate, with a catastrophic surge in loss last year: 6.7 million hectares (Mha) of these forests were cleared in 2024 – an approximately 3 Mha increase compared to the year prior.
Throughout the period 2015-2024, agricultural expansion – both permanent agriculture and shifting cultivation – was the dominant driver of humid tropical primary forest loss across all tropical regions, and remains the top driver in other tropical regions in 2024.
However, the impacts caused by forest fires on the humid tropical primary forests of Tropical Latin America and the Caribbean were tremendous in 2024, and were primarily responsible for the spike observed in that region. Fire even surpassed agriculture as the leading driver of primary forest loss in that region.
Definitions
- GROSS GHG EMISSIONS FROM HUMID TROPICAL PRIMARY FOREST LOSS: estimates of gross greenhouse gas (GHG) emissions resulting from the loss of humid tropical primary forests. These include aboveground carbon, belowground carbon, deadwood and litter carbon, as well as soil organic carbon. Carbon dioxide (CO2), methane (CH4), and nitrous oxide (N2O) emissions from peat drainage are also included (Harris et al., 2021).
Historical data and current trend methodology
- GHG emissions from global deforestation (measured in metric tons of carbon dioxide equivalent per year) are estimated by combining maps of benchmark carbon stocks in 2000, tree cover loss, gain, and extent and additional contextual geospatial data (Harris et al., 2021, updated through 2023, as described by Global Forest Watch). Our estimates of gross GHG emissions include aboveground carbon, belowground carbon, deadwood and litter carbon, as well as soil organic carbon. CO2, CH4, and N2O emissions from peat drainage and forest fires are also included. Emissions are attributed to deforestation using Curtis et al. (2018, updated through 2023) following the same categories used for the global deforestation indicator. Gross GHG emissions from humid tropical primary forest loss (tCO2e/yr) are estimated by overlaying gross emissions from Harris et al. 2021 with humid tropical primary forest extent in 2001 (Turubanova et al., 2018).
- Analysis of current trends in GHG emissions from humid tropical primary forest loss follows methods developed by the Forest Declaration Assessment and relies on data available on Global Forest Watch.
- The regional classification adopted for this indicator is based on the global distribution of forests by climatic domain (FAO, 2020). The geographic distribution of each climatic domain was overlaid with national borders, and each country was assigned to the climatic domain that overlapped with the largest percentage of its area. Regions are defined by a combination of continent and climatic domain. Country boundaries and continent assignments are based on the Database of Global Administrative Areas (GADM), version 3.6. The regional aggregation is detailed in Annex B of the 2024 Forest Declaration Assessment.
Historical data sources
Full description, licensing and other information available at the original data source.
The destruction of humid tropical primary forests releases massive amounts of greenhouse gas (GHG) emissions, posing an alarming threat to global climate progress. This threat greatly intensified last year: in 2024, gross emissions from humid tropical primary forest loss totaled 3.1 billion metric tons of carbon dioxide equivalent (GtCO2e) – a nearly 76% increase relative to average annual GHG emissions from the baseline period, 2018 to 2020. To put the scale of these emissions in perspective, global emissions from humid tropical primary forest loss were greater than the combined emissions of United States’ emissions from energy systems (2.2 GtCO2e), buildings (574.2 MtCO2e), and agriculture (388 MtCO2e) in 2024.
Definitions
- MANGROVE LOSS: the replacement of mangroves with non-intertidal ecosystems at the 30-meter pixel scale, which includes both natural and human-caused losses (Murray et al. 2022)
Historical data and current trend methodology
- To monitor mangrove loss globally (in hectares per year; ha/yr), we used a dataset on tidal wetland change that estimates gross area of loss of tidal flats, tidal marshes, and mangroves from 1999 to 2019 (Murray et al., 2022). Murray et al. (2022) define mangrove loss as the replacement of mangroves with non-intertidal ecosystems at the 30-meter pixel scale, which includes both natural and human-caused losses, and, using this definition, estimated mangrove loss in three-year epochs. To convert these estimates to annual rates, we divided the gross loss for each epoch by the number of years in the epoch to determine the average annual loss rate in hectares per year.
