摘要
在地形复杂、水土流失严重且地质灾害严重的白龙江流域,土壤含水量遥感监测在地质灾害监测预警研究中具有重要意义。为实时掌握白龙江流域的土壤水分含量状况,利用2013年4月的Landsat7 ETM＋影像,采用温度植被干旱指数法,构建Ts-NDVI特征空间,结合野外88个实测样点土壤水分数据,建立0~60 cm土壤深度范围内3个单层（0~20 cm,20~40 cm,40~60 cm）及2个平均层（0~40 cm,0~60 cm）的土壤水分遥感反演回归模型,对比分析了白龙江流域5个深度的土壤水分的空间变化特征,并用未参与建模的16个实测土壤水分数据样点进行相应的精度验证。结果表明：3个单层中20~40 cm土壤水分反演精度相对较高,RMSE值为3.06%,2个平均层中0~40 cm反演精度最高（RMSE）为2.45%,由此说明TVDI更能稳定地反映和指示土壤中层深度（20~40 cm）的水分分布状况。
In the Bailongjiang Basin, complex terrain, serious soil and water loss as well as serious geological disasters therein, remote sensing monitoring of soil moisture plays an important role in monitoring and early warning of geological disasters. Firstly, the Ts - ND- VI feature space was established by using temperature vegetation index method based on the Landsat 7ETM ＋ data in April of 2013, then the soil moisture inversion regression models of five soil depth including the 0 - 20 cm, 20 - 40 cm, 40 - 60 cm and 0 - 40 cm, 0 -60 cm were established using the TVDI model and combining with soil moisture data measured in 88 field samples to construct five soil depth＇ s soil moisture. At the same time, the measured soil moisture data in the other 16 field samples that did not participate in the modeling were used to verify the accuracy of the inversion results. The results show that among soil depths of 0 - 20 cm, 20 - 40 cm, 40 -60 cm, the 20 -40 cm depth where soll moisture estimation had the high precision of 3.06% , and between the 0 -40 cm and the 0 -60 cm averaged soil layers, the 0 -40 em depth where soil moisture estimation had the high precision of 2.45%. So TVDI method based on Landsat ETM ＋ data could be used to monitor the soil moisture in the Bailongjiang Basin.
出处
《干旱气象》
2015年第2期213-219,共7页
Journal of Arid Meteorology
基金
国家科技支撑计划项目(2011BAK12B06
2015BAK15B02)
甘肃省自然科学基金(096RJZA129)共同资助
作者简介
夏燕秋（1989-），女，硕士，研究方向为环境定量遥感．E—mail：1186333103@qq．com
通讯作者：马金辉（1964-），男，副教授，博士，研究方向为环境建模．E-mail：majh@lzu．edu．cn
