基于主动微波遥感的典型黑土区土壤水分反演

Soil moisture inversion based on active microwave remote sensing in typical black soil area

针对野外测量土壤粗糙度因子相关长度不精确问题,提出一种新反演黑土区土壤水分经验模型。以黑龙江省赵光农场为研究区,选择Oh模型作为研究区土壤水分反演基础,通过分析同极化、交叉极化后向散射系数对均方根高度、土壤水分响应特征,建立函数关系式。利用推算交叉极化后向散射系数差消除土壤粗糙度影响,建立黑土区裸露地表土壤湿度测算经验散射模型。结果表明,VV/VH极化后向散射系数模拟精度较高,预测土壤湿度平均绝对误差为0.0530 cm3·cm-3,与组合粗糙度下土壤水分反演结果相比相关性提高15.32%,证明模型有效性。利用C波段Radarsat-2数据反演黑土裸土区土壤水分含量,具有较高精度。结果可为该地区旱涝灾害监测和作物长势提供信息支持,具有重要应用价值。

In order to solve the problem of inaccuracy of measuring correlation length,a new empirical model for retrieving soil moisture in black soil area without correlation was proposed.Taking Zhaoguang Farm in Heilongjiang Province as the research object,and selected Oh model as the foundation of soil moisture inversion in the study area.By analyzing the response characteristics of the same polarization and cross polarization backscattering coefficient to the root mean square height and soil moisture,a functional relationship was established.An empirical scattering model for calculating soil moisture in bare soil was proposed by using the cross polarization backscatter coefficient difference instead of the soil roughness.The result of VV/VH polarization backscattering coefficient simulation was better and the mean absolute error is 0.0530 cm3·cm-3.In addition,the correlation of soil moisture inversion was 15.32%higher than the combined roughness,which proved the validity of the model.It showed that the VV/VH polarization combination could better reverse the soil moisture in the bare soil region.The empirical model could better meet the needs of soil moisture monitoring match with the traditional soil moisture classification method.Thus,the Radarsat-2 data of C band could be used to inverse soil moisture in the bare area of black soil.These results could provide some technical support for farmland soil management.