TAR Mercury Fate
 nd Transport Final
Progress Review Workshop

November 21, 2003
Washington.lJC,

-------
                  STAR Mercury Fate and Transport Final Progress Review Workshop
                                   Table of Contents

Introduction	v

Executive Summary	1

Agenda	13

Participants List	15

Watershed Influences on Transport, Fate, and Bioavailability of Mercury in Lake Superior	19
    James P. Hurley

Factors Controlling the Production, Fate, and Transport of Mercury and Methylmercury
in Aquatic Systems	21
    Robert P. Mason

Response of Methylmercury Production to Changes in Hg Loading: A Comparison of Hg
Isotope Addition Studies	23
    Cynthia C. Gilmour

Methylmercury Sources to Lakes in Forested Watersheds: Has Enhanced Methylation
Increased Mercury in Fish Relative to Atmospheric Deposition?	25
    Edward B. Swain

Cycling of Mercury in Saginaw Bay Watershed	27
    Jerome Nriagu

Chemical and Biological Control of Mercury Cycling in Upland, Wetland, and Lake
Ecosystems in the Adirondack Region of New York	28
    Charles T. Driscoll

Processes Controlling the Chemical Speciation and Distribution of Mercury From
Contaminated Mine Sites	29
    Gordon E. Brown, Jr.

Mercury and Methylmercury Cycling in the Coastal/Estuarine Waters of Long Island Sound
and Its River-Seawater Mixing Zones	31
    William F. Fitzgerald

Pathways, Enzymes, and Trace Metals:  The Importance of Understanding the Biochemistry
of Mercury Methylation	33
    Francois M. Morel

Quantifying the Role of Significant Natural Sources and Sinks in the Biogeochemical Cycle
of Mercury	35
    Mae Sexauer Gustin

Index	37
       The Office of Research and Development's National Center for Environmental Research           iii

-------
                   STAR Mercury Fate and Transport Final Progress Review Workshop
                                      Introduction

    In December 1997, the U.S. Environmental Protection Agency's (EPA) Mercury Study Report to Con-
gress  identified mercury as a critical human health and environmental problem needing  additional scientific
and technical research. Mercury poses risks to humans and wildlife, particularly because it concentrates in the
tissues of animals as it moves up through a food chain. Observed adverse effects in mammals, fish, and birds
have  included behavioral  and  neurological abnormalities,  impaired growth and development, reproductive
abnormalities such as fetal deformities, and in some cases complete reproductive failure. Fish consumption is
the dominant exposure pathway for humans and wildlife. In some cases, mostly in the past, wildlife were killed
by extreme environmental concentrations of mercury, for example, from seed grains treated with mercury or
from unusually severe instances of waste release. The most pervasive wildlife effects involve reduced breeding
success, which now poses severe consequences for water birds  throughout North America and some  en-
dangered species, including panthers, in the Florida Everglades.

    More U.S. waters are closed to fishing because of mercury contamination than any other toxic contami-
nation problem. Because fish consumption can be a significant route of human exposure to toxic chemicals, the
U.S. Food and Drug Administration (FDA) issues guidelines to states on fish contamination levels that warrant
closures of fishing waters to commercial or recreational fishing. The principal sources of fish contamination,
perhaps surprisingly, are air emissions of mercury from coal burning power plants, municipal waste incinera-
tors, and other industrial sources.

    This meeting will be the third and final Science to Achieve Results (STAR) Grant Progress Review for the
grants from the 1999 Request for Applications (RFA) on Fate, Transport, and Transformation of Mercury in
Aquatic and Terrestrial Environments. The goal of the review will be to ascertain the progress of the research
in addressing the science  questions and issues  described in the  RFA  such as: (1) For a given amount of
mercury transported into a watershed, what is the predicted concentration of methylmercury in fish? How do
mercury and methylmercury spatially distribute across the terrestrial and aquatic components of a watershed?
What controls bioavailability of mercury in the food chain? (2) What environmental and biochemical variables
control transformation of mercury to methylmercury? What environmental variables control the reduction of
divalent mercury to elemental mercury in soils, sediments, and surface waters? (3) How does mercury cycling
vary within different geographic regions of the United States (South Florida, Great Lakes, Northeast or West)?
How might the variability be accounted for (resource types [wetlands], temperature regimes, microbial com-
munities)?  STAR researchers are studying the risks created by mercury in our environment so that we  can
better understand how to eliminate them.

    For further information on  the U.S. EPA's STAR Mercury Research Program, please contact William G.
Stelz, ORD/NCER, Assistant Center Director for Multimedia, by telephone at (202) 564-6834, or by e-mail at
stelz.william@epa.gov.
           The Office of Research and Development's National Center for Environmental Research

-------
                 STAR Mercury Fate and Transport Final Progress Review Workshop
                U.S. EPA National Center for Environmental Research

                       STAR Mercury  Fate  and Transport
                        Final Progress Review Workshop

                                    Washington, DC
                                   November 21, 2003

                                Executive  Summary
OVERVIEW

The U.S. Environmental  Protection Agency's (EPA) Science to Achieve Results (STAR) Program's
Mercury Fate and Transport Final Progress Review Workshop was held on November 21, 2003. The
purpose of this final review was to determine the progress of the  research in addressing the science
questions and issues described in the Fiscal Year (FY) 1999 Request for Applications (RFA) on Fate,
Transport, and Transformation of Mercury in Aquatic and Terrestrial Environments. STAR researchers
are studying the risks created by mercury in the environment to better understand how to eliminate those
risks. Approximately 75 people attended the workshop.

WELCOME AND INTRODUCTORY REMARKS

William Stelz, U.S. EPA

Mr. William Stelz, Office  of Research and Development (ORD), Assistant Director for Multimedia at the
National Center for Environmental Research, welcomed participants to the final progress review  of the
STAR Mercury Program.  He asked participants to concentrate on findings, results as related to the 1999
RFA, and future research  in their areas of expertise. He noted that the STAR 1999 Mercury Solicitation
research focused on mercury transport in a watershed context, and that its specific objectives were to:  (1)
increase our ability to trace mercury from its entrance into the ecosystem, through its biogeochemical
cycling, to the concentration of methylmercury (MeHg) in fish tissue; and (2) promote the development of
risk management strategies  based on sound science. Mr. Stelz gave a brief introduction to the  STAR
2001 Mercury Solicitation,  which focused on mercury  emission sources, atmospheric processes, and
deposition. He closed his  comments by noting that ORD is focusing its research efforts on the Mercury
Multi-Year Plan (MYP), which supports Agency goals. The Mercury MYP aids ORD as a planning and
communication tool, and integrates ORD intramural and extramural research efforts. The two long-term
goals defined in the MYP  are to:  (1) reduce and prevent the release of mercury into the environment; and
(2) understand the  transport and  fate of mercury from  release to the receptor and its effects on the
receptor. The plan can be viewed at: http://www.epa.gov/osp/myp/mercury.pdf
       The Office of Research and Development's National Center for Environmental Research

-------
                  STAR Mercury Fate and Transport Final Progress Review Workshop
FORMATION/TRANSPORT OF METHYLMERCURYIN ECOSYSTEMS AND WATERSHEDS

Watershed Influences on Transport, Fate, and Bioavailability of Mercury in Lake Superior
James Hurley, University of Wisconsin

Dr. James Hurley stated that the overall goal of this research is to assess the importance of watersheds in
controlling sources, transport, fate, and bioavailability of mercury (Hg) in the Lake Superior basin. The
specific research objectives are to:  (1) determine the speciation and bioavailability of Hg transported to
Lake Superior by rivers that drain representative and contrasting watersheds; (2) determine the watershed-
specific characteristics (land cover, land use, surficial deposits) that control physical and chemical forms
of Hg transported by rivers; (3) identify key mechanisms controlling Hg bioavailability and speciation in
near shore zones relative to open lake regions; and (4) provide  process-level information to complement
concurrent development of Hg fate and transport models of the Lake Superior ecosystem. The approach
combined field and laboratory studies with modeling efforts. The majority of time was spent in the field
because there was not a lot of data on Hg cycling in Lake Superior. Field work included sampling within
homogenous subwatersheds, riverine mixing zones, and deep water regions of Lake Superior.

This research represents the first comprehensive overview of Lake Superior for Hg. Data indicate that Hg
species concentrations are quite low  in Lake  Superior, similar to Lake Michigan and oceanic waters.
Nearshore mixing zones are  sites of enchanced bioaccumulation of MeHg for  phytoplankton and zoo-
plankton, and in-lake methylation was extremely low for all offshore sediment sites sampled. Watershed
export of MeHg may be a function not only of land use/land cover (e.g., there is  more transport of MeHg
in sandy soils),  but also  of  subsurface flow characteristics. A preliminary  mass balance for the lake
suggests water retention of 190 years, and that mercury is present for 4.6 years.

Future research  will include:   (1) atmospheric deposition to  explore the  effects of local inputs on
nearshore (urban) regions and the direct deposition of MeHg in the basin; (2) mixing zones and watershed
influences to look at mechanisms for algal and zooplankton uptake and the  fate of colloidal MeHg in
nearshore  zones; and (3)  in-lake studies to determine the fate  of changing anoxia on MeHg dynamics,
mechanics of bioaccumulation in offshore  zones, and methylation in deep sediments. The Great Lakes
could serve as an excellent site for coastal studies.

In response to an inquiry from Dr.  Robert Mason about whether there is direct groundwater input into
Lake Superior, Dr. Hurley responded that there have been studies done on the northern shore, although
this may be fracture flow.  There are sandy areas in nearshore zones as well.

Dr. William Fitzgerald remarked on the utility of comparing work in Lake Superior (e.g., how MeHg is
produced) with similar work in regions such as the Long Island Sound.

Understanding the Role of Sulfur in the Production and Fate of Methylmercury in Watersheds
Robert Mason, University of Maryland

Dr. Robert Mason stated that mercury has recently received a great deal  of media attention. Methylation is
the key process in transforming  inorganic Hg that is emitted from various sources into MeHg, the toxic
form of Hg that bioaccumulates in fish. The  goal of this research is to understand the role of sulfur
(especially sulfide) in Hg methylation, and the MeHg fate and transport in watersheds. The  specific
objectives are to investigate: (1) how  Hg complexation with sulfide  effects bioavailability of Hg to
methylating microorganisms; (2) how  MeHg production and accumulation may change  for  a given
change in loading of Hg,  sulfur,  and  other ecosystem constituents;  (3)  the factors  controlling
demethylation in environmental matrixes; and (4) the factors controlling Hg bioavailability, methylation,
and MeHg fate and decomposition, through laboratory studies and modeling  efforts. A novel technique,
       The Office of Research and Development's National Center for Environmental Research

-------
                  STAR Mercury Fate and Transport Final Progress Review Workshop
octanol water partitioning, was refined to measure Hg speciation in situ in field samples. This technique
can now be used in environments where Hg levels are in the pM range.

Understanding bioavailability to methylation and how net methylation changes over time is important.
Factors to consider are speciation effects, the activity of methylating organisms, demethylation (which
microbes are involved and the role of speciation), and how Hg recently added to the ecosystem is less
bioavailable and less well methylated. Dr. Mason presented a summary of the major findings and future
research efforts.  The role of sulfide and sulfate in Hg methylation is  now well  understood. In controlled
studies, the methylation rate appears to  be directly linked to medium sulfide levels. Further studies need
to focus on dissolved organic carbon (DOC) and other parameters and their role in Hg methylation. There
is a correlation between short-term Hg  methylation and in situ MeHg concentrations  that makes sense
from a thermodynamic and kinetic point of view. Thus,  measurements of MeHg in systems where this
relationship has  been established may provide a reasonable surrogate measure of methylation potential.
Modeling suggests that the role of pH in controlling net MeHg in aquatic systems is complex and may be
more   important  than  was previously thought.  Stable  isotope  and  radioactive  addition  incubation
experiments appear to provide a  similar  relative  measure  of methylation potential, but not for
demethylation potential. More effort is needed to  understand the controls over demethylation.  Future
work is aimed at demonstrating in the field the presence of neutral Hg species and a relationship between
methylation rate  and in situ Hg speciation, and developing a model that incorporates speciation and other
factors to predict Hg bioavailability for methylation and MeHg fate.