- There are several limitations in using these data to assess progress toward our target for mangrove loss. Because loss area is estimated for three-year epochs, fewer data points are available from which to derive the historical trendline, and the trendline for this indicator was derived from the area of mangrove loss across four epochs. Furthermore, this dataset may also underestimate changes that occur at smaller scales or in narrow linear features such as waterways due to the limitations of detecting such changes with medium-resolution satellite imagery. A detailed assessment of the accuracy of these data can be found in Murray et al. (2022).
- Global Mangrove Watch, another commonly used dataset on mangrove extent and change, recently released a version 3.0 dataset that contains estimates of mangrove extent from 1996 to 2020 (Bunting et al., 2022). However, Bunting et al. (2022) recommend using only their net change estimates, rather than gross loss or gain, due to misregistration errors with the JAXA L-Band Synthetic Aperture Radar (SAR) data, which can lead to overestimation of individual loss and gain in some areas. JAXA is currently reprocessing all L-band SAR global mosaics, which will likely resolve this limitation in future versions of the Global Mangrove Watch data.
- Learn more about the methods for estimating mangrove loss (including the known limitations) in the most recent State of Climate Action report and State of Climate Action technical note.
Historical data sources
Full description, licensing and other information available at the original data source.
Stretching across nearly 15 million hectares (Mha) of shoreline globally, mangrove forests are among the world’s most carbon-dense ecosystems, holding at least twice as much carbon per hectare as boreal, temperate, and tropical forests. Due to the carbon density of these ecosystems, the loss of even a small area of mangroves, particularly when their soils are disturbed or dredged, can release an outsized amount of greenhouse gas emissions relative to other ecosystems.
Although average annual rates of global gross mangrove loss have slowed dramatically since the late 20th century, they appear to once again be ticking upward. From 1999 to 2019, the world lost an estimated 560,000 hectares (ha) of mangrove forests, with gross losses of these coastal wetlands increasing by an average of nearly 950 ha per year since 2008. Accordingly, global efforts to halt the conversion of mangrove forests have fallen short, and a sharp reversal in action is needed.
Forest and land ecosystem degradation
Definitions
- FOREST LANDSCAPE INTEGRITY INDEX (FLII): An index that tracks the ecological integrity of forests through a combined use of data on observed and inferred human pressures known to cause forest degradation. The FLII uses proxies for degradation, combining observable pressures (agriculture, forest cover loss, and infrastructure), inferred pressures based on distances from observed pressures (e.g., edge effects, overharvest), and losses in forest connectivity to give an aggregate score. However, the FLII is not designed to detect certain categories of human impact such as those related to climate change or changes of natural fire regimes.
Historical data and current trend methodology
- The extent of forests transitioning from a higher to a lower integrity class is based on the FLII score. Three integrity classes were originally defined: low (FLII ≤ 6.0); medium (FLII > 6.0 and < 9.6); and high ecological integrity (FLII ≥ 9.6). These thresholds were selected by benchmarking FLII scores against reference locations worldwide (Grantham et al., 2020).
- In recent years, the underlying algorithm has undergone further improvements and development. As a result, the value ranges for the three classes have been slightly adjusted compared to those listed in Grantham et al. (2020), to ensure consistency in the classification of points between the original and revised datasets. This indicator uses the revised ranges (FLII ≤ 7.0 for low integrity; FLII > 7.0 and < 9.8 for medium integrity; and FLII ≥ 9.8 for high integrity forests), which should be considered provisional until they are validated in a peer-reviewed paper.
- The value of the indicator reflects areas that have moved from a higher to a lower integrity class (i.e., high to low, high to medium, and medium to low) minus any areas that have increased in integrity. The value of the indicator excludes also areas that were permanently deforested. Notably, the indicator only accounts for anthropogenic degradation drivers such as the expansion of agriculture and urbanization, but does not account forest loss due to fires. The Glasgow Leaders’ Declaration calls for a halt to land degradation (including forest degradation) by 2030, and therefore, the 2030 target is set at zero further degradation—meaning no more forests transitioning to a lower integrity class.
- Analysis of current trends in annual change in FLII follows methods developed by the Forest Declaration Assessment and relies on data from Wildlife Conservation Society.
