The National Mercury Monitoring Network: A Report from the May 2008 Workshop

1. The National Mercury Monitoring Network:A Report from the May 2008 Workshop May 5 - 7, 2008, Annapolis, MD Richard Artz1, Mark Cohen1, Jawed Hameedi2, Tony Lowery3, Winston Luke1, David Schmeltz4 on behalf of many! 1. NOAA Air Resources Laboratory 2. NOAA National Centers for Coastal Ocean Science 3. NOAA Fisheries – National Seafood Inspection Lab 4. USEPA Clean Air Markets Division presentation at the2008 National Water Quality Monitoring Conference, Atlantic City, NJ, May 19

2. May 5 - 7, 2008, Annapolis, MD Steering Committee for the National Mercury Monitoring Workshop* * The composition of the Steering Committee may change (e.g., get larger) as this effort goes forward 2

3. May 5 - 7, 2008, Annapolis, MD Participants at the National Mercury Monitoring Workshop (~50) * * Due to late additions, this may not be a complete list 3

4. May 5 - 7, 2008, Annapolis, MD Please Note: • Not an “official” report from the Workshop • Has not been approved by Steering Committee or Participants • Content of this presentation taken directly from: • materials prepared by the Steering Committee for the meeting • materials presented at the meeting by participants. • Impossible to present the full breadth and scope, but hopefully this brief summary will give you a sense of this important effort 4

5. SETAC North American Workshop on Mercury Monitoring and Assessment, Pensacola, FL May 5 - 7, 2008, Annapolis, MD Bills introduced in House and Senate 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 Mason et al., 2005, ES&T, 39:14A-22A Harris, et al., 2007, CRC Press 5

6. May 5 - 7, 2008, Annapolis, MD The workshop was part of an ongoing effort to enhance monitoring of ecological responses to mercury deposition through coordination of existing monitoring efforts and implementation of new monitoring, if funding resources become available. Huntington Wildlife Forest, NY, photo courtesy of Charley Driscoll Workshop participants attempted to refine the scientific and technical basis for designing a national mercury monitoring network and build agreement around implementation approaches. Acadia National Park, photo courtesy of Dave Evers 6

7. May 5 - 7, 2008, Annapolis, MD Specific goals for the workshop were to: • Distill recommendations from previous work on measurement parameters for tracking ecological responses to mercury; • Share information on existing North American sites and programs that monitor ecological endpoints of mercury contamination (e.g., air, water, watershed, sediments, biota); • Identify mercury monitoring data gaps and establish options for filling those gaps. 7

8. May 5 - 7, 2008, Annapolis, MD The meeting was organized around six questions: 1 – What are the goals and objectives of a national mercury monitoring network? 2 – What are the major elements needed to meet network goals and objectives? 3 – What is already in place? 4 – What are the major gaps? 5 – How can gaps be addressed? 6 – How do the pieces fit together? 8

9. May 5 - 7, 2008, Annapolis, MD The meeting was organized around six questions: 1 – What are the goals and objectives of a national mercury monitoring network? 2 – What are the major elements needed to meet network goals and objectives? 3 – What is already in place? 4 – What are the major gaps? 5 – How can gaps be addressed? 6 – How do the pieces fit together? 9

10. Draft goals and objectives of a national mercury monitoring network* Monitoring Goal Establish an integrated network to systematically monitor, assess, quantify, and report on indicators of nationwide changes in atmospheric mercury concentrations and deposition, and concentrations of mercury species in land, water, and biota in coastal and freshwater ecosystems in response to changing mercury emissions over time. Monitoring Objectives • Establish baseline mercury concentrations in multiple ecosystem compartments that document environmental conditions prior to implementation of atmospheric mercury control measures to detect future ecosystem changes that may be attributable to mercury controls over time. • Track spatial patterns and long-term, temporal changes in mercury concentrations in specific ecosystem compartments: airsheds and watersheds, aquatic ecosystems, aquatic biota, and wildlife as mercury emissions controls are implemented. • Provide data to assess the linkages between atmospheric mercury emissions and methyl mercury concentrations in biota and how these change over time. • Document changes in biological indicators of mercury exposure and effects relative to changes in mercury loadings to ecosystems. • Provide mercury and ancillary data to evaluate predictive and diagnostic models and to advance the development of mercury cycling models and models to assess source receptor relationships. • Assess potential ecological harm and linkages to air emission sources for sites and wildlife of conservation concern. • Connect national mercury monitoring efforts to other monitoring programs in North America and adjacent waters where feasible. * Pre-Workshop Draft 10

11. May 5 - 7, 2008, Annapolis, MD The meeting was organized around six questions: 1 – What are the goals and objectives of a national mercury monitoring network? 2 – What are the major elements needed to meet network goals and objectives? 3 – What is already in place? 4 – What are the major gaps? 5 – How can gaps be addressed? 6 – How do the pieces fit together? 11

