PSInSAR在地面沉降监测中的研究分析
|
摘要
我国是地质灾害多发的国家,地面沉降、滑坡等以地表形变为直接特征的地质灾害给国家和人民带来了严重的危害。据不完全统计,我国目前已有96个城市和地区发生了不同程度的地面沉降,其中天津地区的地面沉降很有代表性。
DInSAR具有广覆盖、高分辨率、全天候、全天时等优势,是一种很有潜力的地表形变监测技术,但实际应用中的时间和空间去相干因素、误差、大气相位以及大气延迟等往往会降低其获取形变信息的精度,甚至在某些情况下使监测不能实施。PSInSAR技术提取并只分析呈类点状散射体的PS点的相位,保证了相位质量,通过建立形变模型和时序滤波去除PS上的大气相位等误差,克服了DInSAR的失相干和大气影响等限制因素。但常规的PSInSAR方法大多通过分析像素的亮度值及其变化选择PS候选点,再利用形变的时间模型来识别PS点和去除PS点上的干扰信息,它们对很多自然地貌不能得到可靠的结果,且受形变模型的限制。
本次研究并不是继续探讨非线性形变和更复杂运动模型的构建和使用,而是采用了一种新的PSInSAR技术方法,该方法通过对相位进行空间相关性分析,适用于几乎所有地表情况的PS点的选取,且不受形变模型限制,能获得各个成像时期的形变信息,提高了监测的时间分辨率,对形变规律和形变的分析有重要的指导意义。由于相位解缠技术在InSAR、DInSAR、PSInSAR中都是至关重要的一个环节,涉及到空间规则和非规则格网分布的数据的二维、三维相位解缠,本次研究也对这些相关的算法进行了总结、比较和实践。
我们利用收集的天津地区2003年10月至2005年12月的14景ENVISAT ASAR数据,用基于空间相关性分析的PSInSAR技术进行了PS点的选取、误差去除和形变信息提取。其中相位解缠技术采用的是最小0范实现的伪三维相位解缠算法。针对PSInSAR技术只能得到PS点上的形变信息,又对离散分布的PS点的形变结果进行Kriging插值和形变等值线提取等处理,得到实验区的连续分布的形变场,然后对形变场进行了三维显示、侧视图和剖面图的生成,对受关注的沉降中心进行了形变时间序列分析,最后联系沉降区的地下水开采和经济状况等进行沉降成因的分析,并作了控沉的建议。
China is a country prone to geological disasters, Land subsidence, landslide and other geological disasters with characteristics of surface deformation, has caused serious harm to the country and the people. According to incomplete statistics, there are currently about 96 cities and regions, which have different degrees of subsidence. And the land subsidence in Tianjin is very representative.
Synthetic aperture radar differential interferometry- DInSAR, with advantages of wide-coverage, high-resolution, all-weather, all-time, etc, is a promising monitoring technique of surface deformation. However, in practical application, time and space decorrelation, noise and errors, atmospheric phase and atmospheric delay, etc, have reduced its accuracy for deformation information acquisition, even make it unenforceable in some cases.
PSInSAR (permanent scatter InSAR) is possible to avoid many of the limitations of conventional DInSAR by only analyzing pixels, which behave like point scatters and retain some degree of correlation. However, in almost all of its previous algorithms, an initial set of PS pixels is selected using amplitude variation, which doesn’t work very well in non-urban areas, and the approximate temporal model for deformation required to refine PS, estimate and remove nuisance terms, which is complicated and also the aim of study, is usually assumed to be constant in rate, or periodic in nature.
This study doesn’t continue to explore non-linear or more complex motion model, but introduced a new PSInSAR, which uses spatial correlation of interferogram phase to find a network of stable pixels in almost all terrain without prior knowledge of temporal variations in the deformation. As the new method is not restricted by deformation model, we can get the deformation information for each data-acquisition time, which improves the time resolution and has a significant guide to the analysis of the deformation laws and causes.
