Hierarchical Density-Based Cluster Analysis for Fault Segment Detection: A Case Study of the 2009 M w 6.1 L’ Aquila Earthquake (Central Italy)

Identifying seismogenic faults, their geometries, and associated slip rates is essential for assessing seismic hazard estimates. During earthquake sequences, complex fault systems are often activated, involving the rupture of multiple segments. Identifying these fault segments remains a challenging task. This study proposes a quantitative, statistically based workflow using a cluster analysis to identify the number and dimensions of fault segments based on earthquake hypocenter distributions, an approach particularly suitable for aftershock sequences. A Hierarchical Density-Based Spatial Clustering of Applications with Noise (HDBSCAN) algorithm is employed for its ability to analyse high-dimensional data and identify clusters of arbitrary features. We applied this methodology to the 2009 L’ Aquila seismic sequence (Mainshock = Mw 6.1), clustering the seismicity to identify the geometric characteristics of the individual activated fault segments. The results provide valuable insights into the seismogenic potential and rupture mechanics of complex fault systems, contributing to a more accurate assessment of seismic risk. The methodology is highly transferable, providing a valuable framework for replicating analyses across various seismic sequences and seismogenic fault systems.