Mercury in Maryland: Past, Current and Future Plans

 

April 2006 Table of Contents Workshop Overview Workshop Agenda Major Findings: Atmospheric Deposition of Mercury Mercury in Ambient Air Emissions of Mercury from Maryland Power Plants Mercury Controls in Maryland Modeling Atmospheric Mercury Deposition Maryland's Mercury TMDL Program Biological and Ecological Effects       Workshop Overview Organized by Dr. Mark S. Castro, University of Maryland, Appalachian Laboratory and Dr. John Sherwell, Maryland Department of Natural Resources, Power Plant Research Program to facilitate discussion among mercury researchers active in the State of Maryland. Location: Appalachian Laboratory, University of Maryland Center for Environmental Science, 301 Braddock Road, Frostburg, MD 21532 Date: November 2 and 3, 2005 Purpose: To bring together a diverse group with interests in many aspects of mercury dynamics to develop a long-term (20 to 25 years) research agenda for the state of Maryland. Participants were interested in source characteristics, atmospheric modeling, atmospheric deposition, ambient air measurements, TMDLs, policy issues and natural resource management. The primary goal of the workshop was to develop a plan to satisfy current and future mercury assessment needs of each group and to develop a collaborative network of mercury groups within Maryland. Format: Invited speakers (see attached Agenda ) gave short presentations describing their ongoing mercury programs, highlighting their current and future needs. There was extensive discuss on each topic. Products: This website includes the workshop report and all presentations. Action Items: 1) Implement the first phases of our research agenda, 2) Look for new funding sources, such as Baltimore's LTER project, NPS, EPA and NSF, 3) Prepare to meet with mercury modelers at EPA Region 3 and ORD, and 4) Plan our next annual workshop. Workshop participants: Mark Castro (UMD-AL), John Sherwell (MDNR), Chris Moore (UVA), Eric Prestbo (Frontier Geosciences), Mark Cohen (NOAA), Winston Luke (NOAA), Steve Brooks (NOAA), George Constantz (CVI), Ray Morgan (UMD-AL), Walter Zachritz (NPS), Keith Felts (MDE), Chris Smith (MDE), Brian Hug (MDE), Ann Baines (ERM), Julie Ross (ERM), Mark Garrison (ERM), Joseph Beaman (MDE), Ann Soehl (MDE), Joe Thompson (UMD-AL), Tim Rule (MDE), Ray Downs (UMD-AL), Andrew Heyes (UMD-CBL), Andrew Heltibridle (MDE), Donald Poteat (MDE), Eva Naylor (MDE) and Abel Folarin (MDE)       Workshop Agenda Mercury in Maryland Meeting: Agenda on Nov 2-3, 2005 Day 1: Atmospheric Inputs 11:00 am: Field/Lunch Trip: Field Trip to the Piney Reservoir Air Quality Monitoring Station (MDN station MD08) 1:30 pm: Start Workshop Introductions/Goals: (Mark Castro) Major Goal: Develop a Mercury Research Program for Maryland Measurement, Modeling and Policy Needs 2:00 pm: Emissions Measurements from Power Plants in Maryland (John Sherwell) Summary of results from source emission studies 2:45 pm: Mercury Control Policies in Maryland (Brian Hug) 3:30 to 4:00 pm: Coffee/Chocolate Break 4:00 pm: Wet and Dry Deposition and Ambient Air Measurements of Mercury Past and Future Plans: (Mark Castro): Site-specific presentations about the following locations were: Piney and other location (Mark Castro) Beltsville (Mark Castro) In MD and VA (Steve Brooks) Dry Deposition (Eric Prestbo) Discussion: Summary of Future Measurement Needs Day 2: Measurements/Modeling/Effects 8:30 am: Air Measurements: Data Management, Access and Policies (John Sherwell and Mark Castro) 9:30 am: Modeling Atmospheric Mercury Dynamics (Mark Cohen and Mark Garrison) Discussion : What air measurements are needed by the modeling community? 10:30 am: Coffee Break 11:00 am: Biological and Ecological Effects Mercury in Maryland: Lakes, Streams and Biotic Communities Mercury TMDLs (Tim Rule) SERC (Andrew Heyes) State Wide Toxicology (Joe Beaman) Discussion: What does this group need from the measurement and modeling communities? 12:00 pm: Lunch at AL 1:00 pm: Biological and Ecological Effects Continued Western Maryland (Mark Castro) CVI-Mercury Research (George Constantz) 1:45 pm: Summary and Wrap-Up (Mark Castro and John Sherwell) Research Strategy for Maryland Continuation of Collaborative Working Group Dry Deposition Measurement from Synthetic Media papers Funding opportunities Data access Strategies to make this happen 3:00 pm: Adjourn w/Coffee       Summary of Presentation Topics and Major Findings Links to workshop presentations have been provided. Please click on the author's name to view a presentation. 