Stream temperature/air temperature relationship: a physical interpretation

• Published in: Journal of hydrology (Amsterdam) 1999-05, Vol.218 (3), p.128-141
• Main Authors: Mohseni, O., Stefan, H.G.
• Format: Article
• Language: English
• Subjects: Earth sciences; Earth, ocean, space; Evaporation; Exact sciences and technology; Freshwater; Global warming; Heat transfer; Hydrology; Hydrology. Hydrogeology; Stream; Temperature
• ISSN: 0022-1694; 1879-2707
• DOI: 10.1016/S0022-1694(99)00034-7

In studies of the potential effects of global climate change on freshwater ecosystems, water temperature is a primary factor. Linear regressions of stream temperature versus air temperature are attractive for this purpose, because they require only one input variable (air temperature) which can be simulated by General Circulation Models (GCMs) better than other climate variables. Under a warmer climate scenario, high stream temperatures must be projected by extrapolation. The question arises whether linear extrapolation is valid. To answer the question, the heat exchange processes that contribute to surface water temperature have been analyzed and related to air temperature on a weekly time scale. The equilibrium temperature concept introduced by Edinger has been used. In stream reaches with large drainage area, stream temperature can be approximated by equilibrium temperature. At elevated air temperatures, the vapor pressure deficit above a water surface increases drastically (even in humid regions) causing strong evaporative cooling and hence a flatter stream temperature/air temperature relationship. At low air temperatures, stream temperatures often reach 0°C as an asymptote. If an upstream flow control (dam, reservoir release) or a waste heat input is present, the lower asymptotic value can be larger than 0°C. As a result of these upper and lower constraints for stream temperatures, the stream temperature/air temperature relationship resembles an S-shaped function rather than a straight line. Linear extrapolations to high and low air temperatures are therefore not justified.