Phase-unwrapping is a key point in the workflows of InSAR, DInSAR and PSInSAR, which involves2-D and 3-D phase unwrapping for space regular and non- regular grid distribution data. Some related algorithms are summarized, compared and practiced in this study.
We have collected 14 ENVISAT ASAR SLCs covering from Oct 2003 to Dec 2005 for Tianjin area, and by using the new PSInSAR technique, which is based on phase spatial correlation analyzing, PS selection, noise estimation, phase unwrapping and deformation information extraction are applied to these ASAR images. A L0-norm pseudo-3D algorithm is implemented for phase unwrapping. As the PS are discrete and irregularly distributed, Kriging interpolation and deformation contours are applied to get continuous deformation senses. Monitoring results in 3D views, side views and profile maps are also generated. Some subsidence centers are analyzed by deformation time series. At last, combined with the mining of ground water and economic station of this area, the cause of deformation is analyzed, and suggestions on subsidence control are proposed.
DInSAR具有广覆盖、高分辨率、全天候、全天时等优势,是一种很有潜力的地表形变监测技术,但实际应用中的时间和空间去相干因素、误差、大气相位以及大气延迟等往往会降低其获取形变信息的精度,甚至在某些情况下使监测不能实施。PSInSAR技术提取并只分析呈类点状散射体的PS点的相位,保证了相位质量,通过建立形变模型和时序滤波去除PS上的大气相位等误差,克服了DInSAR的失相干和大气影响等限制因素。但常规的PSInSAR方法大多通过分析像素的亮度值及其变化选择PS候选点,再利用形变的时间模型来识别PS点和去除PS点上的干扰信息,它们对很多自然地貌不能得到可靠的结果,且受形变模型的限制。
本次研究并不是继续探讨非线性形变和更复杂运动模型的构建和使用,而是采用了一种新的PSInSAR技术方法,该方法通过对相位进行空间相关性分析,适用于几乎所有地表情况的PS点的选取,且不受形变模型限制,能获得各个成像时期的形变信息,提高了监测的时间分辨率,对形变规律和形变的分析有重要的指导意义。由于相位解缠技术在InSAR、DInSAR、PSInSAR中都是至关重要的一个环节,涉及到空间规则和非规则格网分布的数据的二维、三维相位解缠,本次研究也对这些相关的算法进行了总结、比较和实践。
我们利用收集的天津地区2003年10月至2005年12月的14景ENVISAT ASAR数据,用基于空间相关性分析的PSInSAR技术进行了PS点的选取、误差去除和形变信息提取。其中相位解缠技术采用的是最小0范实现的伪三维相位解缠算法。针对PSInSAR技术只能得到PS点上的形变信息,又对离散分布的PS点的形变结果进行Kriging插值和形变等值线提取等处理,得到实验区的连续分布的形变场,然后对形变场进行了三维显示、侧视图和剖面图的生成,对受关注的沉降中心进行了形变时间序列分析,最后联系沉降区的地下水开采和经济状况等进行沉降成因的分析,并作了控沉的建议。
China is a country prone to geological disasters, Land subsidence, landslide and other geological disasters with characteristics of surface deformation, has caused serious harm to the country and the people. According to incomplete statistics, there are currently about 96 cities and regions, which have different degrees of subsidence. And the land subsidence in Tianjin is very representative.
Synthetic aperture radar differential interferometry- DInSAR, with advantages of wide-coverage, high-resolution, all-weather, all-time, etc, is a promising monitoring technique of surface deformation. However, in practical application, time and space decorrelation, noise and errors, atmospheric phase and atmospheric delay, etc, have reduced its accuracy for deformation information acquisition, even make it unenforceable in some cases.
