Simulation of spatiotemporal response of baseflow to vegetation variations based on coupled application of ensemble baseflow separation technique and SWAT model

Abstract

[Objective] Baseflow is a relatively stable component of runoff. Accurately characterizing the spatiotemporal distribution characteristics of baseflow in river basins, especially its response to vegetation variations, is crucial for water resource and river basin management in the water-scarce mountainous areas of northern China. [Methods] The SWAT model was coupled with the numerical simulation-based ensemble baseflow separation technique. Taking the Xiaoluan River Basin as an example, the spatiotemporal variations of baseflow in the study area were analyzed, and the controlling effect of vegetation variations on river basin baseflow were investigated. [Results] The result showed that:(1) the coupled application of the SWAT model and the ensemble baseflow separation technique could effectively simulate the hydrological processes in areas with limited baseflow. The baseflow simulation performed well, with a coefficient of determination as high as 0.9, demonstrating strong applicability of this method.(2) From 2006 to 2020, the annual and growing-season runoff and baseflow in the Xiaoluan River Basin showed an overall decreasing trend, although the trend was not significant. In contrast, the runoff and baseflow during the non-growing season increased significantly(p<0.05).(3) During the study period, baseflow remained generally stable in approximately 35.85% of the river basin area, with distribution across the upper, middle, and lower reaches. In the mountainous area near the dam, baseflow in some sub-basins even showed a significant increase(accounting for 22.47% of the river basin area). In contrast, in the alpine meadow area above the dam, baseflow showed a slight decrease(accounting for 18.32% of the river basin area) or a significant declining trend(accounting for 19.82% of the total area).(4) Vegetation was one of the important influencing factors for the improvement of baseflow processes in the study basin. In most parts of the mountainous area near the dam, NDVI and baseflow showed a positive correlation, and some sub-basins even showed a significant positive correlation(accounting for 10.68% of the total river basin), indicating that vegetation restoration was conducive to the improvement of baseflow processes in the study area to a certain extent. [Conclusion] At present, the increase in vegetation cover in the study area has a certain positive controlling effect on baseflow process. However, due to the interaction of processes such as spring snowmelt recharge, the controlling effect of vegetation on baseflow under future climate change requires continuous observation and investigation.

Keywords

SWAT model / ensemble baseflow separation technique / spatiotemporal variations of baseflow / Xiaoluan River Basin / influencing factors