12. Slide from Rob Mason, based on Mason et al. (2005) Proposed Design • Propose a combination of “intensive sites” and “cluster sites”: • Intensive sites are those where detailed studies will be done to track changes and assess the cause of any changes • Cluster sites will allow data from the intensive sites to be extrapolated to a broader area, and extrapolate results of the detailed investigations across ecosystems of similar atmospheric input • Propose 10-20 intensive sites in the U.S. • Each intensive site would have 15-20 cluster sites surrounding it • Intensive sites would be chosen to represent the different ecoregions of the U.S. intensive site 12

13. Indicators in green would be monitored at intensive sites only slide from Rob Mason, based on Mason et al. (2005) Harris et al. (2007) The Indicators Air & Deposition Water and Sediment • Continuous speciated atmospheric concentrations • Total wet and dry Hg deposition &flux • Total Hg weekly wet deposition/flux • Total and methyl Hg in throughfall • Total and methyl Hg in litterfall • Total Hg in snowpack • Mercury evasion/flux • Watersheld inputs/yields • Total and MeHg in soil • Forest floor surveys • Total and MeHg, %MeHg in sediments (seasonal) • Instantaneous sediment methylation/demethylation rate • Total and methyl Hg accumulation in cores • Total and methyl Hg in surface water (seasonal) • Water column Hg & MeHg profiles 13

14. slide from Rob Mason, based on Mason et al. (2005) Harris et al. (2007) The Indicators, cont. Wildlife Aquatic Biota • Total Hg in blood, feathers, eggs (as appropriate) • Potential Indicator Species • Comparison across habitats: Belted kingfisher • Terrestrial: Racoon, Bicknell’s thrush • Riverine: Mink • Lake: Common loon • Lake/coastal: Herring gull, Common tern • Wetland: Tree swallow • Estuarine: Sharp-tailed & seaside sparrows • Marine nearshore: Harbor porpoise • Marine off-shore: Storm petrel • Total and MeHg in phyto/zooplankton • Total and MeHg in estuarine benthic invertebrates • Total and methyl Hg in whole prey fish (YOY) • Total Hg in muscle of piscivorous fish Indicators in green would be monitored at intensive sites only 14

15. current list of (evolving) site selection considerations (Workshop Steering Committee) Baseline data and infrastructure • Longer-term mercury data • Existing facilities and infrastructure to support the monitoring program Will we see and be able to understand a change? • Sensitive to mercury inputs • Expected to exhibit large changes due to changes in Hg deposition • Near emission sources and may receive elevated Hg deposition • Clearly defined response – few if any confounding factors • Useful testbed for evaluation of atmospheric Hg models • Useful testbed for evaluation of ecosystem Hg models Model evaluation • Overall, want nationwide geographical distribution • Overall, want range of characteristic response times • Overall, want some background sites for characterizing global Hg inputs Want a range of site types • Within common loon breeding range • Endangered, threatened or candidate species at risk to Hg Other site issues

16. May 5 - 7, 2008, Annapolis, MD The meeting was organized around six questions: 1 – What are the goals and objectives of a national mercury monitoring network? 2 – What are the major elements needed to meet network goals and objectives? 3 – What is already in place? 4 – What are the major gaps? 5 – How can gaps be addressed? 6 – How do the pieces fit together? 16

17. May 5 - 7, 2008, Annapolis, MD What is already in place? What are the major gaps? “MercNet” meta-database developed • Based on major environmental monitoring databases from EPA, NOAA, USGS, USFWS, Biodiversity Research Institute • Records of approximately 200,000 mercury sampling events across the United States • Various media: Atmosphere, Water, Sediment and soil, Vegetation, Invertebrates, Fish, Birds, Reptiles and Amphibians, Mammals • Time span of records is from 1896 to 2007 17

18. Slide from Colleen Haney, EPA

19. Slide from Colleen Haney, EPA

20. Slide from Colleen Haney, EPA

21. slide from Dave Evers, Biodiversity Research Institute Avian piscivores Common Loon (>4,000 records), Bald Eagle (900 records), Wading birds (450 records) Key Group because risk and injury assessments can be made based on known LOAELs

22. slide from Dave Evers, Biodiversity Research Institute Avian invertivores Songbirds (>3,000 records) Key Group because risk and injury may be greatest and most diverse

23. slide from Dave Evers, Biodiversity Research Institute

24. slide from Dave Evers, Biodiversity Research Institute

25. Large Point Sources of Mercury Emissions Based on the 2002 EPA NEI and 2002 Envr Canada NPRI* size/shape of symbol denotes amount of mercury emitted (kg/yr) 5 - 10 10 - 50 50 - 100 100 – 300 300 - 500 500 - 1000 1000 - 3000 color of symbol denotes type of mercury source coal-fired power plants other fuel combustion waste incineration metallurgical manufacturing & other 2002 U.S. and Canadian Emissions of Total Mercury [Hg(0) + Hg(p) + RGM] 25 * Note – some large Canadian point sources may not be included due to secrecy agreements between industry and the Canadian government.

