Climate change represents one of the greatest challenges of our time. Although the central point of its mitigation is the reduction of greenhouse gas emissions, removal of atmospheric CO2 is also necessary, as technological solutions are not yet available in the necessary scales. Natural climate solutions (NCS), which rely on the ability of plants to capture CO2 and store carbon in tissues or in the soil, are considered a more direct route. Ecosystem restoration, including the restoration of degraded land to its original state, appears to be an important part of these solutions.

Recent global studies have suggested that ecosystem restoration could significantly offset some of the human carbon emissions since the Industrial Revolution. Influential models, such as those by Bastin et al. and Strassburg et al., suggested that restoration could recapture up to two-thirds of the anthropogenic carbon load. However, these models have been criticized for too "tree-centric" view of global ecosystems, which ignored different types of ecosystems and overlooked the negative impacts of afforestation on biodiversity and the functioning of non-forest ecosystems. Open ecosystems, such as grasslands and savannas also sequester significant amounts of carbon, with their carbon stocks mostly underground, protected from fire and drought – processes that can significantly reduce above-ground carbon in forests. Afforestation of open ecosystems, on the other hand, can lead to water scarcity, changes in fire regimes, reductions in biodiversity and albedo, which can outweigh the climate benefits.

New modeling study sought to address these criticisms and provide a more realistic estimate of carbon sequestration potential. The authors applied a modeling workflow to estimate the carbon sequestration potential of restoring forests, shrublands, grasslands, and wetlands by 2100. Their estimates are more accurate because they take into account:

• All major terrestrial ecosystem types.
• Both above-ground and below-ground carbon stocks.
• The rate of carbon sequestration (rather than total stocks, which would often not be developed by 2100).
• Carbon sequestration rate for the current ecosystem type to calculate net carbon gain.
• Biogeographic differences.
• Socioeconomic considerations that exclude built-up and intensive agricultural areas from regeneration.
• Sustainable land use practices.
• Future carbon emissions.
• Recovery implementation timeline.
• Current and future climate conditions (2061–2080) for predicting restoration goals.

According to this model, it is maximum carbon sequestration potential of 96.9 Gt carbon by 2100This represents 17.6 % of the total 640 Gt of carbon emissions since 1750If future emissions up to 2100 are taken into account (according to different Shared Socioeconomic Pathways (SSP) scenarios), this potential drops to 3.7–12.0 %. Of the total realistic potential (85.2 Gt with phased implementation), 49.4 Gt (58.1 %) for forests and 35.8 Gt (41.9 %) for open ecosystemsIncluding open ecosystems in global plans nearly doubles the potential for carbon sequestration while helping to prevent inappropriate tree planting.

The study also identified priority regions for recovery, which includes temperate regions such as the American prairies and the Central Asian steppes, not just the previously preferred tropical rainforests. This nuanced distribution offers more equitable opportunities for both high-income and low-income countries to contribute to large-scale restoration.

Despite this potential, the results suggest that ecosystem restoration has limited potential for climate change mitigation, even in the case of a large-scale transition to sustainable, low-carbon economies, as in the SSP1–2.6 scenario. Furthermore, when future climate conditions and changes in the state of existing natural ecosystems due to climate change are taken into account, the potential for natural climate solutions (NCSs) related to ecosystem restoration is close to zero. The model even predicts the continued loss of carbon sequestered in ecosystems, especially in tropical forest regions, although carbon is expected to increase at higher latitudes.

The EU Nature Restoration Act, adopted in 2024, aims to restore ecosystems on 20% of EU land and sea areas by 2030 and in all degraded habitats in Member States by 2050. The EU initiative is in line with global movements, including the UN Decade of Ecosystem Restoration. While these initiatives often link restoration to carbon storage, a new study recommends that ecosystem restoration was primarily aimed at restoring biodiversity, supporting livelihoods and resilience of ecosystem services. The contribution to climate mitigation will vary depending on the state transitions between vegetation types.

In summary, restoration of terrestrial ecosystems can play a role in mitigating climate change, but only in the near future and only in conjunction with by immediately transitioning to energy-efficient economies based on renewable resourcesIn other scenarios, the most appropriate goal of ecosystem restoration as NCS would be adaptation to climate change at the local to regional scale. Future policies should prioritize adaptation and optimize recovery activities for the benefit of vulnerable people, make mitigation plans more effective through strict emission reduction mechanisms instead of investing in compensation with uncertain results, and support the original goal of ecosystem restoration: fighting the biodiversity crisis and thereby increase the resilience of ecosystem services, instead of solely carbon sequestration. JRi

The study was published in the journal Nature Geoscience.