Historical data sources
Full description, licensing and other information available at the original data source.
2024 Forest Declaration Assessment: Forests under fire – Forest Degradation
2024 Forest Declaration Assessment using data from Wildlife Conservation Society
Visit this Source
The Forest Landscape Integrity Index (FLII) provides annual estimates of global forest degradation. The index tracks changes in forest extent, forest connectivity, direct pressure from human activities, and inferred pressure from edge effects to estimate forest integrity through an FLII score. Higher scores correspond to higher levels of forest integrity, while lower FLII scores correspond to a decrease in forest integrity. In other words, decreasing FLII scores imply increasing forest degradation. Halting and reversing forest degradation translates into no reduction or an increase in the FLII score at the global and regional levels.
According to FLII data available through 2022, extensive forest degradation has occurred globally and within all regions, including both tropical and non-tropical regions of Africa, Asia, Latin America, the Caribbean, Europe, and North America. A net total of 62.6 million hectares of forest fell to a lower ecological integrity class in 2022 – 10 times the area that was deforested in the same year.
However, despite this large absolute area of degradation, in 2022, the rate of degradation as measured by the FLII was lower than baseline levels (38% less degradation than the 2018-20 average) and from the year prior (29% less degradation than 2021). These findings signal that degradation due to human-induced factors is declining, which is good news. That said, the FLII does not account for the impact of intensifying forest fires, which could derail other progress on reducing degradation drivers.
Definitions
- TREE COVER LOSS: a stand-replacement disturbance which is considered to be clearing of at least half of tree cover within a 30-meter pixel. The exact threshold is variable both through space and time, and is biome-dependent (updated from Hansen et al., 2013). Such a change may or may not be permanent. Non-permanent tree cover loss routinely occurs in the context of logging, fire, or shifting agriculture. Tree cover loss data is often analyzed as a first step to measure deforestation.
- INTACT FOREST LANDSCAPES (IFLs): Mosaics of forested and naturally treeless ecosystems that show very few signs of significant human activity or habitat fragmentation and are large enough to maintain all native biodiversity, including viable populations of wide-ranging species (Potapov et al., 2017). They can include temporary treeless areas after natural disturbances, water bodies, or treeless intact ecosystems where climate, soil, or hydrological conditions prevent forest growth.
Historical data and current trend methodology
- The concept of Intact Forests Landscapes was developed by a team of research and environmental organizations such as the University of Maryland, Greenpeace, World Resources Institute, and Transparent World. The global IFL map was first developed for the year 2000 then updated in 2013, 2016, and 2020 by the IFL Mapping Team with the support from Greenpeace, The University of Maryland, Wildlife Conservation Society, Transparent World, WWF Russia, and the World Resources Institute.
- The indicator represents the area of tree cover loss (Hansen et al. 2013) within the boundaries of Intact Forest Landscapes for the year 2000 (Potapov et al. 2017).
- The regional classification adopted for this indicator is based on the global distribution of forests by climatic domain (FAO, 2020). The geographic distribution of each climatic domain was overlaid with national borders, and each country was assigned to the climatic domain that overlapped with the largest percentage of its area. Regions are defined by a combination of continent and climatic domain. Country boundaries and continent assignments are based on the Database of Global Administrative Areas (GADM), version 3.6. The regional aggregation is detailed in Annex B of the 2024 Forest Declaration Assessment.
Historical data sources
Full description, licensing and other information available at the original data source.
The last frontiers of wilderness: Tracking loss of intact forest landscapes from 2000 to 2013
Peter Potapov, Matthew C. Hansen, Lars Laestadius, Svetlana Turubanova, Alexey Yaroshenko, Christoph Thies, Wynet Smith, Ilona Zhuravleva, Anna Komarova, Susan Minnemeyer, and Elena Esipova
Visit this Source
Intact forest landscapes are mosaics of forested and naturally treeless ecosystems that show very few signs of human activity or habitat fragmentation. Occupying a minimum area of 50,000 hectares, they are large enough to play a critical role in helping to maintain native biodiversity. Accordingly, these ecosystems are hotspots for biodiversity and contain large carbon stores. Reducing tree cover loss within these natural terrestrial ecosystems is a key part of halting forest loss and land degradation by 2030.
