Fig. 6: Temporal smoothing analysis on a synthetic model of the March 2017 slow slip event.
From: Multi-station deep learning on geodetic time series detects slow slip events in Cascadia
a Model of the static displacement associated with the March 2017 slow slip event by Itoh et al.32 (black arrows). The red triangles represent the 135 stations used in this study and the points show the tremor location from March 1 to April 30, 2017, color-coded by date of occurrence. b Matrix showing the temporal evolution of Itoh et al.’s model. Each row represents the detrended E-W GNSS position time series for a given station, color-coded by the position value. c Detrended E-W synthetic time series (sum of the Itoh et al.’s model and a realization of artificial noise output by SSEgenerator, color-coded by position value. d Moment rate function associated with the slip evolution of the model by Itoh et al. e The blue curve represents the daily probability output by the SSEdetector from the synthetic time series shown in the c panel. The red curve shows the probability curve associated with the prediction of SSEdetector on the March 2017 slow slip event on real GNSS data (see also Fig. 3d).
