No abstract

No abstract

[1] Equilibrium temperature is the water temperature at which the sum of all heat fluxes through the water surface is zero. It can be calculated from weather data. Mean weekly stream temperature was found to be linearly related to mean weekly equilibrium temperature above 0°C. The slopes and intercepts of the linear relationship were used to identify effects of shading, sheltering, cold water inputs (groundwater, meltwater, and deep reservoir releases) and warm water inputs (wastewater, cooling water, and lake surface water). The linearity hypothesis was confirmed for data from 596 U.S. Geological Survey stream gaging stations in the eastern and central United States. For approximately 15% (89 of 596, NSC ! 0.90) of stream gaging stations, weekly equilibrium temperature was a good estimator of weekly stream temperature with zero wind sheltering and sun shading. With sheltering and shading the number rose to 26% (156 of 596). For these streams the heat exchange through the water surface has the most controlling effect on stream temperatures. For the remaining 74% of streams the relationship between weekly stream temperature and weekly equilibrium temperature was also linear, but its slope was significantly less than 1.0 (even after calibration for shading and sheltering), indicating that processes other than surface heat exchange have an influence. Cold water inflows were dominant for 8% (46 of 596) of stream gaging sites. Less than 5% of all stream gaging stations showed evidence of unnatural heat inputs. Fitting an equilibrium temperature linearly to recorded stream temperatures, e.g., at a weekly timescale, can be of use to project stream temperatures under different weather-climate scenarios or to identify both natural and anthropogenic heat and hydrologic inputs to streams.

IntroductionPotential global warming caused by an increase in some atmospheric gases, especially carbon dioxide and methane, will influence the thermal regime of aquatic environments. These changes will impact the organisms living in the aquatic environments. For example, if maximum water temperature in a stream reach exceeds the thermal tolerance of a fish species, that fish species is likely to disappear in that stream reach [Eaton and Scheller, 1996]. Similarly, if stream temperature drops below a threshold, it may cause the demise of a fish species through osmoregulatory dysfunction. There is also evidence that fish species stop feeding or feed only sporadically at weekly stream temperatures below a threshold, for example, on average, 8øC for warm water fish species [Scheller et al., 1999].To project fish habitat changes under a 2 x CO2 climate scenario, i.e., a climate after doubling of carbon dioxide in the atmosphere, it is necessary to know the thermal constraints of different fish species and to have access to the current and projected stream temperature time series. A fish survival model, based on the stream thermal regime and thermal constraints of warm water fish species, has been developed by Scheller et al. [1999]. The fish survival model has a weekly Regressions between water temperatures at individual stream gaging stations and air temperatures at nearby weather stations provide the easiest practical method to estimate stream temperatures for the entire United States. They are also attractive for climate change effect studies because only one input variable, air temperature, is used, and general circulation models (GCMs) simulate this variable better than they simulate other climate variables [Lau et al., 1996]. Among regression models, linear models are more common and easier to utilize. However, a linear function of air temperature is usually not sufficient to determine stream temperatures yearround. It has been shown that because of evaporative cooling, stream temperature starts leveling off as air temperature exceeds ---20øC. The overall stream temperature-air temperature 3723