Dr. John Sherwell noted that it may not be necessary to know the details of speciation on the input side,
because we can measure the total  Hg entering the  system.  Dr.  Mason responded  that you can get
differences in the methylation rate just by spiking Hg into the system in a different form, suggesting that
the kinetics of the exchange processes are much slower than one might assume.  Thus, measuring only the
total Hg input may not provide the necessary information, although speciation measurement may be more
critical in Hg reduction than in methylation.

Response of Methylmercury Production and Accumulation to Changes
in Hg Loading:  A Whole-Ecosystem Mercury Loading Study
Cynthia Gilmour, The Academy of Natural Sciences

Dr. Cynthia Gilmour stated that this study attempts to answer the question: How do ecosystems respond
to changes in  Hg deposition?  The overall objective  of  this research is to examine the magnitude and
timing of the response in net MeHg production to changes  in Hg loading. The specific objectives are to:
(1) examine the bioavailability of Hg deposited  to  different types of  the watershed, and different
watershed components, for methylation through time; (2) use Hg-stable isotopes to elucidate patterns of
fate, transport, and bioavailability; and (3) examine the dose-response to Hg loading. Stable Hg isotopes
were used to assess old Hg (existing in sediments and  soils) versus new Hg (freshly deposited).  The
analysis was conducted using standard Hg/MeHg techniques, but substituted ICP-MS for CVAF.

In the first study, which is embedded within the Aquatic Cycling of Mercury in the Everglades study, Hg
dose-response experiments were conducted with stable isotopes in 1 m diameter mesocosms. The results
from the Florida Everglades suggest that:   (1) the bioavailability of new Hg  for methylation changed
rapidly through time; (2) methylation and bioaccumulation of new Hg in the warm shallow water systems
occurred in a quick pulse after the Hg spike; and (3) DOC has a strong effect on methylation, and there
are interactions  between  SO4 and the age  of the Hg being methylated. In the Mercury  Experiment to
Assess Atmospheric Loading in Canada and the United  States (METAALICUS) study, mercury stable
isotopes were used to track Hg transport, methylation, and bioavailability for methylation and uptake at
the ecosystem level. The key  STAR  objective of the  METAALICUS  project was to examine the
bioavailability of Hg deposited to different parts of the watershed for methylation through time. Hg stable
        The Office of Research and Development's National Center for Environmental Research

-------
                  STAR Mercury Fate and Transport Final Progress Review Workshop
isotopes have been delivered to the uplands, wetlands, and a first-order drainage lake in northwest Ontario
since 2001. Dosing has continued through 2003, at roughly four times the level of atmospheric Hg
deposition. A key result from this work is that there is a significant difference between new and old Hg in
terms of MeHg production. Other results include:  (1) lake budgets show that most of the native and spike
MeHg is formed within the  lake; (2) surficial sediments, anoxic profundal waters, the adjoining wetland,
and epilithon (to a much lesser extent) are sites of MeHg production; (3) lake Hg spike is more dynamic
than existing Hg pools; and (4) the time to reach a new equilibrium with the higher Hg deposition rate
will  be at least 3 years for the Hg  spike to the lake, and longer for  Hg deposition to the upland and
wetland.

The new hypotheses are that ligand exchange rates in natural systems are much slower than expected and
influence the availability of Hg for methylation  through  time, and equilibrium Hg speciation may be
inadequate to model MeHg production. The METAALICUS project team will continue with additional
years of isotope spike, and would like to address the following questions:  (1) How long will it take until a
new equilibrium with the new Hg dose is achieved? (2) What will be the time course of 200Hg movement
off the wetland, and  198Hg movement  in the wetland? (3) How bioavailable for methylation  and
bioaccumulation will Hg from uplands and wetland be? (4) What biogeochemistry underlies the changes
in Hg bioavailability through time and space? As the project moves forward, the third question will be
key.

Ms. Kritee asked how the errors in stable isotope  measurement were handled in terms of the addition of
the 202Hg isotope, which is  also present in the ambient environment. Also, how do we know that some
isotopes are not preferentially enriched? In response to the first question, Dr. Gilmour stated that the data
represents the concentration of excess 202 over natural abundance, which involves matrix algebra and
solving simultaneous  equations. As to the second question, things  as heavy  as Hg do not undergo
fractionation by enzymatic processes.

Methylmercury Sources to Lakes in Forested Watersheds: Has Enhanced Methylation
Increased Mercury in Fish Relative to Atmospheric Deposition?
Edward Swain, Minnesota Pollution Control Agency
Presented by Edward Nater, University of Minnesota

Dr. Edward Nater stated that the objectives of this research are to: (1)  determine the relative importance
of atmospheric, in-lake, and wetland sources of MeHg to a  lake in a forested watershed;  (2) determine
MeHg net retention and/or source strength of different wetland types; and (3) conduct mesocosm (wetland
and lake) and whole wetland experiments to elucidate methylation-enhancing processes. Specifically, it is
important to answer two questions. First,  in a forested watershed, what is the relative importance of
various MeHg  sources  (in-lake  methylation, transport from  uplands/wetlands, release  from lake
sediments,  atmospheric deposition)? Second,  has MeHg  become a larger fraction of the total  Hg in
aquatic ecosystems because  of enhanced methylation rates  stimulated as a result of higher rates of sulfate
deposition?

Watershed-scale sulfate additions  were conducted in the Marcell Experimental Forest, north of  Grand
Rapids, Minnesota. An irrigation system  was built on one-half of a bog, with the upper half as a control
and the lower half as the treatment area. The lower portion of the bog received sulfate doses four to five
times a year, for a total (addition plus precipitation) of five  times the  atmospheric deposition. Bog well
and weir samples were collected. The following  conclusions resulted from the research: (1) enhanced
sulfate deposition produces significant increases in MeHg in weir outflow and bog wells; and (2) this
ombrotrophic bog is sulfate poor,  so results may  not transfer well to ground water-influenced wetlands
that  contain more sulfate. The significance of the findings are that:  (1) enhanced  sulfate  deposition
appears to significantly enhance net Hg methylation; and (2)  reductions in atmospheric sulfate deposition
        The Office of Research and Development's National Center for Environmental Research

-------
                  STAR Mercury Fate and Transport Final Progress Review Workshop
may cause a decrease in environmental MeHg, but because there is no way to reduce the amount of
sulfate in the system, this hypothesis cannot be tested empirically.

In spring 2003, there was  a 3-week period of high MeHg levels, with hundreds of ng/L MeHg in weir
outflows. A next step will  be to conduct a laboratory study to examine whether the levels were real, and
why this phenomenon occurred. There was an uncommon deep freeze in the bogs, so the hypotheses are
that the system produced  high MeHg  in very cold conditions, or that the freezing resulted in a salt
exclusion effect. Also, there are plans to add sulfate to the uplands rather than the bog, and to finish the
mass balance for the lakes.

Dr. Gilmour commented that in Florida, reconstructed wetlands that were allowed to dry  out produced
enormous amounts of MeHg upon rewetting.  Her laboratory studies suggest that when the soil is dry, the
organic sulfur and inorganic sulfur pools in the soils oxidize; upon rewetting, the sulfate results in rapid
production of MeHg. Dr.  Nater responded that there is a  1/2-meter zone of dry out every year that
refloods in the spring; he has not seen this effect yet, but he will look into it.

BlOGEOCHEMICAL CONTROLS ON MERCURY METHYLATION/DEMETHYLATION RATES

Mercury:  Transport and Fate Through a Watershed
Jerome Nriagu, University of Michigan

Dr. Jerome Nriagu stated that the objective of this research is to investigate the role of photochemistry on
the cycling  of Hg in the Saginaw Bay watershed.  Different parts of the Saginaw Bay watershed have
different ecosystems (agricultural, wetland, and forested areas), and it is a fairly industrialized area. From
the outer bay  to Lake Huron there are well-defined chemical gradients, which may play a role in the
cycling of Hg in the bay. The ambient airborne Hg levels and fluxes of Hg were measured at more than
40 sites throughout the watershed. Sampling sites included forest soils, agricultural soils,  wetlands, city
park soils, beach sands, and water samples from different parts of the bay.

Some of the significant results are that:  (1) levels of total Hg in the watershed are typically very low; (2)
the inner and outer bay differ in their nitrate,  sulfate,  and chloride concentrations,  and this may be
affecting Hg cycling in the watershed; (3) there  are higher concentrations of total Hg  and dissolved
gaseous mercury (DGM) in the forested, western areas compared to the agricultural, eastern areas of the
bay; (4) diel levels increase during the day in response to solar radiation, and water temperature does not
play a significant role; (5) diel changes of DGM vary from day to  day and from location to location,
raising the question of how to model the production of DGM in a given body of water; (6) there is an
increase in DGM  production with sunlight upon the addition of ferric ion compared to the dark control,
suggesting  a link between the organic carbon cycle and the production of DGM; (7) the presence of
humic acid increases the production of Hg(0) in the water column; (8) there is higher production of DGM
in a closed system compared to an open system, and it cannot simply be concluded that most of the Hg is
lost because there are other factors in the lake water that may be affecting production of DGM; and (9)
upon heating, the majority  of Hg is released below 350 °C, and all Hg is in the form of Hg(0).

The measurements lead to the overall conclusion that little of the mercury that is deposited in the Saginaw
Bay is retained in the basin. This leads to the question of what the ultimate sink is for the majority of the
Hg that is released in North America. It is possible that Hg is being deposited in the Arctic, a global sink.
The next steps will include relating the diel  variations to the biological and chemical activities at each
location, and working out the photochemical mechanism (including the role of ferric ion) behind the
reduction of mercuric ion to elemental mercury in the lake water.
        The Office of Research and Development's National Center for Environmental Research

-------
                   STAR Mercury Fate and Transport Final Progress Review Workshop
Dr. Fitzgerald  commented  that although  some  Hg  is  going to the Arctic,  there  are  many other
contaminated systems (e.g., there is plenty of Hg in areas like the watersheds around mines). Dr. Nriagu
agreed, but maintained that this still leaves a lot of Hg to be distributed to the atmosphere, which did not
get deposited in the Saginaw Bay watershed. Dr. Fitzgerald added that the residence time for elemental
Hg in the air is quite long,  so  it will be distributed rather broadly. Dr. Mason  asked  if in the thermal
desoprtion experiments,  Dr. Nriagu measured the sediment to ensure that all of the Hg  had disappeared.
Dr. Nriagu replied that a large percentage of the Hg was lost.

Chemical and Biological Control of Mercury Cycling in Upland, Wetland,
and Lake Ecosystems in the Northeastern United States
Charles Driscoll, Syracuse University

Dr. Charles Driscoll  stated that the objectives of this research are to:  (1) quantify patterns of transport
and transformations of Hg species in an upland northern hardwood forest through adjacent wetlands to the
aquatic environment;  (2) evaluate the processes and mechanisms controlling MeHg concentrations and
transport in pore waters  and surface waters in wetlands; and  (3) develop and apply a lake/watershed Hg
cycling model to  a lake/watershed ecosystem.  The Sunday Lake watershed, located in the Adirondack
region of New York,  was selected because it has different wetland types, and wetlands in the drainage
area.

