Long-term temporal variations of soil water content under different vegetation types in the Loess Plateau, China

- Semivariogram is an effective method to investigate long-term temporal characteristics of soil water content.
- Purple Alfalfa induces greater variations in soil water content in deeper soil layers.
- Land use is the dominant factor affecting long-term temporal variations in soil water content.

The temporal characteristics of soil moisture are critical for the stability and sustainability of rebuilt ecosystems. In China's Loess Plateau (CLP), large areas of unsuitable croplands were restored to forest or grassland in order to control soil and water loss. Thus, knowledge of the temporal pattern and its controlling factors of soil moisture in revegetated areas over long-term is indispensable. In this study, four plots with different land uses (shrubland, grassland, fallow land and cropland) were set up on a loess slope in northern CLP. The profile soil moisture in the 0-4 m soil layer was measured from 2004 to 2015. Semivariogram models were established to analyze the profile temporal characteristics of soil water content (SWC) over 12 years of revegetation. The results suggested that: temporal stability of SWC under cropland was better than that under the three types of revegetated lands. Among the three types of revegetated lands, fallow land had stronger stability and higher value of SWC and was thus recommended as the best revegetation type. Generally, variation of SWC decreased with the increase of soil depth under shrubland, fallow land and cropland. However, under grassland, an inverse trend of SWS variation was observed and huge water absorption by deep grass roots was responsible for this trend. The temporal variation in SWC was under the joint influence of land use, soil depth and slope position, and land use was regarded as the dominant one. The study provided wider insight into scientific and reasonable vegetation restoration strategies in CLP region.