Soil organic matter (SOM) fundamentally alters soil hydrothermal properties, thereby regulating soil temperature and moisture simulations. These hydrothermal dynamics are critical for maintaining permafrost stability, which determines the transition of SOM between carbon sources and sinks, and ultimately feeds back to the regional climate and ecosystem. However, its effects in the permafrost regions of the Qinghai–Tibet Plateau (QTP) remain unclear due to strong spatial heterogeneity and limited field observations. In this study, we conducted sensitivity experiments with CLM5.0, combined with in situ data from two permafrost sites, to evaluate the role of SOM in active layer soil temperature and moisture dynamics. Four sensitivity experiments were conducted, including runs with and without SOM, a modified maximum SOM content, and a virtual surface organic layer. Results showed that excluding SOM substantially overestimates soil temperatures during the thawing period and amplifies seasonal variations, while soil moisture is underestimated. Moreover, SOM delays the onset of soil thawing by up to 15 days. Quantitative results showed that compared with the experiment without SOM (EXP1), the control experiment with SOM (CTL) significantly reduces the root mean square error (RMSE) of soil temperature during the thawing period: by 0.21 °C at the TGL site and 1.21 ℃ at the BLH site. Furthermore, the experiment incorporating a virtual surface organic layer (EXP3) exhibited the best performance, with soil temperature RMSE ≤ 1.60 ℃ at both sites, the lowest among all experiments. SOM substantially reduced errors and improved agreement with observations by increasing soil porosity and reducing thermal and hydraulic conductivities, thereby damping downward heat flux and enhancing near-surface moisture retention. Therefore, SOM plays a critical role in accurately reproducing soil hydrothermal dynamics in permafrost regions of the QTP. In contrast, soil hydrothermal dynamics are more sensitive to the prescribed maximum SOM content at sites with higher near-surface SOM, whereas introducing a thin surface organic layer mainly affects the simulations during the autumn–winter transition at sites with mineral-dominated surface soils and low SOM. This study highlights the importance of accurately representing SOM content and parameterization schemes in land surface models and emphasizes the need to account for regional surface heterogeneity to improve simulation accuracy.
Peng et al. (Thu,) studied this question.