You see, but you do not observe.

“You see, but you do not observe” is a quote by Sherlock Holmes in A Scandal in Bohemia (1891, written by Sir Arthur Conan Doyle). Holmes referred to himself as a ‘consulting detective’. Sketch by Mason Dykstra.

Author: Mason Dykstra, Ph.D., VP Energy Solutions

Wavelets are for Watson (The Doctor; Not IBM) 

When was the last time you interpreted seismic data by looking at the wavelet itself? How often have you pulled out your magnifying glass to examine the wavelet shape? The peaks, the troughs, the amplitude, the curvature, the zero crossing? Learn much from that? I didn’t think so… 

Looking at wavelets is simply not how humans interpret seismic data. We look for and recognize patterns in rendered images of the data. Pattern recognition is one thing we’re really good at. We evolved to do it. We do it every day to recognize faces, and houses, streets, and pets. Those of us who are geologically inclined recognize rocks, mountains, landscapes, seismic facies, shapes and geometries in seismic data. That’s how humans interpret seismic data visually. Observe, don’t just see.

The Mathematical Moriarty 

So why, then, do we use auto-tracking algorithms that, by and large, use seismic wavelet characteristics to do their job? Well, first off, because it’s mathematically relatively easy to do. Waves are well understood, easy to describe mathematically, and we can therefore effectively analyze the characteristics of wavelets and assess similarity between wavelets from nearby traces. Secondly, because it can be a good approach for intervals with acoustic character which doesn’t change much spatially, such as regionally persistent condensed sections. 

However, wavelet similarity will always fail either over larger distances, due to changes in rock facies (and therefore acoustic character), or when the geology becomes complicated. So what then? Well, why not get the computer to do what we do, and recognize patterns?

Elementary, My Dear Watson   

And that’s where artificial intelligence comes into play. With well-crafted networks we can build models that allow artificial intelligence engines to learn patterns and recognize them elsewhere in the data. Not only can this approach expand how much interpretation we can automate, but it can fundamentally change how we approach interpretation. 

We’re no longer constrained to only interpreting horizons or faults. Instead, we can directly interpret seismic facies, intervals, geobodies, or whatever we’re interested in, and the AI will emulate what we teach it. We’re at the beginning of a revolution in the interpretation and analysis of the subsurface. 

It’s an exciting time, and the journey is just beginning. Come see what the ‘consultative detectives’ at Enthought are up to with seismic data and deep learning

About the Author

Mason Dykstra, Ph.D., VP Energy Solutions at Enthought holds a PhD from the University of California Santa Barbara, an MS from the University of Colorado Boulder, and a BS from Northern Arizona University, all in the Geosciences. Mason has worked in Oil and Gas exploration, development, and production for over twenty years, split between oil industry-focused applied research at Colorado School of Mines and the University of California, Santa Barbara; and within companies including Anadarko Petroleum Corporation and Statoil (Equinor).

Share this article:

Related Content

3 Trends for Scientists To Watch in 2023

As a company that delivers Digital Transformation for Science, part of our job at Enthought is to understand the trends that will affect how our…

Read More

Accelerating Science: the Classical Mechanics Perspective

When thinking about enhancing R&D processes, Newton’s second law of motion provides the perfect framework. Classical mechanics teaches us that putting a body into motion…

Read More

Retuning the Heavens: Machine Learning and Ancient Astronomy

What can we learn about machine learning from ancient astronomy? When thinking about Machine Learning it is easy to be model-centric and get caught up…

Read More

Announcing Enthought Academy

Dear Students and Friends of Enthought,  I am pleased to announce Enthought Academy—the culmination of over twenty years of teaching Scientific Python. Since our founding…

Read More

Extracting Target Labels from Deep Learning Classification Models

In the blog post Configuring a Neural Network Output Layer we highlighted how to correctly set up an output layer for deep learning models. Here,…

Read More

True DX in the Pharma R&D Lab Defined by Enthought

Enthought’s team in Japan exhibited at the Pharma IT & Digital Health Expo 2022 life sciences conference in Tokyo, to meet with pharmaceutical industry leaders…

Read More

Exploring Python Objects

Introduction When we teach our foundational Python class, one of the things we do is make sure that our students know how to explore Python…

Read More

Choosing the Right Number of Clusters

Introduction When I first started my machine learning journey, K-means clustering was one of the first algorithms I was introduced to – and it is…

Read More

Prospecting for Data on the Web

Introduction At Enthought we teach a lot of scientists and engineers about using Python and the ecosystem of scientific Python packages for processing, analyzing, and…

Read More

Life Sciences Labs Optimize with New Digital Technologies and Upskilling

Labs are resetting the trajectory for drug development: reducing timelines from years to months; decreasing costs from billions to millions; and gaining an advantage by…

Read More

Join Our Mailing List!

Sign up below to receive email updates including the latest news, insights, and case studies from our team.