Enhanced Landslide Monitoring in Complex Mountain Terrain Using Distributed Scatterer InSAR and Phase Optimization: A...

Area of Science:

• Geosciences
• Remote Sensing
• Geohazards

Background:

• Landslide monitoring in mountainous areas is vital for disaster prevention but challenged by rugged terrain, vegetation, and low coherence, limiting conventional InSAR methods.
• Effective deformation monitoring requires high-density measurements and reliable data, especially in complex geological settings prone to landslides.

Purpose of the Study:

• To develop a robust time-series Interferometric Synthetic Aperture Radar (InSAR) framework for enhanced landslide deformation detection and measurement density in low-coherence, complex mountainous terrain.
• To introduce and validate the Sequential Estimation and Total Power-Enhanced Expectation-Maximization Inversion (SETP-EMI) approach for overcoming coherence loss and improving distributed scatterer extraction.

Main Methods:

• Integration of dual-polarization Sentinel-1 SAR time series within a recursive estimation framework.
• Augmentation with polarimetric coherence optimization and dynamic assimilation of SAR data.
• Application of the SETP-EMI approach to Zhenxiong County, Yunnan Province, for landslide monitoring.

Main Results:

• The SETP-EMI method achieved a 94.1% landslide detection rate in the study area.
• Generated approximately 2.49 million measurement points, significantly outperforming Persistent Scatterer InSAR (PS-InSAR) and Small Baseline Subset InSAR (SBAS-InSAR).
• Validated accuracy with ground-based leveling data, showing a standard deviation of 5.21 mm/year, confirming high robustness.

Conclusions:

• The SETP-EMI method, within a Distributed Scatterer InSAR (DS-InSAR) framework, effectively overcomes coherence loss in vegetated mountainous regions.
• This approach substantially improves landslide monitoring density and accuracy, enhancing early-warning capabilities for geohazards in complex terrains.