Yet, annual rates of tree cover loss across these intact forest landscapes have been on the rise since 2001. In 2024 alone, 6.8 million hectares were lost – a 19% increase compared to 2023 and more than double the average annual loss from 2018 to 2020. Though not all tree cover loss is permanent, the increasing trend likely indicates more degradation and fragmentation of these ecosystems, as well as a rise in human activity. Efforts to address tree cover loss in intact forest landscapes must be accelerated, urgently and rapidly, to reverse this concerning trend.
Definitions
- TROPICAL MOIST FORESTS (TMFs): Include all closed forests in the humid tropics with two main forest types: the tropical rain forest and the tropical moist deciduous forest. Tropical rain forest is found in permanently humid areas, i.e., those with, at most, only limited seasonality in rainfall distribution, while tropical moist deciduous forest, also called monsoon forest, is found in areas with a distinct dry season. The TMFs are characterized by low variability in annual temperature and high levels of rainfall (>200 cm annually). In the Holdridge life zones classification scheme, the TMFs include the moist forest, the wet forest, and the rain forest. The TMFs are located mostly in the tropical moist and humid climatic domains, but they also include small areas of gallery forests in the tropical dry domain.
- DEGRADED FORESTS: moist forest cover where disturbances were observed over a period shorter or equal than 2.5 years (900 days). The methodology assumes that the duration of the disturbance – and, consequently, the period over which the disturbance is detectable in satellite imagery – is a proxy of the disturbance impact, where disturbance of longer duration corresponds to higher impact on the forest and the higher the risk of having a permanent conversion of the TMF. All disturbance events for which the impacts are detectable over a period of more than 2.5 years (900 days) are considered as deforestation processes.
Historical data and current trend methodology
- The indicator presents the extent of degraded tropical moist forests, defined as moist forest cover where disturbances were observed over a period shorter than 2.5 years.
- Country data is retrieved from the TMF Data Portal to calculate regional aggregates.
- The regional classification adopted for this indicator is based on the global
distribution of forests by climatic domain (FAO, 2020). The geographic
distribution of each climatic domain was overlaid with national borders, and
each country was assigned to the climatic domain that overlapped with the
largest percentage of its area. Regions are defined by a combination of
continent and climatic domain. Country boundaries and continent
assignments are based on the Database of Global Administrative Areas
(GADM), version 3.6. The regional aggregation is detailed in Annex B of the 2024 Forest Declaration Assessment.
Historical data sources
Full description, licensing and other information available at the original data source.
In 2024, 8.8 million hectares of tropical moist forests experienced degradation, which is more than twice the annual level necessary to halt the degradation of tropical moist forests by 2030.
Fire-induced degradation in the Amazon Basin significantly contributed to a sharp increase in the extent of degraded tropical moist forests from 2023-24. Amazon fires caused the emission of 791 million metric tons of carbon dioxide equivalent, which is more than the total annual emissions of a high-income country like Germany. The Amazon rainforest is under mounting ecological stress, with 17 to 38% of its area already degraded.
Projections suggest that degraded forest area could reach up to 47% by 2050, driven by a combination of climate-related disturbances – such as extreme drought, variable rainfall, and storms – and human pressures like logging, fires, and land-use change.
Degraded forests, particularly those that have lost over 50% of their canopy structure, face a higher risk of deforestation. In other words, degradation is a good indicator of future deforestation, with the likelihood of total deforestation and land use change increasing as degradation worsens. Data from Latin America, Africa, and Asia indicate that degraded forests that experienced deforestation after 2020 previously had significantly lower canopy heights and above-ground biomass compared to those that were not deforested. On average, degraded forests in Latin America exhibited a higher risk of deforestation than those in Africa or Asia.
Historical data and current trend methodology
- We used data from Conchedda and Tubiello (2020) on the annual change in the area of histosols (i.e., soils comprised primarily of organic matter) drained for agriculture, including for the cultivation of crops and grazing, as a best available proxy for peatland degradation. Using these data, we calculated the total increase in the area of histosols drained for agriculture over the study time period (1993–2018) and divided the total increase in area by the number of years to determine the average annual rate of drainage. Using the Harmonized World Soil Database, Conchedda and Tubiello (2020) define histosols as soils with a thick layer of strongly decomposed acidic organic material (70 centimeters thick) and with continuous rock at 80 centimeters, that develop in environments with a large excess of precipitation (Conchedda and Tubiello, 2020; FAO and IIASA, 2012).
