Monitoring vertical ground deformation in the North China Plain using the multitrack PSInSAR technique

In this paper, we present a method to reveal vertical deformation characteristics through reference transformation, data splicing and data fusion in multiple regions, which can overcome the shortage of traditional single-track PSInSAR techniques. In order to solve the problem that the ground deformation is not comparable, which is derived from various methods with reference difference in multi-tracks, vertical baseline difference in multi-tracks and reference PS difference in phase unwrapping, we calculate accurate transformation parameters and realize seamless connection of adjacent two frames through corresponding PS point analysis of the overlapped region and same point identification using the shortest distance algorithm based on proper weight. On the basis of algorithm research, we deduce the model to extract deformation information through data connection and reference transformation in multi-track PSInSAR technology, and then process the data collected in the North China Plain(115.32 degrees E-118.79 degrees E, 36.81 degrees N-40.58 degrees N) which include 3 different orbits and 12 frames, and obtain the vertical deformation of the study area for the period 2006-2010. The results indicate that the research area is in a sinking status, forming several subsidence centers including Beijing, Langfang, Tianjin, Cangzhou and Botou-Dezhou with average subsidence rates up to -34.7 mm/a, -26.3 mm/a, -64.2 mm/a, -34.6 mm/a and -37.7 mm/a, respectively. The spatial distribution of subsidence shows that groundwater extraction of industry and daily life are the main reasons, meanwhile agricultural irrigation and gas exploitation are the other primary causes. The major subsidence zones, trending NNE, coincide with the orientations of the fault zones, implying the subsidence distribution is controlled by the faults. These research results are verified by leveling measurements with errors about 4.72 mm, which shows that the data processing method in this paper can meet the requirement of large spatial scale deformation monitoring.