MERCURY TRANSPORT
AND FATE THROUGH
k A WATERSHED
JANUARY 2006
United States
Environmental Protection
Agency
-------
PREPARED FOR
U.S. Environmental Protection Agency, Office of Research and Development
National Center for Environmental Research
Washington, DC
PREPARED BY
ICE International
Research Triangle Park, NC, and Fairfax, VA
Under EPA Contract No. 68-C-03-137
Work Assignments 00-01,01 -01, and 02-03
-------
DISCLAIMER
The research described in this document has been funded wholly by the United States Environmental Protection Agency
(EPA) under the Science to Achieve Results (STAR) grants program. The information does not necessarily reflect the
views of the Agency and no official endorsement should be inferred. Mention of trade names or commercial products
does not constitute endorsement or recommendation by EPA for use.
The information presented in this synthesis report is intended to provide the reader with insights about the progress
and scientific achievements of STAR research grants. The report lists the grantees whose research is discussed and also
indicates where more detailed peer-reviewed scientific data can be found. This report is not sufficiently detailed nor is
it intended to be used directly for environmental assessments or decision making. Readers with these interests should
instead consult the peer reviewed publications produced by the STAR grantees and conduct necessary data quality
evaluations as required for their assessments.
-------
-------
CONTENTS
Photo Acknowledgements
Executive Summary
111
1. Introduction
1.1 Purpose of this Report
1.2 Background
1.3 Organization of this Report
2. Overview of STAR Mercury Grants
l
2
2.1 Overview of Mercury in the Environment
2.2 FY99 STAR Mercury Grants
2.3 Relationship of 1999 Mercury STAR Grants to Office of Research and Development's Multi-Year Plans
2.4 Research Themes
3. Study Results
4
7
3.1 Biogeochemical Controls of Mercury Transformations
3.2 Biogeochemical Controls of Mercury Cycling in the Environment
11
3.3 Sources and Distribution of Mercury in Terrestrial and Aquatic Systems
13
3.4 Bioavailability and Bioaccumulation of Mercury in Aquatic Systems
16
3.5 New Methods for Mercury Analysis
4. Applications of Findings from STAR Grants
17
18
4.1 Potential Improvements to Mercury Fate and Transport Models
18
4.2 Understanding Key Variables Affecting Mercury Fate and Transport
19
4.3 Understanding Variation in Key Variables by Ecoregion and Ecotype
5. Future Research Needs
19
22
6. References Cited
24
Appendix: Grant Publication Lists
A-l
-------
'HOTO ACKNOWLEDGEMENTS
Cover page, page 2, and page 7
Experimental Lakes Area, Ontario, Canada (courtesy of Cynthia Gilmour)
Page iii
Florida Everglades (courtesy of Cynthia Gilmour)
Pagel
Adirondack lake (courtesy of Charles Driscoll)
Page 18
Calcine dumps at New Idria mercury mine, CA (courtesy of Gordon Brown)
Page 22
Steam plumes from smokestacks in Baltimore, MD (courtesy of Cynthia Gilmour)
Page A-1
Experimental spraying of sodium sulfate solution at the Marcell Experimental Forest in northern Minnesota
(courtesy of the Minnesota Pollution Control Agency).
-------
Background
The U.S. Environmental Protection
Agency's (EPA's) Science to Achieve
Results (STAR) research grants
program, managed by the National
Center for Environmental Research,
has funded significant research on
the fate, transport, and transfor-
mation of mercury in aquatic and
terrestrial environments. This report
summarizes the research and find-
ings generated through nine grants
awarded under the 1999 Request for
Applications (RFA) entitled "Mercury:
Transport and Fate through a Water-
shed" and two other closely related
STAR grants. The important scientific
findings from these grants, data gaps,
and additional research needs on this
topic are described in this synthesis
report.