- While the area of histosols drained for agriculture represents a best available proxy for peatland degradation, these data may underestimate peatland degradation for several reasons: First, the data estimate drainage of histosols solely for agricultural activities, and although agriculture is a primary driver of peatland degradation globally, other causes of degradation— including road and infrastructure development, forestry, oil sands mining, and peat extraction, among others—are not included in the estimates (Conchedda and Tubiello 2020; UNEP 2022).
- Moreover, the threshold of peat depth used to define peatland varies by country, and some countries have yet to establish a nationally recognized definition of peat altogether (e.g., Myanmar, Lao People’s Democratic Republic, Cambodia) (Sulaeman et al., 2022). In nations where this threshold is lower than the depth of organic material used to define organic soil in Conchedda and Tubiello (2020), peatland degradation may not be included in these estimates of drained organic soils. For example, if the threshold used to define peatlands is two meters of organic matter, but the threshold used to define organic soils is three meters of organic matter, then these peatlands would be excluded from this estimate of organic soils. As a result, the global extent of histosols is significantly lower than most recent estimates for peatland area (e.g., Xu et al., 2018; UNEP 2022), and estimates of the area of histosols drained for agricultural activities (25 Mha) are substantially lower than estimates of the global area of degraded peatlands (57 Mha) (Conchedda and Tubiello, 2020; UNEP, 2022)
- Learn more about the methods for estimating peatland degradation (including the known limitations) in the most recent State of Climate Action report and State of Climate Action technical note.
Historical data sources
Full description, licensing and other information available at the original data source.
Covering just 3.8% of the planet’s land, peatlands – also known as mires, bogs, fens, and swamp forests – are global hotspots for carbon sequestration and long-term storage. They also hold large stores of organic nitrogen as their water-logged soils slow decomposition, allowing carbon and nitrogen-rich peat to accumulate over millennia.
But when these ecosystems’ water tables fall, oxygen enters the upper layers of peat, spurring decomposition and subsequent losses of stored carbon and nitrogen. These degraded peatlands can emit carbon dioxide and nitrous oxide for decades to centuries until all peat is fully lost or their soils are rewetted. Peatlands are threatened by drainage for agriculture, forestry, and peat extraction, which is intensified by industrial development.
An estimated 57 million hectares (Mha) – nearly 12% of the world’s peatlands – are degrading such that they are no longer actively forming peat, and peat accumulated over centuries to millennia is now disappearing. Collectively, these degraded peatlands emit about 1.9 gigatonnes of carbon dioxide equivalent (GtCO2e) each year – roughly equivalent to Russia’s greenhouse gas (GHG) emissions in 2020. This estimate, however, excludes GHG emissions from peat fires that, while highly variable and difficult to measure, likely occur on an order of magnitude from 0.5 to 1 GtCO2e annually.
Halting peatland degradation by 2030 can help to limit global warming. However, despite recent advances in mapping peatlands, significant data gaps such as incomplete coverage, inconsistent quality, and outdated data inhibit efforts to monitor progress. Data estimating the area of organic soils drained for agriculture provide the best available (though still imperfect) proxy, and they indicate that degradation of the world’s peatlands continues.
Forest biodiversity
Definitions
- TREE COVER LOSS: a stand-replacement disturbance which is considered to be clearing of at least half of tree cover within a 30-meter pixel. The exact threshold is variable both through space and time, and is biome-dependent (updated from Hansen et al., 2013). Such a change may or may not be permanent. Non-permanent tree cover loss routinely occurs in the context of logging, fire, or shifting agriculture. Tree cover loss data is often analyzed as a first step to measure deforestation.
- AREAS OF HIGH BIODIVERSITY SIGNIFICANCE: Forests that are “highly significant” for biodiversity; that is, forests that are disproportionately important for the concentration of species that they support. Forest biodiversity significance is determined by the rarity-weighted richness of forest mammal, bird, amphibian and conifer species. “Highly significant” refers to the top 10% of biodiversity significance areas.
