← Back to Gallery

An Updated Review of Ground Motion Flatfiles in the Chilean Subduction Zone

The Chilean-Peruvian subduction zone, one of the most seismically active regions globally, requires extensive ground motion datasets to advance seismic hazard assessment and risk mitigation strategies. This work expands upon Bastias and Montalva's 2016 (BM16) flatfile by introducing a significantly larger database with 43,585 records compared to the previous 3,572 (a 12x increase), 6,650events (a 14x increase), and 484 sites (2.7x increase). Additionally, the dataset includes data from a vertical array. The database is organized into three flatfiles for seismic events, site stations, and strong ground motion records, each containing detailed metadata to optimize usability. The dataset incorporates major megathrust earthquakes from the Perú-Chile subduction. The seismic source is defined through the moment tensor and the relative geographic positioning of each event within the subduction slab context. Site characterization employs measured shear wave velocity (Vs) profiles, then we use a proxy-based methodology to estimate average Vs at 30 m depth. In addition to the prior proxies (i.e., topographic and HVSR) used in BM16, this study is complemented with the P-wave estimation, which demonstrates superior predictive performance. Also, kappa estimation is provided for 47 stations.

Ground motion records are processed using a Signal-to-Noise Ratio (SNR > 3) scheme, defining the low-cut frequency at the passband filter and avoiding artifacts in the processed time series. This allows for balancing the noise of the traces on the whole dataset. Theoretical P- and S-wave arrivals are identified through an automated workflow using a new hypocentral distance-based formula and picking techniques. The quality control of traces is based on the residuals within the usable spectral periods and in the definition of the bandwidth (in frequency domain) with SNR higher than 3. The dataset introduces new orientation horizontal spectral measures (RotDXX) and azimuth data for RotD100, providing insights on the path orientation.

Session: Modern Waveform Processing and Engineering Datasets - Accessibility, Quality Control, and Metadata [Poster]

Type: Poster

Room: Exhibit Hall

Date: 4/17/2025

Presentation Time: 08:00 AM (local time)

Presenting Author: Nicolas Bastias

Student Presenter: Yes

Invited Presentation: 

Poster Number: 49

Additional Authors