1) Atmospheric Deposition of Mercury a. Wet Deposition (see Castro's presentation) There have been ONLY two long-term (4 to 5 years) measurement programs in Maryland; one at CBL and the other at AL. The CBL program is no longer in operation. The AL site is ongoing and is now part of the National Mercury Deposition Network (MDN). Currently, there are two MDN sites in Maryland; one at AL and the other in Beltsville. A third site will be established at SERC in Edgewater, Maryland, on the western shore of the Chesapeake Bay. A fourth site may be established at NOAA's Oxford laboratory, on Maryland's Eastern Shore. It was clear from our discussions that these four sites need to be operational for the next 25 years in order to document changes in wet deposition of mercury across Maryland. Action Items: There was some interest in establishing two more wet deposition monitoring stations in Maryland. One site should be located in the rain-shadow of the Allegheny Plateau. A good location would be in the Rocky Gap State Park. This site could be operated by AL. Preliminary data from Castro showed that Rocky Gap receives about 25% less wet deposition of mercury compared to the Piney Reservoir. There was also some support for a new inner city wet deposition site in Baltimore. However, this was not unanimous for several reasons. We have a site in the Beltsville urban area, it is difficult to determine the sources of mercury in cities, other MDN sites are located in urban areas. Despite these concerns, a new site in Baltimore could lead to a very nice collaboration with the Baltimore LTER project. MSC will try to find ways to bring these two new sites on line within the next two years. Everyone agreed, especially the atmospheric modelers, that we need event based wet deposition measurements. b. Dry Deposition It was very clear from three presentations (Prestbo, Cohen and Brooks) that dry deposition is a significant input of mercury at some locations. It was also clear that dry deposition should be measured at a few locations in Maryland. It would be best to make these dry deposition measurements simultaneously with ongoing (or planned) measurements of ambient air concentrations of mercury species and wet deposition of mercury. Action Items: There are several methods that can be used to estimate dry deposition (see Prestbo's presentation). We should use multiple measurement techniques at the same location(s). Two potential sites for dry deposition measurements include Al's site at Piney Reservoir and NOAA's Oxford Laboratory. These two sites represent different topographic situations (coastal plain vs. mountain top), and may make a nice comparative study of dry deposition. 2) Mercury in Ambient Air Three research groups (Brooks, Laurier/Mason, and Castro) have made ambient air measurements of mercury species: gaseous elemental mercury (GEM), reactive gaseous mercury (RGM) and particulate mercury (HgP) in and around Maryland. All of these measurements were made with the Tekran mercury speciation system (2537A, 1130, 1135). The group from CBL measured RGM and GEM for a few months in Baltimore, one month in Beltsville, and for one year at CBL. This group stopped making these measurements and their Tekran system (2537A and 1130) is being stored at CBL. Steve Brooks from NOAA has made summer time measurements of all three species at several locations in Maryland and Virginia and other places in the eastern US. He is in the process of preparing a comparative paper for the mid-Atlantic region. Castro has been making measurements of these three species at the Piney Reservoir since late June 2005. He plans to operate this system indefinitely and make these data available to others. Action Items: We need to get the CBL system in the field again. Fabien Laurier (CBL) would be glad to operate this system, if we can find funding for this effort. The workshop group did not clearly discuss the specific deployment of this instrument. However, Castro suggested that it be placed at Beltsville site because of the ongoing wet (MDN) and dry (CASTNET) deposition measurement programs. It is also relatively close to CBL, which is in Solomons, MD. NOAA's mercury research group would like to operate, on a full time basis, a Tekran speciation system (2537A, 1130, 1135) at their Oxford Laboratory. It is also likely that George Constantz's group will operate a Tekran mercury speciation instrument at the Canaan Valley Institute in West Virginia in the near future. Aside from these, there are no plans for additional ambient air measurements in Maryland. 