Large ESM ensemble reveals complex responses of carbon and climate feedbacks to forestation across emission pathways

Reaching the Paris Agreement’s climate targets will require the large-scale deployment of Carbon Dioxide Removal (CDR), including Afforestation/Reforestation (AR). Carbon sequestration through AR is driven by plant metabolic processes affected by environmental conditions. However, AR-induced reduction of atmospheric CO₂ levels causes compensating CO₂ fluxes towards the atmosphere across the land and ocean. Further, beyond the CO₂-induced reduction in temperature, AR also affects climate through local and non-local biogeophysical effects caused by changes in albedo, surface roughness, and transpiring leaf area. Given the breadth and interaction of Earth system effects of AR, the amount of CDR achieved depends not only on the scale and spatiotemporal pattern of the application, but also on ambient climate and CO₂ levels, as determined by the emissions pathway, and complex emerging feedbacks. At the same time, understanding whether AR can cause a (non-)local warming that could potentially offset the cooling induced by the AR-driven CO₂ reduction, whether this might hold across different emissions scenarios, and whether this signal can emerge from internal variability, is also crucial.

Here, using the fully coupled Earth System Model MPI-ESM, we create a multi-member ensemble of emission- and concentration-driven AR and reference simulations across different emissions pathways (SSP1-2.6, SSP5-3.4os, SSP3-7.0, SSP5-8.5). Our setup features an unprecedented number of 120 simulations in total, that allows us to robustly capture the impacts on the Earth system and the emerging climatic and carbon feedbacks across spatiotemporal scales. In the AR scenario, forest area increases by 935 Mha by 2100, representing ambitious AR in the range of country pledges (Moustakis et al. 2024).

Our results show that, under higher emissions, AR not only sequesters more carbon over land, but also does so more efficiently. In particular, for  every 100 GtCO₂ sequestered over land (compared to the counterfactual reference scenario), atmospheric reduction reaches 89, 85, 74, and 73 GtCO₂ in SSP5-8.5, 3-7.0, 5-3.4os, and 1-2.6 respectively. The reduction of carbon sequestration due to the AR-induced reduction in atmospheric CO₂ can reach 29% in SSP1-2.6, which is significantly higher than the 7% loss in SSP5-8.5. Despite AR being more efficient under higher emissions, this is not translated to gains in temperature reduction, which is not statistically significantly different between scenarios, averaging at 0.2°C globally. Overall, CO₂-induced cooling dominates biogeophysically-induced warming at both global and regional scales across scenarios, whereas the isolated biogeophysical effects on temperature are insignificant at the global scale.

Our results provide robust, scenario-dependent insights into how large-scale AR works within the Earth system, and how the emerging carbon and climate feedbacks affect sequestration and temperatures across global and regional scales.

 

References:

Moustakis, Y., Nützel, T., Wey, HW. et al. Temperature overshoot responses to ambitious forestation in an Earth System Model. Nat Commun 15, 8235 (2024). https://doi.org/10.1038/s41467-024-52508-x