土壤湿度的空间变异特征及尺度效应分析

基于遥感信息分析了土壤湿度的空间变异及尺度效应问题.利用多尺度遥感数据获得了归一化植被指数和地表温度,并基于此参数提取了干旱/半干旱地区的土壤湿度指数;基于分形理论,利用表面积计算方法,通过计算一系列不同尺度窗口像元的曲面面积,研究了三维曲面面积随尺度变化的规律,即定量描述了不同尺度遥感图像提取的土壤湿度指数值的变化特性,随着尺度的增加,驻点像元数减少,说明其空间异质性降低;并利用半变差函数研究了不同空间分辨率下土壤湿度的异质性特征.结果表明:空间分辨率越大,其整体空间异质性程度越小,与实际情况相符,说明利用分形几何学和地统计学方法分析土壤湿度指数的空间异质性特征是可行的.通过该研究可以有效地确定土壤湿度遥感监测的尺度,更好地了解区域尺度的土壤干旱程度及水资源情况.

Spatial heterogeneity and scale effect of soil moisture

Based on remote sensing information,we mainly studied spatial variation and scale effect of soil moisture,obtained normalized difference vegetation index and surface temperature using multi-scale remote sensing data,and extracted soil moisture index of arid and semiarid areas utilizing the two parameters.And then based on the fractal geometry,and surface area calculation method,we calculated curved surface areas of series of different scale windows and after that calculated surface fractal dimension,so as to research the change rules of 3D surface area with scale alteration,that is,quantificationally describing variation characteristics of soil moisture index extracted from different scales of remote sensing images.Additionally,under different spatial resolutions,we used land statistical methods to calculate semi-variogram in order to quantificationally describe the heterogeneity of soil moisture.The results show that the higher the spatial resolution,the smaller the degree of space heterogeneity caused by spatial self-correlation parts,which accords with actual situation and also explains that it is feasible to analyze spatial heterogeneity of soil moisture index using the fractal geometry and statistical methods.Through this research,we can effectively determine soil moisture’s remote sensing monitoring scale and model,and better understand regional scale soil drought and water level.It has certain scientific significance.