Soil moisture is a key variable in the global water cycle and energy exchange process and plays an important role in hydrology,climate and agricultural production.In desertification habitats,surface vegetation is extremely sensitive to changes in soil moisture.In desertification mining subsidence areas,vegetation ecology is more fragile and vulnerable.At present,research in this field is rather weak.Most existing researches are using optical data to retrieve vegetation canopy and surface soil moisture changes.The research methods are relatively single and there is no strict definition of mining areas and non-mining areas.Therefore,the regularity of soil moisture change caused by mining cannot be obtained,and the inevitable connection between the mining subsidence and the change of soil moisture cannot be established.Taking the arid and semi-arid aeolian sand mining area at the junction of Inner Mongolia and Shaanxi as the research area,the surface soil moisture of the study area was firstly retrieved by Radarsat-2 HH single-polarization data,and the results were compared with the retrieval results of MODIS surface reflectance data.Then,combined with the DInSAR results of Radarsat-2,the spatial-temporal variation of soil moisture in the complex background of the mining area was explored.In order to further verify this law of change,the Sentinel-1A VH/VV dual-polarization data was used to retrieve the surface soil moisture in the study area,and the retrieval results of Landsat-8 surface reflectance data were used for comparison verification;then a spatial-temporal variation law of surface soil moisture in mining subsidence area can also be obtained by combining with the DInSAR results of Sentinel-1A.When retrieving with dual-polarization data(Sentinel-1A VH/VV),it was noted whether the surface roughness in the mining subsidence area was “overrun”,that is whether the surface roughness in the mining subsidence area was within the limits of theoretical model roughness.In order to verify the conjecture,the difference method was used to detect the surface roughness before and after the subsidence and it found that the difference inside and outside the subsidence area was not obvious.Therefore,combined with the Sentinel-1A coherence coefficient image,the dual-polarization surface parameters retrieval model was further improved,and the spatial-temporal variation law of surface soil moisture in mining subsidence area was obtained similarly.The main conclusions are as follows:(1)The retrieval results of Radarsat-2 and MODIS,Sentinel-1A and Landsat-8,the Sentinel-1A improved model and the Landsat-8 are compared respectively.The retrieval results of the two kinds of data all have good consistency,and the retrieval accuracy of the three algorithms is gradually improved.This study has made full use of the advantages of multi-source remote sensing.Under the premise that the retrieval algorithm has been verified by the predecessors,the dual-polarization retrieval algorithm has been improved,which provides technical support for the rapid acquisition and verification of soil moisture under the complex background of mining area.(2)In view of the inaccuracy of surface roughness inversion in mining subsidence area,the interference coherence coefficient was added to the surface parameter inversion model as an evaluation factor of surface roughness.It was found that the inversion results of the soil moisture reconstructed model and the Landsat-8 OLI data were smaller,which proved the effectiveness of the improved model.At the same time,it provides a solution to the retrieval of surface parameters in areas with unnatural surface and roughness changes and improves the universality of the model.(3)The retrieval results of Radarsat-2 single-polarization,Sentinel-1A dualpolarization,and Sentinel-1A dual-polarization improved model simultaneous show that,after excluding the influence of other unrelated factors,in the Shendong Mining Area,due to the influence of underground high-intensity mining activities,the surface soil moisture in the subsidence area will be lower than the unaffected area.Therefore,the connection between mining subsidence and soil moisture change has been established,which is of great significance for ecological protection,restoration and reconstruction of subsidence area in desertification mining area. |