- 1Peking University, College of Urban and Environmental Sciences, Beijing, China (yiqinchen@stu.pku.edu.cn)
- 2Peking University, College of Urban and Environmental Sciences, Beijing, China (zhudan@pku.edu.cn)
Soils represent a massive reservoir of organic matter, storing approximately three times the carbon found in the atmosphere. While over half of this soil organic carbon (SOC) is stored in subsoils below 30 cm, our understanding of its vulnerability is hindered by a major discrepancy: most field experiments utilizing top-down warming techniques show that warming rapidly attenuates with depth, whereas Earth system models (ESMs) project synchronous warming of the entire soil profile. Here, we resolve this conflict by combining a synthesis of depth-specific soil temperature measurements from 579 in-situ monitoring sites, analysis of 322 field warming experiments, and process-based modeling. The observed warming rates across depths demonstrate that ambient climate change drives nearly synchronous warming down to 3.5 m with only a slight attenuation along depth. This starkly contrasts with the strong thermal dampening recorded in field experiments using top-down warming techniques including open-top chambers, infrared heaters and heating cables. By modifying a land surface model to explicitly simulate lateral heat transfer, we show that heat loss from warmed plots to adjacent unheated soils is the primary mechanism for this attenuation in experiments. Crucially, our modeling reveals that lateral heat loss inherent in plot-scale designs leads to an average 23% underestimation of heterotrophic respiration, resulting in a 10-fold underestimation of SOC loss after 20 years. Our findings reveal a critical bias in the widely used experimental frameworks and highlight the urgent need for whole-soil warming designs to more accurately predict soil carbon-climate feedbacks.
How to cite: Chen, Y. and Zhu, D.: Near synchronous warming of deep soils reveals prevailing underestimation in soil carbon loss in warming experiments, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-21607, https://doi.org/10.5194/egusphere-egu26-21607, 2026.