面向内陆峡谷型水库的GACOS时序InSAR大气二次改正方法研究  

GACOS-based Secondary Atmospheric Correction Method for Time-series InSAR in Inland Canyon-type Reservoir

作　　者：刘浩锋 孟令奎[1] 罗志 李靖霖 张镇 李涵曼 李林宜[1] 张文[1] LIU Haofeng;MENG Lingkui;LUO Zhi;LI Jinglin;ZHANG Zhen;LI Hanman;LI Linyi;ZHANG Wen(School of Remote Sensing Information Engineering,Wuhan University,Wuhan 430079,China;Nanjing Hydraulic Research Institute,Nanjing 210029,China)

机构地区：[1]武汉大学遥感信息工程学院,湖北武汉430079 [2]南京水利科学研究院,江苏南京210029

出　　处：《华北水利水电大学学报(自然科学版)》2025年第2期13-22,共10页Journal of North China University of Water Resources and Electric Power:Natural Science Edition

基　　金：国家重点研发计划项目(2021YFB3900603)。

摘　　要：大气延迟是影响时序InSAR精度的主要误差源之一,常见的InSAR大气改正方法主要包括基于干涉图内部特性的自改正法和外部数据辅助改正法。为了进一步削弱大气延迟带来的影响,结合研究区域的地形特征,提出了一种面向内陆峡谷型水库的时序InSAR大气二次改正方法。该方法将回归模型和GACOS数据集结合实现了大气二次改正,发挥了二者在大气改正中的优势。以河口村水库为例,将大气二次改正结果与常规时序InSAR结果进行了对比,并采用地表形变实测数据验证。结果表明,相比常规方法,GACOS改正方法的精度提升5.576%,线性改正、指数改正及指数+GACOS改正方法的精度依次提升,其中线性+GACOS改正方法的精度提升22.538%,说明提出的大气二次改正方法能进一步提高时序InSAR形变结果的精度,为InSAR技术在内陆峡谷型水库地区的地表形变监测提供参考。Atmospheric delay remains a critical challenge in improving the accuracy of time-series interferometric synthetic aperture radar(TS-InSAR).Existing correction approaches primarily fall into two categories:self-correction based on interferogram characteristics and external data-assisted methods.To address the limitations of conventional methods in complex terrains,this study proposes a secondary atmospheric correction framework integrating regression models and the Generic Atmospheric Correction Online Service(GACOS)dataset,specifically tailored for inland canyon-type reservoirs.The proposed method capitalizes on the synergistic advantages of statistical regression and high-resolution atmospheric modeling.A case study at Hekoucun Reservoir demonstrates significant improvements:Compared to conventional TS-InSAR,the GACOS-only correction enhances accuracy by 5.576%,while the combined linear regression and GACOS approach achieves a 22.538%reduction in residual errors.Validation using in-situ deformation measurements confirms the method's effectiveness in mitigating stratified atmospheric artifacts induced by canyon topography.These findings provide a robust technical reference for high-precision deformation monitoring of reservoir slopes in geomorphologically constrained regions using advanced InSAR solutions.

关 键 词：时序InSAR 大气延迟 回归模型 GACOS 河口村水库 

分 类 号：P237[天文地球—摄影测量与遥感]