Ten surface-water monitoring sites were sampled at monthly intervals over 3 years at two upland forest
sites and wetland sites around  the watershed.  The upland forest observations are that:  (1) hardwood
canopy is a sink for total Hg and a source of MeHg; (2) coniferous canopy is a  source for total Hg and
MeHg; (3) total Hg concentrations are elevated in foliage and litter; (4) concentrations of total Hg are
elevated in forest floor solutions and decrease with depth in mineral soil; and (5) there is  no soil pattern of
MeHg. One focus of this study was the inputs of Hg via vegetation, in terms of the ultimate fate of the Hg
(i.e., Is the Hg labile  or retained in soil?). Preliminary results  suggest that leaf litter is a sink for inputs of
inorganic Hg. In the  upland studies, the deciduous site has higher levels of Hg per unit carbon than the
coniferous site, which is consistent with  observations of high inputs of Hg via litter at the deciduous site.
In the wetland studies, the highest total Hg concentration was found in sphagnum spp., and Hg increases
in the growing season were followed by marked decreases in the fall season prior to die off of the wetland
vegetation. Bioconcentration factors in Sunday Lake are slightly lower than what has been reported, most
likely due to the higher concentration of DOC at the site. The following overall conclusions were made:
(1) wet deposition of total Hg and MeHg are similar to other values reported for  eastern North America;
(2) it appears that dry Hg deposition (litter + throughfall; 70%) greatly exceeds wet deposition (30%); (3)
leaf litter retains total Hg over the short term;  (3) the watershed is a sink for total Hg  and a source for
MeHg, and MeHg production is high  in riparian wetlands; (4) the lake is a sink for total Hg and a source
for MeHg; and (5)   lake and  yellow  perch  Hg concentrations decrease in  response to decreasing
deposition, according to  the results of a watershed mercury model. Future work with the  model will
examine  the difference  between changes in mercury  deposition and  changes in sulfate  inputs to the
system.

Dr. Nriagu asked  if the flux of Hg from the ecosystem to the atmosphere was measured, because he did
not see it in the mass  balance model.  Dr. Driscoll responded  that it was not done in this study; the land-
atmospheric exchange will be examined in detail in the next study.

Dr. Gilmour inquired about the controls that were put on MeHg production in the model. The response
was that there is a functional dependence on a number of parameters, and that the work is quite detailed.
        The Office of Research and Development's National Center for Environmental Research

-------
                   STAR Mercury Fate and Transport Final Progress Review Workshop
Dr. Patricia Bresnahan asked about what is happening to the mercury being collected in the litter on the
leaves, and if measurements were reported based on wet and dry biomass, or dry biomass. Dr. Driscoll
responded that measurements were reported based on dry biomass. It is hard to know the ultimate fate of
the Hg. Most of the studies in the literature do a single measurement for wetland vegetation over time; he
was surprised at how much the concentrations decreased in the fall just before die off. A comparison
between wet and dry biomass has not yet been done.

Physical and Chemical Processes Affecting Mercury Cycling

Processes Controlling the Chemical/Isotopic Speciation and Distribution of
Mercury From  Contaminated Mine Sites
Gordon Brown, Jr., Stanford University

Dr. Gordon  Brown stated that mercury  contamination is extensive at  former Hg  and gold mining
operations; there is a huge  input of Hg into the San Francisco Bay from the California Coast Range
mines. An understanding of Hg  speciation and sorption processes is critical to  predicting the fate,
transport, and  bioavailability  of Hg in  mine-impacted regions,  and can  assist in the design and
implementation of effective remediation strategies. The specific objectives of the characterization studies
are to:   (1) develop  a method to determine Hg speciation in natural samples using  extended  X-ray
absorption fine  structure  (EXAFS)  spectroscopy; (2) characterize the effects of a number of variables
(geological background, ore processing, particle  size, distance from source)  on Hg speciation; and (3)
determine the macroscopic  behavior  and  molecular-scale modes of Hg(II)  sorption to goethite as a
function  of pH, choride,  and sulfate.  Different types of Hg deposits (silica  carbonate,  hot spring) and
waste materials (calcines, condenser soot, distributed sediments, floe/precipitates, sieved fractions, waste
rock) were sampled. The samples were analyzed in the laboratory using EXAFS and a linear combination
fitting procedure, ore roasting, and transmission electron microscopy (TEM).

Conclusions  from  the speciation experiments are that:   (1)  EXAFS is an important technique for
determining the in situ  speciation of low-concentration,  Hg-bearing samples; (2)  mine  wastes are
composed mainly  of HgS, but some contain environmentally-relevant proportions of more  soluble Hg
phases; (3) Hg speciation is influenced by the geologic origin of the ore, with  Hg-chloride phases
identified in  samples  from hot spring Hg  ore deposits; (4) the ore roasting process appears to convert
cinnabar to metacinnabar; (5) Hg concentration and % Hg-sulfides increase in  calcines  with decreasing
particle size; and (6) Hg in mine wastes remains primarily in the solid phase during transport.

The conclusions for Hg sorption are that: (1) Fe- and Al-(hydr)oxides can play a significant role  in the
uptake of Hg(II) through direct inner-sphere sorption processes; (2) within the pH  range typical  of the
natural environment, Hg(II)  sorption is relatively constant among the substrates studies; (3) the presence
of  chloride results in reduced Hg(II) sorption,  largely through  the  formation  of stable HgCl2(aq)
complexes; (4) the presence  of sulfate results in enhanced Hg(II) sorption, largely through the reduction
of  electrostatic repulsion at positively-charged  surfaces;  (5)  ternary surface complexation has been
identified in the presence  of chloride and sulfate; and (6) model systems appear to provide good correlates
for understanding Hg(II) sorption processes in environmental aquatic systems.

The overall conclusions are  that:  (1) Hg  speciation as determined by XAFS spectroscopy varies from
mine site to mine  site depending on the mine's geologic history, with some species  more soluble and
potentially bioavailable than  others; (2) laboratory column experiments indicate that colloidal transport of
Hg is an important dispersal mechanism for Hg in the environment; (3) Hg colloids identified by XAFS
analysis include cinnabar and metacinnabar, those  identified by TEM include  hematite, jarosite,  and a
poorly crystalline aluminosilicate gel (the amount of Hg sorbed on these colloids appears to be small); (4)
Hg(II) is strongly sorbed by Fe(III)-hydroxides and less strongly sorbed by Al-hydroxides in laboratory
        The Office of Research and Development's National Center for Environmental Research

-------
                   STAR Mercury Fate and Transport Final Progress Review Workshop
experiments (chloride greatly reduces Hg(II) sorption in both cases, while sulfate slightly enhances Hg(II)
sorption on these hydroxides); (5) sorption processes involving Hg(II) do not appear to be as important in
California Coast Range mine tailings environments as for other heavy metals such as lead or metalloids;
(6) the lightdark ratio is greater in samples containing metacinnabar relative to samples containing just
cinnabar; (7) flux measurements indicate that non-HgS phases such as HgO (montroydite) and eglestonite
may be significantly larger contributors of gaseous Hg than HgS phases; (8) substrate Hg  concentration
and speciation appear to be dominant factors in controlling Hg atmospheric emissions  from mine wastes;
(9) 201Hg/202Hg isotopic ratios among Hg minerals from various mines  show variations, suggesting that
such ratios will not be useful in distinguishing among Hg from different sources; and (1) understanding
mercury  speciation and  sorption processes is  critical to predicting its fate, transport,  and potential
bioavailability in mine-impacted regions and can assist in the design and implementation of effective
remediation strategies.

Research already underway includes an  examination of particulate Hg to determine if it is a significant
source of bioavailable Hg, and the identification of the species of dissolved Hg that are  readily methylated
by sulfate-reducing bacteria and  why these species  are preferred. Planned  research will  examine Hg
uptake by plants in the South San Francisco Bay wetlands restoration project.

Dr. Francois Morel asked why Dr.  Brown believes that the sorption of Hg on oxides would  not stop
methylation by bacteria. Dr. Brown responded that he is not certain whether it will stop methylation. He
suspects  that it will not because Hg is in an ionic form, bound to the surface by  two  bonds,  and a
bacterium could remove the Hg from the surface. Dr. Morel inquired about Hg sulfide, and Dr. Brown
responded that he suspects a bacterium will have a more difficult time removing Hg from the surface of a
sulfide than an oxide, although this could be debated.

Microbiological and Physicochemical Aspects of Mercury Cycling in the
Coastal/Estuarine  Waters of Long Island Sound and Its River-Seawater Mixing Zones
William Fitzgerald, University of Connecticut

Dr. William Fitzgerald stated that consumption offish is the principal route of human exposure to MeHg.
The  overall objective of this research is to examine by field and laboratory investigation the physico-
chemical/microbiological reactions and processes governing Hg cycling, speciation, and bioavailability in
the waters and sediments of the Long Island Sound (LIS) and its watershed/coastal interface. Specifically,
the goals of this research are to examine the bioavailability of Hg and its relationship to methylation, in
situ  Hg° production and emissions, organic complexation of  Hg, sulfur  and  Hg interactions, and
sedimentary Hg methylation. Field collection included an analysis of elemental Hg and sediment samples
by electrochemistry measurements. It was determined that Hg emissions from the sound (80 kg/year) are
a significant source of Hg to the coastal/continental atmosphere, approximately 35  percent  of the Hg
entering the sound is volatilized, and these  emissions are principally anthropogenic (greater than  75
percent).  An expanded spatial  and temporal coverage of Hg distribution and speciation is needed, and
complementary laboratory investigations of elemental Hg are needed to gain insight into processes and
mechanisms.

The  LIS MeHg budget suggests that MeHg is carried away from the sound by migrating organisms like
fish, implying  that the  coastal zone can affect  more than just the  localized  region. The following
conclusions resulted from the research:   (1) Hg  methylation in LIS  sediments  is  influenced by the
availability  of dissolved Hg(II) complexes,  which is largely affected  by  organic matter;  (2)  seasonal
effects on Hg methylation are apparent (e.g., temperature, relative activity of various bacterial  groups);
(3) bioturbation enhances Hg methylation and may be methylating some portion of the legacy Hg; and (4)
much of the MeHg in LIS biota is attributable to sedimentary production. The information and hypotheses
generated in this study are being applied to a comprehensive study of MeHg production and mobilization
        The Office of Research and Development's National Center for Environmental Research

-------
                  STAR Mercury Fate and Transport Final Progress Review Workshop
from sediments in the New York/New Jersey Harbor. Future research will involve an examination of
MeHg  production  in sediments on the continental  shelf, and an investigation into the significance of
MeHg produced in near-shore and shelf sediments to the open ocean and its food web.

Dr. James Hurley asked about atmospheric inputs into the mass balance. Is it just wet bulk deposition, or
are there local effects of reactive gaseous mercury  (ROM)  in the urban  area? Dr. Fitzgerald responded
that bulk collectors were set up for 14 months, presumably collecting ROM along the coast. There are two
sources, local RGM associated with various high-temperature practices and elemental Hg production with
ROM generated as a result of interaction with the bromine system. If it were matching with elemental Hg
production, a very large increase in localized deposition of Hg in those collectors should be observed, but
it was not. If it were relatively modest,  like an RGM contribution to the wet deposition of 30 percent, that
would be lost in the uncertainty. This is a good mass balance that can be constrained with other data, and
the balance suggests that the  RGM contribution is not large. However, a  30 or 40 percent addition could
not be  distinguished. Dr. Hurley said that the waste water  inputs are interesting, and asked if it is just
aqueous phase coming off, or if sludge inputs are considered. Dr. Fitzgerald responded that the number
for the waste water contributions from  the LIS is based on the modeling  in the sound and an estimate of
sewage inputs, and that the scale is huge.

Dr. Arnold Kuzmack commented that he is getting lower values for the amount of Hg in waste water, and
he would like to sort out the  differences. Dr. Fitzgerald responded that the values are not very high, and
the number deals with sewage input into the Long Island Sound and cannot be applied nationally.