Historical data and current trend methodology
- The indicator estimates the area of tree cover loss (Hansen et al. 2013) in forests of high biodiversity significance (Hill et al. 2019).
- For full methodology, visit Global Forest Watch.
Historical data sources
Full description, licensing and other information available at the original data source.
Stretching across roughly 460 million hectares (Mha) as of 2018, forests that are highly significant for biodiversity are disproportionately important for supporting forest-dependent species. Designation of these areas of high significance for biodiversity accounts for both species richness and endemism across forests globally and is complementary to key biodiversity areas, which also include important areas for geographically restricted species.
Loss of forest habitat in areas with high significance for biodiversity, specifically, may have outsized impacts on the species that inhabit these areas, but tree cover loss in these areas continues to occur. In 2023, for example, the world lost 2.6 Mha of tree cover in areas of high significance for forest biodiversity, an 18% increase from 2022.
Note: “Areas highly significant for forest biodiversity” and “Key Biodiversity Areas” are two different classifications with some overlap. Key Biodiversity Areas use standardized IUCN criteria established in 2016, while areas of high biodiversity significance are identified using broader, more flexible criteria.
Definitions
- TREE COVER LOSS: a stand-replacement disturbance which is considered to be clearing of at least half of tree cover within a 30-meter pixel. The exact threshold is variable both through space and time, and is biome-dependent (updated from Hansen et al., 2013). Such a change may or may not be permanent. Non-permanent tree cover loss routinely occurs in the context of logging, fire, or shifting agriculture. Tree cover loss data is often analyzed as a first step to measure deforestation.
- KEY BIODIVERSITY AREAS (KBAs): sites of global significance for the conservation of biodiversity identified through globally standardized criteria and thresholds (IUCN, 2022). There are 11 criteria organized into five categories: threatened biodiversity, geographically restricted biodiversity, ecological integrity, biological processes, and irreplaceability.
- FORESTED KBAs: sites contained in the tree cover in the year 2000 (Hansen et al., 2013), for which the Forest Landscape Integrity Index (Grantham et al., 2020) was available, and that supported at least one forest-dependent species that triggered KBA criteria at the site (Crowe et al., 2023). Season was taken into account for migratory species that are not forest-dependent throughout their annual life cycle.
- FOREST-DEPENDENT SPECIES: any species listed in the IUCN Red List (IUCN, 2023) for which forests are a major habitat.
Historical data and current trend methodology
- The indicator estimates the area of tree cover loss (Hansen et al., 2013`) occurring within forested KBAs (Crowe et al., 2023).
- Analysis of current trends in tree cover loss in forested KBAs follows methods developed by the Forest Declaration Assessment and relies on data from Global Forest Watch and the Key Biodiversity Area Secretariat.
- The regional classification adopted for this indicator is based on the global distribution of forests by climatic domain (FAO, 2020). The geographic distribution of each climatic domain was overlaid with national borders, and each country was assigned to the climatic domain that overlapped with the largest percentage of its area. Regions are defined by a combination of continent and climatic domain. Country boundaries and continent assignments are based on the Database of Global Administrative Areas (GADM), version 3.6. The regional aggregation is detailed in Annex B of the 2024 Forest Declaration Assessment.
Historical data sources
Full description, licensing and other information available at the original data source.
Forest cover loss in key biodiversity areas (KBAs) is particularly concerning, as these areas play an outsized role in conserving biodiversity due to, for example, being ecologically intact or hosting species that live in just a few geographies.
In 2024, nearly 2.2 million hectares of tree cover were lost within forested Key Biodiversity Areas (fKBAs), marking an increase of 23% compared to the baseline level, and 47% from 2023. This is particularly disappointing given significant strides in reducing tree cover loss in KBAs had been made from 2020 to 2022, and because even small amounts of loss within KBAs can significantly harm biodiversity.
Note: “Areas highly significant for forest biodiversity” and “Key Biodiversity Areas” are two different classifications with some overlap. Key Biodiversity Areas use standardized IUCN criteria established in 2016, while areas of high biodiversity significance are identified using broader, more flexible criteria.
