Multi-Attribute Analysis enables Interpretation Below Seismic Resolution
Machine learning, a method of data analysis that automates analytical model building, has moved beyond the realm of self-driving cars, banking fraud detection, and large-scale consumer behavior analysis and has begun to show major successes when applied to interpreting seismic data.
Using algorithms that iteratively learn from data, machine learning allows interpreters to analyze and visualize multiple seismic attributes simultaneously, thereby revealing features below tuning - difficult to visualize by any other method.
Paradise, from Geophysical Insights, is the industry's first and most robust machine learning engine applied specifically to seismic interpretation. Using self-organizing maps within Paradise, interpreters can identify geologic features and stratigraphy well below typical seismic resolution or tuning thickness (1/4 wavelength thickness). In this white paper, "Case Studies in Sub-seismic Resolution", several interpretation case studies outline how this technology was used to gain greater insights.