Sensitivity and attribution analysis based on the Budyko hypothesis for streamflow change in the Baiyangdian catchment, China

Abstract

In the past 50 years, the decrease of river runoff in China has attracted wide attention from government decision-making departments and the public. The Baiyangdian catchment is one of the regions with the most serious water shortage and the most prominent human-water conflicts, and attribution analysis of the runoff change is of considerable interest at a range of spatial scales. The Budyko framework has been widely used to attribute changes in streamflow to the effects of climate and catchment changes. In this study, we used the elasticity method based on the Budyko framework to examine the sensitivity of streamflow to climate and catchment variables, which indicated that a 1-mm decrease in precipitation would induce a 0.3546-mm decrease in streamflow, a 1-mm decrease in potential evapotranspiration would induce a 0.1045-mm increase in streamflow, and an increase of 1 in the catchment characteristic coefficient would induce a 79.6711-mm decrease in streamflow. The absolute sensitivities of streamflow to climate variables decreased with increases in the aridity index, which indicates that the streamflow was more sensitive to climate change in wet regions. Among the total changes in streamflow (−43.41 mm), the effect of climate change was +3.86 mm (accounting for 7.55%), and the effect of the catchment characteristic changes is −45.99 mm (accounting for 89.95%) based on the sensitivity analyses. The results indicate that streamflow changes in the Baiyangdian catchment are mainly caused by catchment changes.