Air - water temperature relationships in the trout streams of southeastern Minnesota's carbonate - sandstone landscape: implications for climate change, brown trout biological processes, and land management

Carbonate - sandstone geology in southeastern Minnesota creates a heterogeneous landscape of springs, seeps, and sinkholes that supply groundwater into streams. Air temperatures have been shown to be effective predictors of water temperature in surface - water dominated streams. However, no published work investigates the relationship between air and water temperatures in groundwater - fed streams across watersheds. We used simple linear regression models to examine air - water temperature relationships for 40 groundwater - fed streams in Southeastern Minnesota. A 40 - stream, weekly time scale, composite linear regression model has an R2 of 0.83, a slope of 0.38, and an intercept of 6.63. Regression models were also combined by common intercept and slope and split into winter and non - winter air temperature regimes to allow approximation of winter water temperature regimes based on non - winter data. The air - water temperature relationships for groundwater - fed streams are different in slope and intercept compared to surface - water dominated streams. The high R2 values demonstrate that air - water temperature regression models for groundwater - fed streams may be useful in predicting the thermal regimes for these systems under future climate scenarios. Climate change is expected to alter the thermal regime of groundwater - fed systems but will most strongly affect streams that are more vulnerable to climate change and will do so at a slower rate than surface - water dominated systems. A regression model of intercept vs. slope can be used to identify streams for which water temperatures are more meteorologically controlled than hydrologically controlled, and thus more vulnerable to climate change, with evidence provided by an investigation into the resulting mean summer water temperature under a moderate climate change scenario for various types of linear regression models. Modeling the possible increases in mean summer water temperature provides insight into the potential effects of climate change on the amount of suitable brown trout habitat as well as the possible effects on brown trout biological processes and behavior. Information on possible future thermal conditions and can be used to guide restoration versus management strategies to protect the thermal integrity of trout streams and ensure the persistence of their stenothermic communities.