Mercury Transport and Fluxes in the Lake Ontario Basin
Total Page:16
File Type:pdf, Size:1020Kb
Syracuse University SURFACE Dissertations - ALL SURFACE May 2014 Mercury Transport and Fluxes in the Lake Ontario Basin Joseph Steven Denkenberger Syracuse University Follow this and additional works at: https://surface.syr.edu/etd Part of the Engineering Commons Recommended Citation Denkenberger, Joseph Steven, "Mercury Transport and Fluxes in the Lake Ontario Basin" (2014). Dissertations - ALL. 117. https://surface.syr.edu/etd/117 This Dissertation is brought to you for free and open access by the SURFACE at SURFACE. It has been accepted for inclusion in Dissertations - ALL by an authorized administrator of SURFACE. For more information, please contact [email protected]. Abstract In this dissertation I assess mercury dynamics in the Lake Ontario Basin. In four interrelated phases, ecosystem mercury concentrations and fluxes were analyzed at increasing scales. Each phase is presented in order of increasing scale, starting with a reach-by-reach analysis in the Seneca River in New York, then moving up to an analysis of the Lake Ontario watershed, followed by an assessment of Lake Ontario, and culminating with an estimate of atmospheric mercury exchange for the entire Great Lakes Basin. Phase 1 of my research is a multi-year study exploring mercury (Hg) dynamics in the Three Rivers system, with particular emphasis on the Seneca River in Central New York. In addition to bi-weekly water sampling, additional field investigations were conducted to estimate elemental Hg (Hg0) volatilization rates from the river, and to assess the impacts of zebra mussel metabolism on Hg0 volatilization. Elemental Hg volatilization was estimated at 1.3 ng m-2 hr-1 for the field season, and was positively correlated to incident solar radiation. It appears that the clearing of the water column caused by zebra mussels may increase Hg0 volatilization rates due to subsequent increased penetration of incident solar radiation, though additional limiting factors are apparent. Reach-by-reach Hg mass balances were developed for three river reaches to assess the effects of an intervening hypereutrophic lake, a zone of intense zebra mussel infestation, and contributions from Onondaga Lake which is contaminated with elevated levels of Hg. It was determined that particulate Hg (THgP) is the dominant form of Hg in the Seneca River, and Hg flux in the ecosystem is governed by flow characteristics. Intervening Cross Lake serves an important role for Hg transport in the Seneca River, reducing total Hg (THg) flux in the river by over 65% due to deposition of THgP. Methylmercury (MeHg) concentration increased over the reach of intense zebra mussel infestation, possibly due to support of anaerobic respiration as a result of the zebra mussel oxygen demand. The river reach receiving input from Onondaga Lake shows a 15% increase in THg flux. Net atmospheric Hg exchange is in the direction of deposition. However, direct atmospheric Hg deposition to the water surface plays a minimal role in Seneca River reaches, representing less than 1% of the fluvial Hg flux of the river. Overall, the Seneca River watershed is a sink for inputs of atmospheric Hg at a rate of 42 kg yr-1, and the watershed efficiently retains >85% of this Hg, exporting approximately 5 kg yr-1 to the Three Rivers confluence. For Phase 2, Hg speciation and concentrations were measured near the mouths of nine tributaries to Lake Ontario during two independent field-sampling programs. Among the study tributaries, -1 mean THg ranged from 0.9 to 2.6 ng L ; mean dissolved Hg (THgD) ranged from 0.5 to 1.5 ng -1 -1 -1 L ; mean THgP ranged from 0.3 to 2.0 ng L ; and mean MeHg ranged from 0.05 to 0.14 ng L . Watershed land-cover, total suspended solids (TSS), and dissolved organic carbon (DOC) were evaluated as potential controls of Hg. Significant relationships between THgD and DOC were limited, whereas significant relationships between THgP and TSS were common. Methylmercury was largely associated with the aqueous phase, and MeHg as a fraction of THg was positively correlated to open water land-cover. Wetland cover was positively correlated to THg and MeHg particle associations. A relation was evident between dense urban land-cover and higher THgP fractions. Phase 3 utilized the results of Phase 2 to estimate Hg flux for ten inflowing tributaries and the outlet for Lake Ontario. Total Hg flux for nine study watersheds that directly drain into the lake ranged from 0.2 kg yr-1 to 13 kg yr-1, with the dominant fluvial THg load from the Niagara River at 154 kg yr-1. Total Hg loss at the outlet (St. Lawrence River) was 68 kg yr-1. Fluvial Hg inputs largely (62%) occur in the dissolved fraction and are similar to estimates of atmospheric Hg inputs. Fluvial mass balances suggest strong in-lake retention of THgP inputs (99%), compared to THgD (45%) and MeHg (22%) fractions. Wetland land-cover is a good predictor of MeHg yield for Lake Ontario watersheds. Sediment deposition studies and coupled atmospheric and fluvial Hg fluxes indicate that Lake Ontario is a net sink of Hg inputs and not at steady-state likely due to recent decreases in point source inputs and atmospheric Hg deposition. In Phase 4, rates of surface-air Hg0 fluxes in the literature were synthesized for the Great Lakes Basin (GLB). For the majority of surfaces, fluxes were net positive (evasion). Annual Hg0 evasion for the GLB was estimated at 7.7 Mg yr-1, whereas Hg deposition to the area is estimated at 15.9 Mg yr-1. This analysis therefore suggests the GLB is a net sink for atmospheric Hg. Land-cover types that contributed most to the annual Hg0 evasion for the GLB are agriculture (~55%) and forest (~25%), and the open water of the Great Lakes (~15%). Most land-cover types displayed similar areal evasion rates, with a range of 7.0 to 21.0 µg m-2 yr-1. The highest rates were associated with urban (12.6 µg m-2 yr-1) and agricultural (21.0 µg m-2 yr-1) lands. Considerable uncertainty was noted in estimates of Hg0 evasion. Methods used to estimate Hg0 evasion vary, and results are affected based on the methods applied. A unified methodological approach could partially remedy uncertainty in estimating Hg0 fluxes. Mercury Transport and Fluxes in the Lake Ontario Basin By Joseph S. Denkenberger B.S. Chemistry, Le Moyne College, 2000 M.S. Environmental Engineering Science, Syracuse University, 2005 DISSERTATION Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Civil Engineering in the Graduate School of Syracuse University May 2014 Copyright © Joseph S. Denkenberger 2014 All Rights Reserved vi Acknowledgements This research was supported by the US Environmental Protection Agency, the Ontario Ministry of the Environment, the New York Energy Research and Development Authority, and the Syracuse Center of Excellence through the Collaborative Activities for Research and Technology Innovation (CARTI) program. Partial funding for this work was also provided by the Wen- Hsiung and Kuan-Ming Li Graduate Fellowship. In the US, field and laboratory support was provided by Mario Montesdeoca and Edward Mason at Syracuse University and Adam J.P. Effler, Bruce A. Wagner, David M. O’Donnell, Michael Spada, Tony R. Prestigiacomo, and many others at Upstate Freshwater Institute. Extra thanks go to Adam who provided technical and moral support on the river and throughout my research activities; Mario who fielded calls in the middle of the night regarding malfunctioning equipment, and also provided more than a fair share of technical and moral support; and Ed who did all the leg work for the US portion of the Lake Ontario tributary sampling, and made sure to get me outside when I needed it. In Canada, field and laboratory support was provided by Ashley Warnock, Tracie Greenberg, Tom Ulanowski, Britney Meyers, and Geoff Stupple at the University of Toronto. Extra special thanks go to Ashley for graciously fielding all my questions and data requests regarding the Canada portion of the Lake Ontario tributary sampling, for very insightful editing efforts, and for support of publication development for the Lake Ontario work. Ram Yerubandi at Environment Canada and Tony Zhang at the University of Toronto increased the accuracy and precision of this work by aiding in development of discharge estimates for the Trent River. vii Brad Blackwell and Colin Fuss donated significant time and effort to support the Seneca River evasion work, staying awake through the middle of the night to monitor sampling equipment so that I could sleep. A special mention and heartfelt thanks also go to AnneMarie Glose, who spent multiple sleepless nights sitting in a lawn chair, surrounded by poison ivy and biting insects, watching sampling equipment occasionally work and occasionally malfunction. AnneMarie also provided laboratory support. Chris S. Eckley, Mark Cohen, and Pranesh Selvendiran provided significant support as coauthors of a publication generated from the Great Lakes Basin evasion synthesis work. Marty Risch, David Gay, Tom Holsen and Kim Driscoll also provided considerable aid with the analysis of Great Lakes Basin evasion estimates. Mark Gravelding and Alain Hebert, of ARCADIS U.S., Inc., provided additional support during my dissertation development. Completion of this work may not have been possible without their aid, and my thanks go out to them for that. I thank all my Syracuse University graduate student confederates that have yet to be mentioned: Afshin Pourmokhtarian, Amy Sauer, Dimitar and Svetla Todorova, Sam Werner, Jason Townsend, Rouzbeh Berton, Habib Fakhraei, Sam Fashu-Kanu, Qingtao Zhou. Special thanks go to Afshin, Amy, and Sam W. for being some of the more supportive friends I’ve had in the past several years at Syracuse University.