26. There are a number of existing atmospheric measurement networks and sites Clean Air Status and Trends (CASTNET ) O3, SO2 , HNO3, major ions in particulate NADP Atmospheric Mercury Initiative (emerging) speciated Hg, dry deposition estimates NAMS / SLAMS state/local criteria pollutants O3, PM, SO2, NOx and/or Pb National Mercury Monitoring Network (intensive + cluster sites) IMPROVE particulate composition and visibility Mercury Deposition Network (MDN) mercury wet deposition other networks, e.g., AIRMoN, WACAP, IADN, etc... National Trends Network (NTN-NADP) major ion wet deposition individual research / monitoring sites not necessarily affiliated with existing networks 26

27. 2008 NADP Atmospheric Mercury Initiative Network Sites 2008 Funded Speciated Hg Network NOAA/EPA—Beltsville, MD NOAA—Canaan Valley, WV NOAA—Grand Bay, MS Ohio Univ—Athens OH ERG—Underhill, VT Univ. Maryland—Frostburg, MD Clarkson Univ./NYSDEC- Rochester, NY Clarkson Univ/Syracuse Univ. Newcomb, NY slide from David Schmeltz, EPA Clean Air Markets Division

28. The presence of existing speciated atmospheric mercury measurements is an important siting consideration always or usually included usually co-located with this measurement sometimes or partially included measurements needed to assess and understand atmospheric mercury 28

29. May 5 - 7, 2008, Annapolis, MD The meeting was organized around six questions: 1 – What are the goals and objectives of a national mercury monitoring network? 2 – What are the major elements needed to meet network goals and objectives? 3 – What is already in place? 4 – What are the major gaps? 5 – How can gaps be addressed? 6 – How do the pieces fit together? 29

30. May 5 - 7, 2008, Annapolis, MD How can gaps be addressed? How do the pieces fit together? Initial Characterization of Potential “Intensive” Sites • In some cases, places with substantial mercury monitoring (historical and/or current) • In some cases, places with some mercury monitoring, but would have to be expanded • In some cases, places with little or no mercury monitoring, but considered to be important for geographical or other reasons • List is not complete... • Characterization of potential intensive sites is not complete... 30

31. Slide from Tim Sharac, EPA

32. PRELIMINARY DATA Fish Sediment Water Slide from Tim Sharac & Colleen Haney, EPA Huntington Forest, NY Underhill, VT Steubenville, OH Piney Reservoir, MD Athens, OH Beltsville, MD Legend Invertebrates Canaan Valley Institute (CVI) Plants Wildlife 32

33. slide from Dave Evers, Biodiversity Research Institute

34. Slide from Tim Sharac, EPA What are the relative advantages and disadvantages of these and other potential intensive sites? • Discussion has begun, but a process… • Sites are not all fully characterized • Some progress was made at the meeting towards focusing in on ~20 potential intensive sites • However, further discussion is needed before advancing even this preliminary set of potential sites • Stay tuned!

35. May 5 - 7, 2008, Annapolis, MD Thanks! 35

36. Extra Slides for background, context, and in case specific questions come up 36

37. May 5 - 7, 2008, Annapolis, MD Few watershed mercury models, and little experience in applying mercury models to different watersheds. Here are some characteristics and data needs for watersheds that would be important to the successful application and the rigorous testing of watershed mercury models* • Relatively uniform and well characterized land cover; • Well-characterized meteorology and hydrology; • Low and relatively uniform concentrations of total suspended solids; • Relatively long time-series of mercury speciation measurements in atmospheric deposition and stream water, and ancillary watershed inputs and stream output measurements collected at an interval that is adequate to characterize seasonal variations in mercury chemistry and water quality; • Well-characterized food chain & total/methyl mercury content of organisms of that food chain; • Well-characterized with respect to soils/vegetation, & their transformations involving mercury; • Information on watershed mercury processes such as evasion, throughfall, litterfall, methylation, demethylation, and soil adsorption. * Draft prepared by Charlie Driscoll, Syracuse University 37

