Variability of the gas transfer velocity of CO2 in a macrotidal estuary (the Scheldt)

Abstract

We report a large data-set of 295 interfacial CO 2 flux measurements obtained in the Scheldt estuary (November 2002 and April 2003), using the floating chamber method. From concomitant measurements of the air-water CO 2 gradient, we computed the gas transfer velocity of CO 2 . The gas transfer velocity is well correlated to wind speed and a simple linear regression function gives the most consistent fit to the data. Based on water current measurements, we estimated the contribution of water current induced turbulence to the gas transfer velocity, using the conceptual relationship of O' Connor and Dobbins (1958). This allowed us to construct an empirical relationship to compute the gas transfer velocity of CO 2 , that accounts for the contribution of wind and water current. Based on this relationship, the spatial and temporal variability of the gas transfer velocity in the Scheldt estuary was investigated. Water currents contribute significantly to the gas transfer velocity in the Scheldt, but the spatial and temporal variability (from daily to seasonal scales) is mainly related to wind speed variability. REFERENCESO'Connor D.

Planktonic and whole system metabolism in a nutrient-rich estuary (the Scheldt estuary)

“…To compute DpCO 2 , we used monthly values of atmospheric CO 2 molar fraction from Weather Station Mike (66.00uN, 2.00uE), representative of the open North Sea waters (from the National Oceanic and Atmospheric Administration, Climate Monitoring and Diagnostics Laboratory air samples network, available at http://www.cmdl.noaa.gov/) to which was added a value of 27 ppm that corresponds to the average observed increase of the atmospheric signal in the Scheldt basin (Borges et al 2004b). The CO 2 molar fraction values were converted to partial pressure values in wet air (pCO 2air ) using the procedure described in DOE (1994).…”

“…DDIP and DDIC (mmol m 22 d 21 ) are the nonconservative fluxes of dissolved inorganic phosphorus and carbon. F is the CO 2 flux (mmol C m 22 d 21 ) at the air-water interface computed using the parameterization of the gas transfer velocity of Borges et al (2004b). NEP (mmol C m 22 d 21 ) was estimated from DDIP (NEP DIP), DDIC (NEP DIC), and direct processes measurements (NEP incubations).…”

The FuGas 2.3 Framework for Atmosphere–Ocean Coupling: Comparing Algorithms for the Estimation of Solubilities and Gas Fluxes

“…For sparingly soluble gases or for CO 2 in alkaline waters, and under low winds, the transfer velocity is enhanced by the hydration reaction (b T ), which is sensitive to temperature. Other simple empirical formulations based only on u 10 [30,31], or also accounting for bottom-generated turbulence [32], used data collected in estuaries under low wind conditions. However, modelling the coastal ocean at finer resolutions requires an enhanced representation of the multitude of processes involved.…”

“…However, modelling the coastal ocean at finer resolutions requires an enhanced representation of the multitude of processes involved. Hence, we updated the framework by Vieira et al [21], with the k w being decomposed into its shear produced turbulence (k wind + k current ) and bubbles from whitecapping (k bublle ) forcings [32,41,42,44]. Sc w was determined from temperature and salinity following Johnson [20]:…”

“…Concomitantly, there emerged empirical formulae best fitting low wind data collected from estuaries. Such are the formulations proposed by Carini et al [30] or Raymond and Cole [31], accounting only for u 10 , or by Borges et al [32], adding current drag with the bottom. Meanwhile, new evidence emerged for the effects of the sea surface cool-skin and warm-layer [33][34][35][36][37][38], atmospheric stability [39,40], wave-breaking [41][42][43][44][45][46][47][48][49][50], water surface renewal [28,[51][52][53][54][55][56][57], surfactants [28,[58][59][60] and rain [61][62][63][64], just to mention a few examples from an extensive list of published works.…”