Southern California Earthquake Data Center

Getting Started With SCEDC AWS Public Dataset

This page describes some quick examples in python to start using the SCEDC dataset.
Python Code Snippet Docker Image Amazon Machine Image



    AWS account: sign up at
        AWS credentials: For programmatic access you need Access Keys.
            boto3 and botocore: AWS SDK for Python


              Python Snippet

              import boto3, botocore
              BUCKET_NAME = 'scedc-pds'
              except botocore.exceptions.ClientError as e:
                      if e.response['Error']['Code'] == "404":
                              print("The object does not exist.")

              Docker Example

              This example shows how to run a Docker image with ObsPy and Jupyter Notebook. You can then run the python snippet and start using classes from the Obspy framework.


              run from os prompt:
              ~ $: docker pull crleblanc/obspy-notebook
              ~ $: docker run -e AWS_ACCESS_KEY_ID=  -e AWS_SECRET_ACCESS_KEY=  -p 8888:8888 crleblanc/obspy-notebook:latest 
              ~ $: docker exec  pip install boto3


              Using an Amazon Machine Image (AMI)

              There is a public AMI image called scedc-python that has a Linux OS, python, boto3 and botocore installed.
            • from your AWS management console, choose "EC2"
            • Under "Instances" choose to launch an instance. Select existing AMI named "scedc-python" (under Community AMIs)


            • Boto 3 Documentation
            • AWS CLI User Guide AWS Command Line Interface. An open source tool for working with AWS via command line.
            • AWS Getting Started Resource Center Starter page for learning about what types of AWS resources are available.
            • Downloading using parallel threads A more sophisticated script which may increase download speeds written by Tim Clements
            • Cactus to Clouds: Processing The SCEDC Open Data Set on AWS A conference paper (2019 SCEC Annual Meeting) by Tim Clements and Marine Denolle at Harvard University detailing their work using the SCEDC dataset