The 1999 RFA funded fundamental
research on the complex chemical
and physical transformations and
movement of mercury through the
environment. The overall goal of the
research solicited on mercury trans-
port in a watershed was summarized
in the original RFA as follows:
"...to develop a better understand-
ing of terrestrial and aquatic fate
and transformation processes
(especially microbial) that medi-
ate ecological and human expo-
sures to mercury. The development
of improved models of the fate of
mercury in aquatic and terres-
trial systems in order to estimate
ecosystem response to decreased
anthropogenic inputs of mercury
is also needed."
The RFA specifically invited grant
applications addressing the following
critical research questions:
(1) For a given amount of mercury
transported into a watershed,
what is the predicted concentra-
tion of methylmercury in fish?
How do mercury and methyl-
mercury 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 environ-
mental 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 re-
gions of the United States (e.g.,
south Florida, Great Lakes,
northeast, or west)? How might
the variability be accounted for
(e.g., resource types (wetlands),
temperature regimes, microbial
communities)?
The eleven STAR grants discussed in
this report examined how methyl-
mercury transports into and within a
watershed and concentrates in fish.
Projects included studies of factors
that control the partitioning and dis-
tribution of different mercury species
in different environmental media in
aquatic and terrestrial ecosystems
(Question 1), factors that influence
bioavailiability of mercury (Ques-
tion 1), and studies of the factors
controlling mercury transformations
(Question 2). Most studies identi-
fied controlling factors that are likely
to vary by geographic region, but a
systematic examination of regional
influences was not attempted (Ques-
tion 3).
Note that the critical research ques-
tions are interrelated. For example,
partitioning of mercury into different
environmental media influences its
availability to various transforming
processes, and transforming processes
can create mercury species that move
through the environment via different
pathways. Transformation processes
also can create "sources" of given
mercury species in the environment.
Exhibit ES-1 illustrates these concepts.
This synthesis report discusses the
STAR grant project results along four
research themes. These themes are
intended to correspond with compo-
nents needed to develop an integrated
multimedia modeling framework for
understanding mercury fate from
source to fish concentrations. These
four themes are: (1) controls of
mercury transformations, (2) controls
of mercury partitioning and transport
(i.e., cycling) in the environment,
(3) sources of mercury species in the
environment, and (4) bioavailability
and bioaccumulation of mercury in
food chains. Key findings of the STAR
grant projects are organized according
to these four themes.
-------
EXHIBIT ES-l: SOURCES AND PATHS OF MERCURY IN THE ENVIRONMENT
Msttiyl Mercurjf
In Fish CHLHg*
CtHmlcal reaetwn in sediments
Study Results
(1) Biogeochemical controls of
mercury transformations.
Net methylmercury (MeHg) produc-
tion in aquatic systems is complex.
Elemental mercury (Hg(0)) can be
oxidized to Hg(II), the form that
is methylated by sulfate-reducing
bacteria (SRB), and Hg(II) is readily
reduced to Hg(0) in surface waters
via abiotic and biotic pathways.
Hg(II) chemical complexes tend to
become bound to sediment par-
ticles, whereas Hg(0) is lost to the
atmosphere via volatilization. In
addition, there are rapid conversions
between inorganic mercury (Hg(II))
and MeHg. While SRB in sediments
methylate Hg(II), MeHg is also
demethylated via both abiotic (pho-
tochemical) and biotic pathways.
STAR grant projects added to the
understanding of these processes as
highlighted below.
• STAR grant research has shown
that, as mercury is deposited to
surface waters from air, Hg(II)
becomes progressively bound in
forms that are less easily meth-
ylated, even with rapid cycling
between Hg(II) and MeHg. Thus,
newly deposited Hg(II) is more
available for methylation than old-
er mercury in some environments.
This may not be true in every
environment, but was observed in
both the Florida Everglades and
the Experimental Lake Area in
Canada.
Mercury that is deposited directly
to surface waters is readily methyl-
ated in the sediments. In contrast,
it is unclear how long it takes mer-
cury deposited to a watershed to
be methylated (e.g., via transport
to sites of methylation in surface
water sediments).
