基于GPS/PS-DInSAR综合技术的城市大气水汽时空分布特征研究

Urban atmospheric water vapor temporal-spatial distribution characteristics is the important direction of research on urban climate, which have serious impacts on society and economy, and human health. The low spatial resolution and the sensitive to cloud and rain are disadvantages of conventional atmospheric water vapor inversion methods, which limit its applications to analyze the atmospheric water vapor temporal-spatial distribution characteristics of cities with complex underlying surfaces. So it is important to exploit a high temporal-spatial resolution, all-weather atmospheric water vapor inversion method. This project focuses on the study on the method of integrating global positioning system(GPS) and permanent scatterer differential synthetic aperture radar interferometry(PS-DInSAR) to derive high temporal-spatial resolution atmospheric water vapor, including the studies on the determined and refined methods of the surface deformation initial model, the conversion method of PS relative atmospheric water vapor to PS absolute atmospheric water vapor and the fusion method of atmospheric water vapor derived respectively by GPS and PS-DInSAR. On the basis of the above studies, GIS and mathematical statistics are used to analyze the temporal-spatial distribution characteristics of the high temporal-spatial resolution atmospheric water vapor and to reveal the relationship between temporal-spatial distribution characteristics of atmospheric water vapor and underlying surfaces. The studies of this project not only provide a new mean for high temporal-spatial resolution atmospheric water vapor inversion but also discuss the urban atmospheric water vapor temporal-spatial distribution characteristics, which provide a theoretical reference for urban planning.

城市大气水汽时空分布特征是城市气候效应研究的重要方向，对社会经济可持续发展和人体健康有重要影响。常规大气水汽反演方法空间分辨率低、受云雨等天气影响大，不能满足城市复杂下垫面条件下水汽时空分布特征研究的需要。因此，研究一种高时空分辨率、能全天候反演大气水汽的方法具有重要意义。本项目以GPS和SAR数据为主要数据源，对综合GPS和PS-DInSAR技术进行高时空分辨率大气水汽反演方法进行研究，重点研究地表形变初始模型确定和精化方法、PS点相对水汽到绝对水汽转化方法、GPS水汽和PS-DInSAR水汽融合方法。在上述研究基础上，利用GIS和数理统计方法对综合GPS和PS-DInSAR技术反演的高时空分辨率大气水汽时空分布特征进行分析，揭示其与城市下垫面类型等因素的关系。本项目的研究，不仅为高时空分辨率大气水汽反演提供新手段，而且，从理论上探讨城市水汽的时空分布特征，为城市规划提供理论参考。

研究一种高时空分辨率、能全天候反演大气水汽的方法对城市复杂下垫面条件下水汽时空分布特征研究具有重要意义。本项目对综合GPS和PS-DInSAR技术进行高时空分辨率大气水汽反演方法进行了研究，主要研究内容和成果如下：（1）针对PS-DInSAR参数解算过程中遇到的病态线性模型和非线性模型参数求解问题，研究了改进的雅可比迭代法和猫群算法，指出其优势；（2）提出了一种GPS辅助InSAR相位解缠综合算法，提高了InSAR解缠精度和范围；（3）提出了顾及纹理信息的NSCT域自适应阈值去噪和基于NSCT与模糊理论的图像增强方法，为更好的融合光学与SAR图像进行地物分类奠定基础；（4）获取了时序水汽，并将其与地物类型进行了符合分析。由于反演的水汽精度不高等原因没能就水汽分布的时空特点给出具体的结论，但分析了可能存在的问题，为后续研究指明了方向。