The Redox Cycle of Mercury in Natural Waters
Francois Morel and Eileen Ekstrom,  Princeton University

Dr. Francois Morel presented a  brief  description of the redox cycle of mercury in natural waters. He
discussed the redox of Hg  in surface waters, the source of MeHg in the atmosphere, and the location of
Hg methylation in the open ocean. The estimated daytime Hg° loss in coastal water is much larger from
photo-induced oxidation than volatilization. Cows produce methane, but there is very little MeHg formed;
therefore, cows are not important to MeHg budgets. The amount of Hg in yellowfin tuna in  1971 was
compared to the amount found in tuna  in 1998; over 30 years later, there was the  same amount of Hg in
tuna in 1998.   Hg in tuna does not  reflect changes in the atmospheric  Hg, suggesting an alternate
conclusion that MeHg in the ocean is coming from the deep ocean, and that Hg in tuna may be natural.

Dr. Eileen Ekstrom presented  a  presentation on the  importance of understanding the biochemistry of
MeHg. Methylation  of Hg by sulfate-reducing  bacteria (SRB)  is the predominant system  for MeHg
production in fresh water and coastal systems, yet little is known about the physiology and biochemistry
of Hg methylation in bacteria. Biochemical processes control the varying capacity of sulfate-reducing
strains  to methylate Hg, and may be the  rate limiting step(s) in Hg  methylation. The objective of this
research is to examine the  metabolic pathways, enzymes, and coenzymes responsible for Hg methylation
in SRB.  The research question to be  answered  is:  Is the acetyl-CoA pathway the major route for Hg
methylation  in SRB?  The research approach was toscreen SRB strains  for the presence of key acetyl-CoA
pathway  enzymes and the  ability to methylate Hg.  Specific inhibitors of the acetyl-CoA pathway were
tested to determine their effects on Hg methylation  in pure culture. The data show that incomplete-
oxidizing strains have CODH enzyme activities below the detection limit, and an inhibitor assay, using an
inhibitor of the acetyl-CoA pathway, confirmed that this pathway is  not used for Hg methylation. The
overall conclusions are  that:   (1) four incomplete-oxidizing SRB strains do not use the acetyl-CoA
pathway  for Hg  methylation;  and  (2) three complete-oxidizing strains that require the  acetyl-CoA
pathway for basic metabolism methylate Hg. In the majority of incomplete-oxidizing SRB strains tested,
Hg methylation is independent of the acetyl-CoA pathway, and may not require vitamin Bi2. It is possible
that S-adenosyl-methionine could be the methylating coenzyme for MeHg production in SRB that do not
        The Office of Research and Development's National Center for Environmental Research

-------
                   STAR Mercury Fate and Transport Final Progress Review Workshop
use the acetyl-CoA pathway. Future research will continue to examine the role of methyl-transferring
coenzymes, other than vitamin Bi2, in MeHg production, and the pathways and enzymes responsible for
Hg methylation.

Dr. Michael Aucott stated that he suspects that global Hg emissions from the 1970s are not very different
from today, and so it is not much of a surprise that the Hg in tuna was the same. Dr. Morel responded that
the global Hg emissions may not have changed much, but that this does not apply to the area of study.

Quantifying the  Role of Significant Natural Sources  and Sinks in the Biogeochemical Cycle of
Mercury
Mae Sexauer Gustin, University of Nevada-Reno

Dr. Gustin stated that the objectives of the  STAR research are to:  (1)  develop an understanding of the
factors influencing emissions (Hg concentration in substrate, Hg speciation in substrate, light including
shading, moisture, temperature) and develop algorithms for scaling; (2) develop a preliminary database on
speciation  of  Hg in air above naturally-enriched sites; (3) develop a database of Hg  fluxes for
representative source areas for a variety of geologic settings; and (4) develop a framework  for scaling up
point source measurements to area estimates. The research was conducted by measuring  Hg flux from
terrestrial sources—including Hg mining districts, recent geothermal areas, sulfur banks, gold  mining
districts,  oil  and  gas   fields,  and  massive  sulfide  deposits in  Nevada  and  California—using
micrometeorological  methods and dynamic field flux chambers, and through controlled laboratory studies
using  a gas  exchange  system. The  results show that:  (1)  light,  temperature,  wind  velocity, and
precipitation, soil moisture,  and shading significantly  influence temporal patterns  in emissions; (2)
geologic factors,  such as substrate Hg concentration, hydrothermal alteration, and the presence  of heat
sources and geologic structures, dictate the magnitude of emissions; (3) Hg emissions applied to natural
Hg-enriched substrate in models are underestimated (a database of emissions currently in development for
background sites shows fluxes similar to that applied to natural belts in models); and (4) reactive gaseous
mercury (RGM)  is approximately 1.5 percent  of the Hg associated with naturally-enriched areas, is
elevated above  background concentrations at most sites, and may be directly emitted from the soil. The
implications of this research are that naturally-enriched substrate is a more significant source of Hg than
previously realized, and that background substrate emissions are similar to that predicted  for global Hg
belts. RGM is  higher in naturally  Hg-enriched areas. Also, with the recognition of the importance of
atmospheric oxidation of Hg, all forms of Hg released to the atmosphere are significant in terms of
potential environmental  impacts. In the  future, it is important to continue to investigate natural source
emissions from geologic sources, fire, and vegetation; the role  of background soils; the potential for
elemental Hg deposition; re-emission; and the mechanisms controlling emissions.

Research funded by the U.S. EPA's Experimental Program to Stimulate Competitive Research (EPSCoR)
was  designed to  investigate Hg cycling between plants, soils, and the atmosphere using ecocells and
ecopods. The data indicate that as plants leaf out whole ecosystem flux declines, and although the plants
are assimilating Hg from the air, shading of the soil is the dominant factor influencing Hg flux. Following
the successful demonstration that ecocells can be used to investigate Hg  cycling, a grant was awarded by
the National Science Foundation to examine vegetation-atmospheric Hg  interactions using a tall grass
pristine ecosystem. The overall conclusions are that:   (1) plants can  act as  a source and sink of
atmospheric Hg; (2)  plants can exacerbate Hg emission from the soil to the air (in the tall grass prairie
ecosystem, flux associated with vegetation was  four to five times greater than  soil  Hg flux); and (3)
shading can significantly reduce emissions and should be included in modeling Hg flux associated with
forested systems.  Future work will further investigate the role of plants as a sink for atmospheric Hg and a
source of Hg to terrestrial and aquatic ecosystems, and the influence of vegetation on air-plant-soil Hg
exchange.
10      The Office of Research and Development's National Center for Environmental Research

-------
                  STAR Mercury Fate and Transport Final Progress Review Workshop
Dr. Morel asked about the mechanism of Hg emission, which is season and light dependent. Dr. Gustin
replied that with Hg sulfide and other minerals it is most likely photoreduction. Dr. Morel clarified that he
would like to know about the mechanisms behind the photoreduction. For example, if there is some sort
of temperature  effect. Dr. Gustin responded that in plants there is a temperature effect, but with  the
emissions as a function of speciation, total temperature was controlled.

ADJOURNMENT

Mr. Stelz thanked everyone for their attendance at the conference.
        The Office of Research and Development's National Center for Environmental Research

-------
                  STAR Mercury Fate and Transport Final Progress Review Workshop
                   U.S. EPA National Center for Environmental Research
           STAR Mercury Fate and Transport Final Progress Review Workshop

                                   Grand Hyatt Washington
                                      1000 H Street, NW
                                   Washington, DC 20001

                                      November 21,2003

                                           Agenda

Friday, November 21, 2003

8:00 - 8:30 a.m.    Registration

8:30 - 8:45 a.m.    Welcome and Introductory Remarks—Goals of the Progress Review
                  William Stelz, U.S. EPA, Office of Research and Development, National Center for
                  Environmental Research

Formation/Transport of Methylmercury in Ecosystems and Watersheds

8:45 - 9:15 a.m.    Watershed Influences on Transport, Fate  and Bioavailability of Mercury in Lake Superior
                  James P. Hurley, University of Wisconsin

9:15 — 9:45 a.m.    Understanding the  Role of Sulfur in the Production and Fate of Methylmercury in
                  Watersheds
                  Robert P. Mason, University of Maryland

9:45 -10:00 a.m.   Break

10:00 - 10:30 a.m. Response of Methylmercury Production and Accumulation to Changes in Hg Loading:
                  A Whole-Ecosystem  Mercury Loading Study
                  Cynthia C. Gilmour,  The Academy of Natural Sciences

10:30 - 11:00 a.m. Methylmercury Sources to Lakes in Forested Watersheds: Has Enhanced Methylation
                  Increased Mercury in Fish Relative to Atmospheric Deposition?
                  Edward B. Swain, Minnesota Pollution Control Agency
                  Presented by: Edward Nater, University of Minnesota

Bioseochemical Controls on Mercury Methylation/Demethylation Rates

11:00 - 11:30 a.m. Mercury: Transport and Fate Through a Watershed
                  Jerome Nriagu, University of Michigan

11:30-12:30 p.m. Lunch

12:30 - 1:00 p.m.   Chemical and Biological Control of Mercury Cycling in Upland, Wetland and Lake
                  Ecosystems in the Northeastern U.S.
                  Charles T. Driscoll, Syracuse University
          The Office of Research and Development's National Center for Environmental Research        13

-------
                   STAR Mercury Fate and Transport Final Progress Review Workshop
Physical and Chemical Processes Affecting Mercury Cycling

1:00 - 1:30 p.m.    Processes Controlling the Chemical/Isotopic Speciation and Distribution of Mercury
                   From Contaminated Mine Sites
                   Gordon E. Brown, Jr., Stanford University

1:30 — 2:00 p.m.    Microbiological and Physicochemical Aspects of Mercury Cycling in the
                   Coastal/Estuarine Waters of Long Island Sound and Its River-Seawater Mixing Zones
                   William F. Fitzgerald, University of Connecticut

2:00-2:15 p.m.    Break

2:15 — 2:45 p.m.    The Redox Cycle of Mercury in Natural Waters
                   Francois M. Morel, Princeton University
                   Presented by: Eileen Ekstrom, Princeton  University

2:45 — 3:15 p.m.    Quantifying the Role of Significant Sources and Sinks in the Biogeochemical Cycle of
                   Mercury
                   Mae Sexauer Gustin, University of Nevada-Reno

3:15 - 4:30 p.m.    Questions and Discussion

4:30 p.m.           Adjournment
14         The Office of Research and Development's National Center for Environmental Research

-------
                  STAR Mercury Fate and Transport Final Progress Review Workshop
                   U.S. EPA National Center for Environmental Research
           STAR Mercury Fate and Transport Final Progress Review Workshop
                                   Grand Hyatt Washington
                                      1000 H Street, NW
                                   Washington, DC 20001

                                     November 21,2003
                                      Participants List
Herbert Allen
University of Delaware

Michael Aucott
New Jersey Department of
Environmental Protection

Tamar Barkay
Rutgers University

David Belluck
U.S. Department of Transportation

Tim Benner
U.S. Environmental Protection Agency

Pamela Birak
Research Triangle Institute

Laura Blake
New England Interstate Water Pollution
Control Commission

Andrea Blaschka
U.S. Environmental Protection Agency

Jean-Claude Bonzongo
University of Florida

John Botts
Aquatic Sciences Consulting

Patricia Bresnahan
University of Connecticut

Gordon Brown, Jr.
Stanford University
Joseph Bushey
Syracuse University

Guillermina Calles
Baker Botts, LLP

Rick Carlton
Electric Power Research Institute

Ruth Chemerys
U.S. Environmental Protection Agency

Rebecca Clark
U.S. Environmental Protection Agency

Jace Cuje
U.S. Environmental Protection Agency

Charles Driscoll
Syracuse University

Stan Durkee
U.S. Environmental Protection Agency

Eileen Ekstrom
Princeton University

William Fitzgerald
University of Connecticut

Barbara Fletcher
University at Albany,
State University of New York

Katherine Fogarty
Menzie-Cura and Associates, Inc.
          The Office of Research and Development's National Center for Environmental Research
                                                                                            15