Studies demonstrated that the rate
at which SRB methylate Hg(II)
depends on a variety of factors,
including bacterial activity levels
in general (e.g., seasonal), bacte-
rial community structure (i.e.,
metabolic pathways of species
present), and the bioavailabil-
ity of Hg(II), which depends on
several factors (e.g., sulfides and
dissolved organic carbon (DOC)
concentrations) as discussed later.
The addition of sulfate to fresh-
water environments can enhance
rates of methylation, which is
expected because sulfate and
organic matter are the primary
substrates used by SRB to produce
energy. However, a build-up of
sulfides as reaction products can
reduce the availability of environ-
mental Hg(II) to the bacteria.
Demethylation is less well studied
than methylation, yet is important
in determining net methylation
rates. It occurs via both biotic and
abiotic (photochemical) pathways.
STAR grant research demonstrated
photochemical demethylation
requires UV radiation and is con-
fined to the upper surface water
layers.
The availability of Hg(II) for
SRB also depends on the cycling
between Hg(0) and Hg(II). Pho-
tochemical reduction of Hg(II)
appears to be an important reac-
tion in surface waters, resulting in
a diurnal cycle of Hg(0) formation
and release from surface water
via volatilization. The rate of net
reduction is related to several
IV
-------
factors including light intensity
and DOC. Oxidation of Hg(0) to
Hg(II) is enhanced by free radicals
in surface waters, which tend to
be formed via photolysis of DOC.
• Recently, understanding of the
major processes that impact Hg
methylation has improved so that
it is now possible to model Hg
methylation in a more compre-
hensive way than accomplished in
previous models. Such approach-
es are now being developed and
will soon be incorporated into
models. However, the current
understanding of biotic demeth-
ylation is not as complete, and
models for this reaction remain
relatively crude.
(2) Biogeochemical controls of
mercury cycling in the environment.
Mercury cycling (i.e., phase partition-
ing and inter-media transport) is
key to the distribution of mercury in
the environment. Most of the STAR
research in this area focused on pro-
cesses controlling mercury transfer
between surface water and sediment,
surface water and soils and the atmo-
sphere, and plants and the soil and
atmosphere.
• STAR grant studies indicated that
inorganic mercury (Hg(II)) in
aquatic systems generally is bound
with DOC. Complexes of reactive
Hg(II) can coagulate and adsorb
to suspended sediment particles
which are deposited to the sedi-
ment bed. The influence of DOC
on mercury partitioning in the
sediment bed is complicated and
is influenced by sulfide levels.
• Studies indicated that the bioavail-
ability of Hg(II) to SRB depends on
partitioning of Hg(II) between par-
ticulate and dissolved phases of the
sediments. As expected, methylation
rates vary inversely with the parti-
tion coefficient, Kp, of Hg(II) and
positively with the concentration of
Hg(II) in sediment pore water.
• A model of Hg(II) partitioning be-
tween sediment particles and pore
water that accounts for adsorption
and sulfide concentrations has
been developed and validated for
two ecosystems (Patuxent River,
Florida Everglades). The model's
predictions of neutral mercury-
sulfide complexes, which are the
bioavailable forms, correlated well
with the methylmercury concentra-
tions measured in field samples.
• STAR projects demonstrated
that volatilization of Hg(0) from
surface waters can be as important
as sediment burial and methyla-
tion of Hg(II) in sediments as a
mechanism of loss of inorganic
mercury from different types of
aquatic ecosystems (e.g., Long
Island Sound, wetlands).
• Mercury in soils can emit to the
air via volatilization of Hg(0) and
light-enhanced emissions of mer-
cury sulfide (and other) complex-
es, perhaps via photoreduction of
Hg(II) to Hg(0) with subsequent
Hg(0) volatilization.
• Studies demonstrated that plants
take up Hg(0) directly from the
atmosphere via their stomata, and
plants can accumulate Hg in their
leaves. The uptake of mercury
from soils via plant roots appears
to be a minor uptake pathway. At
the end of the growing season,
litterfall from deciduous plants to
soils and to surface waters is a sig-
nificant source of mercury input
to those systems.
