PT06 Organic, Metallic, Organometallic Pollutants
Exhibit Hall
8:00 AM - Tuesday
(PT094) Nonparametric analysis of first-flush methylmercury concentrations, loads and bioaccumulations in a Northern Everglades constructed wetland.
King, J1, Adak, P1, Fink, L2, 1 Barnes-Ferland, Orlando, FL, USA2 Private Consultant, Boynton Beach, FL, USA
ABSTRACT- Stormwater Treatment Area 2 Cell 1 in the northern Everglades experienced successively increasing methylmercury (MeHg) anomalies in Fall 2000 and 2001 and Summer 2002 following reflooding after extended periods of winter dryout. The excess MeHg then propagated into the Cell 1 food chain, potentially jeopardizing fish-eating wildlife that foraged there preferentially. Expanded mercury monitoring was instituted to better define and correct the problem. From August 2002 until January 2004, ultra-trace total mercury (THg) was analyzed in weekly integrated rain samples and unfiltered ultra-trace THg and MeHg were analyzed in grab samples of the common inflow and individual outflow of each of the three treatment cells biweekly. At three, roughly equally spaced sites in each cell, ultra-trace THg was analyzed in mosquitofish and THg and MeHg in filtered surface water every 4 weeks, surficial sediment every 12 weeks, and up to 6 species of vegetation every 26 weeks. Water and sediment samples were also analyzed concurrently for other potentially influential constituents. This paper presents the results of the exploratory nonparametric univariate and multivariate regression analyses of the untransformed and log-transformed data. In summary, the concentration of soil THg was most strongly positively correlated with soil TFe, soil MeHg most strongly positively with soil THg and inversely with soil AVS, soil AVS with soil TFe and TS, and fish MeHg with soil MeHg. The inverse relationship between soil MeHg and soil AVS increased over time with the build-up of soil AVS. Lag correlation analysis indicated that the influence of inflow and rainfall chemistry on MeHg production peaks with a delay of about 2-4 weeks, while the influences of sediment, pore water, and surface water MeHg on THg bioaccumulation in mosquitofish peak with delays of roughly 6-12 weeks, 4-8 weeks, and 2-4 weeks, respectively. The multivariate analyses generally supported the univariate results.
Key words: bioaccumulation, methylmercury, fate, nonparametric analysis