Attribution of growing season evapotranspiration variability considering snowmelt and vegetation changes in the arid alpine basins

Abstract. Previous studies have successfully applied variance decomposition frameworks based on the Budyko equations to determine relative contribution of variability in precipitation, potential evapotranspiration (E0), and total water storage changes (ΔS) (σET2) different timescales; however, effects snowmelt (Qm) vegetation (M) not been incorporated into this framework snow-dependent basins. Taking arid alpine basins Qilian Mountains northwest China as study area, we extended decompose growing season σET2 temporal covariance rainfall (R), E0, ΔS,Qm, M. The results indicate that incorporation Qm could improve performance a monthly scale; was primarily controlled by R with mean 63 %, followed coupled M (24.3 %) then E0 (14.1 %). or cannot be ignored because they contribute 4.3 % 1.8 σET2, respectively. By contrast, interaction some factors adversely affected out-of-phase seasonality between had largest effect (−7.6 Our methodology these findings are helpful for quantitatively assessing understanding hydrological responses climate regions finer timescale.