-------
                  STAR Mercury Fate and Transport Final Progress Review Workshop
Cynthia Gilmour
The Academy of Natural Sciences

Iris Goodman
U.S. Environmental Protection Agency

Mae Sexauer Gustin
University of Nevada, Reno

Joel Hansel
U.S. Environmental Protection Agency

Pamela Hartman
ICF Consulting

Dale Hoff
U.S. Environmental Protection Agency

Vincia Holloman
U.S. Environmental Protection Agency

James Hurley
University of Wisconsin

Alan Johnson
Clemson University

John Johnston
U.S. Environmental Protection Agency

Margaret Kerchner
National Oceanic and Atmospheric Administration

Christopher Knightes
U.S. Environmental Protection Agency

Kritee Kritee
Rutgers University

Arnold Kuzmack
U.S. Environmental Protection Agency

Kathy Lambert
Hubbard Brook Research Foundation

Barbara Levinson
U.S. Environmental Protection Agency

Michael Loughran
U.S. Environmental Protection Agency
Elizabeth Malcolm
Princeton University

Robert Mason
University of Maryland

Paul Miller
Maryland Department of Natural Resources

Francois Morel
Princeton University

Edward Nater
University of Minnesota

Sarah Nelson
University of Maine

Jerome Nriagu
University of Michigan

Kim Osborn
ICF Consulting

William Patrick
Louisiana State University

Zahinira Pietri
U.S. Environmental Protection Agency

Jack Puzak
U.S. Environmental Protection Agency

Rob Reash
American Electric Power

Stephan Ryba
U.S. Environmental Protection Agency

Tamara Saltman
U.S. Environmental Protection Agency

Jeffra Schaefer
Rutgers University

Joseph Schmidl
Weston Solutions, Inc.

Kristina Schneider
Limno-Tech, Inc.
16
           The Office of Research and Development's National Center for Environmental Research

-------
                   STAR Mercury Fate and Transport Final Progress Review Workshop
Leonard Schugam
Maryland Department of the Environment

Gregg Serenbetz
U.S. Environmental Protection Agency

Katherine Sheehan
University of Maine

John Sherwell
Maryland Department of Natural Resources

Ashley Smith
U.S. Environmental Protection Agency

Douglas Steele
U.S. Environmental Protection Agency

William Stelz
U.S. Environmental Protection Agency

Jerry Stober
U.S. Environmental Protection Agency

Elsie Sunderland
U.S. Environmental Protection Agency
Michael Twiss
Clarkson University

Ingrid Verstraeten
U.S. Geological Survey

loanna Visviki
College of Mount Saint Vincent

Steven Walters
U.S. Environmental Protection Agency

Ed Washburn
U.S. Environmental Protection Agency

Randall Wentsel
U.S. Environmental Protection Agency

Erik Winchester
U.S. Environmental Protection Agency

Alexander Yeboah
Maryland Department of the Environment
           The Office of Research and Development's National Center for Environmental Research
                                                                                              17

-------
                   STAR Mercury Fate and Transport Final Progress Review Workshop
       Watershed Influences on Transport, Fate, and Bioavailability
                              of Mercury in Lake  Superior

           James P. Hurley1'2, Richard C. Back3, David E. Armstrong2, Helen Manolopoulos2,
                                    and Christopher L. Babiarz2

       Aquatic Sciences Center, University of Wisconsin-Madison, Madison, Wl; 2Environmental
          Chemistry and Technology Program, University of Wisconsin-Madison, Madison, Wl;
               Department of Biology, Lake Superior State University, Sault Ste. Marie, MI

    The goal of this study was to assess the importance of watersheds in controlling sources, transport, fate,
and bioavailability of mercury (Hg) in a northern temperate lake system. Specific objectives of the study were
to: (1) determine the speciation and  bioavailability of Hg transported to Lake  Superior by representative
tributaries/watersheds; (2) determine the importance of watershed-specific characteristics  (soil type, land use,
surficial deposits) that  control physical/chemical forms of Hg transported downstream; (3)  identify  key
mechanisms controlling Hg bioavailability and speciation in nearshore zones relative to open lake regions; and
(4) provide process-level information to complement concurrent development of Hg fate and transport models
of the Lake Superior ecosystem.

    The approach combined field and laboratory studies with modeling to assess the importance of watershed
processes in controlling Hg fate and transport in Lake Superior. Each phase of the field studies, laboratory
studies, and modeling efforts were strongly linked. Techniques developed and adapted by our group during
previous projects ("clean" ultrafiltration, resin techniques, biota processing) were supplemented by newer tech-
niques (stable isotope Hg analysis by ICP-MS; "clean" phytoplankton uptake experiments). Modeling efforts
combine ongoing  GIS-based watershed yield modeling with the Dynamic Mercury Cycling Model (D-MCM;
EPRI, Tetra Tech, Inc.).

    Field research focused on three separate efforts—studies in homogenous land cover sites within the water-
shed, estuarine mixing zone transects, and offshore cruises—to  identify key sites for formation and food chain
uptake of bioaccumulative methyl Hg in Lake Superior and its watershed. Sampling in contrasting homo-
geneous watershed subunits identified the importance of the streambed hyporheic zone  as a site of methyl
mercury (MeHg) formation  to headwater streams and eventually transported to mixing zones of tributaries.
Three cruises on Lake Superior revealed that Hg species concentrations are quite low in Lake Superior (0.49 +
0.22 ng L"1  HgT), similar to Lake Michigan and oceanic waters. Surface water total Hg samples were elevated
along the north shore in April 2000, indicative of river inputs during this high-flow period.  During August
2000, methyl Hg (MeHg) averaged 6.4 + 3.8 pg L"1 elevated levels in the western portion of the lake. Phyto-
plankton Hg analyses showed about a two to threefold enrichment of MeHg in riverine mixing zones versus
offshore regions of the  lake.  Sediment sampling in offshore zones of the lake during 2000 reveal extremely
low methylation rates (< 0.1% per day). These studies suggest that in-lake methylation occurred in the warmer,
shallow sediments, whereas tributary MeHg was primarily supplied during the spring consistent with observed
elevated MeHg in nearshore phytoplankton. Such observations  imply that tributary MeHg  is more available to
the planktonic food Web than is MeHg from in situ sources.

    Laboratory efforts were  dedicated to assessing speciation effects on bioavailability of methyl mercury for
planktonic uptake. A culturing and uptake method was developed successfully using "clean" freshwater media,
Fraquil, and introduced  Hg and MeHg to cultures of Selenastrum in  the presence of various ligands and Lake
Superior nearshore waters. Finally, the research work will continue with Tetra Tech, Inc., as they develop the
EPRI-sponsored Lake Superior Mercury Cycling Model. The  model will be calibrated with our results,  and
simulations will be run to determine long-term trends and various management scenarios for Lake Superior.

    This research has allowed a preliminary mass balance of Hg and MeHg and a better understanding of Hg
sources and sinks  in Lake Superior (see Figure 1). Processes in the Upper Great Lakes appear to be analogous
        Th e Office of Research and Development's National Center for Environmental Research            19

-------
                   STAR Mercury Fate and Transport Final Progress Review Workshop
to coastal and offshore processes affecting Hg cycling and allow easier access for process-level work. The
research  group continues to work with the U.S. Environmental Protection Agency's Great Lakes Program
Office on Hg-related issues in the Great Lakes.  In addition, preparation is being made to summarize the
research  results  on a Web-based interface that features various video clips of the Lake Superior research
cruises.
                                               Atm Dep
                                               725±150
                                               4.7 ±1.5
Evasion
600
                                                                         Photodeg
                                                                         6,0
                                 Sedimentation
                                 (by difference)
                                           790
                                           8
Figure 1. Rolfhus K.R., H.E. Sakamoto, L.B. Cleckner, R.W.  Stoor, C.L. Babiarz, R.C. Back, H. Manolo-
          poulos and J.P. Hurley. The distribution of mercury  in Lake Superior. Reprinted with permission
          from Environmental Science and Technology 2003;37(5):865-872.
20
        The Office of Research and Development's National Center for Environmental Research

-------
                   STAR Mercury Fate and Transport Final Progress Review Workshop
   Factors Controlling the Production, Fate, and Transport of Mercury
                      and  Methylmercury in Aquatic  Systems

                Robert P. Mason , Andrew Heyes , Carrie Miller , and Cynthia Gilmour

       Chesapeake Biological Laboratory, University of Maryland, Solomons, MD; 2 The Academy
                   of Natural Sciences, Estuarine Research Center, St. Leonard, MD

    The goals of this project were the careful and detailed examination of the controls over Hg methylation
(Hg) and methylmercury (CH3Hg) fate and transport and demethylation in aquatic systems. Specifically, the
role of ligands and Hg complexation, especially the role of sulfide, in influencing Hg methylation was exam-
ined in detail. To improve the ability to understand the role of sulfide at low sulfide concentrations found in the
environment, it was necessary to develop a new and novel technique of measuring low levels of sulfide (nM-
uM) in natural waters. This was accomplished during this study. Overall, experiments have assessed how Hg
complexation in solution, and with the solid phase, influences Hg methylation in sediments and, subsequently,
how it influences the fate of the CH3Hg produced (see Figure 1). To measure Hg speciation in situ in field
samples,  the  technique of octanol-water partitioning of Hg that had been  used previously in  estimating
bioavailability in laboratory studies at high concentration was refined. By the use of clean techniques and other
modifications, it can now be used in environments where Hg levels are in  the pM range. In addition, studies
have examined the importance of reaction kinetics, in addition to thermodynamic controls, over Hg methyla-
tion as aquatic systems are often not at thermodynamic equilibrium. Laboratory studies have focused on deter-
mining the binding constants for Hg  and CH3Hg to solid phases, but field studies have measured the distri-
bution of Hg across a diverse  range of ecosystems, and the relative methylation potential of these  systems.
This project's research results have  shown that there  is a relationship  between the measured  short-term
methylation rate upon spiking samples with Hg and the in situ CH3Hg concentration. This relationship can be
demonstrated from consideration of thermodynamic relationships and first-order  kinetics. Furthermore, the
relationship between Hg methylation and demethylation has been examined from a kinetic standpoint, and this
has led to insights into the relative bioavailability of inorganic Hg(II) and  CH3Hg(II)  to these microbial pro-
cesses. The importance of microbial community structure in Hg methylation is another important variable that
was investigated using studies across ecosystems, and both spatially  and temporally within an ecosystem. The
overall focus of the study has been geared toward assessing if and why added ("new") Hg is more bioavailable
than the exisiting Hg pool. A model that incorporates speciation and other factors to predict Hg bioavailability
for methylation and CH3Hg fate is being developed.
        Th e Office of Research and Development's National Center for Environmental Research            21

-------
                   STAR Mercury Fate and Transport Final Progress Review Workshop
          105
        OJ1CP
        re
          102
                           CH^Hg production
                                                            Hg2+ + xS2- + yH+ = HgS H 2-2*-
HgS°
Hg(SH)2<
             0.1           1           10
                    HgSPGonc.(pgcni3)

                          (a)
                                                                         (b)

 Figure 1. (a) Comparison of the estimated rate of uptake of neutral Hg sulfide complexes, estimates based on
          measured KQW values and assuming that the complex is either HgS , or as has been suggested,
          HgSHOH . Solid symbols indicate estimated rate and concentration based on thermodynamic mod-
          eling. Methylation rates are plotted as the open circles. The uptake rate (kD) is significantly greater
          than the methylation rate (kM); (b) a diagrammatical representation of the principal processes that
          govern the rate of methylation by sulfate reducing bacteria exposed to Hg under sulfidic conditions.
          In this figure, the pathway and rate, kB, is associated with Hg sequestration reactions within the cell
          that do  not lead to Hg methylation. The overall rate of methylation is determined by the relative
          rate of these processes. Taken from Benoit et al. (2003).
22
        The Office of Research and Development's National Center for Environmental Research

-------
                   STAR Mercury Fate and Transport Final Progress Review Workshop
         Response of Methyl mercury Production to Changes in Hg
          Loading:  A Comparison  of Hg Isotope Addition Studies

      Cynthia C. Gilmour1, Andrew Heyes2, Robert P. Mason2, Carrie Miller2, and Michael Rearick2

            1 The Academy of Natural Sciences, Estuarine Research Center, St. Leonard, MD;
               2 Chesapeake Biological Laboratory, University of Maryland, Solomons, MD

    This study attempts to answer the question, "How much does methylmercury in ecosystems change in
response to a change in mercury loading?" Imbedded in this question are the issues of timing of response,
magnitude of response, the influence of scale, and the differences in responses across watershed components.
To address this question, two powerful techniques have been used that are novel to the  Hg research commu-
nity: (1) the use of stable Hg isotopes, and (2) the manipulation of systems of various sizes up  a boreal lake
and its watershed. The use of Hg stable isotopes allows the tracking of Hg spikes separately from Hg already
accumulated in ecosystems.