PSInSAR (permanent scatter InSAR) is possible to avoid many of the limitations of conventional DInSAR by only analyzing pixels, which behave like point scatters and retain some degree of correlation. However, in almost all of its previous algorithms, an initial set of PS pixels is selected using amplitude variation, which doesn’t work very well in non-urban areas, and the approximate temporal model for deformation required to refine PS, estimate and remove nuisance terms, which is complicated and also the aim of study, is usually assumed to be constant in rate, or periodic in nature.
This study doesn’t continue to explore non-linear or more complex motion model, but introduced a new PSInSAR, which uses spatial correlation of interferogram phase to find a network of stable pixels in almost all terrain without prior knowledge of temporal variations in the deformation. As the new method is not restricted by deformation model, we can get the deformation information for each data-acquisition time, which improves the time resolution and has a significant guide to the analysis of the deformation laws and causes.
Phase-unwrapping is a key point in the workflows of InSAR, DInSAR and PSInSAR, which involves2-D and 3-D phase unwrapping for space regular and non- regular grid distribution data. Some related algorithms are summarized, compared and practiced in this study.
We have collected 14 ENVISAT ASAR SLCs covering from Oct 2003 to Dec 2005 for Tianjin area, and by using the new PSInSAR technique, which is based on phase spatial correlation analyzing, PS selection, noise estimation, phase unwrapping and deformation information extraction are applied to these ASAR images. A L0-norm pseudo-3D algorithm is implemented for phase unwrapping. As the PS are discrete and irregularly distributed, Kriging interpolation and deformation contours are applied to get continuous deformation senses. Monitoring results in 3D views, side views and profile maps are also generated. Some subsidence centers are analyzed by deformation time series. At last, combined with the mining of ground water and economic station of this area, the cause of deformation is analyzed, and suggestions on subsidence control are proposed.
引文
Andrew John Hooper, Persistent Scatter Radar Interferometry for Crustal Deformation Studies and Modeling of Volcanic Deformation, A dissertation submitted to the department of geophysics and the committee on graduate studies of Stanford University in Partial fulfillment of the requirements for the degree of doctor of Philosophy, 2006
A. Hooper, R.Pedersen and F.Sigmundsson, Constraints on Magma Intrusion at Eyjafjallaj?kull and Katla Volcanoes in Iceland, From Time Series SAR Interferometry. The VOLUME Project, VOLcanoes: Understanding subsurface mass movement (eds. C.J.Bean et al.), 2009, 13-24.
Crosetto M., A.Arnaud, J.Dura,E.Biescas, and M.Agudo, Deformation monitoring using remotely sensed radar interferometric data, Proc., 11th FIG Symposium on Deformation Measurements, Satntorini, 2003.
Dehls, J.F.,M.Basilico and C.Colsesanti, Ground deformation monitoring in the Ranafjord area of Norway by means of the permanent scatters technique, IEEE Int. Geoscience and Remote Sensing Symposium, IGARSS 2002,2002:203-207.
Dennis C.Ghiglia and Mark D.Pritt. Phase Unwrapping Theory, algorithms, and software [M],1998.
Ferretti A, PratiC, RoccaF, Permanent Scatters in SAR Interferometry, IEEE Transactions Geoscience and Remote Sensing, 2001, 39(1):8~20.
Lyons,S. , and D.Sandwell, Fault creep along the southern San Andreas from interferometric synthetic aperture radar, permanent scatterers, and stacking, Journal of Geophysical Research, 2003,108(B1):2047~2070.
Werner, C., U.Wegmüller, T.Strozzi, and A.Wiesmann, Interferometric point target analysis for deformation mapping, Geoscience and Remote Sensing Symposium, IGARSS, 2003.
Zebker H A, Villasensor J, Decorrelation in Interferomatric radar echoes, IEEE Trans Geoscience Remote Sensing, 1995, 22:2517~2520.
