Date of Award

2018-01-01

Degree Name

Master of Science

Department

Computational Science

Advisor(s)

Diane Doser

Abstract

Data visualization is an effective way to analyze large amounts of spatial information to identify correlations, trends, outliers, and patterns. In this Thesis I test the application of supervised machine learning algorithms to render a series of 3D visualizations designed to highlight general traits of the Hueco Bolson, a geologic basin located east of the Franklin Mountains in far west Texas and northern Mexico. A geologic basin is one of the most common inland places where sediments are collected. The geology of basins is of much interest to geophysicists, hydrologists, paleontologists, and oil prospectors. Here the task of 3D geologic modeling is approached as a classification problem. The 3D models constructed from this study are built from interpretation data (geologic cross sections) developed from previous studies conducted on the basin.

This new approach to geomodeling addresses some of the limitations associated with surface based modeling in densely faulted areas (using traditional 3D interpolation schemes) and volume based modeling. The 3D models produced for this Thesis have given geoscientists a general understanding of the geometry and structure of Hueco Bolson, which is the principal aquifer for the Greater El Paso Region. The basin is positioned in southwestern Texas and south-central New Mexico on the U.S./Mexican border. El Paso and its surrounding urban area currently relies on groundwater for over half of its water supply, and Ciudad Juarez relies entirely on groundwater from the Hueco Bolson aquifer, supporting the 2.5 million inhabitants of the region.

Language

en

Provenance

Received from ProQuest

File Size

48 pages

File Format

application/pdf

Rights Holder

Joel Gerardo Castro

Download

Included in

Geology Commons

Share

COinS