Non-floodplain ponded waterbodies (NPWs) are key nodes in upland-stream hydrological continuums, playing essential roles in flood and drought mitigation and water quality improvement. The hydrological connectivity between NPWs and downstream water systems determines their ecological and hydrological function. Understanding NPW dynamics through the lens of hydrologic connectivity provides valuable insights for managing the resilience of water resources. However, this aspect has received limited attention in previous research. In the present study, NPWs in the upper Yangtze River Basin were identified using the Global Surface Water dataset. They were then classified into three hydrologic connectivity classes-shallow flow path NPWs (NPWSFP), middepth flow path NPWs (NPWMFP), and deep flow path NPWs (NPWDFP)-based on the depth of their flow paths to the nearest stream. Thereafter, the spatial distribution and changes of these waterbodies from 1990 to 2021 were analyzed. Further, the driving factors, including natural and human induced changes, were investigated across different hydrologic connectivity classes using Partial Least Squares Structural Equation Modelling. NPWSFP and NPWMFP were primarily distributed in the Sichuan Basin, while NPWDFP were mainly concentrated in the Jinsha, Mintuo, and Wu River basins. Between 1990 and 2021, the total area covered by NPWs increased significantly, with notable growth in NPWSFP and NPWDFP. The area of NPWSFP increased primarily due to reservoir construction to meet increasing water demand. In contrast, the area of NPWDFP increased as a result of greater water replenishment driven by climate change. Climate change affected NPW dynamics indirectly through its influence on vegetation. Human activities also shaped NPW changes, mainly by altering vegetation and modifying local climate conditions. These findings offer insights for the targeted conservation and restoration of surface waterbodies.