Dry Deposition of Semivolatile Organic Compounds to Lake Michigan

Abstract

The Lake Michigan Urban AirToxics Study(LMUATS) was carried out from July 8 to August 9, 1991, in the southern Lake Michigan basin. The investigation was the first aimed at assessing the impact of the Chicago/Gary urban plume on the deposition of hazardous air pollutants (HAPs) to Lake Michigan. The concentrations of a large number of semivolatile organic compounds (SOCs) were measured at four locations (Kankakee, IL ; Chicago, IL ; over Lake Michigan onboard the R/V Laurentian ; and South Haven, Ml). These sites were chosen as they are often linked by air mass transport with air masses reaching South Haven, after passing through Chicago, often having an over-water path to the downwind site. The University of Michigan R/V Laurentianwas also deployed off-shore of Chicago to measure the concentrations of HAPs as they were transported out over the water. A hybrid-receptor deposition (HRD) modeling approach, which utilized measured atmospheric concentrations ata a receptor site together with observed meteorological data in a lagrangian dispersion modeling framework, was used to estimate both gas exchange across the air-water interface and particle dry deposition of SOCs to Lake Michigan. The atmospheric concentrations of selected pesticides, PCBs, and PAHs measured during the LMUATS were used as input to the HRD model. The comparison between the measured and calculated ambient concentrations gives reasonable results with measured/calculated concentration ratios in the range of 0.3-1.9 for pesticides, 0.9-2.6 for PCBs, and 0.5-3.8 for PAHs. A comparison of deposition fluxes obtained in this study with those found in literature reveals large differences for some compounds, especially for those that are primarily in the particle phase. This finding suggests that assuming constant values of critical parameters controlling the transfer processes of gaseous and particulate SOCs from the atmosphere to the water surface may result in estimates with large uncertainties.