EGU26-2483, updated on 13 Mar 2026
https://doi.org/10.5194/egusphere-egu26-2483
EGU General Assembly 2026
© Author(s) 2026. This work is distributed under
the Creative Commons Attribution 4.0 License.
Oral | Thursday, 07 May, 14:55–15:05 (CEST)
 
Room 1.34
Global Patterns of Soil Freeze-Thaw States Below 0 °C
Zhaoyu Dong and Yuanchao Fan
Zhaoyu Dong and Yuanchao Fan
  • Tsinghua University, Shenzhen, China (dongzy24@mails.tsinghua.edu.cn)

Accurate prediction of permafrost thaw requires understanding soil freeze-thaw dynamics, yet current climate models assume pure water freezing points that may be inadequate for real soils. Traditional freeze-thaw theory suggests that salt in soil water increases Gibbs free energy, lowering the soil freezing point and enabling sub-zero unfrozen states (SUS). While these physical and chemical interactions have been documented in laboratory studies, they have not been evaluated by large-scale Earth observations. Here, by integrating satellite-derived freeze-thaw states with soil property datasets, we investigated thermal dynamics and mechanisms of SUS events across seasonally frozen soils of the Northern Hemisphere using machine learning and statistical analyses. Using satellite observations, we estimate a representative soil freezing point of 271.83 K, characterized by the median temperature of first seasonal freezing, which is nearly 1.3 °C lower than that of pure water. SUS events showed a wide occurrence temperature range (interquartile: 269.04–272.74 K), with 37.67% occurring at temperatures below the median freezing point. These low-temperature events are mainly driven by soil salinity and spontaneous entropy-increasing thawing processes. We compared observed freezing point depression with a simple physicochemical model incorporating soil salinity and moisture effects, which demonstrated strong agreement. These findings suggest that climate models assuming a 0 °C freezing point may underestimate active layer thawing rates and permafrost degradation risks. Our results indicate an urgent need to incorporate freezing point depression in Earth system models to accurately predict permafrost stability, with important implications for climate projections and ecosystem management.

How to cite: Dong, Z. and Fan, Y.: Global Patterns of Soil Freeze-Thaw States Below 0 °C, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-2483, https://doi.org/10.5194/egusphere-egu26-2483, 2026.