(3) Sources and distribution of mercury
in terrestrial and aquatic systems.
Several STAR grant studies identified
major sources and means of distribu-
tion of mercury in given study areas.
Researchers identified and investi-
gated important processes influenc-
ing movement of mercury through-
out a watershed (e.g., production of
methylmercury, predominant mer-
cury fate and transport processes). In
addition, several researchers per-
formed general mass balance studies
to quantify sources and inputs within
a given watershed.
• Mass balance models of different
watersheds indicated that atmo-
spheric deposition is the primary
source of total mercury to most
aquatic systems. As indicated by
mercury loading from tributaries,
in some watersheds, runoff and
erosion also can be large sources
of mercury to a water body. Mass
balance models indicated that
watersheds and lakes serve as sinks
for total mercury, while wetlands
and lakes serve as sources for
methylmercury. These models also
indicated that dry deposition (i.e.,
via litterfall) and throughfall (i.e.,
via precipitation passing through
the plant canopy) can be more im-
portant than wet deposition. Total
ecosystem mercury deposition is
a combination of wet deposition,
litterfall, and throughfall.
• Studies indicated that in situ
methylation of Hg(II) within
aquatic systems is the dominant
source of MeHg, at least for fresh-
water lakes, wetlands (e.g., Florida
Everglades), and for estuarine
systems (e.g., Long Island Sound
and the Chesapeake Bay).
• Mercury deposited to the terrestri-
al portions of watersheds is slow
to move to sites of methylation via
erosion or runoff.
• Data from one study suggested
that methylation may also occur in
estuarine and coastal areas, per-
haps in deep waters, sediments, or
hydrothermal vents where mercury
concentrations are not greatly af-
fected by human activities.
• There is some indication from
studies in the Adirondacks that the
retention capacity of watersheds
for mercury has decreased over
the last century, although possible
reasons remain obscure.
-------
• STAR grant studies also demon-
strated the importance of subsur-
face transport and transformation
of Hg into MeHg in overall mercu-
ry cycling. The interface between
ground water and streams appears
to be an important methylation
site and source of MeHg, and may
be a more important source of
MeHg from forested ecosystems to
surface waters than expected.
• Transport of MeHg through water-
sheds may be strongly mediated
by the transport of colloids.
(4) Bioavailability and bioaccumulation
of mercury in aquatic systems.
Several STAR grants investigated
factors that influence the bioavail-
ability of inorganic mercury to
SRB for methylation and transport
up the food chain. Some of these
factors have been noted briefly
above. These and other findings are
described below.
• Newly deposited mercury can be
more bioavailable than "older"
mercury.
• Mercury appears to be less bio-
available to algae in surface waters
with higher DOC levels, perhaps
because mercury complexed with
DOC is less bioavailable to SRB for
methylation.
• Mercury concentrations in biota
in lakes are generally higher in the
spring than summer, a trend that
might result from bio-dilution by
growing phyto- and zooplankton
during the summer.
• Studies showing higher concen-
trations of methylmercury in
zooplankton from surface waters
than deeper waters suggest that
mercury in surface waters depos-
ited from the atmosphere may be
more bioavailable than mercury in
deeper waters.
• In a study of fish in an Adiron-
dack lake, bioconcentration factor
(BCF) values offish increased
with each trophic level, indicating
bioaccumulation through the food
chain consistent with previous
studies.
• When sulfide levels are high, neu-
tral mercury-sulfide complexes,
which are the bioavailable forms,
are more likely to form if the pH
is low than if the pH is high. This
relationship may help explain why
fish tend to have higher tissue
mercury residues in waters of
lower pH.
• Bioavailability of inorganic
mercury varies depending on its
chemical form and solubility, with
substantial variation apparent
across mining sites in different
geographic regions.
