Date of Award

Summer 8-1-2021

Degree Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

School

Computing Sciences and Computer Engineering

Committee Chair

Dr. Dia Ali

Committee Chair School

Computing Sciences and Computer Engineering

Committee Member 2

Dr. A. Louise Perkins

Committee Member 2 School

Computing Sciences and Computer Engineering

Committee Member 3

Dr. John Harris

Committee Member 3 School

Mathematics and Natural Sciences

Committee Member 4

Dr. Bo Li

Committee Member 4 School

Computing Sciences and Computer Engineering

Committee Member 5

Dr. Brian S. Bourgeois

Abstract

Due to the difficulty and expense of collecting bathymetric data, modeling is the primary tool to produce detailed maps of the ocean floor. Current modeling practices typically utilize only one interpolator; the industry standard is splines-in-tension.

In this dissertation we introduce a new nominal-informed ensemble interpolator designed to improve modeling accuracy in regions of sparse data. The method is guided by a priori domain knowledge provided by artificially intelligent classifiers. We recast such geomorphological classifications, such as ‘seamount’ or ‘ridge’, as nominal data which we utilize as foundational shapes in an expanded ordinary least squares regression-based algorithm. To our knowledge we are the first to utilize the output of classifiers as input into a numerical model. This nominal information provides meta-knowledge about seafloor creation and growth into our models implicitly.

We performed two suites of experimental studies designed to clarify when these techniques add value. In our first study, we utilized the MergeBathy software for DBM construction to extensively investigate existing interpolators for feature-favoritism on different synthetic, idealized morphologies. This study reduced the possibility that the interpolators were a significant source of error in sparse data regions. Two feature-favoring interpolators then served as our nominal-informed interpolators and ensemble members. In our second study we utilized Friedman’s hypothesis testing to verify that our nominally informed ensemble method outperforms splines-in-tension in the presence of sparse data. To our knowledge this is the first comparison study of interpolation over sparse bathymetric data to verify statistically significant improvement in sparse-data regions.

Share

COinS