| The desert oasis ecotone in the middle reaches of Heihe river is located in the northwest arid area of inland China, In the backdrop of the global climate change, and under the control of special geographical position and climate environment, this research area the precipitation is scarce and distributed uneven with the season change, but evaporation is larger, so serious shortage of water resources is the most obvious environmental characteristics in arid regions. And it is one of major bottleneck restricting economic and social development in the region. natural vegetation show patch shape distribution pattern in this desert oasis ecotone. In recent years, along with the Heihe water diversion project implementation, industrial and agricultural water demand increased and destruction of human activities, this area appeared vegetation degradation, withered, even death phenomenon.And ecological environment gradually deteriorating,surface soil wind erosion is serious, surrounding the sandstorm frequency increasing. The fragility and instability of this ecological system, determines the status of soil moisture is an important ecological factor in this desert oasis ecotone. Soil moisture is the bond and central link between the earth and the atmosphere, participate in the water cycle, energy and momentum exchange land surface process, also is an indispensable and key parameter of the climate system and the ecological system. Especially in arid desert area, the soil moisture is a crucial indicator of soil condition, Affects the growth and recovery of vegetation, distribution pattern and evolution process, at the same time can reflect and explain the degree of soil wind erosion in a region.Soil moisture influenced by numerous factors, have a high degree spatial and temporal heterogeneity. The factors include the underlying surface types, climate, soil texture, topography, human activities and so on. This study used three methods from three different dimensions study the spatial and temporal heterogeneity of soil moisture in the middle reaches of Heihe desert oasis ecotone vegetation patches. First, in the quadrat scales(The area is 340 m * 300 m,40m * 40 m spacing pattern), combined with statistics and statistical methods, quantitative analysis the soil moisture spatial heterogeneity level, scale, composition, variation function and correlation of typical samples 10 cm, 20 cm, 40 cm, 60 cm, 80 cm soil. Second, in the single point scale, under certain initial and boundary conditions, using Hydrus-1D model to simulate the dynamic changes in the time series of soil moisture content in the transition zone of the typical samples(bare soil sample and vegetation sample). Explore the each layer variation characteristics of soil moisture content with time of typical point, and tests it by the measured soil moisture data. Explore the simulation results of Hydrus-1D model. And obtain the corresponding model parameters, Which lay the foundation for the popularization and application of the regional scale. Third, in the regional scale, using the MODIS data and the method of thermal inertia estimate surface soil water content of patches vegetation. Explore the regional scale surface soil moisture variation characteristics in time and space. And tests it by the measured soil moisture data. The study results are as follows:(1) Through the descriptive statistical analysis, with the increase of soil depth, average soil moisture content of the patches vegetation in desert oasis ecotone shows the tendency of increase gradually. Variation of the five layers soil water content are the strong variability, especially the 20 cm soil moisture content, the coefficient of variation is the largest, which reached 159.62%.(2) Through the geostatistical analysis, the spatial heterogeneity of 5 layers soil moisture content has obvious difference. The change range of 5 layers soil moisture of spatial variability between 45.5m ~ 119.4m. The best fitting theoretical model of 10 cm, 20 cm, and 40 cm soil moisture content was exponential model, However, the best fitting theoretical model of 60 cm, 80 cm soil moisture content was spherical model. The coefficient of nugget of 5 layers soil moisture content are less than 25%, show that the spatial heterogeneity of 5 layers soil moisture of the patches vegetation in desert oasis in smaller proportion by random factors, mainly caused by structural factors, Quadrat area is small, the climate and site conditions are basically the same, patchy distribution pattern of vegetation is the main driving factors of the typical spatial heterogeneity of soil moisture.(3) In the vertical gradient of typical quadrat, The Moran's coefficients of global spatial autocorrelation of 5 layers soil moisture content were positive, Shows each layer of soil moisture has positive spatial correlation, and has the characteristics of spatial agglomeration; Soil 60 cm soil moisture has relatively stronger spatial agglomeration and its characteristics of spatial agglomeration is more obvious than other soils.(4)The each lay soil moisture content spatial prediction results are basically the same trend of the interpolation method of IDW and Kriging. But the IDW interpolation method is affected by the soil moisture extreme appeared "bull's-eye" phenomenon. By comparing the evaluation index and the interpolation prediction map, Kriging interpolation accuracy much higher than IDW interpolation accuracy, and interpolation map is smooth.(5) In the single point scale, Through the Hydrus-1D model simulated different soil layers soil moisture content dynamic change with time of two typical sample points under the two types of underlying surface. The results show that, whether in the bare single point or in a single point of vegetation area, the change trend of stimulated different layers soil moisture content are basically the same with the measured values. The correlation coefficient R2 are greater 0.52. The root mean square error(RMSE) and relative error(RE) are controlled in a small range. So the Hydrus-1D model can better simulate soil moisture dynamic change with the time of the patches vegetation in desert oasis ecotone.(6) In the regional scale, using the MODIS data and the thermal inertia method estimated surface soil water content of patches vegetation. Based on the establishment of thermal inertia and the soil moisture relation model. Compared linear, exponential and power function model, found that the exponential function has a high correlation of soil moisture estimation. And the error between estimated values of surface soil moisture and measured values is lower. The application of the thermal inertia to estimate surface soil moisture, providing an effective and feasible method for the arid desert area to obtain a wide range and long time series soil moisture data.The research results can be more in-depth understanding spatial distribution and time variation characteristics of the typical regional soil moisture. And lay the foundation for the study of distribution pattern of vegetation and soil wind erosion control of typical region. And also provide a theoretical basis for the regional vegetation restoration, maintenance of environmental security and the rational use of water resources. At the same time, the research results also provide an effective and feasible method for the typical region to obtain a wide range and long time series soil moisture data. |
PDF Full Text Request