IRIS-EarthScope USArray Data Processing and Analysis Short Course, 2015
Indiana University, Bloomington, USA

Gary Pavlis (Host Instructor, Indiana University), Chuck Ammon (Penn. State University), Mitchell Barklage (NodalSeismic), Mike Brudzinski (Miami University), Andy Frassetto (IRIS), Alex Hutko (IRIS), Wenjie Lei (Princeton), Tobias Megies (Munich University), Danielle Sumy (IRIS), Emily Wolin (Northwestern)

The primary goal of this short course is to provide training in the foundations of seismic data processing for the next generation of geoscientists. Our aim is to inspire the participants to become leaders in developing more effective ways to handle data from large seismic datasets, such as the USArray. This year's course will focus on power methods for fetching, processing, analyzing, and visualizing data locally. Attendees are expected to already have some experience with seismological data and scientific computing. 

Participants will be selected into each short course by the program's steering committee based on the information provided in their applications. Travel, lodging, and most meal costs for selected applicants will be covered by financial support from NSF through the "Seismological Facilities for the Advancement of Geoscience and EarthScope" award to IRIS. Please note: Selection and financial support will be focused on applicants who have not attended previous USArray short courses and are based at institutions within the U.S. International participants may also apply, but are responsible for their expenses and logistics if coming from outside the U.S.

For more information contact Danielle Sumy.

The course will be held on the campus of Indiana University, 75 minutes southwest of Indianapolis. IRIS-sponsored participants will be expected to share rooms. Information for arranging travel, lodging, and reimbursement will be made available to selected participants.

Recommended Reading Materials

Read the two short Nature articles (under 'Materials' section) on Python.
Software Carpentry Tutorials - Specifically, Unix shell, programming basics with Python, or version control with git
The Art of the Command Line - at least skim
Explain Shell - valuable Unix resource for novices to advanced
Makefiles - good resource for compiling
We'll be using Apple workstations throughout the course, so the Apple shell scripting guide (under 'Materials' section) may also be helpful. (Don't read the whole thing unless you feel inspired, it's...long.)

Pavlis Recommended Reading Materials for USArray Data Processing Workshop 2015

This document contains a set of reference material on the web that will be useful for the workshop. We encourage all students to skim these sources to get a feel for where we are going. After the workshop the hope is these will provide useful background as you move forward. Items with an * preceding them are things we recommend you try to read before the workshop if at all possible. We also encourage you to help us expand this list by recommending sites you find that might be particularly useful.

Database Concepts

*Database concepts (Wikipedia)

*Database concepts (Antelope) 

Analysis of Earthquake SeismogramsLook at Lesson 1 and 2. You might also look at Lesson 6.

General seismic data concepts

SEED Manual - Do NOT try to read this whole monster. It is a reference you should glance at to appreciate the complexity and teach you to show some respect for the poor schmucks who had to write the various SEED readers and writers.

*SAC Manual

3D Visualization

Paraview Tutorial

USArray Advanced Short Course, August 3-7, 2015

Monday, August 3: Introduction to IRIS and USArray, Data Services Topics

Monday, August 3rd, 2015, 7am–7pm

7:00 am
8:00 am

BREAKFAST - State Room East, 2nd floor of Biddle Hotel

8:15 am
8:45 am

Welcome and logistics - Gary, Andy, and Danielle

  • Review of resources and computer labs
  • Introduction to group project

8:45 am
9:15 am

Introduction to IRIS and USArray, the facility and the science - Andy

9:05 am
9:15 am

Introduction to IRIS Early Career Activities - Danielle

9:15 am
10:30 am

Foundational Data Services Topics from the IRIS DMC (Part 1) - Alex

  • Webservices tutorials and examples (what does XML mean? timeseries, event, station metadata)

10:30 am
10:45 am

BREAK

10:45 am
12:00 pm

Foundational Data Services Topics from the IRIS DMC (Part 2) - Alex

  • Data vs. Data Products (raw waveforms vs. Ground Motion Visualization (GMVs))
  • Contributed examples of data access tools
  • Quality Assurance/Quality Control (QA/QC) Tools - MUSTANG

12:00 pm
1:00 pm

LUNCH

1:00 pm
3:00 pm

Introduction to Best Practices for Coding (Part 1) - Chuck and Gary

A Selection of software development options and tools

  • Scripting languages
  • Compiled languages
  • Modern code editors
  • An introduction to source code management (SCM) with Git
  • Some other tools and practices to explore later

Procedural versus object oriented languages and algorithm design:

  • Examples of procedural and object oriented code
  • Group design of a simple object with critical review

3:00 pm
3:30 pm

BREAK

3:30 pm
4:30 pm

Introduction to Best Practices for Coding (Part 2) - Chuck

  • Do I need a GUI?
  • Integrated Development Environments (IDEs)
  • Building a simple seismic moment calculator in XCode

4:30 pm
5:00 pm

Review and discuss topics covered, expectations, concerns, etc. - Andy/Danielle

5:00 pm
6:00 pm

Pop-up talk introductions

6:00 pm
6:15 pm

Meet in lobby of IMU

7:00 pm

Welcome Dinner at Casa de Pavlis (catered by Carson's Barbecue)