Definitions
- FOREST-DEPENDENT SPECIES: Otherwise known as forest specialists, i.e., species dependent on forest habitats for their survival or reproduction. Defined as vertebrate species that only occur in forest habitats, per the IUCN Red List.
- FOREST SPECIALISTS INDEX: A tool developed to measure trends in the size of populations of threatened and non-threatened forest specialist vertebrate species (Green et al., 2020). It was created by applying the Living Planet Index (LPI) methodology specifically to forest specialists.
Historical data and current trend methodology
- Data and underlying definitions are presented in the 2023 Forest Declaration Assessment, drawing from the Forest Specialist Index, a subset of the Living Planet Index.
- Full description, licensing and other information available at the original data source.
Historical data sources
Full description, licensing and other information available at the original data source.
Monitoring shifts in species populations offers an important and complementary measure of forest biodiversity because changes in forest cover do not always directly correlate with impacts on species living in these ecosystems.
The Forest Specialists Index (FSI) tracks changes in vertebrate forest specialist populations. Vertebrate forest specialists are species of birds, mammals, reptiles, and amphibians that rely on forest habitats for their survival or reproduction. In 2022, FSI found that between 1970 and 2018, the abundance of vertebrate forest specialist populations declined an average of 79%. Habitat loss, habitat degradation, overexploitation, and climate change are major threats to forest biodiversity.
Forest and land ecosystem restoration
All modeled pathways limiting global temperature rise to 1.5°C with no or limited overshoot rely on carbon removal. Reforestation is a relatively cost-effective, readily available approach to carbon removal that, when implemented appropriately (i.e., by focusing on recovering forests’ ecological functions rather than solely on replanting trees), can also generate additional benefits for adaptation, sustainable development, and biodiversity conservation. Restoring terrestrial ecosystems, including forests, is also a standalone goal of the Glasgow Leaders’ Declaration (GLD).
While data limitations pose significant challenges to monitoring reforestation globally, remotely sensed data on the gross area of tree cover gain offers the best available proxy. However, this data may include tree cover gains that, while potentially beneficial for climate mitigation, do not meet common definitions of reforestation, such as afforestation across historically non-forested lands or regrowth after harvesting within already established plantations. In addition, increases in tree cover occur gradually as these plants grow, making it more challenging to reliably estimate using satellite remote sensing methods over short timescales.
Still, historical cumulative data suggests that worldwide, a total of 130 million hectares (Mha) experienced tree cover gain from 2000 to 2020. However, this average annual rate of tree cover gain (6.5 Mha per year) will need to accelerate to help limit warming to 1.5°C; reverse forest loss as pledged in the GLD; achieve Target 2 of the Global Biodiversity Framework; and deliver the Bonn Challenge pledge to bring 350 Mha of deforested and degraded land into restoration by 2030.
Critically, while reforestation is needed to meet climate and biodiversity goals, it cannot serve as a substitute for protecting standing forests. For example, it may take decades (if not longer) for these ecosystems to regain species diversity, ecosystem structure, and ecological functions, all of which may impact carbon cycling and greenhouse gas fluxes within these ecosystems.
Note: We used tree cover gain (total gross area gained from 2000 to 2020) as the best available proxy indicator for reforestation. Potapov et al. (2022) define tree cover gain as the establishment or recovery of tree cover (woody vegetation with a height greater than or equal to five meters) by the year 2020 in areas that did not have tree cover in the year 2000. Historical data was estimated using maps derived from remotely sensed data, and accordingly, they contain a degree of uncertainty.
Learn more about the methods for estimating reforestation (including the known limitations) in the most recent State of Climate Action report.
Full description, licensing, and other information are available at the original data source (Potapov et al. 2022).
Historical data and current trend methodology
- Although data are insufficient to assess progress of peatland restoration worldwide, there is evidence of rewetting, for example, in Canada, Indonesia, and Russia (UNEP 2022; Sirin 2022; BRGM 2023).
- Learn more about the methods for estimating peatland restoration (including the known limitations) in the most recent State of Climate Action report. Data is compiled from multiple sources.
Historical data sources
Full description, licensing and other information available at the original data source.
