How to predict gas transfer in lakes

Posted on October 1, 2014


Exchange of gases across the air – water boundary layer has an important influence on the aeration and carbon dynamics of lakes. Because the rate of gas transfer depends mainly on turbulence near the lake surface, which is affected by wind strength, gas exchange is usually predicted by reference to wind speed. However, measuring rates of gas exchange is a challenging business, partly because wind speeds measured over land don’t closely reflect those over water, and because gas transfer is affected by wave energy, which in turn depends on fetch length. Canadian researchers investigated whether gas transfer predictions can be improved by using more information than just wind speed. They collected data on meteorological variables, surface turbulence and CO2 flux from a large hydroelectric reservoir and from eight lakes of size 0.19 – 4.00 km2 in the Quebec region. Although there was a significant overall relationship between gas transfer velocity and in situ wind speed, the separate relationships for the reservoir and the lake data were significantly different. This showed that lake size had a considerable impact on gas exchange, which was predicted to be over 50% higher in a 10 km2 lake than in a 1 km2 lake at a windspeed of 5 m/sec . A much stronger, and more broadly applicable, predictive equation was one that included the interaction between wind speed and lake area, as well as wind speed itself. This model explained 68% of the variation in gas transfer velocity. Surprisingly, adding information on fetch length didn’t improve predictive power, probably because fetch length was effectively subsumed in the lake area variable. The fact that gas exchange was lower in small lakes may account for the hitherto unexplained observation that CO2 concentrations tend to be inversely related to lake size.

Reference: Vachon, D. & Prairie, Y.T. 2013. The ecosystem size and shape dependence of gas transfer velocity versus wind speed relationships in lakes. Canadian Journal of Fisheries & Aquatic Sciences 70, 1757–1764. http://www.nrcresearchpress.com/doi/pdf/10.1139/cjfas-2013-0241

Advertisements