Date of Award:


Document Type:


Degree Name:

Master of Science (MS)


Watershed Sciences

Committee Chair(s)

Sarah E. Null


Sarah E. Null


Timothy E. Walsworth


W. Carl Saunders


Water resources are managed for a variety of human needs, including agriculture, industrial and municipal consumption, hydropower generation, and recreation. There has been a recent push to incorporate habitat needs of aquatic wildlife into water management models alongside these other uses, particularly as competition for limited water resources in a changing climate has reduced instream flow and contributed to declining native fish populations. Habitat models are used to estimate species distributions and differentiate between suitable and unsuitable habitat based on variables important to a given species, but are not usually incorporated into water management models. Because there are many ways of modeling habitat and no standard way to compare model accuracy, for this research I used three methods of comparing the accuracy of three commonly used habitat modeling approaches to identify best methods for estimating Bonneville Cutthroat Trout and Bluehead Sucker habitat in the Bear River Watershed (UT, ID, WY). I also explored how well variables used in making each model’s predictions compared with real-world conditions based on field observations. I determined total upstream catchment area was the most important large-scale variable for predicting both Bonneville Cutthroat Trout and Bluehead Sucker habitat suitability, and nearby land use was also important for Bonneville Cutthroat Trout. I showed none of the models’ variables reflected real-world conditions observed in summer 2022, which suggests data commonly used to build habitat models like these can be outdated, incorrect, or over-simplified. Finally, I determined simple habitat models which incorporated aspects of water quality or species biology, rather than simply available water quantity, best predicted both Bonneville Cutthroat Trout and Bluehead Sucker presence, though performance metrics chosen to evaluate model accuracy influenced results. Simpler methods that incorporate species-specific biological criteria are best to include in water management models so fish conservation can be easily and accurately included as a demand for water resources alongside other uses.