Correlation of geomorphological features, geomechanical properties, and clustering of passive seismic recordings for translational rock block slide zoning

Rock block slides in soft rocks are often challenging to characterize due to their structural complexity and hard-to-detect geomorphological features. This is especially true for ancient landslides, where remnant rock blocks are largely obliterated by slopereshaping processes. In this context, passive seismic data can provide useful information that can be correlated with rock block geomechanical properties and geomorphological features , to assess unstable sectors and landslide zoning. However, the lack of clear morphological evidence often makes this correlation difficult and non-trivial. We therefore selected a recently reactivated rock block slide in northwestern Italy to extensively test passive seismic techniques in an environment where geomorphological characteristics are well defined. A dense grid of passive seismic stations was deployed, supported by an active seismic survey and a geological-geomorphological survey. Passive seismic data, analyzed using the HVSR (horizontal-to-vertical spectral ratio) method, revealed clear resonance contrasts between in-place and displaced rock mass. The landslide body shows internal heterogeneity, increased joint density, and reduced mechanical integrity, enhancing seismic impedance contrasts and causing high amplification and strong signal polarization in its most degraded areas. A clustering-based analysis approach through k-means was also employed to assess the spatial distribution of passive seismic signal behaviours. Resonance frequency, amplitude, and polarization direction were analyzed together with several spectrum curve shape parameters, adopted to enhance the clustering-based classification. This approach enabled the landslide area to be zoned into four sectors with different hazard scenarios. The resulting classification is well correlated with geomorphological observation, correctly highlighting areas of increased potential instability within the landslide body. Thanks to the correlations evidenced in the test site between passive seismic data evidence and geomorphological and geomechanical features, the proposed survey and analysis framework can be widely applied to rock block slide contexts supporting landslide hazard mapping and seismic microzonation activities.