Masters Theses

Orcid ID

https://orcid.org/0009-0004-1538-5035

Date of Award

5-2025

Degree Type

Thesis

Degree Name

Master of Science

Major

Computer Science

Major Professor

Michela Taufer

Committee Members

Michela Taufer, Sai Swaminathan, Jack Marquez

Abstract

The GEOtiled framework facilitates the scalable and efficient computation of high resolution terrain parameters using digital elevation models (DEMs) across the Continental United States (CONUS). These parameters are essential in Earth science applications. The original GEOtiled framework, based on the Geospatial Data Abstraction Library (GDAL), was limited to computing only three terrain parameters. This thesis presents significant advances to GEOtiled, focusing on expanding its functionality and optimizing its performance. By integrating the System for Automated Geoscientific Analyses (SAGA) GIS library, we extend the framework to support the computation of over 15 terrain parameters compared to the initial three provided by the original version powered by the GDAL library. We introduce three key optimizations to mitigate SAGA’s higher computational costs: concurrent cropping, efficient mosaic operation, and unified concurrency between the compute, crop, and mosaic steps. These optimizations improve data distribution and handling, reducing computation times while maintaining accuracy. We perform performance evaluations using data at varying resolutions, including a 30-meter DEM covering the State of Tennessee and CONUS. Our results demonstrate that the optimized framework using SAGA performs similarly to the original version using GDAL for small tile sizes while offering a more extensive generation of computed parameters. GEOtiled software, openly available on GitHub, and its data, published as open access in Dataverse for community use, can potentially advance reproducible, data-driven scientific discovery in Earth science.

Download
Files over 3MB may be slow to open. For best results, right-click and select "save as..."

Share

COinS