Key points are not available for this paper at this time.
Many freshwater lakes are supersaturated in CO 2 with respect to the atmosphere. This concentration gradient implies a net flux of CO 2 from the water to the air. The actual rate of gas exchange is governed by both this concentration gradient and the gas transfer coefficient, k . To directly measure k , we added the chemically and biologically inert gas, sulfur hexaflouride (SF 6 ), to the epilimnion of Mirror Lake, New Hampshire, a small (15 ha), low‐wind softwater lake. k was independent of wind speed over the 50‐d summer stratification period and averaged 2.65 ± 0.12 cm h −1 (95% CI; normalized to a Schmidt number of 600); k 800 , was better correlated to precipitation events than it was to wind speed. Our data support the idea that gas exchange across the air‐water interface is largely independent of wind at low wind speeds. The surface water of Mirror Lake was persistently supersaturated in CO, with respect to the atmosphere. During a 3.5‐‐year period the partial pressure of CO 2 in the surface waters of the lake averaged 726 ± 39 µatm (95% CI) and showed substantial seasonal variation (360–2,000 patm). Diel and day‐to‐day variation in CO 2 were very small compared to the CO 2 pool. We combined our estimates of k with weekly measurements of the partial pressure of CO 2 to estimate CO 2 gas exchange in the lake. Mirror Lake released from 26 to 50 g C m −2 to the atmosphere each year, depending on the method of calculating k . Atmospheric CO 2 exchange is a large term in the C economy of the lake—the most conservative gas flux estimate is about four times as large as outflow plus seepage of total dissolved inorganic carbon and 1.5 times as large as the export of dissolved organic C from the lake.
Cole et al. (Mon,) studied this question.