Assessing Variation in Water Balance Components in Mountainous Inland River Basin Experiencing Climate Change

Quantification of the changes of water balance components is significant for water resource assessment and management. This paper employed the Soil and Water Assessment Tool (SWAT) model to estimate the water balance in a mountainous watershed in northwest China at different spatial scales over the past half century. The results showed that both Nash-Sutcliffe efficiency (NSE) and determination coefficient (R2) were over 0.90 for the calibration and validation periods. The water balance components presented rising trends at the watershed scale, and the total runoff increased by 30.5% during 1964 to 2013 period. Rising surface runoff and rising groundwater flow contributed 42.7% and 57.3% of the total rising runoff, respectively. The runoff coefficient was sensitive to increasing precipitation and was not significant to the increase of temperature. The alpine meadow was the main landscape which occupied 51.1% of the watershed and contributed 55.5% of the total runoff. Grass land, forest land, bare land, and glacier covered 14.2%, 18.8%, 15.4%, and 0.5% of the watershed and contributed 8.5%, 16.9%, 15.9%, and 3.2% of the total runoff, respectively. The elevation zone from 3500 to 4500 m occupied 66.5% of the watershed area, and contributed the majority of the total runoff (70.7%). The runoff coefficients in the elevation zone from 1637 to 2800 m, 2800 to 3500 m, 3500 to 4000 m, 4000 to 4500 m, and 4500 to 5062 m were 0.20, 0.27, 0.32, 0.43, and 0.78, respectively, which tend to be larger along with the elevation increase. The quantities and change trends of the water balance components at the watershed scale were calculated by the results of the sub-watersheds. Furthermore, we characterized the spatial distribution of quantities and changes in trends of water balance components at the sub-watershed scale analysis. This study provides some references for water resource management and planning in inland river basins.