37-24 Quantitative Seismic Geomorphology of Deep-Water Depositional Elements in the North Carnarvon Basin, Australia

Session: Geoscience Research Poster Showcase by 2YC and 4YCU Undergraduate Students

Presenting Author:

Authors:

Abstract:

The North Carnarvon Basin, offshore Western Australia, contains well-preserved deep-water depositional systems that provide an excellent opportunity to study submarine geomorphology using 3D seismic data. Understanding the geometry and distribution of these depositional elements helps improve interpretations of sediment transport, depositional processes, and reservoir architecture. This study develops a workflow that combines seismic attributes and unsupervised machine learning to identify, characterize, and measure deep-water depositional elements.

Several geometric and amplitude-based seismic attributes are used to enhance features that are difficult to recognize on conventional seismic amplitude data. Unsupervised machine learning techniques, including K-means clustering, Self-Organizing Maps (SOM), Principal Component Analysis (PCA), and Independent Component Analysis (ICA), are applied to multi-attribute datasets to assist in the interpretation of submarine channels, channel-levee complexes, basin-floor lobes, mass-transport deposits, and other deep-water depositional elements. The different methods are compared to evaluate how well they highlight depositional features and support seismic interpretation.

After interpretation, the identified depositional elements are measured to quantify characteristics such as channel width, thickness, sinuosity, levee dimensions, lobe extent, and spacing between features. These measurements are compiled into a statistical database that will be used to describe the variability of deep-water depositional elements and provide quantitative input for future three-dimensional geomodeling efforts. Preliminary results indicate that integrating seismic attributes with machine learning improves the identification of depositional elements and provides an efficient workflow for quantitative seismic geomorphology. Ongoing work focuses on expanding the morphometric database and evaluating the consistency of the different machine learning approaches

© Copyright 2026 The Geological Society of America (GSA), all rights reserved.