Soil moisture variability of typical mixed forests in the north-south climatic transitional zone of China

Soil moisture plays crucial role in forest hydrological processes. However, modeling soil moisture for different mixed forests lacks well developed framework. This study aims to address this gap by identifying the dominant drivers of soil moisture and establishing simulation framework for different mixed forests, taking the conifer-broadleaved mixed forests (CBMF), deciduous broadleaved-dominated mixed forests (BDMF), and conifer-dominated mixed forests (CDMF) in the north-south climatic transitional zone of China as a case. Our results demonstrate that, compared to BDMF and CDMF, CBMF significantly enhanced soil moisture via increases in saturated hydraulic capacity, total porosity, and soil organic carbon (SOC), while reducing bulk density. PCA-VIF analysis showed that SOC, particle size, and pH were the most significant independent drivers of variation in soil moisture. Based on these results, we developed a framework based on CEEMDAN-LSTM for modeling soil moisture at various depths across different mixed forests. Simulation performance was excellent, with R² values ranging from 0.80 to 0.92 across all soil layers, though greater accuracy was observed in simulations in deeper soil layers (R² > 0.92). Based on this framework, we simulated additional scenarios and found that CBMF exhibited the highest soil hydrological conservation capacity under precipitation variation. These results establish that our framework effectively simulated the impacts of extreme precipitation on soil moisture under climate change, providing a scientific basis for adaptive management of forest ecosystems and water resources.