The influence of climate change and human interference bring high nonstationary characteristics that are different from the original long-term stable state. So that, investigating runoff variation and attribution under changing environments is important for implementing adaptive management of global water resources, especially on the interannual scale. In this study, the attribution of measured runoff change in the Wuding River Basin (WRB) of the Loess Plateau, China, was studied using the elasticity method based on the Budyko framework and groundwater storage change analysis on an interannual scale. The change in the measured runoff driven factors was divided into three categories: climate change, human-induced land use/land cover (HI-LULC) change, and water withdrawal change. The results show that the contribution rate of climate change was the largest (with the mean value 54.07%), followed by the HI-LULC change (with the mean value 43.85%), and the minimum is WW change (with the mean value 2.08%). Climate change had a maximum contribution to runoff change, of which precipitation was dominant, followed by relative humidity, net radiation, and wind speed; air temperature had a minimal impact. In addition, the ecological change tendency was analyzed by combining HI-LULC and NDVI data. The analysis results of ecological change tendency indicate that the policy implementation of soil and water erosion control caused the ecological restoration to some extent, but also led to a decrease in runoff. This study provides scientific and useful information for the conservation of water and soil, runoff prediction, and sustainable water resource management in the WRB, and enriches the knowledge of contribution analysis of runoff variation which can be referenced by other basins across the globe.