Effect of Dissolved Organic Carbon on the Photoproduction of Dissolved Gaseous Mercury in Lakes: Potential Impacts of Forestry N. J. O’DRISCOLL,* ,† D. R. S. LEAN, † L. L. LOSETO, † R. CARIGNAN, ‡ AND S. D. SICILIANO § Biology Department, Faculty of Science, University of Ottawa, P.O. Box 450, Station A, Ottawa, Ontario, Canada K1N 6N5, De ´partem ent de Sciences biologiques, Universite ´ de Montre ´al, Box 6128, Montre ´al, Que ´bec, Canada H3C 3J7, and Department of Soil Science, University of Saskatchewan, 51 Campus Drive, Saskatoon, Saskatchewan, Canada S7N 5A8 The production of dissolved gaseous mercury (DGM) in freshwater lakes is induced by solar radiation and is also thought to be linked to processes mediated by dissolved organic carbon (DOC).Studies investigating these processes using comparisons between lakes are often confounded by differences in DOC content and structure. In this study, we investigated the link between DOC concentrations and DGM production by using tangential ultrafiltration to manipulate DOCconcentrations in water samples taken from a given lake. In this way, a range of samples with different DOC concentrations was produced without substantial changes to DOC structure or dissolved ions. This was repeated for four lakes in central Quebec: two with highly logged drainage basins and two with minimally logged drainage basins. On two separate days for each lake, water samples (filtered to remove >99% of microorganisms) with varying DOC concentrations were incubated in clear and dark Teflon bottles on the lake surface. DGM concentrations were measured at 3.5-h intervals over the course of 10.5 h. Levels of DGM concentrations increased with increasing cumulative irradiation for all lakes until approximately 4000 kJ m -2 (400-750 nm, photosyn- thetically active radiation (PAR)), when DGM concentrations reached a plateau (between 20 and 200 pg L -1 ). When we assumed that DGM production was limited by the amount of photoreducible mercury, reversible first-order reaction kinetics fitted the observed data well (r 2 ranging between 0.59 and 0.98, p < 0.05 with the exception of N70 100% DOC, 0% DOC, and K2 0% DOC w ith p ) 0.06, 0.10, and 0.11, respectively). The DGM plateaus were independent of DOC concentrations but differed between lakes. In contrast, photoproduction efficiency (DGM prod ) (i.e., the amount of DGM produced per unit radiation (fg L -1 (kJ/m 2 ) -1 ) below 4000 kJ m -2 PAR) was linearly proportional to DOC concentration for both logged lakes (r 2 ) 0.97, p < 0.01) and nonlogged lakes (r 2 ) 0.52, p ) 0.018) studied. Furthermore, logged lakes had a lower DGM prod per unit DOC (p < 0.01) than the nonlogged lakes. In these four lakes, the rate of DGM production per unit PAR was dependent on the concentration of DOC. The DGM plateau was independent of DOC concentration; however, there was a significant difference in DGM plateaus between lakes presumably due to different DOC structures and dissolved ions. This research demonstrates an important mechanism by which logging may exacerbate mercury levels in biota. Introduction Dissolved gaseous mercury(DGM)is believed to be primarily composed of dissolved elemental mercury (Hg 0 )(1). Produc- tion of DGM is an important process in lakes, as DGM is the primary form of mercury that can volatilize from the water surface to the atmosphere. The production and evasion of DGM from lakes is an important means by which lakes can reduce their mercury pool. Our recent research observed a direct link between the formation of DGM in lake water and water-to-air mercury flux over several diurnal cycles (2). Existing models, however, fail to predict accurately the measured DGM flux (2), which may reflect the uncertainty surrounding factors that affect rates of mercury photo- reduction and photooxidation in freshwaters. Solar radiation induces both chemical (3) and microbial (4)mercuryreduction in lake water.These reactionsprobably involve other variables, including dissolved ions, availability ofphotoreducible mercury,and DOCconcentration (3, 5-8). DOC is widely acknowledged to be important in the pho- toreduction ofmercury(9, 10),buttheexactmannerin which DOC affects DGM production is not known. Watras et al. (11)sampled surface water from 23northern Wisconsin lakes and found that increasingDOCconcentration was related to an exponentialdecrease in the ratio ofDGM to totalmercury. Xiao et al. (12) spiked water samples with 100 nM mercury- (II) as Hg(OH) 2 and HgCl2 and found that the presence of humics and fulvic acids (DOC)corresponded with significant increasesin DGMphotoproduction.Amyot et al.(5)sampled water from low DOC Artic lakes, spiked it with 1-8 mg of fulvic acids L -1 , and exposed it to varying levels of solar radiation without observinganysignificant changes in DGM. Yet, in another study Amyot et al. (3) compared a low DOC lake (2.2 m g L -1 ) with two higher DOC lakes (8.7 and 5.0 mg L -1 ) and found that hourly DGM production was 1.8 and 7.7 times higher in the low DOC lake than in the two high DOC lakes. In contrast to these results, O’Driscoll et al. (2) report DGM concentrations in surface water in southern Nova Scotia that were 2-4 tim es higher in a high DOC (10.5 m g L -1 ) lake than in a low DOC (3.6 mg L -1 ) lake over a 48-h period (n ) 576). Part of the reason for such conflicting reports on the role of DOC in mercury photoreduction processes is that DOC maydiffer between lakes, not onlyin concentration but also in structure and composition. Thus, comparing DGM production in lakesofdifferingDOCcontentsisconfounded by inherent differences in DOC structure and the unknown impactthatthesemayhaveon theprocessesbeingexamined. Consequently, to study the effects ofDOC concentration on DGM production accurately, the DOC within a lake must be changed in concentration without changingother variables. Such an approach would allow for the assessment of photoproduction rateswhileavoidingtheconfoundingeffects of DOC structure and dissolved ions. *Corresponding author phone: (613)562-5800, ext. 6655; fax: (613)562-5486; e-mail: nodrisco @ science.uottawa.ca. † University of Ottawa. ‡ Universite ´ de Montre ´al. § University of Saskatchewan. Environ. Sci. Technol. 2004, 38, 2664-2672 2664 9 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 38, NO. 9, 2004 10.1021/es034702a CCC: $27.50  2004 American Chemical Society Published on Web 03/27/2004