The Application Of Improved SBAS-InSAR Technology In The Monitoring Of Surface Subsidence In Mining Areas

The longer the mining is,the more prone to a large number of surface subsidence in the mining area,which endangers the safety of human life or another person property.InSAR is developing rapidly with its advantages of all-day,all-weather.SBAS-InSAR overcomes the effects of spatio-temporal decoherence,terrain and atmospheric noise,increasing the sampling frequency by freely combining short baseline interference pairs.Therefore,it is suitable for monitoring the mining areas'surface in which pixels have experienced spatio-temporal decoherence.The paper mainly studies the approach of SBAS-InSAR to monitor ground subsidence in mining areas.Based on coherent grid,an algorithm of extracting points was developed to improve the number of PS points in mining areas.And a multi-stage filtering method was proposed to optimize the quality of PS points for the interferograms of serious atmospheric phase and unreasonable mathematical model.The effectiveness of the improvement is verified through test experiments.Finally,the improved SBAS-InSAR is applied to monitor the mining areas'surface.The main research work in this paper includes:(1)Based on the traditional method for extracting PS points of SBAS-InSAR technology(coherence coefficient method,amplitude dispersion method),an extracting algorithm rely on medium coherent grid is used to increase the number of extraction points in a large number of subsidence areas.In the simulation experiment,the effectiveness of the algorithm improvement is demonstrated through comparative analysis of quantity,density and coherence related to PS points.(2)Considering that atmospheric delay phase is highly correlated with the elevation,that adaptive filter inherits the good edge preservation characteristics while removing speckle noise,the paper proposes the atmospheric phase optimization models related to elevation.As for the verifying the reliability of the model,we compare the interferograms in the test experiments.Taking into account the capability of the model to estimate the phase while using weighted filter and iterative ideas,the paper applies regression analysis model to estimate the phase.In the test experiments,RMS of residual phase obtained by iteration are compared to verify the effectiveness of the improvement.(3)Choosing the Panlong area in Changzhi,Shanxi Province as the research area,the paper applies the modified SBAS-InSAR,traditional PS-InSAR,and traditional SBAS-InSAR to 10 scenes C-band RADARSAT-2 data to achieve cross-validation.The reliability of results are verified through the comparison of results and field conditions.Finally,the cumulative deformation of PSs are superimposed on their own actual elevation to realize the 3D reconstruction of the goaf subsidence area.The results show that a large number of land subsidence occurred in two areas of the study area.The subsidence area in the north is the largest,with a maximum settlement rate of-50.948mm/a and a settlement range of 1.1 km~2.In the south,the second subsidence area is relatively moderate,and its maximum settlement rate reach to-37.842 mm/a.However,both of goafs affected the surrounding villages and should be noted.