微波遥感反演土壤水分中构建粗糙度参数的新方法

A New Method for Constructing Land Surface Combined Roughness Parameter in the Process of Soil Moisture Retrieval by Microwave Remote Sensing

地表粗糙度是影响雷达后向散射系数的重要因素。该文在基于SAR影像反演地表土壤水分的过程中,考虑到地表粗糙度的野外测量误差、取值范围和雷达入射角等方面的影响,统计了裸土、农用地和草地等几种典型地表的粗糙度测量数据,以此限定AIEM模型的输入参数范围。首先,利用AIEM模型模拟雷达后向散射系数与粗糙度、土壤水分之间的关系,构建了基于曲面拟合思想的、与入射角相关的组合粗糙度参数,并以此为基础利用Envisat ASAR双极化数据(VV、VH)建立了土壤水分反演模型。经实测数据验证,在不同入射角范围内,基于该文建立的模型得到的土壤水分反演结果与实测值都有良好的相关性。与其他形式的组合粗糙度参数进行对比,该文提出的模型反演精度较高,能够适用于入射角范围在(5°,65°)内的SAR影像的反演。

Surface roughness,which influences the scattering coefficient of radar,is an important factor in soil moisture retrieval.In this process of estimating the soil moisture by SAR images,the effects of various aspects were considered,such as field observational error and the range of surface roughness,and the radar incident angle.The range of roughness,which could reflect the actual land surface,was determined based on the statistics of effective roughness value of several typical surface types（bare soil,farmland,grassland,et al.）.Then the simulated scattering coefficient could be obtained by limiting the range of input parameters of the AIEM model.Unlike other studies,the curved surface among simulated backscattering coefficientσ0 pq,root-meansquare（rms）height Sand correlation length l was established.By fitting the equation of the curved surface and analyzing the relationship amongσ0 pq,Sand l for different radar incident angles,the new combined roughness parameter with better applicability and precision was constructed.Then the soil moisture retrieval model was established based on Envisat ASAR C-band dual polarization（VV,VH）data.The validation results revealed that the estimated soil moisture had a good correlation with the observed values in different range of incident angle.Compared with other forms of combined roughness parameters,the results based on the model proposed in this paper showed a higher accuracy and it was suitable for soil moisture retrieval by SAR images with the incident angle of 5°-65°.