Inference of creep landslide slip surface by InSAR technology and improved particle swarm optimization

Determination of a slip surface is pivotal for understanding failure mechanisms and implementing effective mitigation strategies. Traditional approaches, including borehole investigations, numerical simulations, and geophysical surveys, often entail substantial costs and time and may not be feasible for remote or difficult-to-access slopes. This work suggests a novel methodology that integrates Interferometric Synthetic Aperture Radar (InSAR) with an Improved Particle Swarm Optimization (IPSO) algorithm to infer slip surface based on surface deformation observations and the principle of mass conservation. The mega and high-locality Xishancun landslide is selected as a case to calculate the slip surface by the InSAR-IPSO method. The inferred basal slip surface depth and geometry align exceptionally closely with borehole-derived data, demonstrating a correlation coefficient of 0.998 within a tight error margin of ± 12.717 m, validating the method’s high accuracy. Furthermore, the IPSO’s performance, in terms of finding the optimal solution, significantly outperforms traditional convex optimization (CVX), reducing computation time from 1015 to merely 7 s. This innovative InSAR-IPSO approach thus offers a combination of precision, efficiency, and accessibility, especially valuable for investigating landslides in challenging terrains.