In this paper, we summarize the data and methods used to estimate atmospheric exchange of polychlorinated biphenyls (PCBs) and trans-nonachlor with Lake Michigan. This work was conducted as part of the Lake Michigan Mass Balance (LMMB) study. For the atmospheric component of the LMMB, more than 400 gas- and particulate-phase samples were collected at eight sites on the shore around the lake (shoreline) and at 14 sites on the lake (over-water). We review the quality of the data set; describe the concentrations in atmospheric gas and particulate phases; report local, instantaneous, net gas fluxes; and estimate annual deposition of the particle-associated compounds. The quality of the data set is high except for a subset of over-water samples where PCB contamination is suspected. Gas-phase trans-nonachlor concentrations (although not the resulting gas fluxes) are inversely correlated with latitude and positively correlated with temperature. Gas-phase sigmaPCBs (sum of 98 congener groups) are highest in concentration at the Chicago site and lowest at the Sleeping Bear Dunes site. The resulting sigmaPCB gas fluxes exhibit a seasonality that reflects elevated summertime gas-phase concentrations not compensated by temperature-corrected Henry's law coefficients. Particulate-phase deposition is much smaller in magnitude than gas fluxes, for either compound. Gas and particulate fluxes are comparable only at the Chicago site and only when large (> 10 microm) particulates are considered.

Atmospheric deposition of anthropogenic air pollutants contributes to degradation of water quality in the Great Lakes and other water bodies and is indicative of atmospheric pollution. In this paper, we discuss deposition of three air pollutants:  atrazine; total phosphorus; and nitrogen (total Kjeldahl nitrogen and nitrate) to Lake Michigan. Throughout 18 months in 1994−1995, over 600 atmospheric samples (gas, particulate, and precipitation combined) were collected and analyzed for persistent organic pollutants and nutrients. Here the measurements and modeled deposition estimates are presented for atrazine, phosphorus, and nitrogen. Results indicate that concentrations of atrazine in precipitation have remained constant over 5 years (0.10−0.40 μg L-1), consistent with the nearly constant sales rate of the herbicide over that time period. Actual loading of atrazine to the lake was less in 1994−1995 (1.04 × 103 kg y-1) than in 1990−1991 (2.6 × 103 kg yr-1). This difference in loading is due to lower overall precipitation in 1994−1995. Phosphorus concentrations in precipitation, on the other hand, have decreased from an average of 57 μg as P L-1 in 1976 to an average of 6.36 μg as P L-1 for 1994−1995. Nitrate deposition has decreased by a small, but not statistically significant, amount since the late 1970s.