Colorado River cutthroat trout, Oncorhyncus clarki pleuriticus , occupy less than 12% of their historic range. Restoration and conservation of this species are currently under way across the upper Colorado River basin, but guidance to inform management decisions related to the impacts of climate change on cutthroat is lacking. Shifts in the thermal distribution of freshwater fish have been documented, and will continue to occur as cold water habitat is threatened by warming water temperatures. Coupled air and water temperature data allow for an estimation of potential resistance and resilience to warming, determining the effect that local air has on stream temperature. The United States Forest Service, cooperating with federal agencies, state agencies and private landowners, placed temperature loggers in the water and two air locations at 50 sites. To select a representative subset of sites, six habitat characteristics of each Colorado River cutthroat trout core conservation population were considered. These characteristics include solar input, elevation, watershed area, riparian vegetation, groundwater input, and the 30-year mean maximum July air temperature. Results from coupled temperature loggers indicate that the relationship between air and water temperature in the upper Colorado River basin is neither linear, nor one-to-one. Using Mohseni's (2003) equation, the relationship between air and water temperature was fit to a nonlinear regression curve. Analysis shows that the median rise in daily maximum water temperature is only 0.41°C for a 1.0°C increase in the median daily maximum air temperature. Air temperature exerts the most influence over water temperature; however, these results indicate that there are other characteristics that influence stream temperature. To determine these characteristics, analysis of the six habitat characteristics used for site selection in addition to aspect, slope, and latitude were used to model multiple temperature metrics. The best model, nonlinear water to air temperature relationship, had an R2 between actual and predicted values of 0.71. It also became clear that using multi-metric analysis would provide a much more robust indicator of resistance. This work will allow managers to consider potential climate change resistance or resilience in project prioritization, by understanding potential habitat characteristics to buffer stream warming.