GEA Group, Germany

Abstract

Afforestation has significantly increased vegetation cover and has slowed soil erosion on China’s Loess Plateau (CLP). However, intensive afforestation activities in semi-arid regions that are in excess of local water availability can cause imbalances in the soil water budget. This can cause soil desiccation, which in turn builds up dry soil layers (DSL) in the soil profile. The occurrence of DSL can threaten the sustainable development of restored ecosystems in water-scarce regions. Knowledge of the spatial characteristics of DSL in artificial forests and their response to land-use change is critical for optimal soil water management and ecological sustainability in the CLP. Therefore, the spatial characteristics of DSL under artificial forest in a typical grassland area on the CLP were investigated. The temporal (2015–2040) dynamics of soil water content (SWC) in the 0–5 m soil profile after the land-use change were simulated and analysed using the Hydrus-1D software. Based on the investigation, DSL existed at some 81% of the sampling sites in the typical grassland area on the CLP. The occurrence of DSL was graded as severe, with a mean thickness of 297 cm and mean SWC of only 9%. The model simulation showed that two phases of SWC dynamics developed after land-use change. One of the phases is the gradual recovery of SWC after the conversion of artificial forests into grassland. The other is that of SWC fluctuations with rainfall under grassland conditions. It requires some 2–13 years (mean of ~7 years) for the soil water to recover in the 0–5 m soil profile after artificial forest is converted to grassland across the semi-arid grass zone of the CLP. Structural equation modelling showed that the duration of DSL recovery was driven directly or indirectly by the degree of dryness of the DSL, soil hydraulic properties, and slope. It is therefore concluded that DSL in the 5-m soil profile can fully recover if there is a vegetation shift in the semi-arid grassland area of the CLP. This is critical for prediction studies of DSL dynamics and sustainable soil water management in water-scarce plateau regions.