Probablistic mapping of seismic regime instability hotspots for risk assessment and early warning

This study explores global patterns of seismic regime transitions using a probabilistic and topological framework, emphasizing fault zone differences in earthquake dynamics. To characterize these shifts, we apply Topological Data Analysis (TDA) to geocoded earthquake time series, identifying robust regime clusters via persistent homology. Markov transition matrices are then used to quantify the likelihood of fault zones moving between regimes, with off-diagonal probabilities serving as indicators of seismic instability.