Tuesday, August 4: ObsPy and Research Ready Data

Tuesday, August 4th, 2015, 8:15am–5:45pm

7:00 am
8:00 am

BREAKFAST - Food Court

8:15 am
9:15 am

Introduction to Python - Emily

9:15 am
12:00 pm

Introduction and In-Depth Look at Seismic Data with ObsPy - Tobias

  • fetch event metadata (via fdsn ws client)
  • fetch stations that recorded the event (fdsn ws)
  • fetch waveform data for the event (fdsn ws)
  • how to do instrument correction in ObsPy using previously fetched waveforms and station metadata
  • show bode plot for an instrument response
  • two mechanisms in ObsPy to make deconvolution stable (water level, freq domain pre-filter like in SAC)

12:00 pm
1:00 pm

LUNCH

1:00 pm
3:00 pm

Research Ready Data (Part 1) - Gary

  • Data processing models in seismology
    • Stream processing (seismic reflection)
    • Files processing (SAC)
    • Database processing
  • Discussion of real work flows students are using today
  • Large data processing scaling issues:
    • Compute time of an algorithm
    • IO performance
    • Granularity
    • How it mixes together
  • Using Antelope
    • Fundamentals
    • Constructing dbmaster tables with dbbuild
    • Building dbmaster tables from a dataless SEED volume
    • Building a wfdisc table with miniseed2db or sd2db
    • Extracting segmented event data from continuous data

3:00 pm
3:30 pm

BREAK

3:30 pm
4:30 pm

Research Ready Data (Part 2) - Gary

  • In this session we will use a python script that combines multiple concepts covered to this time:  obspy, antelope, and web services. We will use this script to fetch waveform data from DMC using database tables available from the ANF.

4:30 pm
5:15 pm

Review: Working with Instrument Responses - Chuck

5:15 pm
5:45 pm

Review and Discuss topics covered, expectations, concerns, etc. - Andy/Danielle

  • Topic: Research Ready Data, Prepared Data Archives, Projects

5:45 pm

ADJOURN (Dinner on own)

Wednesday, August 5, 2015: High-Performance Computing and Analysis Examples

Wednesday, August 5th, 2015, 8:15am–5:30pm

8:15 am
11:45 pm

High Performance Computing and Parallel Processing - Ray Sheppard and Robert Henschel (Indiana University)

11:45 am
12:00 pm

Demonstrations by former students of this course:  

  • Ian Wang, array iterative deconvolution
  • Xiaotao Yang, receiver function editor program

Both applications are examples of how we hope you will follow up on what you learn from this course.

12:00 pm
1:00 pm

LUNCH

1:00 pm
3:00 pm

Large E - Mike

  • TAR Ball and Exercise Prep
  • Template Matching (Matlab)
  • Example of Template Matching
  • Discussion of Approaches and Examples

3:00 pm
4:00 pm

Large N - Mitch

  • Sweetwater dataset
  • Intro to dataset, and past Nodal/Large N arrays and capabilities

4:00 pm
5:00 pm

Visualization Concepts and Examples - Chuck

5:00 pm
5:30 pm

Review and Discuss Topics covered, expectations, concerns, etc.

  • Topic: High Performance Computing and Big Data

5:30 pm

ADJOURN

Thursday, August 6, 2015: Scientific Visualization and Project Time

Thursday, August 6th, 2015, 8:15am–5:30pm

8:15 am
9:15 am

Visualization Concept - Danielle/Chuck

9:15 am
10:15 am

Paraview and data import software - Gary

  • Getting started with paraview
  • Slicing and dice tomography models and 3D scattered wave imaging results
  • Tools for providing georeferenced objects to insert in a 3D scene:
    • primitive graphic conversion to vtk (lines, points, vectors)
    • surfaces to vtk (gocad2vtk and research code examples)

10:15 am
12:00 pm

SpecFem Introduction and Example - Wenjie

12:00 pm
1:00 pm

LUNCH

1:00 pm
5:00 pm

Projects and one-on-one interactions

5:00 pm
5:30 pm

Review and Discuss topics covered, expectations, concerns, etc.

  • Topic: Visualizations

5:30 pm
6:30 pm

ADJOURN

6:30 pm

DINNER - Irish Lion

Friday, August 7, 2015: Project Work and Summary

Friday, August 7th, 2015, 8:15am–2:00pm

8:15 am
11:00 am

Group Presentations (data, methods, observations, analysis, results: 15-20 minutes per group)

11:00 am
12:15 pm

LUNCH

12:15 pm
1:00 pm

Surveys - Danielle

1:00 pm
2:00 pm

Wrap-up Discussion and Summary

2:00 pm

DISMISSAL

Notice The application period for this course closed on Sun, May 31, 2015 - 11:59:00 PM.

Important Dates
  • Short Course Dates:
    Aug 3rd – 7th
  • Registration:
    May 7th – 31st
  • Location
    Indiana University