Land-use types significantly influence the quantity and quality of dissolved organic matter (DOM) in soil. However, the long-term effects of land-use practices on soil DOM dynamics remain poorly understood. A two-year field study was conducted in a hilly agricultural region in the upper reaches of the Yangtze River, Southwest China, to investigate the effects of land use on soil DOM. Five land-use types were examined: forest, grassland, orchard, sloping farmland, and paddy field. Dissolved organic carbon (DOC) concentrations and DOM fluorescence indices-fluorescence index (FI) and biological index (BIX) showed significant seasonal variations, generally characterized by an initial increase followed by a decline throughout the year. DOC concentrations were significantly higher in forest, orchard, and paddy soils during the relatively wet year (2023, 750 mm precipitation) compared with the relatively dry year (2022, 563.2 mm precipitation), whereas grassland and sloping farmland soils displayed greater DOC stability under drier conditions. Structural equation modeling (SEM) revealed that soil temperature had a significant effect on DOM concentration and composition in the upper layers, while bulk density played a more prominent role in deeper layers. Empirical models based on soil properties effectively predicted DOM concentrations and fluorescence indices (FI and BIX) across soil depths and land-use types. This study demonstrates that land use significantly affects the seasonal dynamics and drought responses of soil DOM in a hilly agricultural region of Southwest China. These findings improve the understanding of DOM dynamics and provide scientific insights into land management strategies under climate change.