Data-driven Earthquake Detection
Qseek is a an automatic, data-driven earthquake detection and localisation tool designed for large seismic data sets. It combines neural network phase annotations with a stacking-and-migration and an adaptive octree localisation approach.
Key features are:
- Earthquake phase detection using machine-learning model from SeisBench, pre-trained on different data sets:
- Travel time calculation:
- 1D Layered velocity model (Pyrocko Cake and Fast Marching)
- 3D fast-marching velocity model (NonLinLoc compatible)
- Constant velocity
- Earthquake magnitudes and other features:
- Local magnitudes (ML) with different attenuation models
- Moment Magnitudes (MW) based on modelled attenuation curves (Dahm et al., 2024)
- Ground motion attributes (e.g. PGA, PGV, ...)
- Station Corrections
- SST: station specific corrections
- SSST: source specific station corrections
Qseek is built on top of Pyrocko.
Online documentation is available at https://pyrocko.github.io/qseek/.
From PyPi.
pip install qseekInstallation from GitHub.
pip install git+https://github.com/pyrocko/qseekPrint the default config with
qseek configEdit the my-project.json
Start the earthquake detection with
qseek search search.jsonThe simplest and recommended way of installing from source:
Local development through pip.
cd qseek
uv pip install -e .The project utilizes prek for clean commits, install the hooks via:
prek installPlease cite Qseek as:
Isken, M., Niemz, P., Münchmeyer, J., Büyükakpınar, P., Heimann, S., Cesca, S., Vasyura-Bathke, H., & Dahm, T. (2025). Qseek: A data-driven Framework for Automated Earthquake Detection, Localization and Characterization. Seismica, 4(1). https://doi.org/10.26443/seismica.v4i1.1283
Contribution and merge requests by the community are welcome!
Qseek was written by Marius Paul Isken and is licensed under the GNU GENERAL PUBLIC LICENSE v3.