"The driving objective of AASPI is to try and reveal and see more patterns in the seismic data than we can see just looking at the seismic amplitude data."

Heather Bedle, Principal Investigator at Attribute Assisted Seismic Processing and Interpretation (AASPI) at the University of Oklahoma, joins Seismic Soundoff. 

In this episode, you will discover how AASPI reveals hidden patterns in seismic data, pushes the boundaries of geologic interpretation, and reshapes our understanding of the Earth using cutting-edge research and technology.

Heather shares insights into how machine learning has been utilized in geophysics for decades, emphasizes the importance of critical thinking when interpreting algorithmic outputs, and discusses the potential biases inherent in machine learning models.

Listeners will be intrigued by AASPI's innovative research, including Heather's favorite attribute, aberrancy, which is pushing the boundaries of seismic detail. She also highlights AASPI's drive towards transparency and how it works with its sponsors to move the industry forward.

📋 EPISODE HIGHLIGHTS
* 1:43 - The core objectives of AASPI
* 6:05 - Simplifying the complex world of machine learning in geosciences
* 9:20 - AASPI's unique approach to machine learning applications
* 13:55 - Key papers that embody AASPI's work
* 17:45 - The intrigue of aberrancy in seismic data analysis
* 20:10 - The challenge of understanding algorithmic focus and biases
* 24:25 - The balance between academia and industry in advancing machine learning
* 27:20 - The importance of transparency and documentation
* 29:05 - Upcoming challenges and projects for AASPI
* 31:15 - A challenge for listeners: consider the biases in machine learning
* 33:25 - Heather's journey from industry to academia

✍️ EPISODE LINKS
Visit https://seg.org/podcasts/209-thinking-like-an-algorithm-utilizing-machine-learning-in-seismic-data/ for the complete interview transcript and all the links referenced in the show.

💬GUEST BIO
Dr. Bedle is the Principal Investigator for Attribute Assisted Seismic Processing and Interpretation (AASPI) and is an Assistant Professor at the University of Oklahoma in the School of Geosciences. Her research and teaching focus on applying and developing advanced seismic interpretation. Some research stems from her time as a petroleum geophysicist using rock physics models coupled with seismic reflection data interpretation, and some from her experience as a whole-earth seismologist. All of the projects are based on her interest in understanding what lurks out of sight beneath our feet.

Seismic reflection data is currently the main investigative data being utilized. Current projects in her research group combine seismic interpretation with developments in machine learning, seismic attributes, and rock physics. These methods work as compliments to extract additional information from the seismic waveforms. 

Dr. Bedle also researches how scientists learn to interpret seismic data, as well as how to improve the learning process at both the academic and professional levels. New projects are being developed to improve education in active learning spaces and test the capability of virtual reality to improve seismic interpretation education and research.

SHOW CREDITS
Seismic Soundoff showcases conversations with geoscientists addressing the challenges of energy, water, and climate.

SEG creates these episodes to celebrate and inspire the geophysicists of today and tomorrow.

If you have episode ideas or feedback for the show or want to sponsor a future episode, email the show at podcast@seg.org.

This episode was hosted, edited, and produced by Andrew Geary at TreasureMint. The SEG podcast team is composed of Jennifer Cobb, Kathy Gamble, and Ally McGinnis.

Seismic Soundoff

209: Thinking like an algorithm - utilizing machine learning in seismic data

Seismic Soundoff showcases conversations addressing the challenges of energy, water, and climate. Produced by the Society of Exploration Geophysicists (SEG), these episodes celebrate and inspire the geophysicists of today and tomorrow. Three new episodes monthly.