    During the course of this study, Hg  addition experiments were conducted at different scales and in two
different ecosystems, the Florida Everglades and boreal watersheds  at the Experimental  Lakes Area in north-
west Ontario. Some of the  studies also examined the influence of sulfate, dissolved organic carbon,  and the
drying and rewetting of soils on methylation. The main METAALICUS study examined the response to Hg
addition at the whole-ecosystem level. Hg stable isotopes were delivered to the entire  catchment, including
uplands, wetlands, and a first-order drainage lake. The L658 whole-ecosystem Hg addition began in 2001, and
has continued with Hg stable isotope dosing at roughly four times the atmospheric deposition through 2003.
Upland, wetland, and lake  are dosed with separate  isotopes to study behavior separately. In both the Ever-
glades and METAALICUS studies, the U.S. Environmental Protection Agency's Science to Achieve Results
(STAR) Program has funded the Hg methylation component of large, multi-investigator programs.

    A general conclusion from all of the studies is that Hg newly deposited to zones of methylation  is more
readily converted to methylmercury  (MeHg) than existing Hg pools. Mercury spiked into aquatic enclosures,
and into L658 directly, was rapidly transported  to the primary zone of methylation at the sediment surface. In
all of the aquatic studies, a significantly larger fraction of this "new" Hg  in surface sediments was found as
MeHg relative to existing pools (See Figure  1 for an example from L658). The geochemical behavior of "new"
Hg in sediments also is different than that of existing Hg pools, supporting the idea that recently deposited Hg
is more  bioavailable for methylation. A key question in the L658 study is the timing to reach a new equili-
brium concentration of MeHg in sediments  and biota. Through 2003, the concentration  of Me202Hg (derived
from the 202Hg added directly to the lake surface) has continued to increase  in surficial sediments, young of the
year perch, and other matrices. However, the  rate of increase in surface sediments appears to be slowing,
suggesting that equilibrium with the new Hg load to the lake surface may be approaching.

    The aquatic additions studies at various size scales have shown that the eventual amount of MeHg formed,
and the  timing of its formation,  is highly dependant on the rate and magnitude of Hg  delivery to zones of
methylation. It appears that the more rapidly Hg is delivered to zones of methylation (generally sediments), the
higher the fraction  converted to MeHg. In studies where a range of Hg dosing rates were examined, the re-
sponse in sediment MeHg production to Hg stable isotope doses was generally linear. However, the slope of
the response varied dramatically among study sites.

    Hg deposited to uplands and wetlands  with dry surfaces  is deposited away from zones of methylation.
Further, Hg deposited to these areas at ELA has been very immobile in the 2 to 3 years since initial dosing.
There appears to be little methylation  of this Hg,  and little or no transport of methylated isotope downslope.
Mercury delivered to the surface of the wetland has not yet penetrated into the zone of methylation at the top
of the water table, often 10-50 cm below the plant surface. However, Hg delivered to the  lake surface has been
           The Office of Research and Development's National Center for Environmental Research         23

-------
                  STAR Mercury Fate and Transport Final Progress Review Workshop
extensively methylated in the L658 wetland, during times when lake level rise forced lake water horizontally
into methylation zones in the wetland.

    Overall, these studies show that the change in MeHg production in response to a change in Hg loading
depends in large part on the timing and magnitude of transport to sites of methylation. Although little of the
isotopic Hg delivered to the L658 uplands and wetlands has moved to date, budgets for ambient Hg strongly
suggest that Hg deposited to uplands and wetlands are important contributors to MeHg in lake food webs. It is
anticipated that transport and methylation of the Hg  isotopes will be delivered to the upland and wetland in
future years. The timing and magnitude of the movement, and subsequent bioavailability for methylation in the
wetland and in the lake, will be critical to modeling the effects of changing Hg deposition on aquatic food
webs.
   0.10
0.08
O)
i
0.06
X
0
0.04-
0.02-
o.oo
                            0  2001
                            •  2002
                            •  2002 very shal low sites
                                           °
                                                     0.4
                                                  en
                                                  I
                                                     0.3-
                                                  ^0.2
                                                       0.1 -
                                                     0.0
                                                                    O
                                                                       0
                2468
                   V\foter Depth, m
                                           10
12
                                                                0     2      4     6     8     10
                                                                          V\teter Depth, m
                                                    12
            Native MeHg/Hg
         2001 avg = 0.023 +/- 0.009
         2002 avg = 0.032 +/- 0.014
                                                                      Me202Hg/202Hg
                                                                 2001 avg = 0.091  +/- 0.074
                                                                 2002 avg = 0.073 +/- 0.048
                                          202
     Figure 1.  Methylation of newly deposited   Hg vs. existing Hg pools in L658 surface sediments.
24
          The Office of Research and Development's National Center for Environmental Research

-------
                   STAR Mercury Fate and Transport Final Progress Review Workshop
         Methyl mercury Sources to  Lakes  in Forested Watersheds:
            Has Enhanced  Methylation Increased  Mercury  in Fish
                       Relative to Atmospheric Deposition?

  EdwardB. Swain1, B.A. Monson1, J. Jeremiason2, P.L. Brezonik3, E. Nater3, J. Cottier3, B.M. Johnson3,
                      N.A. Nines3, D. Engstrom4, J. Almendinger4, and R. Kolka5

     1 Minnesota Pollution Control Agency, St. Paul, MN; 2Gustavus Adolphus College, St. Peter, MN;
          University of Minnesota,  St. Paul, MN; Science Museum of Minnesota, St. Paul, MN;
       U.S. Department of Agriculture, Forest Service, North Central Research Station, St. Paul, MN

    This project hypothesizes that after WWII there was increased atmospheric deposition of sulfate enhanced
mercury  methylation and exacerbated Hg  concentrations  of fish in  sensitive ecosystems. The  specific
objectives of this  study are to:  (1) conduct microcosm (wetland and lake sediment) and whole  wetland
experiments to  elucidate the effect of increased sulfate loading; (2) establish the relative importance  of
atmospheric, in-lake,  and wetland sources of MeHg to  lakes in forested watersheds;  (3) determine the net
retention  and source  strength of different wetland types; and (4) delineate and classify wetlands  in other
watersheds, and estimate Hg and methylmercury watershed loads to the receiving lakes.

    The hypotheses are being addressed in the mercury-sensitive region of north-central Minnesota, where
more than 80 percent of assessed lakes are on the U.S. Environmental Protection Agency's list of "impaired
waters" due to mercury contamination offish.  Utilizing the facilities of the Marcell Experimental Forest (MEF,
USDA Forest Service), net methylation of mercury was compared in a variety of wetlands and in the sediments
of Spring Lake. Field sampling  was  conducted at Spring Lake related to THg and MeHg mass budgets. The
exports of THg and MeHg were quantified from  six different  wetland regimes. Enhanced methylation
experiments were conducted in the laboratory. In addition, an irrigation system was constructed in one-half of
S6, a black spruce-tamarack wetland  to increase atmospheric sulfate deposition four times relative to the
control half,  and  construction  of a GIS wetland database  was completed  in the MEF and  in 12 other
watersheds.

    Sulfate Addition Experiment to Wetland S6. Prior to the first sulfate (S04) addition, MeHg concentra-
tions were similar between the control (0.47±0.10 ng*L-1, n=12; mean ± one standard error) and experimental
0.52±0.05 ng*L-1,  n=18) halves. Following the first addition in May 2002, MeHg porewater concentrations
increased to 1.63 ± 0.27 ng*L-1 in 2 weeks after the addition, resulting in a threefold increase.  Subsequent
additions in 2002 did not further raise porewater MeHg, but MeHg in the experimental half remained  elevated
relative to the control. MeHg concentrations in flow from the wetland increased following each S04  addition
(see Figure 1). A second full year of sulfate treatments was completed in 2003.

    Spring Lake Mercury Budgets. Quantification and  modeling of evasion rates of elemental mercury were
made during the daytime, yielding an average loss of 1.1 ng m^h"1 for 2001-2002. Photodegradation rates  of
MeHg  were quantified and yielded loss rates of 0.4 y"1 in mid-July and 0.04 y"1 in mid-October. Profiles  of
MeHg in sediment porewater suggest that early in the year, there may be a small diffusive flux of MeHg from
sediment to water (0.06 |ag m"2 yr"1).  Several serum bottle experiments were performed with homogenized
sediments to further investigate the relationship between net Hg methylation and factors such as carbon, nitro-
gen, and sulfur. These experiments indicate that in these lake sediments, mercury availability is a major factor
in net methylation, based on the addition of mercuric ion or the reduction of Hg availability by the binding
with sulfides.

