青藏高原交通廊道沿线Sentinel-1影像几何畸变与升降轨适宜性定量分析

Quantitative Analysis of Sentinel-1 Imagery Geometric Distortion and Suitability Along Qinghai-Tibet Plateau Traffic Corridor

  • 摘要: 青藏高原交通廊道作为中国意义非凡的重大工程,也是世界铁路史上地形最复杂、地质安全问题最严峻的铁路。合成孔径雷达干涉测量技术(interferometric synthetic aperture radar,InSAR)凭借其广覆盖、高精度、高效率等特点已成为地质灾害隐患早期识别与监测的重要手段,但是其在青藏高原地形起伏剧烈的高山峡谷区域会面临较为严重的几何畸变问题。以覆盖面最广、获取时间最密的全球完全开放获取的哨兵1号卫星(Sentinel-1)数据为例,将青藏高原交通廊道沿线作为研究区域开展卫星雷达影像几何畸变与适宜性分析。综合考虑合成孔径雷达(synthetic aperture radar,SAR)成像几何、Sentinel-1影像斜距向入射角差异及被动几何畸变3个方面,提出了一套基于局部入射角的Sentinel-1影像几何畸变区域精准定量识别方法,并对全线几何畸变进行了定量计算与升降轨适宜性分析。结果显示,青藏高原交通廊道所在极陡高山峡谷区单轨道几何畸变区域达到31%~35%,升降轨联合观测可以有效将几何畸变区域减小至1.5%,同时约有35%区域适宜升降轨同时观测进行联合分析。

     

    Abstract:
    Objectives The Qinghai-Tibet Plateau traffic corridor is not only a state key project in China, but also the most difficult super project in railway construction history due to the most complex terrain and the active geological structure. Interferometric synthetic aperture radar (InSAR) has proved to be a critical tool for detecting and monitoring geohazards because of its wide coverage, high precision, and high efficiency. Howe-ver, due to its side-looking geometry, InSAR faces serious geometric distortions in mountainous areas, particularly in extremely steep high mountain and gorge areas.
    Methods Utilizing the full open-access Sentinel-1 data, which has the wide coverage and a short revisiting time, we carry out satellite radar geometric distortion and feasibility analysi. Taking into account synthetic aperture radar(SAR) imaging geometry, the variations of the incidence angles across the whole SAR image, and the passive geometric distortions, we present an integrated approach to accurately determine the areas with SAR geometric distortions.
    Results Our results for Sentinel-1 data show that the areas with geometric distortions for observations with a single-orbit (i.e. ascending or descending) reach 31%—35%, whilst the geometric distortion areas reduce to 1.5% for observations with both ascending and descending orbits; furthermore, about 35% of the whole area is suitable for joint analyses with both ascending and descending observations.
    Conclusions The quantitative result also reveals the feasibility to use the Sentinel-1 data to detect and monitor geohazards from the perspective of geometric distortions. It can not only improve the overall understanding of the applicability of Sentinel-1 based InSAR and the InSAR results interpretation in this area, but also provide a guideline for efficiently selecting other satellite SAR datasets for detecting and monitoring geological disasters.

     

/

返回文章
返回