基于温度植被干旱指数的黄河源区土壤表层含水量反演

摘要

For the inversion and spatial-temporal analysis of the surface soil moisture content in the source region of the Yellow River, the temperature vegetation drought index (TVDI) was calculated in this study. The linear relationship between the LST/ NDVI slopes for different pixel scale windows and the measured surface soil moisture content was constructed. The fitted results for different pixel scale windows were also compared. The results show that: the fitting result for the 9 ×9 pixel window (regressionrncoefficientR2 = 0.76) is better than that for the other pixel scale windows; the inter-seasonal fitting accuracy (regression coefficient R2 = 0.76) is better than the long time series simulation results; since the value of the RMS error between the simulated results and the measured data in summer is 0.09, which satisfies experimental requirements, the surface soil water content can be estimated based on the short time series soil moisture inversion in the source region of the Yellow River.%为了对黄河源区的土壤表层含水量进行反演与时空分析,利用黄河源区2006年第65-321天MODIS温度和植被指数产品数据,计算了温度植被干旱指数(TVDI),构建了不同像元尺度窗口的LST/NDVI斜率与实测表层土壤相对含水量之间的线性关系,比较了不同尺度像元窗口的拟合效果.分析结果表明:9×9像元窗口的拟合效果(回归系数R2为0.65)优于其他像元窗口尺度；季节尺度的拟合精度(回归系数R2为0.76)优于长时间序列的拟合精度；夏季模拟值与实测值的均方根误差为0.09,模拟精度满足实验要求,可用于黄河源区短时间序列的土壤表层含水量反演.