Forest water dynamics and hydro-chemical characteristics are essential for understanding forest hydrology and ecological processes. Yet, such understanding is limited by a lack of long-term monitoring data and observations from specialized forest ecosystems such as those from high elevation. Here, we analyze the precipitation redistribution including interception, stemflow, and throughfall and hydro-chemical characteristics by using a 15 year (2005–2019) precipitation dataset in a high elevation, evergreen broad-leaved forest in Southwest China. The forest experienced an obvious seasonal variation in precipitation with a monthly average of 117.31 ± 91.21 mm. The precipitation redistribution was influenced by precipitation intensity and leaf area index and differed inter-annually and intra-annually, with a general pattern: throughfall > canopy interception > stemflow. Throughfall rate increased significantly from 2015 to 2019 after experiencing the January 2015 snowstorm. The majority of water within the study site was retained in the soil and apoplastic materials. The primary means of water output was evapotranspiration, with minimal surface runoff. Quality of surface water was affected by the weathering of rocks, resulting in a lower pH than that of atmospheric precipitation. During the rainy season, elemental Ca and Mg showed negative correlation with precipitation due to plant mediation. The other elements, pH, total dissolved solids, precipitation, air temperature, and water temperature showed different degrees of correlation with each other. Overall, while the water balance fluctuated over the past 15 years, the water-holding capacity remained relatively stable. Alkali cations such as Ca2+, Mg2+, and K+ in the water body showed a decreasing trend during 2005–2019, which is a potential threat to ecological stability.