The Industrial Revolution and the associated large-scale combustion of fossil fuels, as well as human activities associated with land use, have led to a significant increase in atmospheric levels of greenhouse gases, particularly CO₂. Terrestrial Ecosystems, the world's major carbon sink, have absorbed approximately 20 % of anthropogenic carbon dioxide (CO₂) emissions over the past three decades. Land cover change (LCC), caused by anthropogenic interventions and natural forces, represents one of the most dynamic components of global environmental change and has profound impact on carbon sequestration dynamicsDeforestation and forest degradation have accounted for about a third of anthropogenic carbon emissions since the Industrial Revolution. Conversely, mitigation measures such as afforestation initiatives have the potential to increase terrestrial carbon sinks and enhance the absorption of atmospheric CO₂.

New study, which quantified changes in net ecosystem productivity (NEP) from 1981 to 2019 using the BEPS model and high-resolution HILDA+ data, yielded surprising findings. Despite global forest loss and the expansion of agricultural land and urban areas, land cover changes have led to net carbon gain of 229 Tg CThis underlines the complexity of interactions between humans and the natural environment.

Key findings on the impact of land cover changes:

• Afforestation and reforestation (so-called "afforestation" and "reforestation") significantly increased NEP by 1559 Tg C, thereby largely compensated for losses caused by deforestation (–1544 Tg C). This means that the gains from newly created forests in the Northern Hemisphere were able to offset emissions from tropical deforestation.
• Although they occupied a smaller area, the newly formed forests demonstrated higher sequestration efficiency than degraded older forests, highlighting the key role of forest age in shaping the dynamics of the global carbon sink. Younger forests increased cumulative NEP by an average of 281 g C/m², while loss of older forests decreased it by 192 g C/m².
• During the initial phases (1980s–early 2000s), land cover changes predominantly reduced NEP. However, in the following decades, this trend gradually reversed to a net improvement in NEP, which is likely related to the maturation of newly regenerated forests, whose carbon sequestration capacity has surpassed losses from older, degraded forests.
• Regional differences:
  • Carbon gains were concentrated mainly in East Asia, North America and Europe. The conversion of pastures to forests contributed 64.5 % of national NEP gains in China, 63.7 % in the USA, and 30.5 % in Europe. In Russia, the gains were largely the result of the abandonment of agricultural land after 1990, leading to the conversion of grassland/shrubland to forests (56.7 %) and agricultural land to forests (40.1 %).
  • Losses occurred mainly in Amazonia and Southeast Asia, mainly due to deforestation. Indonesia experienced the largest NEP loss due to LCC (278 Tg C) due to extensive deforestation, including peatland drainage and conversion of forests to agricultural land. Brazil lost 157 Tg C, mainly due to conversion of Amazonian forests to pastures.

Overall, it was found that forest dynamics dominated LCC-induced NEP changes, with contrasting hemispheric patterns: afforestation/reforestation in the north versus deforestation in the tropics. In addition to forests, changes in agricultural land also had an impact, with its expansion negatively correlated with NEP changes.

These findings highlight the critical importance of afforestation, forestry and spatially informed land use strategies to enhance carbon sinks and support global carbon neutrality goals. To achieve carbon neutrality, countries, especially those in tropical rainforest areas with high deforestation rates (such as Indonesia, Brazil), need to prioritize measures to restore forests and manage existing stands. In addition, improving productivity within existing agricultural land can alleviate the need for agricultural expansion at the expense of forests. Spring

Glossary of key terms

• Land Cover Changes (LCC): Transformations in the type of vegetation or other material covering the Earth's surface, often caused by human activities (e.g. deforestation, urbanization, afforestation) or natural disturbances.
• Carbon sequestration: The process of removing carbon dioxide from the atmosphere and storing it in a carbon sink (e.g., in forests, soil, or geological formations).
• Net Ecosystem Productivity (NEP): A key metric of the carbon cycle that represents the net flow of carbon between an ecosystem and the atmosphere. It is the difference between gross primary productivity (GPP) and total ecosystem respiration (autotrophic and heterotrophic respiration).
• Biosphere-atmosphere Exchange Process Simulator (BEPS): A process-based, remote sensing-driven diagnostic model used to estimate carbon fluxes, such as NEP, based on meteorological data, vegetation parameters, and soil characteristics.
• HILDA+ (High-resolution Historic Land Dynamics Assessment+): A high-resolution land cover dataset (1 km) that synergistically integrates multi-source open data streams including remote sensing, land use reconstructions, and statistical records to track land cover changes over decades.
• Afforestation: Establishing a forest on land that has not been recently forested (e.g., agricultural land, pasture).
• Reforestation: Reforestation on land that was previously forested but the forest has been removed (e.g. due to deforestation or logging).
• Deforestation: Clearing forests for other land uses, such as agriculture, which results in the release of stored carbon into the atmosphere.
• Gross primary productivity (GPP): The total amount of organic matter (carbon) produced by photosynthesis in an ecosystem.
• Autotrophic respiration (RA): Respiration carried out by plants to maintain life processes, which releases CO2.
• Heterotrophic respiration (RH): Respiration carried out by soil microbial communities and other heterotrophs during the decomposition of organic matter, releasing CO2.
• TRENDYv10 ensemble: A suite of dynamic global vegetation models (DGVMs) used to simulate the carbon cycle and assess carbon budgets, often as the primary basis for global carbon budget estimates.
• Leaf Area Index (LAI): A dimensionless quantity defined as the total area of plant leaves per unit area of land. It is a key parameter for modeling ecosystem processes such as photosynthesis and carbon sequestration.
• Carbon neutrality: Achieving zero net release of carbon dioxide into the atmosphere, usually by balancing emissions with carbon removals or offsets.