Spatiotemporal deformation trend assessment based on ICOPS and ICA methods using multitemporal InSAR data in...

Area of Science:

• Geosciences and Remote Sensing
• Geotechnical Engineering
• Environmental Monitoring

Background:

• Industrial expansion necessitates robust infrastructure and monitoring.
• Remote sensing, specifically Synthetic Aperture Radar (SAR), is crucial for analyzing surface deformation in urban and industrial zones.
• Previous studies highlight the need for advanced techniques to interpret complex deformation patterns.

Purpose of the Study:

• To assess surface deformation in the Bugok industrial area, South Korea, using advanced interferometric techniques.
• To evaluate the effectiveness of the improved combined scatterer interferometry with optimized point scatterer (ICOPS) methodology.
• To identify contributing factors to observed deformation and their implications for infrastructure monitoring and risk management.

Main Methods:

• Implementation of the improved combined scatterer interferometry with optimized point scatterer (ICOPS) methodology.
• Utilization of Sentinel-1 SAR data (ascending and descending tracks) and Cosmo SkyMed (CSK) data from 2014 to 2022.
• Application of machine learning algorithms for time-series refinement and independent component analysis (ICA) for deformation data decomposition.

Main Results:

• The ICOPS method successfully visualized linear and nonlinear spatiotemporal deformation.
• Time-series analysis revealed significant land subsidence ranging from 0.76 to 2.83 cm/year, with a maximum cumulative deformation of approximately 20 cm.
• Deformation data exhibited seasonal, linear-seasonal, and exponential trends, linked to industrial structures and intensive groundwater extraction.

Conclusions:

• The study confirms substantial subsidence in the Bugok industrial area, driven by anthropogenic factors.
• The ICOPS methodology, enhanced by machine learning, proves reliable for analyzing complex deformation patterns.
• InSAR technology offers significant potential for post-construction infrastructure monitoring and risk management in industrial regions.