Spatial and temporal variation of soil moisture in dependence of multiple environmental parameters in semi-arid grasslands

Grazing of grasslands changes soil physical and chemical properties as well as vegetation characteristics, such as vegetation cover, species composition and biomass production. In consequence, nutrient allocation and water storage in the top soil are affected. Land use and management changes alter these processes. Knowledge on the impacts of grazing management on nutrient and water fluxes is necessary because of the global importance of grasslands for carbon sequestration. Soil water in semi-arid areas is a limiting factor for matter fluxes and the intrinsic interaction between soil, vegetation and atmosphere. It is therefore desirable to understand the effects of grazing management and stocking rate on the spatial and temporal distribution of soil moisture. In the present study, we address the question how spatio-temporal soil moisture distribution on grazed and ungrazed grassland sites is affected by soil and vegetation properties. The study took place in the Xilin river catchment in Inner Mongolia (PR China). It is a semi-arid steppe environment, which is characterized by still moderate grazing compared to other regions in central Inner Mongolia. However, stocking rates have locally increased and resulted in a degradation of soils and vegetation also in the upper Xilin River basin. We used a multivariate geostatistical approach to reveal spatial dependencies between soil moisture distribution and soil or vegetation parameters. Overall, 7 soil and vegetation parameters (bulk density, sand, silt and clay content, mean weight diameter, mean carbon content of the soil, vegetation cover) and 57 soil moisture data sets were recorded on 100 gridded points on four sites subject to different grazing intensities. Increasing stocking rates accelerated the influence of soil and vegetation parameters on soil moisture. However, the correlation was rather weak, except for a site with high stocking rate where higher correlations were found. Low nugget ratios indicate spatial dependency between soil or plant parameters and soil moisture on a long-term ungrazed site. However, the effect was not found for a second ungrazed site that had been excluded from grazing for a shorter period. Furthermore the most important soil and vegetation parameters for predicting soil moisture distribution varied between different grazing intensities. Therefore, predicting soil moisture by using secondary variables requires a careful selection of the soil or vegetation parameters.