3) Emissions of Mercury from Maryland Power Plants The Power Plant Research Program (PPRP) in the Department of Natural Resources has had an active mercury research program for a number of years. John Sherwell presented summary results from four field studies. The first was a plume dilution chamber study at coal- and municipal waste-fired facilities. These results showed the effect of plume chemistry on mercury speciation especially the potential oxidation role for chloride and reduction role for sulfur dioxide. The second study was an effort to study wet and dry deposition upwind and downwind of the stacks evaluated in the first study. Complex meteorology and a dry summer complicated the study, but significant dry deposition was shown as well as gradients in upwind to downwind wet deposition. The third study was a characterization of mercury in the combustion fluegas of four coal-fired units owned by Constellation. Each unit was different, with different boilers and air pollution control devices. The objective was to assess the possibility of improving mercury removal by adjusting operating procedures and to assess retrofit technologies for mercury removal. The fourth study was a statistical analysis of data from the Mercury Deposition Network and heat input (as a surrogate for mercury in coal) to geographically isolated power plants. An unexpected, but interesting, result showed that there might be significant emissions associated with start up and/or shut down activities at power plants. Further work is needed on characterizing the fate of mercury in combustion systems. This can provide important emissions information for modelers and assist operators in minimizing the cost of future mercury control requirements. Action Items: In MD FY06, MDNR will study the effects of air pollution control devices and operating conditions on mercury emissions from the Mirant Dickerson coal-fired power plant. Additional studies of mercury emissions during start up and shut down will also be planned. 4) Mercury Controls in Maryland (Hug, MDE) Maryland currently regulates mercury emissions from municipal waste incinerators. However, Maryland does not regulate mercury emissions from power plants and other sources. Currently, Maryland's total mercury emissions are 2.98 tons per year, with 1.1 tons per year from power plants. Power plants in the entire US release about 50 tons per year. Maryland expects reductions in power plant mercury emissions as a co-benefit from the federal CAIR program, which is designed to reduce emissions of NOx and Sox from power plants. At full implementation in 2015, CAIR is expected to reduce Sox and NOx emissions 73% and 61% below the 2003 levels, respectively. CAIR is also expected to lower mercury emissions from power plants. The amount of the co-benefits is uncertain. The Clean Air Mercury Rule (CAMR), which will affect mercury emissions from Maryland's power plants establishes "standards of performance" limiting mercury emissions from new and existing coal-fired power plants and creates a market-based cap-and-trade system expected to reduce nationwide utility emissions of mercury in two distinct phases. The first phase sets a 38-ton per year cap on mercury emissions from coal-fired power plants in the US by 2010. Most of this reduction is expected as a co-benefit of the CAIR program. The second phase, due in 2018, will subject coal-fired power plants to a second cap of 15 tons per year. The CAIR and CMAR are expected to reduce mercury emissions from Maryland coal-fired power plants from 1.1 tons per year to 0.49 tons per year between 2010 and 2017 and to 0.193 tons per year after 2018. Thus, long-term monitoring of atmospheric deposition inputs (wet and dry) and ambient air concentrations are needed to document the reduction in mercury emissions from coal-fired power plants in Maryland and the entire US. 5) Modeling Atmospheric Mercury Dynamics Mark Cohen (NOAA) and Mark Garrison (ERM, representing MD DNR) each provided a summary of ongoing modeling activities related to mercury transport, transformation, and deposition with a focus on determining source-specific contributions to mercury impacts at locations important to Maryland. Mark Cohen's model is based on the NOAA HYSPLIT model