| Soil water content is an important component of water resources in the Loess Plateau. The research on the spatial variability of soil moisture on small scale watershed helps to promote the sustainable management of water resources and ecological restoration in arid and semi-arid area and it also has very important significance for the further understanding of the rules of soil moisture and quantifying ecohydrological process of specific area. The study takes a small watershed called Sanyanjing of Shouyang County in Shanxi province as the study area, using classical statistics, geostatistics, and GIS method. It discusses the spatial variation and the spatial structure of soil moisture in horizontal and vertical direction (0-300cm) in the basin before and after the rainy season, and evaluates the influence of selected environmental factors on soil moisture content of different soil layers (0-20cm,20-160cm,160-300cm) (the type of land use, topography and geomorphology), so as to provide scientific basis for valuation of soil moisture and ecological construction in this area and other similar terrain conditions. We can draw the following conclusions:(1) Analysis of the characteristics of horizontal variation of soil moisture show:based on the classical statistical analysis of soil moisture content in the small watershed called Sanyanjing, the result is that before and after the rainy season, soil moisture content in different depth showed moderate variation; Geostatistical analysis results show that the sill and the range in different seasons have certain differences. The sill and the range showed a trend of increase as the soil water content increases. Increased soil moisture was masked by enhanced effect of land use and terrain; The layered distribution of terraces and the slope terrain weakened the spatial autocorrelation of soil moisture in horizontal direction; The variation of which ranged from 19 to 102m before and after the rainy season. And the spatial structure information of soil moisture could be accumulated by increasing the sampling density.(2) Classical statistical analysis results show that 0-300cm soil moisture distribution showed weak and moderate variability; Geostatistical analysis results show that soil moisture was influenced by different land use types and slope position by the sequence of grassland> woodland> grassland> cultivated land, and beam> valley respectively before the rainy season, but had no correlation with the degree of slop. At the same time, the sill and ran-ge under different land use and conditions before rainy season are less than th-at after rainy season; Cluster analysis showed that soil profiles with high-level soil moisture in a stable trend (from top to bottom) were most commonly present in the catchment, especially in the gully related to terrace. Woodland soil profiles had low-level soil moisture with vertical variations in a descending or stable trend. Most abandoned farmland and grassland soil profiles had medium-level soil moisture with vertical variations in varying trends. No soil profiles had low-level soil moisture with vertical variations in an ascending trend before and after the rainy season.(3) Multi-regression analysis showed that average soil moisture was significantly affected by land-use type in different soil layers (0-20,20-160, and 160-300cm), generally in descending order of terrace, abandoned farmland, grassland, and woodland. There was a significant negative correlation between average soil moisture and gradient along the whole profile (P<0.05). Landform significantly affected soil moisture in the surface soil layer (0-20cm) before the rainy season but throughout the whole profile after the rainy season, with lower levels on the ridge than in the gully. Altitude only strongly affected soil moisture after the rainy season. |
PDF Full Text Request