• Sorption of Hg(II) onto mineral
particles may effectively seques-
ter mercury in mineral soils,
particularly at mining sites. Iron
and aluminum (hydr)oxides are
particularly effective in sequester-
ing Hg(II) in colloids.
• Mixing zones between two water
masses (e.g., river/lake interfaces,
estuaries) demonstrate enhanced
bioaccumulation. Although
coagulation and settling of larger
suspended particles occur in these
zones, increased concentrations
of smaller particles (with sorbed
methylmercury) are available for
ingestion by zooplankton.
Major Remaining Tasks/Uncertainties
Future research needs related to
mercury fate and transport were
identified by comparing the STAR
research results to the goals out-
lined in the ORD Mercury Research
Multi-Year Plan (MYP). The STAR
research summarized in this report
identifies fate and transport issues
and provides information on mercury
cycling in complex ecosystems.
Other ORD Mercury MYP Research
Goals and Measures that are related
to the grants described here, but
may require further work to fully
accomplish, include development of
a model for mercury in fish, identifi-
cation of sources of mercury emis-
sions, and the eventual creation of
an integrated multimedia modeling
framework for mercury in the envi-
ronment.
The research provides information
that relates to all three of the STAR
mercury RFA questions, but it is most
responsive to parts of Question 1 and
Question 2 by identifying variables
that control partitioning of mercury
in the environment and transforma-
tion of mercury to methylmercury
as well as transformations between
inorganic species of mercury. Fur-
ther research, which builds on the
findings presented here, into the
relationship between mercury in the
watershed to mercury in biota (Ques-
tion 1) may be necessary to close
information gaps regarding uptake
by fish and other organisms. Also,
further research into how mercury
cycling and transformation pro-
cesses vary by region and ecosystem
(Question 3) is needed to support
regional modeling of mercury in the
environment.
The STAR grant investigators also
identified general areas where follow-
up work would be beneficial as well
as specific tasks that would advance
the science addressed by their proj-
ects. Some of the topics identified as
important for follow-up work include
processes such as demethylation and
photooxidation and factors affecting
methylation such as sulfate concen-
tration, DOC, intracellular sequestra-
tion of mercury, and other bacterial
processes, as well as observations
that newly deposited mercury con-
tributes more substantially to produc-
tion of methylmercury than "older"
mercury.
-------
1.1 Purpose of this Report
In 1999, the U.S. Environmental
Protection Agency (EPA) published
a Request for Applications (RFA)
entitled "Mercury: Transport and
Fate through a Watershed" under the
Science to Achieve Results (STAR)
program. The purpose of the RFA
was to solicit proposals to research
fate, transport, and transformation
of mercury (Hg) in aquatic and ter-
restrial environments. Nine grants
were awarded in 1999 based on this
RFA, and the work under these has
recently concluded. The purpose
of this report is to summarize the
important scientific findings from
these and two related grants, to
describe how these findings have
improved understanding of mercury
behavior in the environment, and
to convey remaining data gaps and
research needs on this topic. This
report is designed to provide EPA
program managers and staff, as well
as state environmental agencies,
Alignment of ORD's Mercury
Research with EPA's Strategic Goals
Directly contributes to:
Goal 4: Healthy Communities
and Ecosystems
Supports achievement of:
Goal 1: Clean Air and Global
Climate Change
Goal 2: Clean and Safe Water
Goal 3: Land Preservation
and Restoration
Goal 5: Compliance and
Environmental Stewardship
Source: EPA 2003a
researchers, and the public, with
an overview of advances in science
achieved through these grants and a
description of areas where additional
research may be needed.
This report is not intended to be
used directly for environmental
assessments or decision making.
Readers with these interests should
instead consult the peer reviewed
publications produced by the STAR
grantees and conduct necessary data
quality evaluations as required for
their assessments.
