Using Improved PS Detection Method To Investigate Land Subsidence In Tianjin

Geographical disasters, man-made or natural, like land subsidence, are influencing people's living environment and development of human society and economy. With the speeding up of urbanization, large quantities of population flooded into cities, where land subsidence became more serous. China detected land subsidence in 1920s, and now this problem has developed into a region-distributed phenomenon. Tianjin City has become the center of subsidence in North China due to the over-drained groundwater.Compared with traditional leveling and GPS, InSAR technique has more advantages like better coverage and lower cost, and it can work for all-weather condition to monitor land subsidence. But traditional D-InSAR is liable to be influenced by decorrelation and atmosphere delay influence in mapping land subsidence, thus unable to be applied to slow deformation measurement of surface for a long time. On the other hand, PSInSAR technique is able to extract long-time stable permanent scatterers with good-qualitied phases to acquire time series of surface deformation information, which has overcomed the disadvantage of traditional D-InSAR. But PSInSAR technique needs to be perfected in the choosing of PS points, net-shaped construction method and nonlinear deformation estimation and so on. Meanwhile, Interferomtric technology of PS based on high resolution TerraSAR images is a new subject too.This paper aims to acquire the value of land subsidence in Tianjian area by analyzing high resolution images, with some improvement to PSInSAR technique. The job includes the following parts:1.This paper is to analyze D-InSAR technique, its data-processing procedure and its application in practice in details based on a large amount of reviewed references. Then the paper is to introduce the representative ways based on high coherence point target analysis to monitor long-time series of surface deformation.2.This paper is based on the experimental research on TerraSAR-X high resolution data, which explores regular PS detection method. The results show that high resolution data can dramatically increase the number of PS points, and in practice the choice of thresholds should be adjusted. In order to improve the reliability of PS points, this paper has improved the PS detection method on the basis of initial experiment.3.As to the isolated PS points caused by restricted Delaunay triangle side, this paper is to adopt the method of closest connection method(CCM) which connects these isolated points with their closest PS so as to achieve the PS network closure. 4.After analyse the disadvantage of traditional unweighted Least Squares, phase gradient minimum criterion being used to set the weight, the weighted least squares phase unwrapping methods were studied through the simulation. Simulation experiment based on different levels of noise has been a reliable test conclusion.5.This paper is to extract the land subsidence data of Tianjian by analyzing 17 TerraSAR-X images scanned from March 2009 to Nov.2010 and adopting improved PSInSAR technique. The data are compared with the ground-level data, which indicates that this method works and the obtained subsidence data is reliable.