neptuneprojects / riscluster

Deep clustering for seismic signals (icequakes and earthquakes)

RISCluster

RISCluster is a package that implements deep embedded clustering (DEC) and Gaussian mixture model (GMM) clustering of seismic data recorded on the Ross Ice Shelf, Antarctica from 2014-2017. This package is an accompaniment to the paper published in the Journal of Geophysical Research: Solid Earth.

Figure 1. 34-station passive seismic array deployed on the Ross Ice Shelf, Antarctica from 2014-2017.

---

Installation

Pre-requisites: Anaconda or Miniconda

The following steps will set up a Conda environment and install RISProcess, and have been tested on MacOS 11.1 and Red Hat Enterprise Linux 7.9. If you have a CUDA-enabled machine (i.e., not MacOS), you can install the CUDA version of RISCluster. Unfortunately, PyTorch GPU & RAPIDS libraries are not implemented for MacOS, so you will need to install the CPU version if you use a Mac, or if your Linux machine is not CUDA-capable. This package has not been tested on Windows.

CUDA-enabled RISCluster (Linux)

1. Open a terminal and navigate to the directory you would like to download the RISCluster_CUDA.yml environment file.
2. Save RISCluster_CUDA.yml to your computer by running the following:
   wget --no-check-certificate --content-disposition https://raw.githubusercontent.com/NeptuneProjects/RISCluster/master/RISCluster_CUDA.yml
3. In terminal, run: conda env create -f RISCluster_CUDA.yml
4. Once the environment is set up and the package is installed, activate your environment by running conda activate RISCluster_CUDA in terminal.

CPU-based RISCluster (Mac or Linux)

1. Open a terminal and navigate to the directory you would like to download the RISCluster_CPU.yml environment file.
2. Save RISCluster_CPU.yml to your computer by running the following:
   a. Mac:
   curl -LJO https://raw.githubusercontent.com/NeptuneProjects/RISCluster/master/RISCluster_CPU.yml
   b. Linux:
   wget --no-check-certificate --content-disposition https://raw.githubusercontent.com/NeptuneProjects/RISCluster/master/RISCluster_CPU.yml
3. In terminal, run: conda env create -f RISCluster_CPU.yml
4. Once the environment is set up and the package is installed, activate your environment by running conda activate RISCluster_CPU in terminal.

---

Usage

Please refer to the RISWorkflow repository for detailed instructions on how to implement the workflow.

---

References

William F. Jenkins II, Peter Gerstoft, Michael J. Bianco, Peter D. Bromirski; Unsupervised Deep Clustering of Seismic Data: Monitoring the Ross Ice Shelf, Antarctica. Journal of Geophysical Research: Solid Earth, 30 August 2021; doi: https://doi.org/10.1029/2021JB021716

Dylan Snover, Christopher W. Johnson, Michael J. Bianco, Peter Gerstoft; Deep Clustering to Identify Sources of Urban Seismic Noise in Long Beach, California. Seismological Research Letters 2020; doi: https://doi.org/10.1785/0220200164

Junyuan Xie, Ross Girshick, Ali Farhadi; Unsupervised Deep Embedding for Clustering Analysis. Proceedings of the 33rd International Conference on Machine Learning, New York, NY, 2016; https://arxiv.org/abs/1511.06335v2

---

Author

Project assembled by William Jenkins
wjenkins [@] ucsd [dot] edu
Scripps Institution of Oceanography
University of California San Diego
La Jolla, California, USA