陈强.基于永久散射雷达差分干涉探测区域地表形变的研究:[博士学位论文].四川:西南交通大学,2006
程璞等.INSAR相位解缠算法研究[J].大地测量与地球动力学,2007(6)
董克刚,周俊,等.天津市地面沉降的特征及危害[J].地质灾害与环境保护, 2007,18(1):67~70
董克刚,周俊,于强,等.天津市地面沉降的特征及其危害[J].地质灾害与环境保护,2007,18(1):67~70
范景辉,郭华东,等.基于相干目标的干涉图叠加方法监测天津地区地面沉降[J].遥感学报, 2008,12(1):111~118
范景辉.基于相干目标的DInSAR技术地表形变监测研究与应用:[博士论文].北京:中国科学院遥感应用研究所,2007
范景辉,李梅,等.基于PSInSAR方法和ASAR数据监测天津地面沉降的试验研究[J].国土资源遥感,2007,4:23~17
付进朋.InSAR相位解缠算法的研究:[硕士论文].四川:西南交通大学,2006
傅文学,田庆久.等,PS技术及其在地表形变监测中的应用现状与发展[J].地球科学进展,2006,11:119~1198
傅文学.低相关区滑坡移动的DInSAR监测方法研究-以树坪滑坡为例:[博士论文].南京:南京大学,2007
葛大庆,郭小方,等.基于序列差分干涉图的地表形变速率提取[J].国土资源遥感, 2007,1:24~31
葛大庆,王艳,等.基于相干点的多基线D-InSAR技术与地表形变监测[J].遥感学报,2007,11(4):574~560
葛大庆,王艳,等.利用短基线差分干涉图集监测地表形变场[J].大地测量与地球动力学,2008,28(2):61~66
郭良迁,薄万举,陈宇坤,等.天津地区的垂直形变与构造活动研究[J].大地测量与地球动力学,2009,29(5):1~5
李平湘,杨杰.雷达干涉测量原理与应用[M].北京:测绘出版社,2006
刘广.长时序DInSAR技术中的空间基线校正和图像配准方法研究.[博士论文],北京:中国科学院遥感应用研究所,2007
刘艳华.合成孔径雷达干涉(InSAR)测量技术测量地表变形的理论和算法研究,[博士后研究工作报告].北京:中国测绘科学院研究院,2007
罗小军,黄丁发,刘国祥.基于永久散射体雷达差分干涉测量的城市地面沉降研究[J].测绘通报,2009,4:4~8
马锋,杨发俊,陈润桥,等.天津市地下水开采对地面沉降影响的多元回归分析[J].中国地质灾害与防治学报,2008,19(2):63~66
母景琴,姚国清.干涉雷达时间序列分析方法在地面沉降监测中的应用[J].国土资源遥感,2007,3:28~30
潘云,潘建刚,宫辉力,等.天津市区地下水开采与地面沉降关系研究[J].地球与环境,2004,32(2):36~39
任立生,黄立人,等.天津GPS沉降监测的十年试验结果[J].测绘科学, 2006,31(4):17~20
舒宁.雷达影像干涉测量原理[M].武汉:武汉大学出版社,2003
舒宁.微波遥感原理[M].武汉:武汉大学出版社,2003
孙占义.InSAR&DInSAR中相位解缠方法的研究,[硕士论文],武汉:武汉大学,2003
汤益先,张红,王超.基于永久散射体雷达干涉测量的苏州地区沉降研究[J].自然科学进展,2006,16(8):1015~1020
王超,张红,刘智.星载合成孔径雷达干涉测量〔M〕.北京:科学出版社,2002
王艳.利用相干目标分析方法的长时间地表形变研究.[博士论文],武汉:武汉大学,2006
夏耶,郭小方.地面沉降和山体滑坡的雷达图像差分监测分析.遥感科技论坛中国遥感应用协会2005年年会论文集
谢酬.基于永久散射体的青藏高原冻土形变研究.[博士论文],北京:中国科学院遥感应用研究所,2008
许才军.王华.INSAR相位解缠算法比较及误差分析[J].武汉大学学报.信息科学版,2004(1)
薛跃明,郭华东,等.基于D-InSAR技术的矿区地表形变检测研究[J].遥感应用,2008,33-36
闫大鹏.合成孔径雷达差分干涉测量在监测地表沉降应用中的方法研究.[硕士论文],四川:西南交通大学,2008
姚国清,母景琴.D-InSAR技术在地面沉降监测中的应用[J].地学前缘,2008,15(4):239~243