    Research results clearly demonstrate that: (1) wetlands are relatively strong sources of methylmercury to
lakes, and (2) sulfate additions to wetlands enhance methylation and increase MeHg export from the wetland,
suggesting that decreasing sulfate deposition rates would lower MeHg export from wetlands. Lower MeHg
export would likely result in lower mercury concentrations in the fish of receiving waters.
        The Office of Research and Development's National Center for Environmental Research           25

-------
                   STAR Mercury Fate and Transport Final Progress Review Workshop
    The next steps for this research project are to:  (1) quantify the effect of chronically increased sulfate
loading on net methylation rates; can we predict how much less methylmercury is exported due to post 1990
Clean Air Act reductions in atmospheric sulfate deposition? (2) study the details of how sulfate stimulates net
methylation, including methylation, demethylation, sulfate reduction, and the fate of the added sulfur; and (3)
utilize the new knowledge on in-lake methylation and methylmercury export from wetlands in computer mod-
els of mercury in a variety of lakes; does our new knowledge allow us to better predict which lakes will have
elevated mercury in fish?
                                     SCX
                       SCX
                     SCX
        0)
        c
        o
        CD
        E
             0
             Mar
May
Jul
Sep
Nov
                                                  2002
Figure 1.   Concentration of methylmercury in water flowing out of the S-6 wetland, half of which had been
           experimentally treated with sodium  sulfate to mimic a 4x increase in atmospheric sulfate deposi-
           tion. The sulfate doses (vertical dotted lines) were added with an irrigation system.
26
        The Office of Research and Development's National Center for Environmental Research

-------
                   STAR Mercury Fate and Transport Final Progress Review Workshop
                 Cycling of Mercury in Saginaw Bay Watershed

   Jerome Nriagu1, Hong Zhang2, Steve Lindberg3, Gerald Keeler1, John Lehman', andXia-Qin Wang1

   Department of Environmental Health Sciences, School of Public Health, University of Michigan, Ann
 Arbor, MI; Tennessee Tech University, Cookeville, TN; Oak Ridge National Laboratory, Oak Ridge, TN

    This project is designed to evaluate the role of photochemistry on the cycling of mercury in Saginaw Bay
watershed. Focus of the investigations includes photodependent biotic, abiotic, physical, or chemical process-
ses, and the reaction mechanisms of interest may be homogeneous or heterogeneous. The transect from Sagi-
naw Bay to Lake  Huron system has well-defined gradients in chemical parameters (including chlorophyll-a,
dissolved organic carbon, phosphorus, nitrogen, chloride, and  suspended particulates), trophic conditions, tax-
onomic composition and biomass, and provides a unique site for the study. The ambient airborne Hg levels and
fluxes of Hg were measured at many sites covering forest soils, agricultural soils, wetlands, city park soils,
beach sands, and industrial/municipal waste dumps. In  addition, the Hg levels were measured in soil samples
from the stations and in water samples from the rivers and various parts of the Saginaw Bay. From the survey
data, a number of sites were selected, which were  intensively monitored over a period of time. This research
project focuses  on the  following: (1) inventory of tributary  input of mercury into Saginaw Bay; (2) Diel
changes of dissolved gaseous mercury (DGM) in nearshore surface waters of Saginaw Bay; (3) dark oxidation
of DGM in the water from surface shorewater of Saginaw Bay; (4) photoreduction of Hg(II) in water samples
from Saginaw Bay studied in the laboratory with and without spikes  of photoactive chemicals; (5) estimated
rates of air/water exchange of Hg in aquatic environments in the watershed; and (6) thermal desorption of Hg
from Saginaw Bay area in relation to the likely effects of forest fires on release  of Hg from the area. The
measurements led us to conclude that little of the mercury that is deposited in Saginaw Bay is retained in the
basin. This raises  the question, "What is the ultimate sink for much of the mercury that is released in North
America?"
        The Office of Research and Development's National Center for Environmental Research            27

-------
                  STAR Mercury Fate and Transport Final Progress Review Workshop
     Chemical and Biological Control of Mercury Cycling in Upland,
         Wetland, and  Lake Ecosystems in  the Adirondack Region
                                       of New York

                Charles T. Driscoll1, Robert Newton2, Ron Munson3, and Joseph Yavitf*

        Department of Civil and Environmental Engineering, Syracuse University, Syracuse, NY;
   2Clark Science Center, Department of Geology, Smith College, Northampton, MA; 3Tetra Tech, Inc.,
           Pittsburgh, PA; 4Department of Natural Resources, Cornell University, Ithaca, NY

   A detailed study was conducted on the biogeochemistry  of mercury (Hg) at Sunday Lake, a forested
wetland/lake/watershed in the Adirondack region of New York. The objective of this study was to improve
understanding of the inputs, transport, transformations, and fate of Hg in upland forest, wetland, and lake
ecosystems. To accomplish this objective, a detailed watershed mass balance study was conducted, including
measurements of wet deposition, throughfall, litterfall, soil, soil water, vegetation, wetland porewater, hydrol-
ogy, surface water chemistry, and aquatic biota. Wet Hg deposition was 10.8 |_ig/m -yr, with 0.6 percent occur-
ring as methyl Hg. Forest vegetation was important in mediating the inputs of Hg to the forest floor. Inputs of
total Hg from litterfall and throughfall greatly exceeded wet Hg deposition, suggesting that dry deposition is
the dominant Hg input to this forest ecosystem. Concentrations and fluxes of total Hg were elevated in forest
floor leachate, with soil solution concentrations decreasing in the mineral soil. Likewise, soil concentrations of
Hg were highest in the forest floor (13-188 ng/g), and  decreased with increasing depth in the mineral soil.
Concentrations of Hg in surface waters ranged from 1.9 to 4.6 ng/L, with methyl Hg concentrations from 0.2 to
2.53 ng/L. Atmospheric Hg deposition was retained in the watershed.

   Sunday Pond is a sink for inputs of total Hg. However, the watershed, particularly riparian wetlands, and
the lake were net sources of methyl Hg to downstream  surface waters. Concentrations of Hg increased with
each trophic level in the aquatic food chain. Mercury bioconcentration factors were lower in zooplankton and
in fish than have been reported in other studies, probably due to binding of methyl Hg with high concentrations
of organic solutes. The Hg biogeochemistry data have been used to  calibrate the Mercury in Adirondack
Wetlands Lakes and Terrestrial Systems  (MAWLTS) Model for the site. The project team is currently com-
pleting the final report and a series of articles for a peer-reviewed journal.
28      The Office of Research and Development's National Center for Environmental Research

-------
                   STAR Mercury Fate and Transport Final Progress Review Workshop
     Processes Controlling the Chemical Speciation  and  Distribution
                    of Mercury From Contaminated  Mine Sites

           Gordon E. Brown, Jr.1, Christopher S. Kim1'2, Gregory V. Lowry1'3, Samuel Shaw1'4,
             Aaron J. Slowey1, Stephen B. Johnson1, Mae S. Gustin5, and James J. Rytuba6

       1 Department of Geological and Environmental Sciences, Stanford University, Stanford, CA;
      Earth Sciences Division, Lawrence Berkeley National Laboratory, Berkeley, CA; Department of
     Civil and Environmental Engineering, Carnegie Mellon University, Pittsburgh, PA;  Department
        of Earth Sciences, University of Oxford, England; Environmental and Resource Sciences,
           Fleischmann Agriculture, University of Nevada, Reno, NV; 6U.S. Geological Survey,
                        Geologic Division, Middlefield Road, Menlo Park, CA

    The major objectives of this project are to:  (1) determine  the chemical speciation and relative abundance
of different forms of Hg in mine wastes from the California Coast Range; (2) investigate the role of colloidal
transport as a mechanism for dispersal of Hg from waste sites; (3) identify the mode of Hg sorption in down-
stream sediments and fine-grained precipitates; (4) determine the effects of complexing ligands on the desorp-
tion and sorption inhibition of Hg; and (5) measure and correlate the emission of Hg into the atmosphere with
Hg speciation, climate, and geologic factors. A combination of synchrotron-based x-ray absorption fine struc-
ture (XAFS) spectroscopy, scanning (SEM) and transmission  electron microscopy (TEM), x-ray powder dif-
fraction, and selective chemical extractions have been used to  determine the molecular-scale speciation of Hg
in different size fractions from mining wastes and in colloidal  fractions from column experiments designed to
examine the colloidal transport and  speciation of mercury (see  Figure 1).  The speciation of Hg has been
determined in mine wastes from a number of abandoned Hg-mines in the California Coast Range, finding that
highly insoluble cinnabar and metacinnabar are the primary minerals present. In addition, a number of highly
soluble Hg-containing species have been identified, including mercuric chloride (HgCl2), montroydite (HgO),
schuetteite (Hg302S04), and terlinguite (Hg2OCl), which may be disproportionately larger contributors of ionic
Hg to the surrounding environment than the mercuric sulfides.

    One key finding from the laboratory column experiments, coupled with XAFS spectroscopy and TEM, is
that most of the colloids are cinnabar or metacinnabar, which is surprising in light of the widely held belief that
most heavy metals  attach to colloidal particles  as  sorption complexes. Although Hg(II) forms  strong, inner-
sphere adsorption complexes on Al- and  Fe-oxyhydroxide particles  in laboratory sorption/XAFS studies,
significant amounts of sorbed Hg(II) have not been detected in natural samples from these mine sites. Potential
connections have been explored between Hg speciation and the  extent of gaseous Hg emissions from mine
wastes employing the spectral databases  that we have generated using  XAFS spectroscopy (Stanford Group)
and controlled flux measurements at a number of localities (University of Nevada, Reno Group). Hg emissions
were measured on the same samples on which XAFS measurements were performed  at constant temperature
under both light and dark conditions based on the known ability of light to increase Hg flux rates.  Significant
results from the comparisons of Hg speciation and Hg emissions include the finding that the lightdark ratio is
greater among samples containing metacinnabar (HgS, cubic) relative to samples containing just cinnabar
(HgS, hexagonal) (i.e., light  enhancement is more pronounced  in samples containing  metacinnabar). This
finding has implications for the atmospheric release  and  cycling of Hg from specific types of mine wastes,
particularly since our previous XAFS studies have shown that calcines (roasted ore) typically contain high
proportions of metacinnabar, which is generated by the ore roasting process. The flux measurements indicate
that non-HgS phases such as HgO and Hg3Cl302H may be significantly larger contributors of gaseous Hg than
HgS phases. Therefore, regions with appreciable  Hg-oxides  and Hg-chlorides, such  as hot-spring type  Hg
deposits, may be of more concern regarding atmospheric Hg emissions than areas where HgS dominates. The
most recent and continuing focus of this project has been on determining the effect of simple low molecular
weight organic  acids generated by plants  in revegetated Hg-mine environments on  the dissolution of Hg-
containing minerals.
        The Office of Research and Development's National Center for Environmental Research            29

-------
                   STAR Mercury Fate and Transport Final Progress Review Workshop
                      Figure 1.  Speciation and distribution of Hg in mine wastes.
30      The Office of Research and Development's National Center for Environmental Research

-------
                   STAR Mercury Fate and Transport Final Progress Review Workshop
  Mercury and  Methyl mercury Cycling in the Coastal/Estuarine Waters
        of Long  Island Sound and Its River-Seawater Mixing Zones

         William F. Fitzgerald, Pieter T. Visscher, Prentiss H. Balcont, ChadR. Hammerschmidt,
                       CarlH. Lamborg, Chun-Mao Tseng, and Ming Z. Chung

                Department of Marine Sciences, University of Connecticut, Groton, CT

    Consumption of marine fish and seafood products is the principal pathway by which humans are exposed
to the very toxic organo-mercurial, methylmercury (MeHg). Thus, biologically productive nutrient-rich near-
shore regions, which  support major commercial and recreational fisheries, are of special interest. This U.S.
Environmental Protection Agency's Science to Achieve Results  Hg research project is focused on one of the
"Great Waters" of the United States, Long Island Sound (LIS). A comprehensive field and laboratory investi-
gation has been conducted of the physicochemical/microbiological reactions and processes controlling Hg cy-
cling, speciation, and  bioavailability in the waters and sediments of LIS and its watershed/coastal water inter-
face (i.e., Connecticut River; East River). The specific objectives are focused on major features of the aquatic
biogeochemistry  of Hg, particularly MeHg production, elemental mercury (Hg ) cycling and emissions, inter-
actions between terrestrial watersheds, rivers and nearshore marine waters, and the role of organic matter in
governing the bioavailability Hg for  competing methylation/reduction reactions. Several important accom-
plishments will be emphasized, as they affect our understanding of the behavior and fate of methyl and inor-
ganic Hg in the marine environment.

    This project hypothesized and confirmed that:  (1) in situ sedimentary production is the primary source of
MeHg to the waters and biota of LIS, and (2) rates of MeHg production depend on the activity of methylating
bacteria and availability of inorganic Hg. Potential rates of Hg methylation were greatest in mid-summer (i.e.,
greatest bacterial activity) and varied  spatially in LIS sediments. MeHg production was related inversely to
sedimentary organic matter, which is a major control on the partitioning and subsequent bioavailability of in-
organic  Hg in the low-sulfide pore waters of LIS. Potential rates of MeHg  production were related inversely
with the distribution coefficient (KD) of inorganic Hg and positively with the concentration of inorganic Hg in
pore waters. Based on diffusional flux estimates,  in situ sedimentary production is  the dominant source of
MeHg to LIS (see Figure 1). This flux is likely to be enhanced by infaunal mixing of sediment; a process that
we suspect also promotes Hg methylation. Because in situ production is the major source of MeHg to LIS, then
by extension, it should be the principal source in other comparable near shore systems. In addition, sedimen-
tary methylation can account for most of the MeHg in primary producers of LIS.

