Analysis of soil moisture variation and its influencing factors in semi-arid steppe watershed

Soil moisture is an important part of the lithosphere-biosphere-atmosphere-hydrosphere, and which is the main limiting factor for plant growth in watersheds of the semi-arid steppe as well. This study aimed to explore the characteristics of soil moisture variation and its potential environmental impact factors. This study had a significant effect on maintaining steppe ecosystem health, such as promoted the understanding of the eco-hydrological cycle of grassland and degraded grassland restoration. The Xilin River Basin located in the middle east of Inner Mongolia was chosen as the study area. The experiments had been performed in May 2019. Two meters of soil moisture detection tube was buried at each sampling point after removing the ground cover and floating soil. Besides, the soil moisture measurement depth was set from 0 cm to 180 cm. Soil moisture data were observed by using the Time Domain Reflectometry (TDR) at 55 sampling points from July 22 to October 20 in 2019 after the soil layer was stabilized. Three replications were performed at each sample point. Such a soil survey was conducted within 3 days by twice a month in July and August, and once a month in the remaining months. Meanwhile, vegetation investigations were executed once a month, including vegetation height, vegetation coverage, and aboveground biomass. Besides, 3 original soil samples were brought back from the field for calibrating the TDR in the laboratory. It was indicated that the TDR was reliable with guaranteed that the soil moisture data used in the calibration formula was ranged from saturation to dryness completely. Based on the observed soil moisture and affiliated data in the field the characteristics of soil moisture changed in the vertical profiles, as well as its potential environmental impact factors were analyzed and compared among 4 soil types by using Principal Component Analysis (PCA) and Redundancy Analysis (RDA). The results indicated that the soil moisture in the Xilin River Basin was low overall. The changes in soil moisture at different soil depths over time were similar in kastanozems and arenosols while they had a great difference in solonchaks and phaeozems. Moreover, the soil moisture content of Phaeozems, Arenosols, Kastanozems, and Solonchaks increased sequentially of the basin. The variability of soil moisture in each layer was greatly affected by the soil type, and the variability weakens as the depth of the soil layer increases. At the same time, great differences in the variability of soil moisture among 4 soil types were found after the changes in soil moisture in different sampling periods were compared and analyzed. PCA indicated that 3 principal components containing information such as altitude, vegetation height, and slope, etc., which were sufficient to explain 68.50% of the soil moisture variation where PC1 and PC2 accounted for 40.00% and 15.60% respectively. In detail, the PC1 included altitude, potential evapotranspiration, air temperature, and vegetation coverage, and the PC2 integrated information of vegetation height and aboveground biomass, and the PC3 was slope and aspect. Consequently, the meteorological factors that affected soil moisture evaporation, vegetation factors, and slope and aspect factors were summarized as the main potential environmental impact factors of soil moisture variations in the Xilin River Basin. Altitude and vegetation height was defined as the key above-ground potential environmental factors driving the soil moisture variation by RDA. Meanwhile, the impact factors that affected soil moisture changes at different depths were different. The results of this study were conducive to understanding the soil moisture conditions and its distribution of different soil types in the Xilin River Basin, and also would provide data and theoretical support for the study of eco-hydrological processes in the semi-arid grassland watersheds.