Assessment of vegetation carrying capacity at the basin-scale using an improved SWAT model

Vegetation restoration represents a highly effective strategy for offsetting carbon emissions, mitigating climate change, and enhancing the quality of the eco-environment. Despite the ecological benefits, the large-scale ‘Grain-for-Green’ program in China has caused unintended hydrological consequences, threatening the sustainability of regional water resources in this arid and semi-arid region. Therefore, it is crucial to understand the competing water demands of ecosystems and humans. In this study, we investigated the effects of vegetation change on the water cycle and the vegetation carrying capacity in a semi-arid loess-gully basin using our previously modified Soil and Water Assessment Tool (SWAT), which focuses on simulating forest growth from young to mature stages. Model validation demonstrated that the improved SWAT model can well simulate eco-hydrological processes, including leaf area index (LAI), streamflow, evapotranspiration (ET), and soil water content. Simulations under various vegetation restoration scenarios revealed that low- and medium-intensity revegetation (converting sloping farmland (slope > 15°) to grass or forest) resulted in slight changes (≤4.84 %) in mean annual streamflow, soil water, and ET. In contrast, high-intensity revegetation caused significant water stress, with streamflow and soil moisture reduced by 19.35 % and 15.14 %, respectively. In addition, the vegetation carrying capacity of the watershed was evaluated, indicating that the basin can sustainably support a maximum LAI increase of 0.20 (15.63 %). Overall, this study can be valuable for vegetation restoration and forest management in the Chinese Loess Plateau, and the proposed methods can be applicable in other areas.