1.2 Background
As part of its mission to improve pub-
lic health and increase the reliability
with which risks to public health and
the environment are identified and
measured, EPA's Office of Research
and Development (ORD) funds
extramural research through its STAR
program. The program is managed by
ORD's National Center for Environ-
mental Research (NCER). It provides
grants and fellowships to scientists
to address research goals in various
environmental science and engineer-
ing disciplines through a competitive
solicitation process and independent
peer review: Many of these research
goals are outlined in ORD's Multi-year
Plans (MYPs), which ORD develops to
communicate research proposed for
the next five to eight years. The STAR
program engages the nation's best
scientists and engineers in targeted
research that complements EPA's own
intramural research program and
research of EPA's partners in other
federal agencies.
NCER issues REAs for topics based on
ORD's Strategic Plan (EPA 2001) and,
more specifically, on the research
goals described in the MYPs. These
REAs are prepared in cooperation
with other parts of the Agency and
concentrate on areas of special
significance to EPA's mission. For
example, ORD's mercury research
directly contributes to meeting one
strategic goal "Goal 4: Healthy Com-
munities and Ecosystems" in the EPA
Strategic Plan (2003a) and supports
achievement of the four other goals
(see text box).
1.3 Organization of this Report
This report describes the research
performed under each of the nine
grants awarded under this RFA as
well as research performed under
two additional, closely related
STAR grants1. Section 2 provides
a brief overview of mercury fate
and transport, identifies the STAR
research grants that addressed this
topic, and describes the research
themes covered by their results. Sec-
tion 3 describes key findings from
these grants according to research
theme, emphasizing how they have
advanced general understanding of
mercury fate and transport. Section
4 describes how this information can
be practically applied to address EPA's
goals. Section 5 discusses outstand-
ing research needs and data gaps
identified by the grantees and other
sources. Section 6 lists the cited
references, and the Appendix to this
report provides the lists of publica-
tions derived from the research
conducted under these grants.
Two research grants lead by Mae S. Gustin are presented here because they involve mercury fate and transport although they were not funded
under the FY99 mercury RFA. These grants were instead funded under the Exploratory Research - Air Chemistry & Physics (1996) and Experimental
Program to Stimulate Competitive Research (1998) STAR RFAs.
-------
2. OVERVIEW OF STAR MERCURY GRANTS
This section first provides a brief
overview of issues and information
needs concerning mercury in the
environment at the time ORD issued
its RFA on this topic (Section 2.1). It
then describes the 1999 STAR pro-
gram mercury RFA and lists the nine
STAR grants awarded under this RFA
in 1999 and two other related STAR
grants awarded under two other
REAs (Section 2.2). The relation-
ship of those grants to ORD's MYPs
is highlighted next (Section 2.3).
Finally, the overarching research
themes covered by the results of the
grants are described (Section 2.4).
2.1 Overview of Mercury in the
Environment
As summarized in the introductory
material included in the original so-
licitation for these 1999 STAR grants,
the presence of mercury in the
environment poses potential risks
to human health and wildlife. Both
natural and anthropogenic sources
contribute to mercury in the envi-
ronment, with a substantial increase
in the contribution from anthropo-
genic sources since the beginning of
the industrial age. Mercury can cycle
between various environmental me-
dia, including air, land, water, and
biota, through deposition, volatiliza-
tion, and other fate and transport
processes (see Exhibit 1). It is a
metal with complicated chemistry
and can transform between different
chemical species, including elemen-
tal mercury liquid and vapor, inor-
ganic salts, and organic forms (e.g.,
methylmercury), through a series
of complex chemical and physical
transformations. Methylmercury is
toxic and can bioaccumulate in the
tissues of animals. Consumption of
contaminated fish is a predominant
exposure pathway for human and
wildlife populations. A number of
adverse effects have been linked
to mercury exposure, including
behavioral abnormalities, impaired
growth and development, reduced
reproductive success, and death. Of
particular concern are the potential
neurotoxic effects of methylmercury
exposure on the developing fetus.