张景发等.PSINSAR技术在地壳长期缓慢形变监测中的应用[J].国际地震动态,2006(6):1193-1198
张微.基于D-InSAR的杭州地区地壳形变监测及机理研究.[博士论文],浙江:浙江大学,2008
赵超英.差分干涉雷达技术用于不连续形变的监测研究,[博士论文],陕西:长安大学,2009
赵景民,谌华,甘卫军,等.影响重轨INSAR地表微量形变监测精度的误差分析[J].交通运输工程与信息学报,2009,7(3):22~26
A. Hooper, R.Pedersen and F.Sigmundsson, Constraints on Magma Intrusion at Eyjafjallaj?kull and Katla Volcanoes in Iceland, From Time Series SAR Interferometry. The VOLUME Project, VOLcanoes: Understanding subsurface mass movement (eds. C.J.Bean et al.), 2009, 13-24.
Crosetto M., A.Arnaud, J.Dura,E.Biescas, and M.Agudo, Deformation monitoring using remotely sensed radar interferometric data, Proc., 11th FIG Symposium on Deformation Measurements, Satntorini, 2003.
Dehls, J.F.,M.Basilico and C.Colsesanti, Ground deformation monitoring in the Ranafjord area of Norway by means of the permanent scatters technique, IEEE Int. Geoscience and Remote Sensing Symposium, IGARSS 2002,2002:203-207.
Dennis C.Ghiglia and Mark D.Pritt. Phase Unwrapping Theory, algorithms, and software [M],1998.
Ferretti A, PratiC, RoccaF, Permanent Scatters in SAR Interferometry, IEEE Transactions Geoscience and Remote Sensing, 2001, 39(1):8~20.
Lyons,S. , and D.Sandwell, Fault creep along the southern San Andreas from interferometric synthetic aperture radar, permanent scatterers, and stacking, Journal of Geophysical Research, 2003,108(B1):2047~2070.
Werner, C., U.Wegmüller, T.Strozzi, and A.Wiesmann, Interferometric point target analysis for deformation mapping, Geoscience and Remote Sensing Symposium, IGARSS, 2003.
Zebker H A, Villasensor J, Decorrelation in Interferomatric radar echoes, IEEE Trans Geoscience Remote Sensing, 1995, 22:2517~2520.
陈强.基于永久散射雷达差分干涉探测区域地表形变的研究:[博士学位论文].四川:西南交通大学,2006
程璞等.INSAR相位解缠算法研究[J].大地测量与地球动力学,2007(6)
董克刚,周俊,等.天津市地面沉降的特征及危害[J].地质灾害与环境保护, 2007,18(1):67~70
董克刚,周俊,于强,等.天津市地面沉降的特征及其危害[J].地质灾害与环境保护,2007,18(1):67~70
