| As a growing city in the period of accelerated industrialization and urbanization,Xuzhou has a large amount of abandoned industrial and mining land to be developed and reused.Xuzhou as the largest coal energy base in East China in the early days,underground coal mining in Xuzhou has inevitably formed a large-scale ground subsidence area,and it hinder the development and utilization of abandoned mining areas.Therefore,it is of great significance to carry out the continuous long-term time series surface subsidence monitoring in Xuzhou abandoned mining areas for the comprehensive treatment of Xuzhou mining subsidence area as well as the transformation and upgrading of mining area.At present,time series In SAR technology has been widely used in surface deformation monitoring.However,facing the geographical environment surface type with few abandoned mining buildings and sparse vegetation coverage,conventional time-series In SAR technology often has the deficiency of low observation point density and uneven distribution.In addition,during the rainy season in Xuzhou,there is often a confluence of cooling and heating flows.The significant atmospheric difference caused by it also seriously affect the accuracy of time series In SAR moitoring.Although there are many correction methods for atmospheric delay,none of the existing methods has universal applicability,which may not be able to effectively remove the atmospheric delay error of Xuzhou abandoned mining areas.In view of the above problems,and in order to fully exploit the potential of multipolar SAR data in surface deformation monitoring,based on time-series In SAR technology,this paper compares and analyzes the surface deformation monitoring results of mining areas under different polarizations,and on this basis,develops the study of atmospheric correction methods in Xuzhou abandoned mining areas;finally it is applied to the long time series surface deformation monitoring of Xuzhou abandoned mining areas the study.The research results are as follows:(1)The main principle of Distributed Scatterer Interferometric Synthetic Aperture Radar(DS-In SAR)is described.In addition,the VV polarization and VH polarization Sentinel-1A images covering the mining area around Xuzhou from January 2018 to December 2019 are used as the data source to carry out comparative experimental research on different polarization data based on five homogeneous pixel recognition methods and phase optimization method of eigenvalue decomposition based on coherence matrix.And Finally,the cross-validation with the conventional PS-In SAR monitoring results shows the reliability of the monitoring results under different polarization data.The result shows:(1)In terms of average amplitude,the average backscattering coefficient of VV polarization image is larger;(2)Under the same processing parameters,for different recognition methods of the same polarization data,the GLR inspection method has a better recognition effect of statistically homogenous pixels;From the perspective of the same recognition method of different polarization data,the VH polarization data set can identify more statistically homogeneous pixels,but it is inferior to the VV polarization data set in terms of phase optimization effect;(3)The deformation monitoring results of DS-In SAR technology with the two polarized data are basically consistent,and their spatial distribution trends are basically consistent with those obtained by PS-In SAR.However,DS-In SAR technology can extract a larger number of measurement points and more uniform spatial distribution for more comprehensive and reliable surface deformation results.(2)Based on the characteristics of frequency division of wavelet transform,a method of atmospheric delay correction based on wavelet transform is proposed,Taking the monitoring of abandoned mining areas in Xuzhou from 2018 to 2019 as an example,the differences between the atmospheric correction results based on wavelet and those based on GACOS and principal component analysis are compared and analyzed.Results show that the three methods of correction for atmospheric correction of xuzhou mining area has certain limitation,but the atmospheric correction method based on wavelet transform to wrap interference figure atmospheric correction effect is better,and go up in space and time in addition to the adverse effects of high frequency atmospheric error,improve the accuracy of deformation monitoring of the xuzhou abandoned mine area.(3)A long-term time series wide-area surface deformation monitoring in Xuzhou mining area from July 2015 to September 2020 was carried out based on the conventional PS-In SAR technology and DS-In SAR technology with atmospheric correction technology based on wavelet transform.The result shows:(1)The results obtained by the DS-In SAR can effectively overcome the shortage of targets with high coherence selected by the conventional PS-In SAR technology,and improve the coherence of the surface deformation phase in the spatial distribution to a certain extent,which is more beneficial to the extraction of large magnitude surface deformation information;(2)The long time series monitoring with a span of 5 years can more completely reflect the development course of the surface morphology of the mining area and its evolution law in time series;(3)There are significant spatial differences in surface deformation of abandoned mining areas in Xuzhou,and most of the surface uplifted areas tend to be stable with time,the reason is analyzed combined with the time accumulation deformation,which shows that abandoned mining areas have been closed for a long time,and the backflow of groundwater will make the surface of the goaf gradually rise and stabilize.Some mined-out areas are affected by surface subsidence during mining and appear as surface subsidence in the results of monitoring period.In addition to the mining area,the surface deformation caused by urban construction also presents significant nonlinear characteristics.There are 39 figures,4 tables and 85 references. |
PDF Full Text Request