Soil moisture is a key factor that induces the deterioration of ecological environments in mining subsidence areas.In this study,a method of soil moisture inversion using Sentinel-1A dual polarized SAR and OLI surface reflectance data is proposed.First,the water content of vegetation is determined based on the normalized difference water index.Second,the effect of vegetation on the Sentinel-1A backscattering coefficient is eliminated by using the water cloud model,which has been transformed into the backscattering coefficient of bare soil.Third,based on the empirical model established by AIEM and Oh models,the surface parameters of the study area are determined and verified by the OLI optical inversion results.Finally,the soil moisture contents in the subsidence and non-subsidence areas are compared.(1) Compared with the results of soil moisture based on the soil moisture monitoring index,the VH-polarized inversion results exhibit better agreement between two polarization patterns.Moreover,the inversion results of two polarization patterns simultaneously show that the inversion result of the desertification grassland area is better than those of loess hilly and gully regions,which indicates that the effect of terrain on the backscatter cannot be ignored.(2) In the 72 sets of data during 2016,the VH-polarized inversion results show that the soil moisture content in the comparative area is larger than that in the subsidence area; it accounts for 41 cases,and the proportion is 57％.Meanwhile,the VV-polarized inversion results show that the soil moisture content in the comparative area is larger than that in the subsidence area; it accounts for 36 cases,and the proportion is 50％.In addition,subsidence in different mining areas is affected differently.Results indicate that the increase in surface roughness caused by mining subsidence exerts a negative effect on surface soil moisture.However,differences exist between different mining areas.%开采沉陷地质灾害诱发矿区生态环境恶化的关键因子是土壤水分变化.研究提出了一种利用Sentinel- 1A双极化SAR和OLI地表反射率数据联合反演土壤含水量的方法,即基于归一化水体指数(NDWI)反演植被含水量;采用Water-Cloud Model(WCM)模型消除植被对Sentinel-1A后向散射系数产生的影响,将其转化为裸土区的后向散射系数;利用基于AIEM模型和Oh模型建立的经验模型反演研究区地表参数,并用OLI光学反演结果进行验证;最后比较了开采沉陷区内外土壤水分含量.研究表明:(1)与基于OLI的土壤水分监测指数(SMMI)的土壤水分含量反演结果相比,两种极化方式中VH极化反演的水分结果具有更好的一致性,且两种极化方式反演结果也表明荒漠化草原区比黄土丘陵沟壑区反演效果更好,说明地形对后向散射的影响不可忽略.(2)在2016年内72期数据中,VH极化反演结果对比区土壤水分含量大于沉陷区的有41期,所占比例为57％;VV极化反演结果对比区土壤水分含量大于沉陷区的有36期,所占比例为50％,且不同矿区内的沉陷区受到的影响不同.说明开采沉陷造成的地表粗糙度的增加会对地表土壤水分产生负面影响,但不同矿区之间又有差异.
展开▼
