Ensemble Deep Learning to further mine Cascadia offshore data

This project uses an ensemble deep learning algorithm, ELEP (Yuan et al, 2023) to detect and pick P and S waves from continuous data in coastal and offshore Cascadia using the 2011-2015 Cascadia Initiative Experiment Ocean Bottom Seismometers.

The workflow also includes association, location, and relocation using : PyOcto (Munchmeyer 2024), HypoInverse (Klein 2002), and HypoDD (Waldhauser and Ellsworth 2000) for relocation.

The repository also provides scripts to compare the new catalog with established catalog in the region (USGS ComCat and Morton et al, 2023).

Project "CoolTeam" at UW:

• Hiroto Bito: hbito@uw.edu
• Zoe Krauss: zkrauss@uw.edu
• Qibin Shi: qibins@uw.edu
• Yiyu Ni: niyiyu@uw.edu
• Marine Denolle: mdenolle@uw.edu
• Nate Stevents: ntstevens@uw.edu

Project team at Stanford:

• Ian MacBreatry
• Yifan Yu

Repository Structure

📜README.md
📜LICENSE
📜.gitignore
📜environment.yml
📦0_availability
📦1_picking
📦2_associate
📦3_location
📦4_relocation
📦workflow_simplified
 ┣ 📜0_data_availability_7D.ipynb
 ┣ 📜1_parallel_detect_picks_elep.ipynb
 ┣ 📜2_format_pick2associate.ipynb
 ┣ 📜3_associate.ipynb
 ┗ 📜4_quality_control.ipynb
📦data
 ┣ 📜vel_*.csv # velocity profiles
 ┣ 📜stations_*.csv # station locations in regions
 ┣ 📜nodes_*.csv # regions for velocity profiles
 ┣ 📜ds03.xlsx
 ┣ 📜jgrb52524-sup-0002-2017jb014966-ds01.csv # Morton et al, 2023 data
 ┗ 📜jgrb52524-sup-0003-2017jb014966-ds02.csv # Morton et al, 2023 data
📦figures
📦old
📦utils

Installation Guide

Prerequisites

Before you begin, ensure you have the following installed on your machine:

• Python 3.8+
• pip

Setting Up the Environment

1. Clone the repository:

   git clone https://github.com/Denolle-lab/cascadia_obs_ensemble.git
   cd cascadia_obs_ensemble

2. Create a virtual environment:

   conda env create -f environment.yml

3. Activate the virtual environment:

   • On macOS and Linux:

     conda activate seismo_cobs

   • On Windows:

     .\venv\Scripts\activate

4. Install the required packages:

   If you have a requirements.txt file:

   pip install -r requirements.txt

   If you have an environment.yml file (for Conda environments):

   conda env create -f environment.yml
   conda activate your-environment-name

Running the Notebooks

1. Start Jupyter Notebook:

   jupyter notebook

2. Open the notebooks:

   In the Jupyter Notebook interface, navigate to the following notebooks and open them:

Pick data

After running ELEP, the picks are now available on Gdrive (private access until we upload to zenodo)

• Picks