Temporal stability of soil water storage and its influencing factors on a forestland hillslope during the rainy season in China’s Loess Plateau

Abstract

Large-scale vegetation restoration in China’s Loess Plateau has been initiated by the central government to control soil and water losses since 1999. Knowledge of the spatio-temporal distribution of soil water storage (SWS) is critical to fully understand hydrological and ecological processes. This study analysed the temporal stability of the SWS pattern during the rainy season on a hillslope covered with Chinese pine (Pinus tabulaeformis Carr.). The soil water content in eight soil layers was obtained at 21 locations during the rainy season in 2014 and 2015. The results showed that the SWS at the 21 locations followed a normal distribution, which indicated moderate variability with the coefficients of variation ranging from 14 to 33%. The mean SWS was lowest in the middle slope. The spatial pattern of SWS displayed strong temporal stability, and the Spearman correlation coefficient ranged from 0.42 to 0.99 (p < 0.05). There were significant differences in the temporal stability of SWS among different soil layers (p < 0.01). The spatial patterns of SWS distribution showed small differences in different periods. The best representative locations of SWS were found at different soil depths. The maximum RMSE and MAE at 0–1.6 m soil depth for the rainy season were 4.27 and 3.54 mm, respectively. The best representative locations determined during a short period (13 days) can be used to estimate the mean SWS well for the same rainy season, but not for the next rainy season. Samples of SWS collected over a fortnight during the rainy season were able to capture the spatial patterns of soil moisture. Roots were the main factor affecting the temporal stability of SWS. Rainfall increased the temporal stability of the soil water distribution pattern. In conclusion, the SWS during the rainy season had a strong temporal stability on the forestland hillslope.