with algorithms added to address mercury. The modeling domain for Cohen's work is North America, and includes emissions from anthropogenic sources in the U.S. and Canada. The model estimates wet and dry deposition of mercury species at modeled receptors representing water bodies and/or watersheds of interest, and outputs include detailed estimates of source-specific contributions. Cohen's model has been published in the peer-reviewed literature and Mark Cohen has participated in international forums designed to test, compare and ultimately validate modeling approaches for estimating contributions of anthropogenic emissions to mercury deposition. Mark Garrison's model is based on EPA's CALPUFF model with algorithms that have been added to address mercury, based in part on the approaches used in Mark Cohen's version of HYSPLIT. The DNR/Garrison modeling domain includes the northeastern U.S. and is designed to focus in great detail on specific sources and specific locations. Both models (CALPUFF and HYSPLIT) are based on the Lagrangian formulation and are capable of producing detailed estimates of source-specific contributions that can subsequently be used for scenario testing to evaluate potential emissions control options. Comparing and contrasting results of both models should provide additional insights into the capabilities of models to estimate mercury impacts from anthropogenic sources. Both modelers made a plea for more data to use in model validation. Although weekly data from the mercury deposition network (MDN) are invaluable in terms of examining long-term trends, finer detail (i.e., hourly data and speciated measurements over a long time period) concentration data are badly needed to evaluate and improve mercury modeling. Action Items: There is a critical need for long-term, highly time-resolved, speciated data sets to provide model validation (currently underway at Piney and other locations). There is also an important need to generate dry deposition measurements to assess how the models perform in this area that current and planned measurements do not address. Mark Cohen and Mark Garrison will maintain communications to develop means of sharing inputs and results. Both Mark Cohen and Mark Garrison continue to work towards improving the models' simulation of the behavior of atmospheric mercury. 6) Maryland's Mercury TMDL program (Based on Tim Rule's presentation) In 2002, Maryland started to develop mercury TMDLs for selected water bodies, including numerous small recreational impoundments, larger reservoirs, such as Liberty, Prettyboy and Loch Raven and Deep Creek Lake. These water bodies were selected because tropic level four fish (small mouth and largemouth bass, walleye, etc.) contained > 235 ug/kg of mercury. The EPA approved Maryland's TMDLs for all systems except the Liberty Reservoir. Liberty's TMDL was not approved because the EPA wanted a better source assessment and plan to assure its implementation. Apparently, increased efforts by MDE and EPA to develop a better source assessment using the EPA approved approaches (REMSAD and RELMAP models) were not appropriate for the Liberty Reservoir. MDE is now collaborating with other organizations (MDNR and NOAA) to develop suitable source assessments for the Liberty reservoir, such as CALPUFF or HYSPLIT. Action Items: It appears that the Maryland TMDL program could be improved by working with the mercury modeling groups at ERM and NOAA. Based on the discussion, it was not clear if the TMDL program could use other data sets, such as mercury concentrations in surface waters or wet deposition inputs. It seems that their program constrained by EPA protocols and is based on the mercury concentrations in fish tissue, rather than surface waters. 7) Biological and Ecological Effects (Heyes, Beaman, Castro, Constantz) Andrew Heyes summarized his ongoing work associated with the METAALICUS Project at ELA in Canada, which is designed the track to movement of mercury within a terrestrial landscape and a lake ecosystem using stable isotopes of mercury. Hg200, Hg202 and Hg198 were added to upland forests, the lake surface, and fringing wetlands, respectively. After 4 years of addition, preliminary data indicate that 50% of the Hg in the lake water column is from direct additions to the lake, 30% of the Hg in young fish is from the lake isotope, upland runoff delivers 80% of the Hg deposited on the watershed to the lake and there is a large pool of old