【摘要】 利用遥感手段监测土壤湿度有利于分析大尺度区域的土壤干湿状况。比对分析不同植被指数计算的温度植被干旱指数(TVDI)的精度能够提高TVDI反演土壤湿度的实际应用价值。以三江平原为研究区,基于2013年5—9月的四期MODIS影像,利用归一化植被指数(NDVI)、增强型植被指数(EVI)、修正土壤调节植被指数(MSAVI)、比值植被指数(RVI)分别计算TVDI,并以地面实测土壤湿度数据及降水数据进行精度验证。结果表明:(1) 4种植被指数计算的TVDI与土壤湿度数据均具有一定的负相关关系,即TVDI值越高,土壤湿度值越低;(2)不同植被指数计算的TVDI在5月、6月、9月与土壤湿度回归分析的R~2数值相近,均适合用来反演这3个时间段的土壤湿度,在7月份,相较于NDVI和RVI计算的TVDI结果(R~2均在0.15左右),基于EVI和MSAVI计算的TVDI(R~2均在0.35左右)更适合反演该时期的土壤湿度;(3) 5—9月期间,干旱现象主要发生在三江平原的中部及西南部,干旱程度主要为轻旱,东部及东北部在不同时期基本保持在正常或轻微湿润状态。更多还原

【Abstract】 The use of remote sensing to monitor soil moisture is conducive to the analysis of soil moisture conditions in large-scale areas. Comparing and analyzing the accuracy of the Temperature Vegetation Dryness Index(TVDI) calculated under different vegetation indices can increase the practical value of TVDI on retrieval of soil moisture. There are four vegetation indices, including NDVI(Normalized Difference Vegetation Index), EVI(Enhanced Vegetation Index), MSAVI(Modified Soil Adjusted Vegetation Index), RVI(Ratio Vegetation Index), were respectively used to construct LST-Ⅵ feature space and calculate TVDI by combining with land surface temperature data. The results indicated that:(1) all the TVDIs calculated by different vegetation indices presented the negative correlations with soil moisture data, that is, the higher the TVDI value, the lower the soil moisture value;(2) the values of R~2 of regression analysis between soil moisture data and TVDI based on different vegetation indices were similar in May, June and September, and they were all suitable for retrieving soil moisture in these three periods; in July, compared to the TVDI results calculated by NDVI and RVI(R~2 is around 0.15), TVDI based on EVI and MSAVI(R~2 is around 0.35) was more suitable for retrieving soil moisture in this period;(3) from May to September, the drought mainly occurred in the central and southwest parts of the Sanjiang Plain, and the degree of drought was mainly mild drought. The eastern and northeastern parts remained basically normal or slightly humid in different periods.更多还原