Monitoring of urban subway lines subsidence Using satellite radar interferometry method (Study area: Part of Tehran Metro Line 7)

Nowadays, satellite radar interferometry method plays an important role in calculating natural and artificial displacements, including displacements caused by floods, earthquakes, landslides, drilling of unauthorized wells, drilling of urban tunnels, etc. Metro tunneling is very important due to the importance of preventing irreparable financial and human losses. In Tehran, especially due to the drilling of subway tunnels and lack of attention to underground aqueduct routes and groundwater aquifers, we are witnessing large-scale land subsidence. The use of Multi-Temporal InSAR in determining the interferometry model is very well understood today and it is possible to use these methods to calculate subsidence with very high accuracy in different areas. Of course, these methods also have their own limitations, such as access to appropriate images in the desired time period, changes in coverage of the desired area over time.   In this paper, using satellite images of the Sentinel-1 low-pass in a period of six months (April to October 2019), the subsidence of the Tehran Metro Line 7 tunnel (between Hafdeh Shahrivar St. and Ghiam Square) using the method Radar interferometry is calculated and then compared with field sedimentation performed using sedimentation and precision mapping pins that have a very high accuracy (in the tenth of a millimeter). The number of images used in this period was 22 images, of which seven images with shorter base line length and higher degree of correlation were selected and the interferometry measurement operation was performed in pairs. In the coregistration operation, first two images are precisely aligned on each other and the phase difference between them is obtained by multiplying the conjunction of the first image in the second image which This obtained phase difference is called the interferometric model and at this stage the flat ground phase is eliminated. After obtaining the interferometric model, Deburst operation is performed to integrate the interferometric model and to reduce the volume of calculations, the desired spatial range is specified on the model And the topography phase is eliminated using the strm 3sec DEM. Then, for better performance and phase unwrapping operations, Goldstein filter and SNAPHU package are used, respectively. SNAPHU implements statistical cost and network-flow algorithms to unwrapp phase. The amount of displacement is then calculated from the unwrapped phase. Finally, using these images, the RMSE values ​​for images with VH and VV polarization compared to field data were 2.19 and 20.79 mm, respectively. We will see that the use of VH-polarized images performs better in determining subsidence in urban areas.