As required under section
112(n)(l)(B) of the Clean Air Act, as
amended in 1990, EPA has pre-
pared the Mercury Study Report to
Congress (EPA 1997a), which sum-
marizes the magnitude of mercury
emissions in the United States, the
health and environmental effects
of the emissions, and the cost and
availability of control technologies.
This and other major EPA reports
(e.g., the Great Waters Second
Report to Congress (EPA 1997b), the
Utility Air Toxics Report to Con-
gress (EPA 1998)) stress the adverse
effects of mercury exposure on hu-
mans and wildlife. Those and other
publications indicate that additional
research is needed to better under-
stand mercury fate and transport in
the environment.
EPA has taken and is planning dif-
ferent actions to address the health
and environmental concerns about
mercury, and the STAR research re-
sults should be useful to the Agency
in these important and high-profile
efforts. For example, most of the
research results from these STAR
grants have been published in time
to be considered as the Agency
finalized its regulations to reduce
mercury emissions from coal-fired
power plants.
2.2 FY99 STAR Mercury Grants
The "Mercury: Transport and Fate
through a Watershed" RFA was
issued to fund fundamental re-
search on the complex chemical
and physical transformations and
movement of mercury through the
environment. The outcome of this
research was intended to increase
EPA's ability to trace mercury from its
entrance into an ecosystem through
its biogeochemical cycling to the
concentration of methylmercury in
fish tissues in particular. The overall
goal of the research solicited on
mercury transport in a watershed
context was summarized in the origi-
nal RFA as follows:
"The goal of this solicitation is to
develop a better understanding of
terrestrial and aquatic fate and
transformation processes (espe-
cially microbial) that mediate
ecological and human exposures
to mercury. The development of
improved models of the fate of
mercury in aquatic and terres-
trial systems in order to estimate
ecosystem response to decreased
anthropogenic inputs of mercury
is also needed."
The RFA specifically invited grant ap-
plications addressing the following
critical research questions:
-------
EXHIBIT l: MAJOR ROUTES INTO THE ENVIRONMENT
(a) Major ecosystem inputs and outputs of mercury
Evasion from ,
soil and vegetation
Wet and dry deposition of gaseous and particulate mercury
Evasion
\
T
and transport/runoff of
mercury and CH Hg+
(b) Major aquatic mercury pathways
Biota Demethylation
(i)
(2)
(3)
Diffusion
Resuspension
Methylation
Sedimentation
Hg"
For a given amount of mercury
transported into a watershed,
what is the predicted concentra-
tion of methylmercury in fish?
Haw do mercury and meth-
ylmercury spatially distribute
across the terrestrial and aquatic
components of a watershed?
What controls bioavailability of
mercury in the food chain?
What environmental and
biochemical variables control
transformation of mercury to
methylmercury? What environ-
mental variables control the
reduction of divalent mercury
to elemental mercury in soils,
sediments, and surface waters?
How does mercury cycling vary
within different geographic re-
gions of the United States (e.g.,
south Florida, Great Lakes,
northeast, or west)? How might
the variability be accounted for
(e.g., resource types (wetlands),
temperature regimes, microbial
communities)?
Finally, the RFA presented three
specific research objectives:
Evasion
Reduction
Hg
Bioaccumulation
of CH Hg+
Burial
in sediments
^^
Jon
Oxidation
Diffusion
Resuspension
(1) The performance of theoretical
and laboratory investigations
focused on understanding the
behavior of mercury in the
environment, including mer-
cury cycling models; the role
of biogeochemistry especially
mercury sulfide complexes;
interactions among nutrients,
carbon, and sulfur on meth-
ylation processes; the role of
microorganisms; and the role of
macrophytes, periphyton, and
their interactions with hydro-
logical processes.
(2) The development and evalua-
tion of biogeochemical models
of the microbial transformations
of mercury in ecosystems in or-
der to interpret the sources and
distributions of total mercury
and methylmercury in terrestrial
and aquatic systems.
