为研究变化环境下渭河流域多时间尺度径流变化特征及影响因素，基于渭河流域咸阳水文站1961—2015年实测径流量数据，分析不同时间尺度上径流的变化趋势，并选取多个气象要素与归一化植被指数（normalized difference vegetation index，NDVI）构建向量自回归（vector autoregression，VAR）模型，量化各影响因素对径流变化的贡献程度，探讨全年、汛期和非汛期径流变化主控因子的差异性。结果表明，1961—2015年渭河流域径流整体呈下降趋势，汛期径流量占比为65.5 %。在年代尺度上径流的丰枯变化存在差异，1970—1979年和1990—1999年径流量呈显著减少趋势；影响汛期和全年尺度径流变化的主控因子均为降水，贡献率达50%，其次是太阳辐射，在汛期和全年的贡献率分别为14.37%、18.45%。非汛期影响径流变化的主控因子为NDVI（贡献率30.08%）与太阳辐射（贡献率30.05%），且NDVI对径流变化的滞后效应更明显，其他因子贡献大小依次为降水、空气比湿、气温。

Runoff is an important part of the hydrological cycle and can respond to environmental changes. The Weihe River is the first-class tributary of the Yellow River basin and is also the most important water resource supply area, agricultural irrigation, and industrial water use in Guanzhong. Due to natural reasons such as large topographic relief, loose and porous soil, uneven distribution of precipitation within the year, and the intensification of human activities, the basin is characterized by frequent alternating water and drought disasters, serious soil erosion, and is a resource-type water-scarce area. In recent years, with climate change and the intensification of human activities, the ecological environment of the Weihe River basin has become more fragile, and the hydrological situation has also changed to different degrees. Therefore, analyzing the change characteristics of runoff in the Weihe River basin on multiple time scales and its influencing factors can help to understand the change rules of the hydrological situation of the basin and water resources planning and management.Based on the measured runoff data from 1961 to 2015 at Xianyang Hydrological Station in the Weihe River basin, a linear regression method was used to analyze the trend of runoff on different time scales. Several meteorological factors such as temperature, specific humidity, precipitation, solar radiation, wind speed, and Normalized Difference Vegetation Index (NDVI) were selected. The factors were screened by the smoothness test and Granger causality test. Vector Autoregression models are constructed to quantify the contribution of each influential factor to the runoff changes, and the differences of the main controlling factors of the runoff changes throughout the year, flood season, and non-flood season are explored.The runoff showed a decreasing trend, with decreasing rates of 0.76, 0.44, and 0.31 mm/（10a）respectively during the whole year, flood season and non-flood season from 1961 to 2015 in the Weihe River basin. Flood season runoff accounted for 65.5% of the total runoff, which was highly similar to the trend of the whole-year runoff. On the time scale, the runoff decreased significantly in the 1970s and 1990s. There were differences in the degree of runoff dispersion in different eras, with greater intra-annual dispersion of runoff in the 1970s and 1980s, and a more concentrated distribution of runoff from 2000 to 2009. The variance decomposition based on the vector autoregression model is used to analyze the effects of different lag times of meteorological elements and NDVI on runoff. On the annual scale, the influences of runoff changes were dominated by precipitation and supplemented by solar radiation and NDVI when the influences of the elements and runoff reached a steady state over time, with a contribution of 49.04% for precipitation and 18.45% and 13.60% for solar radiation and NDVI respectively. The contribution rate of precipitation to runoff is 49.04%, and the contribution rates of solar radiation and NDVI are 18.45% and 13.60% respectively, and the influence of each factor on runoff has different degrees of lag. The distribution of contributing factors to the runoff changes in the flood season and the annual scale were similar. During the flood season, precipitation was the main controlling factor for runoff changes, and the contribution of each factor was in the following order: precipitation＞specific humidity＞solar radiation＞NDVI＞ temperature. During the non-flood season, the main controlling factors were NDVI and solar radiation, with a contribution of more than 30%. The contribution of precipitation decreases significantly, the influence of NDVI on runoff increases significantly, the contribution of temperature increases to some extent, and the lag effect of the influencing factors becomes more obvious.During the flood season and the non-flood season, the main controlling factors affecting the runoff changes are different in the Weihe River basin. During the flood season, runoff changes are mainly influenced by precipitation, while the influence of precipitation is significantly reduced during the non-flood season, and the main controlling factors affecting runoff become NDVI and solar radiation.