Date of Award:


Document Type:


Degree Name:

Doctor of Philosophy (PhD)


Civil and Environmental Engineering

Committee Chair(s)

David G. Tarboton


David G. Tarboton


Daniel Watson


Luis Bastidas


Fred Ogden


David Rosenberg


In hydrology we straddle the domains of science and engineering. As hydrologists our goal is to predict the movement and volume of water. As scientists we seek to improve our understanding of water-related processes and how to model them. As engineers we seek to be able answer specific water-related questions to provide protection and an essential resource for the people we serve. Underlying all of our work is a body of knowledge that we have developed and continue to develop. This knowledge involves many aspects, such as the role of various hydrologic processes, how to obtain data, computational models that have been developed, and many other things. Knowing this body of knowledge is the key to being a hydrologist. The goal of this work is to enable a computer to begin to think as we do, to reason over hydrologic processes, to deduce what tasks need to be accomplished to answer the hydrologic questions we are asked. To do that we must be able to model how we think as hydrologists, to capture the concepts and procedures we use in a form that a computer can understand.

This dissertation creates a reasoning engine that is able to include both semantic (concept) knowledge as well as procedural knowledge. This new form of knowledge model is called a "functional ontology". To demonstrate the utility and power of the reasoning engine several functional ontologies are created that capture knowledge about delineating watersheds, knowing how to set up and run computational models, as well as how to create a chain of models to answer the "what-if" questions we are asked. The work shown in this dissertation demonstrates how, through a reasoning engine that combines semantic and procedural knowledge, we can actually model many of the concepts and simple tasks we do as hydrologists in a form that enables the reasoning engine to use deductive logic and automate many of the tasks we do as hydrologists. The focus of this work has been to enhance the use of the tools we as hydrologist use now to examine and engineer solutions to hydrologic problems. We hope that in the future the reasoning tools and knowledge models are further developed to enable a wide range of automated watershed analysis and model creation processes.