Long-term trends in atmospheric concentrations of pesticides, PCBs, and PAHs near the Great Lakes

We used concentration data from the Integrated Atmospheric Deposition Network (IADN) to explore the spatial and long-term temporal trends, as well as loadings, of polychlorinated biphenyls (PCBs), polycyclic aromatic hydrocarbons (PAHs), and several organochlorine pesticides in air near the Great Lakes. Results indicated that while the concentrations of most banned compounds have declined, urban effects can still be seen for PCBs and PAHs. Recent basinwide loadings estimates indicated that for hexachlorocyclohexanes (HCHs) and PCBs, inputs and releases of these compounds to and from the lakes, respectively, were decreasing over time.; In a more detailed study, we compared atmospheric concentrations of PAHs, PCBs, and several pesticides at Brule River, WI and Eagle Harbor, MI, two rural sampling sites on Lake Superior. Day-by-day regression analyses of the data showed that PAH concentrations, an indication of urban contamination, were significantly higher at Brule River than Eagle Harbor. Based on wind patterns, proximity to other cities, and further analyses, we concluded that pollution from Duluth, MN, a nearby industrial city, is influencing PAH concentrations at Brule River.; Temporal trends in gas-phase PCBs near Lakes Michigan and Superior were also studied in greater detail. Decreasing trends were found at both sites, but half-lives estimated from a first-order rate constant were four times as long as those found in previous studies. When historical data were added to the IADN data, the half-lives decreased and agreed well with half-lives in other environmental compartments. A similar, historical analysis of α- and γ-HCH data indicated that IADN data were the best predictor of trends.; We also investigated the causes of variability in gas-phase PCB and pesticide concentrations measured near Lakes Michigan, Superior, and Erie. A multiple linear regression model that incorporates temperature and time was used along with autocorrelation analyses of the residuals. Similar environmental behaviors were seen for most banned compounds, while an agricultural application cycle was present in the residual data for γ-HCH. PCBs and hexachlorobenzene showed strong residual autocorrelation patterns near Lake Michigan of an unknown origin.