Numerical Modelling of Late Pleistocene to Holocene Earthquake‐Induced Progressive Rock Slope Damage in the Paraglacial Serpentine Valley, Prince William Sound, Alaska

Résumé fourni par la source

ABSTRACT Landslides in deglaciating fjords pose a potential tsunami threat to nearby communities; however, processes contributing to long‐term progressive rock damage and landslide conditioning remain poorly constrained in paraglacial settings. Here, we analyse the role of earthquake‐induced rock mass damage over late Pleistocene to Holocene time scales, as a conditioning factor for modern landslides, using distinct element numerical modelling to assess spatial and temporal patterns of fracture propagation influenced by varying glacier thickness. Conceptualised numerical models were parameterised by in situ rock mass, glacial and topographic conditions in Serpentine valley, located in Prince William Sound, Alaska, where several large landslides are actively developing along the western valley wall. Results show that although rigid glacier buttressing reduces co‐seismic rock mass damage, it does not suppress it completely, and damage occurs both above and below the glacier surface elevation. We demonstrate that topography, and especially steep slopes with topographic convexity, as well as preexisting damage of joint networks and faults inherited from tectonic and exhumation induced stresses, exert primary control on the location of new co‐seismic damage. Simulations representing a simplified deglaciation sequence over the past 25 ka, with a series of 10 evenly spaced earthquakes, generated rock mass damage patterns that qualitatively match in situ landslide structural and kinematic observations at one instability in Serpentine valley. Our conceptual study helps clarify the role of repeated seismicity over glacial timescales as a long‐term conditioning process for paraglacial rock slope failure, highlighting spatial and temporal patterns of progressive damage accumulation, with outcomes relevant for modern landslide hazard assessment.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.