Ecosystem Service Restoration after 10 Years of Rewetting Peatlands in NE Germany
Zerbe, S., P. Steffenhagen, K. Parakenings, T. Timmermann, A. Frick, J. Gelbrecht, and D. Zak
Visit this SourceGelar Refleksi Akhir Tahun, BRGM Ingin Capaian Tahun 2021 Berlanjut
BRGM (Badan Restorasi Gambut Dan Mangrove)
Visit this SourceAkhir Tahun, BRGM Gelar Evaluasi Sekaligus Persiapkan Kegiatan Restorasi Gambut dan Rehabilitasi Mangrove Tahun 2023
BRGM (Badan Restorasi Gambut Dan Mangrove)
Visit this Source
Even if peatland degradation ended today, degraded peatlands could continue emitting roughly 1.9 gigatonnes of carbon dioxide equivalent (GtCO2e) per year for decades to centuries because, unlike forests, peatlands store carbon primarily within their waterlogged soils rather than in aboveground vegetation. Carbon and nitrogen losses following land-use changes continue until the soil is rewetted or all peat is lost. The efficacy of restoring peatlands to avoid these greenhouse gas (GHG) emissions, however, will depend in part on what form of degradation the wetland ecosystems experienced (e.g., drainage, burning, or cutting). Rewetting peatlands drained by agriculture, for example, can significantly reduce or even halt carbon losses, as well as enable carbon sequestration. Because drained peatlands will emit carbon dioxide and nitrous oxide for up to hundreds of years, restoring these ecosystems’ water tables should occur as quickly as possible to maximize avoided GHG emissions.
Although data is insufficient to assess global progress made in restoring peatlands, available evidence suggests that current efforts are occurring but likely not at the pace and scale required across many countries. From 2010 to 2013, for example, the Russian government implemented one of the largest-scale peatland rewetting projects in the Northern Hemisphere across more than 73,000 hectares (ha) near Moscow. During the early 2000s, Germany rewetted more than 20,000 ha of peatlands in one of its northeastern states. Likewise, Indonesia reported that it rewetted a total of 1.8 million ha from 2016 to 2023.
Restoring mangrove forests enhances their ability to sequester and store carbon and may also reduce greenhouse gas emissions after certain disturbances – such as the loss of soil organic carbon following drainage for aquaculture ponds – that otherwise would have continued for decades. Mangrove restoration in suitable areas could capture 0.93 GtCO2 at just $11.49 per ton of CO2, per a 2025 study. This falls well below the social cost of carbon emissions, even without accounting for the extensive ecological and economic benefits that thriving mangrove ecosystems generate. Monitoring mangrove restoration, however, remains challenging. Mangroves grow gradually, and therefore, restoration is challenging to monitor on shorter timescales, as gain may not be detected until mangrove trees reach a certain level of maturity. Moreover, the establishment of mangrove trees does not always indicate restoration of the ecological functions of these ecosystems, and in some cases, the addition of mangroves can lead to negative consequences (e.g., the loss of other coastal ecosystems) or only short-lived gains if tree-planting is not implemented appropriately.
Still, available global estimates indicate that the world gained approximately 180,000 hectares (ha) of mangrove forests from 1999 to 2019, but only 8% of these gains (15,000 ha) can be attributed to direct human interventions, such as mangrove planting or other restoration activities. Although mangrove gain due to direct human interventions does not indicate whether the establishment of these mangroves restored the ecological function of these ecosystems, it does provide the best available proxy for mangrove restoration.
Note: Murray et al. (2022) estimated that a gross area of 180,000 ha (9% confidence interval of 0.09 to 0.30 Mha) of mangrove gain occurred from 1999 to 2019, only 8% of which can be attributed to direct human activities, such as mangrove restoration or planting. We estimated the most recent data point for mangrove restoration by taking 8% of the total mangrove gain from 1999 to 2019 (15,000 ha). Historical data were estimated using maps derived from remotely sensed data, and accordingly, they contain a degree of uncertainty.
Learn more about the methods for estimating mangrove restoration (including the known limitations) in the most recent State of Climate Action report.
A detailed assessment of the accuracy of these data can also be found in Murray et al. (2022).