    This project's research results indicate that most Hg is  associated with organic matter in river water and
seawater. New and innovative measurements have been conducted to establish the nature and bonding strength
of these Hg-organic matter associations. Results  indicate that Hg complexing agents are present in the dis-
solved-phase  (< 0.2 |im) at ca. < 1-30 nN, with log conditional stability constants (log K') in the range of
22-25.  Such  ligand characteristics explain why the majority of ionic inorganic Hg dissolved in fresh and
saltwaters is associated with organic complexes (> 99% in freshwater, > 50% in seawater). Further, terrestrial-
ly derived and marine derived complexing ligands appear to  be equally important in LIS.

    In addition, the prominent role of Hg° production and evasion has been confirmed in an important coastal
ecosystem. Annual Hg emissions from LIS are estimated at 80 kg, which is comparable  to fluvial inputs. In-
deed, about 35 percent of the Hg input (240 kg/y) to LIS is remobilized to  the atmosphere. This flux is prin-
cipally anthropogenic, as most (> 70%) Hg entering LIS has  a pollution origin. Production  of Hg° competes
with the in situ biological synthesis of MeHg. This work is  designed to increase understanding of in situ Hg°
production and emissions and their role in aquatic and atmospheric Hg cycles, which will help constrain Hg
biogeochemical cycling and mixing models.
        The Office of Research and Development's National Center for Environmental Research

-------
              STAR Mercury Fate and Transport Final Progress Review Workshop
                             Atmospheric Deposition
   Salt Marshes
        +0.3
                     WPCFs
                    +14 + 02
                                Rivers
                               +3+1
                                    +0.7 + 0.2
Tidal Exchange
      -1  + 1       Bioaccumulation
                        -10 + 3
                Burial
                 -5 + 5
In Situ Methylation
       +11+4
                                                            Atlantic Ocean
Figure 1.  Estimated current methylmercury mass balance for Long Island Sound. In: Balcom, et al. Mercury
        sources and cycling in the Connecticut River and Long Island Sound. Marine Chemistry (Sub-
        mitted).
      The Office of Research and Development's National Center for Environmental Research

-------
                  STAR Mercury Fate and Transport Final Progress Review Workshop
          Pathways, Enzymes, and Trace Metals:  The Importance
                of Understanding the Biochemistry of Mercury
                                        Methylation

                    Francois M.M. Morel1, Eileen B. Ehstrom2, and J.M. Benoi/

       Department of Geosciences, Princeton University, Princeton, NJ; 2Department of Civil and
      Environmental Engineering, Princeton University, Princeton, NJ; 3Department of Chemistry,
                                  Wheaton College, Norton, MA

    Mercury methylation by sulfate-reducing bacteria (SRB) is the predominant mechanism for methylmer-
cury production in freshwater and coastal systems, yet little is known about the physiology and biochemistry
of Hg methylation in SRB. Understanding the metabolic pathways for mercury methylation, the enzymes re-
sponsible, and the coenzyme requirements is critical to understanding the basis for methylmercury formation
and to designing successful strategies for its mitigation in the environment.

    The biochemistry of SRB can  vary significantly between groups, with some SRB genera (incomplete-
oxidizers) only capable of the catabolic conversion of carboxylic acids to acetate, while complete-oxidizing
SRB oxidize reduced carbon substrates all the way to C02 via the citric acid cycle or the acetyl-coenzyme A
(CoA) pathway. Previous research on one incomplete-oxidizing SRB strain, Desulfovibrio desulfuricans LS,
implicated the corrinoid-containing enzyme in the acetyl-CoA pathway, presumably a minor biosynthetic path-
way in this strain, as being key to mercury methylation. To assess whether the acetyl-CoA pathway is the con-
trolling biochemical process for methylmercury production in SRB, five incomplete-oxidizing SRB strains and
two Desulfobacter strains that do not use the acetyl-CoA pathway for major carbon metabolism were assayed
for methylmercury formation and acetyl-CoA pathway  enzyme  activities (see Table  1). Three of the SRB
strains also were incubated with chloroform to  inhibit  the corrinoid-containing enzyme in the acetyl-CoA
pathway. So far, all species that have been found to have acetyl-CoA activity are complete oxidizers that re-
quire the acetyl-CoA pathway for basic metabolism, as well  as methylate mercury. Chloroform inhibits Hg
methylation in these species either by blocking the methylating enzyme or by indirect effects on metabolism
and growth.  However, four incomplete-oxidizing strains have been identified that clearly do  not utilize the
acetyl-CoA pathway either for metabolism or mercury methylation (as confirmed by the absence of chloroform
inhibition). Hg methylation is thus independent of the corrinoid-containing enzyme in the acetyl-CoA pathway
and may not require vitamin B]2 in some and perhaps many incomplete-oxidizing SRB strains.

    Vitamin B]2 has been long considered to be the Hg  methylating coenzyme in the enzymatic formation of
methylmercury,  because it was believed to be the only major methylating coenzyme that  would donate a
methyl-carbanion group able to bind with inorganic mercury.  In addition, vitamin  B]2 can methylate mercury
abiotically, and a corrinoid compound was found to be  important in the methylation of Hg by Desulfovibrio
desulfuricans LS. Currently, investigation of the role of vitamin B]2 in mercury methylation is being conducted
in the incomplete-oxidizing strain D. africanus.
        The Office of Research and Development's National Center for Environmental Research

-------
                   STAR Mercury Fate and Transport Final Progress Review Workshop
      Table 1. Relationship between Hg methylation capacity and presence of the acetyl-CoA pathway.
       Genus
Species
Hg Methylation
Acetvl-CoA Pathway"
Incomplete Oxidizer
Desulfovibrio

Desulfobulbus

Desulfobacter
africanus
desulfuricans LS
desulfuricans subsp.
desulfuricans
vulgaris subsp. vulgaris
Marberg
propionicus Ipr3
propionicus MUD
BG-8
yes
yes0

-------
                   STAR Mercury Fate and Transport Final Progress Review Workshop
      Quantifying the Role of Significant Natural Sources and Sinks
                     in the Biogeochemical Cycle of Mercury

   Mae Sexauer Gustin1, Jay Arnone2, Gordon Brown3, James Coleman2, Mark Engle1, Jody Ericksen1,
       Toby Frescholtz, Tony Giglini1, Dale Johnson1, Steve Lindberg4, Daniel Obrist1, Chris Kim3,
                   David Nacht1, Jim Rytuba5, Dave Schorr an2, and Richard Zehner1

       Department of Environmental and Resource Sciences, University of Nevada-Reno, Reno, NV;
       Desert Research Institute, Reno, NV; Department of Geological and Environmental Sciences,
           Stanford University, Stanford, CA; Oak Ridge National Laboratory, Oak Ridge, TN;
                       U.S. Geological Survey,  Geologic Division, Menlo Park, CA

    Mercury (Hg) is considered a global pollutant because of its presence as a volatile phase, which facilitates
transport and distribution via the atmosphere pathway.  To assess the local, regional, and global  impacts of
atmospheric Hg, significant sources and sinks must be identified and their role  quantified. Work associated
with two research projects at the University of Nevada (UNR) funded by the U.S. Environmental Protection
Agency's (EPA) Science to  Achieve Results (STAR) Program on "Light enhanced Hg volatilization from
substrate: mechanisms responsible and in situ occurrence" and "Processes controlling the chemical/isotopic
speciation and distribution of Hg from contaminated mine sites," focused on developing a quantitative un-
derstanding of the factors influencing Hg emissions from HG-enriched substrate; developing a database of
emissions from representative source areas; developing a preliminary database on the speciation of Hg in the
air above Hg-enriched sites; and providing a framework for scaling up Hg emissions for area sources. Research
questions were addressed by measuring Hg flux from terrestrial sources using micrometeorological methods
and dynamic field flux chambers, and through controlled laboratory studies using a gas exchange system. This
work has led to the recognition that broad areas of natural  Hg enrichment are a more significant source of
atmospheric Hg than previously realized (see Figure 1). Other research has indicated that there are  previously
unrecognized diffuse sources of atmospheric Hg  (landfills, forest fires, sewage sludge amended soils). With
this new information on sources, the question arises, "Are there unrecognized sinks?" Research funded by the
U.S. EPA's Experimental Program to Stimulate Competitive Research (EPSCoR)  at UNR and Desert Research
Institute focused on  addressing the role of plants in the biogeochemical cycle of Hg using two state-of-the-art
7.3 x 5.5 x 4.5 m (L x W x D) open flow enclosed mesocosms. This project demonstrated that plants can act as
a sink for atmospheric Hg and as  a pathway by which Hg derived from the atmosphere may be transferred to
terrestrial ecosystems.

    These three research projects have developed important information that may be applied towards assessing
the role of natural sources in the global Hg budget and have advanced the understanding of the biogeochemical
cycle of Hg. This information is important for determining the need for and the  effectiveness of controls for
major anthropogenic sources of Hg such as coal-fired utility boilers.

    This work may be and has been used to impact regulatory and management decisions at local and regional
scales. In addition, it has been used for assessment of the health hazards for Hg exposures via the atmospheric
pathway at Hg-contaminated mine sites; assessment of the impact of releases from Hg-enriched substrate on
local ecosystems; and  development of remediation strategies for Hg-enriched areas based on quantified Hg
releases. The better understanding of the significance of plants  as a pathway for Hg to be transferred to ter-
restrial ecosystems is important information for watershed management decisions associated with  Hg-impacted
ecosystems and watersheds with fish consumption advisories for Hg.

    Research from these projects has significantly furthered the understanding of the role of natural sources in
the biogeochemical  cycling of Hg and revealed significant data gaps that still need to be addressed. To fill
some data gaps, current research at UNR funded by the U.S. EPA STAR program is focusing on  generating
data that will allow us to develop a natural source emission estimate for the United States and a more complete
understanding of atmospheric Hg-lithosphere interactions. Additional work funded by the National Science
Foundation is focused on furthering the understanding of plant-soil-atmosphere Hg exchange.


        Th e Office of Research and Development's National Center for Environmental Research           3 5

-------
                            STAR Mercury Fate and Transport Final Progress Review Workshop
    Medicine Lake

       Long Valley



         Knoxville

     Peavine Peak

     Flowery Peak



         Ivanhoe

        New Idria



Steamboat Springs

     Sulphur Bank
  Areas of recent volcanic
  activity
Hg mining districts

                         Areas including
                         precious metal
                         mineralization
                              Geothermal
                                 areas
                                                10                             100

                                              Area averaged mercury flux, ng/m2h
                                                   1000
          Figure 1.  Mercury emissions for a variety of types of natural source areas. The "models" value represents
                    that estimated for global belts of Hg-enriched substrate.
                 The Office of Research and Development's National Center for Environmental Research

-------
           STAR Mercury Fate and Transport Final Progress Review Workshop
                                    Index
                               Brown Jr., G.E., 29
                               Driscoll, C.T., 28
                               Fitzgerald, W.F., 31
                               Gilmour, C.C., 23
                               Gustin Sexauer,  M, 35
                               Hurley, J.P., 19
                               Mason, R.P., 21
                               Morel, P.M., 33
                               Nriagu, J., 27
                               Swain, E.B., 25
The Office of Research and Development's National Center for Environmental Research          37

-------