反演策略对SMOS土壤水分反演算法的影响研究  

Study on Retrieval Strategy of SMOS Soil Moisture Retrieval Algorithm

作　　者：劳从坤 杨娜[1] 徐少博 汤燕杰 张恒杰 Lao Congkun;Yang Na;Xu Shaobo;Tang Yanjie;Zhang Hengjie(School of Surveying and Land Information Engineering,Henan Polytechnic University,Jiaozuo 454000,China)

机构地区：[1]河南理工大学测绘与国土信息工程学院,河南焦作454000

出　　处：《遥感技术与应用》2020年第1期65-73,共9页Remote Sensing Technology and Application

基　　金：国家自然科学基金青年基金项目(41501363)。

摘　　要：为降低SMOS土壤水分反演算法的复杂度、提高土壤水分反演精度,对SMOS土壤水分反演策略进行调整:将多参数反演改为单参数反演以简化观测与模拟亮温的代价函数,以固定步长(0.001 m^3/m^3)代替不定步长从而避免复杂的矩阵运算,将围绕土壤水分先验值的少量局部搜索调整为全土壤水分区间(0~0.05 m^ 3/m^ 3)的密集全局搜索。利用美国USCRN 44个站点实测土壤水分分别与SMOS官方反演的土壤水分和SMOS调整算法反演的土壤水分进行对比分析。结果表明:与SMOS相比,算法调整后土壤水分的平均绝对偏差MAD、均方根误差RMSE和无偏均方根误差ubRMSE分别降低了0.012、0.018和0.020 m^3/m^3。In order to reduce the complexity of SMOS official soil moisture retrieval algorithm and improve the accuracy of soil moisture retrievals,a new retrieval strategy on SMOS soil moisture retrieval algorithm was developed.In the new retrieval strategy on SMOS soil moisture retrieval algorithm,the fixed step size(0.001 m^3/m^3)was used to replace the flexible step size obtained by the SMOS matrix operation.The multi-parameter was changed to a single-parameter in the cost function.The data from 44 USCRN sites in the United States were compared with the soil moisture retrieved from SMOS official algorithm as well as the adjustment of SMOS algorithm.The results show that compared with the SMOS official algorithm,the average absolute deviation,root mean square error,and unbiased root mean square error of the adjustment of SMOS algorithm are reduced by 0.012 m^3/m^3,0.018 m^3/m^3,and 0.020 m^3/m^3,respectively.

关 键 词：土壤水分 SMOS 反演算法 亮温模拟 

分 类 号：TP701[自动化与计算机技术—检测技术与自动化装置]