Soil moisture response to land use and topography across a semi-arid watershed: Implications for vegetation restoration on the Chinese Loess Plateau

Soil moisture is a limiting factor of ecosystem development in the semi-arid Loess Plateau. Characterizing the soil moisture response to its dominant controlling factors, such as land use and topography, and quantifying the soil-water carrying capacity for revegetation is of great significance for vegetation restoration in this region. In this study, soil moisture was monitored to a depth of 2 m in three land use types (native grassland, introduced grassland, and forestland), two landforms (hillslope and gully), and two slope aspects (sunny and shady) in the Nanxiaohegou watershed of the Loess Plateau, Northwest China. The MIKE SHE model was then applied to simulate the soil moisture dynamics under different conditions and determine the optimal plant coverage. Results showed that the average soil moisture was higher in native grassland than in introduced grassland and Platycladus orientalis forestland for a given topographic condition; however, a high soil moisture content was found in Robinia pseudoacacia forestland, with a value that was even higher than the native grassland of a sunny slope. The divergent results in the two forestlands were likely attributed to the differences in plant coverage. Gully regions and shady slopes usually had higher soil moisture, while lower soil moisture was usually distributed on the hillslope and sunny slope. Furthermore, the mean absolute relative error and Nash-Sutcliffe efficiency coefficient of the MIKE SHE model ranged between 2.8%–7.8% and 0.550–0.902, respectively, indicating that the model could effectively simulate the soil moisture dynamics. The optimal plant coverage was thus determined for hillslope P. orientalis by the model, corresponding to a leaf area index (LAI) value of 1.92. Therefore, for sustainable revegetation on the Loess Plateau, selecting suitable land use types (natural vegetation), controlling the planting density/LAI, and selecting proper planting locations (gully and shady slope regions) should be considered by local policy makers to avoid the over-consumption of soil water resources.