Humans have depleted global terrestrial carbon stocks by a quarter

[Geoengineering News]

‎https://www.sciencedirect.com/science/article/pii/S2590332225002180

‎

Authors: Raphael Ganzenmüller, Wolfgang A. Obermeier, Selma Bultan, Seth A. Spawn-Lee, Florian Zabel, Julia Pongratz 

https://doi.org/10.1016/j.oneear.2025.101392

15 August 2025

Highlights

•Global terrestrial carbon stocks have been depleted by 24% (344 PgC)

•Our estimates of the carbon deficit are at unprecedented resolution and consistency

•Pasture expansion, cropland expansion, and forest management are major drivers

•Dynamic global vegetation models underestimate the carbon deficit by 37% on average

Science for society

Natural ecosystems play a crucial role in regulating the climate, storing more carbon than all fossil fuel reserves combined. However, land-use activities such as agricultural expansion and forest management are massively altering vegetation and soil carbon stocks, releasing vast amounts of carbon into the atmosphere, which substantially contribute to global warming. Yet, quantifying carbon loss from natural ecosystems globally at high resolution has been challenging.

Our research addresses this gap by providing a consistent estimate of this deficit: globally, carbon in vegetation and soil has been reduced through land-use activities by 24% (344 PgC), comparable to all fossil fuel emissions over the past 50 years. We also find that current global vegetation models considerably underestimate this loss. Our findings offer valuable insights for policymakers to prioritize conservation and restoration efforts, guide sustainable land-use decisions, and improve global vegetation models.

Summary

Mitigating global climate change requires massive greenhouse gas emission reductions and carbon removal efforts. Although terrestrial ecosystems store large amounts of carbon, land-use change has substantially diminished these stocks in many regions. However, a consistent, high-resolution approach to quantify the differences between actual and potential carbon stocks in vegetation and soils—the terrestrial carbon deficit—remains elusive, limiting the evaluation of global climate models. Here, we combine semi-empirical data with machine learning to estimate the terrestrial carbon deficit to be 344 (251–393) PgC, equivalent to a 24% depletion, predominantly driven by pasture expansion (30%), cropland expansion (24%), and forest management (23%). We reveal that dynamic global vegetation models (DGVMs) underestimate the terrestrial carbon deficit by 37% on average (range: 2%–58%), highlighting critical limitations. Our findings support assessments of anthropogenic impacts on ecosystems and help constrain global climate models to better evaluate nature-based solutions and climate mitigation policies.

Source: One Earth