We are happy to share SCMLPick, a new open source SeisComP module developed by the TexNet team to integrate machine learning phase picking directly into real time SeisComP workflows.
The current version of SCMLPick uses EQCCT, the deep learning phase picker developed at TexNet, for P and S phase picking. The module is currently running in production at TexNet and provides an operational framework for integrating machine learning based phase picking within the existing SeisComP environment. The framework is designed to be extended in future versions to support additional machine learning picking methods.
The project is publicly available on GitHub and is now included in the official SeisComP Community Add Ons:
https://www.seiscomp.de/addons/
GitHub repository:
https://github.com/ut-beg-texnet/SCMLPick
The development and its application to real time seismic monitoring are described in our recently published paper in Seismological Research Letters (SRL):
SCMLPick: A SeisComP Module Implementing Machine Learning Phase Picking for Real Time Seismic Monitoring
https://doi.org/10.1785/0220250368
We hope SCMLPick can be useful to other SeisComP users and seismic networks interested in incorporating machine learning picking into their operational workflows.
From the TexNet team, we will be happy to provide support to users interested in installing, testing, or using SCMLPick. We also welcome feedback and collaboration from anyone interested in contributing to the project, implementing support for additional machine learning picking methods or other new features, or helping identify and resolve bugs.
Please feel free to contact us through the GitHub repository or this forum for questions, feedback, bug reports, or contributions. You can also contact me directly at camilo.munoz@beg.utexas.edu for additional information or potential collaborations.
We look forward to hearing about your experience with SCMLPick and to continuing its development together with the SeisComP community.