Resolution limitations of conventional seismic methods prohibit the subsurface investigation of high-frequency depositional sequences and small-scale fluvial systems. However, modern high-frequency 3D seismic methods provide sufficient resolution for such analyses. This study utilized high-resolution 3D seismic data to investigate Quaternary high-frequency depositional sequences and incised valley evolution on the northwestern Gulf of Mexico’s (GoM) inner shelf. Through an extensive seismic attribute analysis and the employment of machine learning algorithms, fluvial systems along three major unconformities were delineated and determined to be incised valleys which are known to have formed during eustatic lowstand events. The evolution of these incised valleys and their architectural elements in response to rapid transgression was investigated, and various seismic attributes were employed in a principal component analysis (PCA) which facilitated an understanding of lithologic distribution throughout the seismic volume. Through use of a tripartite seismic stratigraphic analysis, systems tracts—lowstand systems tracts (LSTs), transgressive systems tracts (TSTs), and highstand systems tracts (HSTs)—were identified, and each sequence boundary was correlated to a lowstand event on a eustatic sea level curve. This study underscores the significance of high-resolution seismic data, seismic attributes, and machine learning in understanding high-frequency sea level cycles and provides insights into the evolution of the GoM’s inner shelf over the past 300,000 years. An understanding of these high-frequency sea level cycles and small-scale fluvial features may provide valuable insight into the small-scale heterogeneity of Cenozoic hydrocarbon reservoirs throughout the northern GoM.

This content is only available via PDF.
You can access this article if you purchase or spend a download.