No abstract

Lake Ontario, bordering both Canada and the United States, is the only Great Lake with persistent, significant levels of mirex in its biota. Some models suggested that it would take hundreds of years before mirex disappeared from the ecosystem. From 1977 to 1996 the mirex concentrations in coho and chinook salmon greater than 2 kg in weight exceeded the 0.1 mg/kg Food and Drug Administration (FDA) action level for mirex. To determine temporal trends in salmonine mirex levels, slopes and elevations of the regression lines of mirex concentration versus fish weight were compared for each of the six sampling years (1977, 1982, 1986, 1992, 1996, and 1999) by ANCOVA with weight as a covariate. Within 24 years of mirex being banned, mirex least-squares mean concentrations in salmon fillets had decreased significantly. ANCOVA revealed that the slope of the 1999 regression line was significantly flatter (P < or = 0.014) than the slopes of all other regression lines except 1996 (P = 0.966). A Tukey test revealed that the elevation of the 1999 regression line was also significantly lower than all other years (P < 0.001). Based on our results, mirex concentrations in the fillets of most salmon under the size of 12 kg are now below the 0.1 mg/kg United States FDA action level for human consumption. Models suggest that mirex reductions in biota are most likely due to the settling of mirex-contaminated organisms to the sediments and the loss of mirex from the lake through the St Lawrence River. A third mechanism is suggested as the cause of the higher rate of reduction observed in the mid to late 1990s--the control and removal of contaminated groundwater at the former Hooker Chemical site on the Niagara River, the major source of mirex in the watershed of Lake Ontario.