Characterizing the plate interface with microseismicity: Central Chile

Central Chile may not have experienced a major earthquake that ruptured the entire megathrust since 1730. At the same time, this stretch of the Chilean margin hosts major population centers which are endangered by such a future earthquake and the following tsunami. While geodetic techniques are most commonly used to constrain the state of the megathrust and thus the possible extent of future large earthquakes, the analysis of background microseismicity can also deliver valuable information. In previous studies, it was suggested that highly coupled regions on the megathrust exhibit very low seismicity rates and low b-values, but are surrounded by half-ellipses or ‘rims’ of enhanced seismicity. Regions of lower coupling generally exhibit higher seismicity rates and higher b-values, and in some cases show seismicity clusters with swarm-like behavior.

To extend this type of analysis in Central Chile, we applied state-of-the-art automatic approaches, i.e. a deep-learning based picker (EQTransformer) and a novel associator (PyOcto) to available continuous seismic data from Central Chile, covering the years 2014-2023. We thus retrieve a seismicity catalog that comprises >350,000 events, to which we apply relative relocation via hypoDD to retrieve high-resolution sharpened features. We classify the events into different populations corresponding to the main seismogenic regions (plate interface, downgoing slab, upper plate).

In this contribution, we mainly analyze the plate interface seismicity in the catalog in terms of statistical properties and temporal evolution, evaluate the spatiotemporal detection capacity of our approach by retrieving completeness magnitudes, and correlate the retrieved features to existing evidence from other disciplines.