Rocky Mountain Section - 75th Annual Meeting - 2025

Paper No. 28-8
Presentation Time: 8:00 AM-5:30 PM

PERFORMANCE OF MACHINE LEARNING TECHNIQUES IN RECOVERING GEOMORPHIC HISTORY FROM STRATIGRAPHY


BATES, Robert, Computer Science, Western Washington University, 516 High Street, Bellingham, WA 98225, FOREMAN, Brady Z., Geology, Western Washington University, 516 High Street, Bellingham, WA 98225 and STRAUB, Kyle, Earth and Environmental Sciences, Tulane University, 6823 St. Charles Avenue, New Orleans, LA 70118

Stratigraphy preserves information on geomorphic activity within depositional basins. However, due to erosion and stasis events, information loss is inevitable. Experimental basin studies of fluvio-deltaic systems illustrate this loss. They show: (1) bed thicknesses do not map one-to-one with depositional events; (2) a disconnect between the structure of erosional surfaces and the topography that generated them; and (3) changes in the underlying driver of unconformities. These experiments quantify aspects of the classic inversion problem faced by stratigraphers. Herein, we assess if machine learning techniques can quantitatively solve aspects of this inversion problem.

Our dataset is derived from fluvio-deltaic experiment TDB-17-1, performed at Tulane University’s Sediment Dynamics Laboratory. This experiment occurred in a 4.2 m x 2.8 m x 0.65 m basin, with a constant sediment and water discharge, and a steady, uniform base level rise. The sediment mixture included analogs for coarse and fine grain sizes along with minor admixtures to reproduce cohesive behaviors. The main phase of the experiment was purely aggradational. Frequent laser scans of the experimental surface captured a point cloud of topography that was converted into a digital elevation model with mm-scale resolution horizontally and < 1 mm scale resolution vertically. This provides a precise time series of the erosion, deposition, and stasis geomorphic events embedded behind each stratigraphic bed as well as the knowledge of the depositional sub-environment (e.g., channel, lobe, wet/dry overbank).

A variety of machine learning techniques (i.e., random forests, neural networks, support vector machines) were applied with the goal of predicting the proportion of time represented by deposition, erosion, and stasis behind a given stratigraphic bed. Predictor variables included sedimentation rate, bed thickness, depositional sub-environment, and binary “tags” for a bed’s association with deep scours or long stasis. We find prediction errors generally of ~10% for the proportion of time spent depositing, eroding, or in stasis behind a given bed. We hypothesize, if appropriately applied, these techniques could potentially improve basin age models on the scale of kyrs to tens of kyrs.