Responses of soil moisture at different topographic positions to rainfall events along a steep subtropical forested hillslope

Understanding the dynamic response of soil moisture to rainfall is crucial for describing hydrological processes at the hillslope scale. However, because of sparse monitoring coupled with the complexity of water movement and steep topography, the findings of rainfall‐related soil moisture dynamics have not always been consistent, indicating a need for systematic investigations of soil moisture dynamics and infiltration patterns following rainfall inputs at multiple topographic positions along a hillslope. This study aimed to examine the nature of these responses by characterizing and quantifying the response amplitude, rate and time for 37 large rainfall events at 25 combinations of topographic positions and soil depths along a steep forested hillslope. Our results showed that soil moisture responses under different rainfall patterns could be attributed to one or the other rainfall characteristics, such as rainfall intensity and amount. However, soil moisture dynamics at different hillslope positions after rainfall varied widely due to the controls of soil properties, topography, and non‐equilibrium flow. Preferential flow was more evident under dry initial soil conditions than under wet initial soil conditions. Findings of this study reveal that the dynamic response patterns of soil moisture to rainfall do not always follow topographic controls, which can improve our understanding of water cycling related to the infiltration process at the hillslope scale, and support water resources management in subtropical mountain ecosystems. Both matrix infiltration and preferential flow may be dominant mechanisms of wetting front propagation through soil profile in upslope areas, and matrix infiltration is likely the dominant flow mechanism in downslope areas. Second, the lower positions on the hillslope generally showed a shorter response time and faster wetting front rates. Third, soil moisture did not systematically increase moving from uphill to downhill. Notably, the soils at S2 site on the ridge exhibited a high moisture content, which was only lower than that of the S5 site near stream channels. Among the five monitoring sites, the third layer (20–40 cm depths) displayed the highest proportion of SWS within the profile.