范景辉,郭华东,等.基于相干目标的干涉图叠加方法监测天津地区地面沉降[J].遥感学报, 2008,12(1):111~118
范景辉.基于相干目标的DInSAR技术地表形变监测研究与应用:[博士论文].北京:中国科学院遥感应用研究所,2007
范景辉,李梅,等.基于PSInSAR方法和ASAR数据监测天津地面沉降的试验研究[J].国土资源遥感,2007,4:23~17
付进朋.InSAR相位解缠算法的研究:[硕士论文].四川:西南交通大学,2006
傅文学,田庆久.等,PS技术及其在地表形变监测中的应用现状与发展[J].地球科学进展,2006,11:119~1198
傅文学.低相关区滑坡移动的DInSAR监测方法研究-以树坪滑坡为例:[博士论文].南京:南京大学,2007
葛大庆,郭小方,等.基于序列差分干涉图的地表形变速率提取[J].国土资源遥感, 2007,1:24~31
葛大庆,王艳,等.基于相干点的多基线D-InSAR技术与地表形变监测[J].遥感学报,2007,11(4):574~560
葛大庆,王艳,等.利用短基线差分干涉图集监测地表形变场[J].大地测量与地球动力学,2008,28(2):61~66
郭良迁,薄万举,陈宇坤,等.天津地区的垂直形变与构造活动研究[J].大地测量与地球动力学,2009,29(5):1~5
李平湘,杨杰.雷达干涉测量原理与应用[M].北京:测绘出版社,2006
刘广.长时序DInSAR技术中的空间基线校正和图像配准方法研究.[博士论文],北京:中国科学院遥感应用研究所,2007
刘艳华.合成孔径雷达干涉(InSAR)测量技术测量地表变形的理论和算法研究,[博士后研究工作报告].北京:中国测绘科学院研究院,2007
罗小军,黄丁发,刘国祥.基于永久散射体雷达差分干涉测量的城市地面沉降研究[J].测绘通报,2009,4:4~8
马锋,杨发俊,陈润桥,等.天津市地下水开采对地面沉降影响的多元回归分析[J].中国地质灾害与防治学报,2008,19(2):63~66
母景琴,姚国清.干涉雷达时间序列分析方法在地面沉降监测中的应用[J].国土资源遥感,2007,3:28~30
潘云,潘建刚,宫辉力,等.天津市区地下水开采与地面沉降关系研究[J].地球与环境,2004,32(2):36~39
任立生,黄立人,等.天津GPS沉降监测的十年试验结果[J].测绘科学, 2006,31(4):17~20
舒宁.雷达影像干涉测量原理[M].武汉:武汉大学出版社,2003
舒宁.微波遥感原理[M].武汉:武汉大学出版社,2003
孙占义.InSAR&DInSAR中相位解缠方法的研究,[硕士论文],武汉:武汉大学,2003
汤益先,张红,王超.基于永久散射体雷达干涉测量的苏州地区沉降研究[J].自然科学进展,2006,16(8):1015~1020
王超,张红,刘智.星载合成孔径雷达干涉测量〔M〕.北京:科学出版社,2002
王艳.利用相干目标分析方法的长时间地表形变研究.[博士论文],武汉:武汉大学,2006
夏耶,郭小方.地面沉降和山体滑坡的雷达图像差分监测分析.遥感科技论坛中国遥感应用协会2005年年会论文集
谢酬.基于永久散射体的青藏高原冻土形变研究.[博士论文],北京:中国科学院遥感应用研究所,2008
许才军.王华.INSAR相位解缠算法比较及误差分析[J].武汉大学学报.信息科学版,2004(1)
薛跃明,郭华东,等.基于D-InSAR技术的矿区地表形变检测研究[J].遥感应用,2008,33-36
闫大鹏.合成孔径雷达差分干涉测量在监测地表沉降应用中的方法研究.[硕士论文],四川:西南交通大学,2008
姚国清,母景琴.D-InSAR技术在地面沉降监测中的应用[J].地学前缘,2008,15(4):239~243
张景发等.PSINSAR技术在地壳长期缓慢形变监测中的应用[J].国际地震动态,2006(6):1193-1198
张微.基于D-InSAR的杭州地区地壳形变监测及机理研究.[博士论文],浙江:浙江大学,2008
赵超英.差分干涉雷达技术用于不连续形变的监测研究,[博士论文],陕西:长安大学,2009
赵景民,谌华,甘卫军,等.影响重轨INSAR地表微量形变监测精度的误差分析[J].交通运输工程与信息学报,2009,7(3):22~26