Application of an Internal Boundary Layer Transport Exchange Model to Gas Transfer of Semivolatile Organic Contaminants in Lake Superior

A two-dimensional Lagrangian internal boundary layer transport exchange (IBLTE) model was developed that is capable of predicting the modification of temperature, humidity, and trace gas concentrations and fluxes as a function of fetch in offshore flow. This model was specifically designed to complement over-water flux measurement campaigns in several ways: design and improvement of experimental methods, selection of sampling times and locations, and interpretation of results. The IBLTE model incorporates the NOAA COARE bulk flux algorithm and gas transfer model to calculate turbulence scaling parameters and gas transfer velocity, respectively. The internal boundary layer growth rate was calibrated using a large data set of air temperature and humidity modification over Lake Ontario collected during International Field Year for the Great Lakes, 1973. Current methods to estimate air-water exchange of SOCs in the Great Lakes have rarely been verified with over-water measurements. Given temporal and spatial variability of these fluxes and complexity of flux measurements, simultaneous development and improvement of models and measurement methods helps to reveal shortcomings of each approach. The IBLTE model is applied to micrometeorological SOC flux measurements performed along a transect experiment in Lake Superior in July, 2006. Measurements were collected at three over-water stations at 16, 28, and 59 km fetch under off-shore flow conditions. Measured and modeled temperature and humidity modification and vertical temperature gradients were compared for validation of the modeled vertical transport. Modification of concentration, temperature, and flux with fetch resulted from fetch-dependent equilibration, growth of the IBL, and variation of the lake surface temperature during the transect experiment. Modeled hexachlorobenzene fluxes were greater in magnitude than measured fluxes at the 16 and 28 km stations but nearly equal at the 59 km station. Both model and measurements indicated vertical gradients in HCB concentration were smaller in magnitude at the first two stations than at the third station. Measured fluxes are sensitive to uncertainty in HCB concentration measurements when the vertical gradient is small. The IBLTE model is a useful tool that allowed consideration of the entire experimental data set and enabled hypothesis-testing regarding the sources of discrepancies between model and measurement.