Detecting the dominant contributions of runoff variance across the source region of the Yellow River using a new decomposition framework

Quantifying the contributions of climatic variables to runoff variance is still a great challenge to water resource management. This study adopted an extended Budyko framework to investigate the effects of terrestrial water storage changes (ΔS) on runoff variance across the source region of the Yellow River, China, during the period of 2003–2014. A new decomposition framework based on the extended Budyko framework was proposed to effectively quantify the contributions of different climatic variables including precipitation, PET and ΔS to runoff variance. The results demonstrated that the extended Budyko framework showed a better performance in presenting the water and energy balance than the original Budyko framework, especially at fine time scales. Meanwhile, the variance in runoff estimated by the new decomposition framework was close to that of runoff observations, indicating that this framework can effectively capture the variation in runoff during 2003–2014. It was also found that precipitation was the most important factor that contributed to runoff changes, while PET made a slightly smaller contribution compared to precipitation. Notably, the results also emphasized the important effects of ΔS on runoff variance at fine time scales, which was useful to better understand the interactions between atmospheric and hydrological processes for regions. HIGHLIGHTS A new decomposition framework was proposed to effectively quantify the contributions of different climatic variables to runoff variance.; Precipitation was the most important factor that contributed to runoff changes, while PET made a slightly smaller contribution compared to precipitation.; The important effects of ΔS on runoff variance at fine time scales can not be neglected.;