mercury in the lake sediments. As a result, we might expect lake ecosystems to have short-term and long-term responses to changes in atmospheric mercury deposition. He suggested that lakes with very large surface areas relative to the watershed areas are likely to exhibit changes in fish mercury concentrations relatively quickly after changes in atmospheric mercury deposition. These changes are likely to be first observed in the young of the year size class. In contrast, fish in lakes with large watershed areas relative to the surface water areas are not likely to see rapid changes in fish mercury concentrations due to reductions in mercury deposition because most of the mercury in the lake is derived from the watershed. However, this generalization may not be true if there are significant places for mercury methylation in watersheds. Action Items: Synthesize existing land-use and water quality databases for Maryland's reservoirs, including ratios of watershed to water areas. Identify reservoirs likely to respond quickly and slowly (the extremes) to reductions in atmospheric mercury deposition. Continue monitoring mercury concentrations in adult fish in these systems, but start measuring mercury concentrations in the young of the year fish communities. This cohort is likely to respond to the mercury reductions before other cohorts. Joseph Beaman has an ongoing monitoring program of mercury concentrations in fish in Maryland's water bodies, including the Chesapeake Bay. Between 1995 and 2000, mercury concentrations in largemouth bass and bluegill were measured in most, if not all of Maryland's freshwater reservoirs. These data were used to set consumption advisories and to identify water bodies in need of mercury TMDLs. In addition, he measured mercury concentrations in white perch and rockfish from the Chesapeake Bay and its tributaries. Currently, there are no mercury consumption advisories for white perch or rockfish. However, these fish have consumption advisories because of elevated concentrations of PCBs. Action Items: This program is critical and funding must be continued for several years in order to detect changes in fish mercury concentrations. It is also important to start measuring mercury concentrations in young of the year cohorts. Mark Castro summarized several of his mercury research projects in western Maryland. He has studying mercury dynamics in watersheds, lakes and stream ecosystems. He is particularly interested in the linkages between atmospheric mercury deposition and watershed runoff into lakes and the subsequent bioaccumulation in fish communities. His lake and bioaccumulation studies have been conducted in Piney reservoir, Deep Creek Lake and Savage River reservoir. He reported some interesting patterns in mercury bioaccumulation between water bodies. He is also looking at mercury in brook trout in streams that drain different watersheds, with and without wetlands. Brook trout in streams that drain watersheds with wetlands have significantly more mercury. He also has a six-year record of monthly total mercury concentrations in six streams that drain into the Piney reservoir, including the outlet of the reservoir. He is in the process of trying to find funding to extend this record for many years into the future. Action Items: It is very important to continue these measurements for several more years. Long-term records are needed to assess the impacts of reductions in atmospheric deposition. In addition, we also need to conduct selected studies to determine the lake and stream systems most likely to respond to changes in atmospheric mercury deposition. Heyes and Castro are interested in a comparative study of mercury dynamics in selected reservoirs in Maryland and will try to find funding for this work early next year. George Constantz described a comprehensive mercury research program for the CVI region (this presentation is not available on the website). He presented some preliminary data of mercury concentrations in different ecosystem pools, such as stream water, soils and vegetation. He is expected to expand this program in the near future. Action Items: It is important to help the CVI program with their initial measurements. It would be great to do some inter-laboratory comparisons for quality control purposes.

Return to the Maryland DNR Home Page Return to the Power Plant Research Program Page Return to the PPRP Program Activities Page This page was updated on August 3, 2006.