基于MODIS昼夜地表温度的广西土壤相对含水量反演  

Relative Soil Moisture Retrieval Based on MODIS Day and Night Surface Temperature Data in Guangxi

作　　者：卢献健 黎振宝 晏红波 周斌 LU Xianjian;LI Zhenbao;YAN Hongbo;ZHOU Bin(College of Geomatics and Geoinformation,Guilin University of Technology,Guilin 541006,China;Augur Intelligence Technology Co.,Ltd.,Guangzhou 510663,China)

机构地区：[1]桂林理工大学测绘地理信息学院,广西桂林541006 [2]奥格科技股份有限公司,广东广州510663

出　　处：《无线电工程》2022年第12期2186-2195,共10页Radio Engineering

基　　金：广西自然科学基金(2022GXNSFBA035639);国家自然科学基金(42064003,42267040);广西空间信息与测绘重点实验室开放基金(桂科能19-050-11-23)。

摘　　要：基于遥感影像的土壤相对含水量(Relative Soil Moisture,RSM)反演是遥感技术应用的重要内容。高植被覆盖区地形复杂多变引起的白天与夜晚温度波动,带来了RSM反演的不确定性。利用MODIS白天、夜晚的地表温度分别与增强植被指数构造了EVI-LST特征空间,利用线性函数与多项式函数分别拟合干湿边方程并进行了拟合精度比较;在此基础上,根据干湿边方程计算增强温度植被干旱指数并研究了它们的差异,同时利用气象站实测的RSM数据对反演结果进行评价,对反演的广西2008年10月上半月土壤相对含水量进行干旱等级划分与评估。结果表明:①从相关系数及其变化来看,针对广西高植被覆盖度、地形复杂多样的特性,利用多项式函数拟合EVI-LST特征空间干湿边所取得的效果较线性函数拟合更优;②基于ETVDI模型反演所获得的RSM通常小于实测值,且基于夜晚温度计算的RSM相较于利用白天温度计算的结果更接近实测值;③广西2008年10月上半月干旱程度较轻,重旱、轻旱、正常、湿润和极湿润的面积分别为0.02%,5.40%,37.66%,52.26%,4.65%。Relative Soil Moisture(RSM)reversal based on remote sensing images is an important element of remote sensing technology application.The temperature fluctuations during daytime and nighttime caused by the complex and variable topography in high vegetation covered area bring uncertainties for RSM reversal.The EVI-LST feature space is constructed by using MODIS daytime and nighttime surface temperature and enhanced vegetation index respectively,and the dry and wet side equations are fitted with linear functions and polynomial functions respectively after which the fitting accuracies are compared;on this basis,the enhanced temperature vegetation drought indices are calculated based on the dry and wet side equations and the difference among indices is studied,and the RSM data measured by meteorological stations are used to evaluate the reversal results.Also,the RSM in the first half of October 2008 is further used to classify and evaluate the drought level in Guangxi.The results show that:①from the perspective of correlation coefficients and their variations,for the characteristics of high vegetation cover and complex and diverse terrain in Guangxi,the results achieved by using polynomial functions to fit the dry and wet edges of EVI-LST feature space are better than those achieved by using linear functions;②the RSM based on model reversal is usually smaller than the measured values,and the RSM calculated with night temperature is closer to the measured values than those calculated with daytime temperature;③the drought level in the first half of October 2008 in Guangxi is relatively low,and the percentage of area for severe drought,light drought,normal,wet,and very wet is 0.02%,5.40%,37.66%,52.26%,and 4.65%respectively.

关 键 词：土壤相对含水量 ETVDI 多项式拟合 干旱等级 

分 类 号：P407.6[天文地球—大气科学及气象学]