Tropical moist forests account for 70 per cent of global living biomass. Deforestation and degradation – that is, the partial damage to tree stands – as well as the subsequent regeneration of forests therefore play a pivotal role in the global carbon cycle. Whilst the effects of large-scale tropical deforestation are well understood, the impacts of forest degradation have remained highly uncertain until now. A large-scale international study led by Dr Viola Heinrich from the GFZ Helmholtz Centre for Geosciences in Potsdam (Germany) and Dr Amelia Holcomb from the University of Cambridge in the United Kingdom now synthesises data from 146 studies on tropical moist forests since 1988 and offers one of the most comprehensive assessments to date of carbon dynamics following forest disturbances and regeneration. Among other things, the researchers show and quantify that forest degradation also leads to significant carbon losses in above-ground biomass. However, forests that retain some of their structural integrity sequester carbon much more quickly than areas that have been completely cleared by deforestation. The study is published in the journal Science Advances. The database created as part of the study and the quantitative meta-analysis will help to improve forest carbon modelling, national greenhouse gas inventories and reporting to the United Nations Framework Convention on Climate Change (UNFCCC).
Background: The importance of tropical moist forests for the global carbon cycle
Tropical moist forests play a pivotal role in the global carbon cycle. They account for around 70 per cent of the world's living biomass and historically account for approximately one-third of the global terrestrial carbon sink. Deforestation and forest degradation – that is, the partial damage to forests caused by factors such as selective logging, understory forest fires, windthrow or drought – reduce the potential of forests as carbon sinks and simultaneously lead to carbon emissions, as CO₂ is released when dead biomass is burned or decomposes.
"An accurate and consistent accounting of carbon losses and gains is of great importance for national and international reports such as the National Greenhouse Gas Inventories (NGHGI), Forest Reference Emissions Levels (FREL) and within the framework of the IPCC," explains Dr Viola Heinrich, a researcher in Section 1.4 "Remote Sensing and Geoinformatics" at the GFZ and co-lead author of the new study.
Whilst carbon losses associated with large-scale tropical deforestation are generally well documented and quantified, this has so far proved difficult in the context of forest degradation. Consequently, carbon gains from forest regeneration are currently presented in carbon cycle models and national reports with varying levels of detail. Carbon losses due to forest degradation are also either reported inconsistently or omitted entirely.
"Advances in satellite-based remote sensing since around 2015, combined with data from field plots and aerial photographs, are increasingly enabling a distinction to be made between deforestation and degradation and the associated carbon losses and gains," explains Viola Heinrich.
New, large-scale meta-analysis synthesises data from 146 studies
To better understand the climatic role of tropical forests and improve the representation of forest degradation and regeneration in carbon accounting and climate reporting, a harmonised database and a synthesis have now been developed as part of the study published in Science Advances. Led by Dr Viola Heinrich from the GFZ Helmholtz Centre for Geosciences in Potsdam (Germany) and Dr Amelia Holcomb, formerly at the University of Cambridge in the UK and now at the University of Maryland (USA), an international consortium of 41 authors from 34 institutions collaborated on the project.
The research team compiled data from 146 studies on tropical rainforests since 1988, thereby providing one of the most comprehensive and differentiated assessments to date of carbon dynamics following forest disturbances and regeneration, both – human-induced and natural.
Results I: Fires, logging and fragmentation drive major carbon losses
The new meta-analysis shows that immediate losses of above-ground carbon following human-induced disturbances can be substantial, and it identifies and quantifies the most severe drivers:
- Forest fires caused average carbon losses of 49%
- Selective logging led to average losses of 34%
- Edge-of-forest effects led to average losses of 31%
- More intense and frequent disturbances significantly increase carbon loss.
"Our synthesis highlights that degradation and regeneration processes must be taken into account more comprehensively in carbon reporting and modelling," concludes Viola Heinrich.
Results II: Degraded forests recover faster than fully cleared land
The analysis also revealed important differences in forest regeneration pathways.
After 20 years of regeneration, recovering degraded forests accumulated substantially more aboveground carbon than forests regenerating after complete deforestation. Carbon stocks were higher in recovering degraded forests (41–117%) than those in secondary forests regrowing after full clearing (1–74 %).
The results suggest that preserving forest structure, soils, seed sources, and ecological connectivity greatly enhances the capacity of tropical forests to recover carbon.
"Forest degradation occurs on a spectrum, and a degraded forest is not a lost forest—every action we can take to reduce the intensity and frequency of disturbance helps maintain the structural integrity these ecosystems need to bounce back," added Dr Amelia Holcomb, co-lead author from the University of Cambridge, UK and now at the University of Maryland, USA.
Significance of the study for improving carbon accounting and climate policy
The authors emphasise that the compiled database and the quantifying meta-analysis will help to improve the modelling of forest carbon and both better understand the overall climatic role of tropical forests and standardise and refine national greenhouse gas inventories and reporting to the United Nations Framework Convention on Climate Change (UNFCCC).
Marieke Sandker, from the Food and Agriculture Organisation (FAO), who was not involved in the study, added: "The values from this study can be combined with countries' carbon stock estimates from their national forest inventories to get degradation emission factors, filling a frequent gap in reporting. This not only helps understand carbon fluxes better but potentially opens the door to climate finance."
Countries are already expressing interest in applying the findings and underlying database to Forest Reference Emission Levels (FRELs) and other national reporting frameworks. Nigeria already used it to estimate emissions from forest degradation in its 2026 FREL submission.
The synthesis arrives at a critical moment as governments seek more accurate estimates of emissions and removals from forests to support climate mitigation targets, the implementation of the UN forest conservation framework REDD+, and biodiversity conservation.
"Substantial carbon losses and gains from degradation and recovery are now better characterised," concludes co-author Prof. Luiz Aragão from the National Institute for Space Research (INPE) in Brazil. "This evidence base can support policy measures to halt forest degradation and foster recovery as part of global climate mitigation efforts."
A global scientific collaboration
The study is the result of a large-scale international scientific collaboration involving researchers from tropical forest regions and research institutions around the world. It was developed through collaboration, starting in March 2024 as part of the workshop "Quantifying Regrowth and Recovery from Deforestation and Degradation" ( R2D2 ), which was funded by the European Space Agency (ESA) and the World Resources Institute (WRI) and supported by the GFOI R&D programme, whose aim is to improve the scientific basis for national forest monitoring and carbon assessment.
Original publication:
V. Heinrich, A. Holcomb et al., A meta-analysis of carbon losses and gains from tropical moist forest degradation and regeneration. Science Advances (2026) DOI: 10.1126/sciadv.adz1923
