Abstract: Industrial parks, as the core areas of urban economic development, are particularly important for monitoring surface deformation. Currently, there is limited research on the deformation mechanism of industrial parks, and traditional monitoring methods are costly and inefficient. Therefore, this study proposes the use of time-series interferometric synthetic aperture radar (InSAR) technology to construct a comprehensive monitoring model for industrial parks, which improves monitoring efficiency while reducing costs. Taking the Yinxi industrial park in Baiyin district as an example, based on 34 scenes of Sentinel-1A data from June 2018 to April 2021, the deformation information of the park’s surface was obtained using the StaMPS-PS (stanford method for persistent scatterers-permanent scatterers) and SBAS-InSAR (small baseline subsets-interferometry synthetic aperture radar) techniques. The deformation information obtained from the two techniques was cross-validated from a spatio-temporal distribution perspective. The results show that the deformation features obtained by both techniques correspond to the deformation locations in field survey photos. Additionally, using 585 identical latitude and longitude points for accuracy verification, a good correlation between the two techniques is found, with a coefficient of determination (R²) of 0.82 and a root mean square error (RMSE) of 2.20 mm/a. The deformation rates are highly consistent as well. Since the StaMPS-PS technique identifies 47% more deformation points than the SBAS-InSAR technique, it is more applicable for the industrial park. Finally, the geological conditions and factors inducing surface deformation in the industrial park are analyzed and discussed, providing reference for better understanding the deformation mechanism and early warning of disasters in the park.

Key words: StaMPS-PS, SBAS-InSAR, ground deformation, WOA-BP neural network, Yinxi industrial park