Validation of the Epidemic-Type Aftershock Sequence (ETAS) Models for Simulation-Based Seismic Hazard Assessments

Abstract Moderate-to-large-magnitude earthquakes induce considerable short-to-medium-term increases in seismic hazard, due to the subsequent occurrence of aftershocks. Most studies on aftershock models focus their performance as part Operational Earthquake Forecasting (i.e., real- or near-real-time predictions ongoing sequences). Aftershock model validation a longer timescale is scarcely dealt with literature, however. This study addresses this research gap by validating medium-term months years) an advanced formulation epidemic-type sequence (ETAS) for potential future implementation simulation-based probabilistic hazard analyses (PSHAs). The aim determine whether considered ETAS can forecast adequate numbers aftershocks, and consistent spatial magnitude–frequency distributions, three years after given moderate-to-large-magnitude mainshock. Two different calibration procedures (region wide averaged) are used investigate ability resulting describe common characteristics sequences. region-wide methodology based entire catalog large geographic area conventional approach calibrating models. second instead sequence-specific data. procedure employs only out-of-sample retrospective) testing, its evaluation metrics catalog-based scores recently proposed Collaboratory Study Predictability framework. It demonstrated using New Zealand (1990–2020), including seven sequences period 1990–2020, two case studies. Resulting test indicate that not suitable building use PSHA. Sequence-averaged generally more acceptable, terms both aftershocks distributions. findings be guide implementations formulation, possibly combination mainshock