Slow slip detection with deep learning in multi-station raw geodetic time series validated against tremors in Cascadia

Nature Communications Earth & Environment
Reference: G. Costantino, S. Giffard-Roisin, M. Radiguet, M. Dalla Mura, D. Marsan and A. Socquet (2023). Slow slip detection with deep learning in multi-station raw geodetic time series validated against tremors in Cascadia, Nature Communications Earth & Environment, submitted.

Slow slip events (SSEs) originate from a slow slippage on faults that lasts from a few days to years. A systematic and complete mapping of SSEs is key to characterizing the slip spectrum and understanding its link with coeval seismological signals. Yet, SSE catalogues are sparse and usually remain limited to the largest events, because the deformation transients are often concealed in the noise of the geodetic data. Here we present the first multi-station deep learning SSE detector applied blindly to multiple raw geodetic time series. Its power lies in an ultra-realistic synthetic training set, and in the combination of convolutional and attention-based neural networks. Applied to real data in Cascadia over the period 2007-2022, it detects 78 SSEs, that compare well to existing independent benchmarks: 87.5% of previously catalogued SSEs are retrieved, each detection falling within a peak of tremor activity. Our method also provides useful proxies on the SSE duration and may help illuminate relationships between tremor chatter and the nucleation of the slow rupture. We find an average day-long time lag between the slow deformation and the tremor chatter both at a global- and local-temporal scale, suggesting that slow slip may drive the rupture of nearby small asperities.

Overview of the performance of SSEdetector on real raw GNSS time series. The blue curves show the probability of detecting a slow slip event (output by SSEdetector) in 60-day sliding windows centered on a given date. Grey bars represent the number of tremors per day, smoothed (gaussian smoothing, σ = 2 days) in the grey curve. Red horizontal segments represent the known events catalogued by Michel et al., 2019. The (a) panel shows the global performance of SSEdetector over 2007-2022. The red arrow indicates the time window analyzed by Michel et al., while the green arrows describe the two periods from which the synthetic training samples have been derived. The grey rectangle indicates the period which was not covered by the PNSN catalog. In this period, data from Ide, 2012 has been used. The (b), (c) and (d) panels show zooms on 2016-2017, 2019-2021 and 2017 (January to July), respectively.

Updated on 25 April 2023