W4 AM Environmental Fate of Mercury
Wednesday, 16 November 2005: 8:00 AM - 11:40 AM in Ballroom 4

(AAA-1113-333544) Gross photo-reduction kinetics of mercury in temperate freshwater lakes and rivers: Development of a general model for DGM dynamics.

O'Driscoll, N¹, Siciliano, S², Lean, DR³, Amyot, M¹, ¹ University of Montreal, Montreal, Quebec, Canada² University of Saskatchewan, Saskatoon, Saskatchewan, Canada³ University of Ottawa, Ottawa, Ontario, Canada

ABSTRACT- Previous published methods to quantify mercury photo-reduction only provide data for net-photo-reduction, since photo-oxidation processes occur simultaneously. In this research we combine continuous dissolved-gaseous mercury (DGM) analysis with a LUZCHEM ICH-2 photo-reactor and quartz sparger. Using this novel technique, DGM gross photo-reduction kinetics for constant UVB and UVA irradiations were measured every 5 minutes over a period of 23 hours for sterile freshwater from diverse environments. Gross photo-reduction proceeded for the initial 200 minutes after which total mercury was depleted in the sample and photo-reduction decreased to negligible amounts. Photo-reduction was not linked to a loss of DOC fluorescence, with substantial losses in DOC fluorescence observed during the incubations for UVA radiation but not for UVB. Pseudo first order reaction kinetics fit the data very well (r2 > 0.87 for all samples) and rate constants were derived by curve fitting. Rate constants were 10 times lower for lakes (mean k_UVB = 6.91 x 10-5 s^-1; SE = 1.20 x 10^-6 and mean k_UVA = 6.09 x 10^-5 s^-1; SE = 3.97 x 10^-7) compared to rivers (mean k_UVB = 2.46 x 10^-4; SE = 3.91 x 10^-6 and mean k_UVA = 2.23 x 10^-4 s^-1; SE = 8.19 x 10^-6). Here we propose a generalized model for mercury photo-redox dynamics for temperate lake and river systems using a reversible first order reaction model. The lake model was validated using principal axis analysis to compare observed and predicted DGM data (n=279) from a variety of lake sites in Nova Scotia and Central Quebec. Principal axis analysis found a linear fit (correlation = 0.80; slope = 2.04) between predicted and observed environmental DGM values when log normalized. Error on the predicted values was attributed to estimates of available reducible mercury and the effect of DGM volatilization on observed data.

Key words: Mercury, Photo-reduction, Modeling, volatilization