现代农业研究2026，Vol.32Issue(3)：48-53,6.
基于多极化SAR数据的黑土区农田土壤水分反演研究
Research on Soil Moisture Retrieval in Black Soil Region Based on Multi Polarimetric SAR Data
摘要
Abstract
Farmland soil moisture is an important parameter to characterize surface drought.The accurate acquisition and long-term monitoring of soil moisture spatio-temporal distribution information are closely related to black land ecological protection and national food security.Microwave remote sensing is an effective means to monitor soil moisture.It can achieve remote access to the required information and is not affected by external weather.However,the scattering con-tribution of vegetation and soil surface roughness greatly reduces the accuracy of soil moisture estimation.In order to solve the above problems,this paper takes the friendship farm in Shuangyashan City of Heilongjiang Province as the re-search area,constructs the radar backscatter parametric model based on the multi polarization SAR data,and develops the soil moisture retrieval algorithm suitable for the farmland in the black soil area of Northeast China.Firstly,The Water Cloud Model suitable for crop growth period was selected as the forward model.The vegetation water content was estimated by vegetation index(GNDVI,NDVI,EVI),and then the contribution to vegetation backscattering coefficient was separated;At the same time,the combined roughness parameter Zg was introduced to weaken the influence of soil surface roughness on backscattering coefficient.Finally,the simulated backscatter coefficient and radar backscatter coeffi-cient were used to construct the cost function to achieve high-precision retrieval of soil moisture.The results showed that:(1)using GNDVI to establish vegetation water content estimation model had better effect,and the determination coefficient R2 reached 0.73 for corn and 0.83 for soybean;(2)The precision of soil moisture retrieval and surface roughness retrieval was high,and the root mean square errors of corn field were 0.016cm3/cm3 and 0.168cm,respective-ly;The root mean square errors of soybean fields were 0.031cm3/cm3 and 0.201cm,respectively.This study has impor-tant reference value for monitoring drought and flood disasters and field management in black soil region.关键词
合成孔径雷达/土壤水分/土壤表面粗糙度/后向散射耦合模型Key words
synthetic aperture radar/soil moisture/soil surface roughness/backscattering coupling model分类
农业科技引用本文复制引用
郭君达,陈思,孙晨,王荣彬..基于多极化SAR数据的黑土区农田土壤水分反演研究[J].现代农业研究,2026,32(3):48-53,6.基金项目
国家自然科学基金项目"削弱表面粗糙度耦合影响的农田土壤水分光学和雷达遥感协同反演研究"(项目编号:42201435) （项目编号:42201435）
吉林省科技发展计划项目"东北地区农田土壤水分多源遥感协同反演算法研究"(项目编号:YDZJ202301ZYTS230). （项目编号:YDZJ202301ZYTS230）