38. Issues relating to ability to use data from the site for atmospheric model evaluation • (a) Degree to which the terrain is atmospherically "simple" -- at least for initial sites -- so that uncertainties in meteorological data do not overwhelm models. For example, if there are sub-grid scale weather phenomena important to the Hg model that cannot be resolved practically by a meteorological model, then it will be difficult if not impossible to use the data at the site for model evaluation or improvement • (b) Nature and extent of existing efforts to simulate meteorology in the local and regional environment, e.g., if some one is already running, say, MM5 on a fine grid in the region and is willing to collaborate and share data • (c) Degree of ability to characterize emissions sources in region contemporaneously with the measurements. Since an atmospheric model relies on emissions inventories as an input, preferred sites are those where the local and regional sources are -- or can be -- well characterized. An example of this would be a site in the region of a coal-fired power plant that has elected to install a continuous, speciated mercury emissions monitor 38

39. Issues relating to ability to use data from the site for atmospheric model evaluation • … continued • (d) Tendency to get well defined "episodes”, including source-related episodes. In general, good to have a wide range of concentrations of each atmospheric mercury form to evaluate the model against, as opposed to a site with relatively constant concentrations of atmospheric mercury • (e) Relative tendency to get "simple" plume impact episodes -- e.g., from one single well defined source at at time -- as opposed to getting "complex" plume impact episodes from multiple, diverse sources. • (f) Not too close to sources (e.g., less than 5-10 km and meteorological uncertainties may be too dominant). And, if too close to tall-stack source, plume will not even have hit the ground yet. • (g) Not too far from sources. At distances greater than ~100 km or so, may be difficult to "see" a source, even if wind blowing directly from the source to the site. • (h) Degree of atmospherically relevant interfering activities at site, e.g., amount / proximity of onroad or offroad traffic • (i) Ability to erect a stable tower, e.g., at least 10 meters tall. Ability to erect or existence of taller tower for the possibility of making measurements aloft (e.g., 100 m tower). 39

40. current list of (evolving) overall site selection considerations (Workshop Steering Committee) • Longer-term mercury data • Sensitive to mercury inputs • Expected to exhibit large changes due to changes in Hg deposition • Existing facilities and infrastructure to support the monitoring program • Overall, want range of characteristic response times • Overall, want some background sites for characterizing global Hg inputs • Useful testbed for evaluation of atmospheric Hg models • Useful testbed for evaluation of ecosystem Hg models • Near emission sources and may receive elevated Hg deposition • Clearly defined response – few if any confounding factors • Within common loon breeding range • Endangered, threatened or candidate species at risk to Hg

42. USGS Mercury Sensitivity Map for Aquatic Ecosystems in the Contiguous 48 United States Myers et al. Science 2007

43. Water Chemistry pH DOC Sulfate ANC Total Hg MeHg Others Hydric Soils Hg Dep. Metrics used to create the map

44. Controls on bioaccumulation: • Food web structure • length of food chain • benthic vs. pelagic • Water chemistry – DOC, Cl • Lake stratification • Transport of MeHg from sites of methylation Deposition Transport Deposition Watershed retention Methylation Methylation Controls on deposition and transport: Deposition: Strength and proximity of sources Source type Atmospheric chemistry Transport: Landuse Geology/soil type Catchment:lake area ratio Bioaccumulation • Controls on net methylation: • “Age” and complexation of Hg • Basin morphometry - • extent of shallow sediments and wetlands • surface to volume ratio • Water and sediment chemistry – • sulfate, DOC, nutrients • Temperature • Drying and rewetting of soils/sediments Conceptual Diagram of the Controls on MeHg in Fish slide from Cindy Gilmour, Smithsonian Environmental Research Center, MD

45. The Mercury Cycle in the Chesapeake Bay Watershed slide from Cindy Gilmour, Smithsonian Environmental Research Center, Maryland

46. Fish consumed by people YOY fish Amphibians Fish and insect-eating birds Chesapeake Bay Program Recommended Key Indicators slide from Cindy Gilmour, Smithsonian Environmental Research Center, Maryland Atmospheric Hg speciation Wet and dry deposition Watershed yields Hg and MeHg profiles in water Sediment and soil accumulation rates Hg and %MeHg in sediments

47. We need to consider multiple dietary pathways for MeHg biomagnification (i.e., documented locations where invertivore >> piscivore Hg levels) slide from Dave Evers, Biodiversity Research Institute

48. Hg HgBA Hg slide from David Krabbenhoft, USGS The Mercury Cycle Emissions Hg(II) Hg0 Deposition Methylation Methylation Sulfate CH2O SRB Bacteria MeHg Bioaccumulation CO2 Sulfide Demethylation

49. DLoad DMeHg Using Mercury Sensitivity to Anticipate Response slide from David Krabbenhoft, USGS

50. Biotic Data Summary Statistics (not including fish) Data origin: USFWS – 19,600 records USGS – 4,200 records (CEE-TV) USEPA - 800 records NOAA - 3,600 records BRI – 11,600 records slide from Dave Evers, Biodiversity Research Institute