(3) Investigation of hypotheses
about the regional behavior of
mercury, extrapolating micro-
biological and biogeochemical
process data from experimental
scales to ecologically meaning-
ful scales and time periods.
EPA awarded nine STAR grants
under this solicitation. The grant
titles and principal investigators are
listed in Table 1 along with the in-
formation for two additional grants
related to this topic. This document
describes the results of these 11
research projects. Throughout this
report, we use the last name of the
principal investigator to refer to a
specific research project or research
team and/or the grant reference
number that is designated in the first
column of Table 1.
2.3 Relationship of 1999 Mercury
STAR Grants to ORDMYPs
As described in Section 1, one of
the primary purposes of the STAR
program is to provide grants and
fellowships to scientists for research
that addresses the goals outlined in
the ORD MYPs. The mercury fate and
transport STAR grants pertain primar-
ily to the Mercury Research MultiYear
Plan (EPA 2003b). The research results
obtained from these STAR grants
relate most directly to Long-term Goal
(LTG) number two of the Mercury
Research MYP, which states:
-------
EXHIBIT 2: MERCURY MYP RESEARCH ASSISTS OTHER MYPS
Global
Change
4
Human
Health
Ecosystem
Protection
Pollution
Prevention
•
Contaminate
Sites
d
"to understand the transport and fate
of mercury from release to the recep-
tor and its effects on the receptor."
There are seven general Annual
Performance Goals (APGs) defined
by ORD under this LTG, with several
specific Annual Performance Mea-
sures (APMs) for each APG. Specific
laboratories and centers within ORD
are designated as the lead for each
APM. One of the APMs assigned to
NCER under LTG number 2, "hold
workshop/state-of-the science on
mercury with emphasis on fate and
transport in watersheds and ecosys-
tem impacts," was accomplished by
NCER when it held the STAR Mer-
cury Fate and Transport Final Prog-
ress Review Workshop in November
21, 2003, which was attended by all
FY99 mercury STAR grant recipients
and other scientists.
In addition, results from these STAR
grants will support four key APGs
and three key APMs from the Mer-
cury MYP (as listed in Table 2).
Research directed by the Mercury
MYP will influence and advance the
goals in other multi-year plans (see
Exhibit 2). For example, the results
of the mercury STAR research should
also play a role in accomplishing the
general goal of the Air Toxics Multi-
Year Plan (EPA2003c) of reducing
the uncertainty associated with risk
assessment of air emissions of an im-
portant toxic air pollutant, mercury.
2.4 Research Hiemes
Several overarching research themes
were embodied by the specific re-
search questions and objectives list-
ed in the original REA and presented
in Section 2.2. Each investigator
developed preliminary objectives for
their proposed research that could
be linked to one (or more) REA
research objective(s). The actual
research results generated through
the funded projects are more easily
discussed according to their contri-
butions to four somewhat more spe-
cific research areas or themes, which
are described below. The first three
of these relate to critical research
question 2, while the last one relates
to critical research question 1.
(1) Biogeochemical controls of
mercury transformations.
Research on factors that influ-
ence transformations among
mercury species, including
biologically and chemically me-
diated processes involving both
organic (i.e., methylation and
demethylation) and inorganic
forms of mercury (i.e., reduc-
tion and oxidation of inorganic
mercury).
(2) Biogeochemical controls of
mercury cycling in the envi-
ronment. Research on factors
that affect phase partitioning and
inter-media transport of mercury
in both aquatic and terrestrial
systems, such as transfer be-
tween sediment and surface wa-
ter, surface water and the atmo-
sphere, soil and the atmosphere,
and interchange between the
atmosphere and plants through
various processes.
4
-------
TABLE 1. FY99 STAR GRANTS ON TRANSPORT AND TRANSFORMATION
OF MERCURY IN AQUATIC AND TERRESTRIAL ENVIRONMENTS
Grant
Ref. No.
in this
Report
[i]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]"
[11]"
EPA STAR
Grant No.
R827629
R827630
R827631
R827632
R827633
R827634
R827635
R827653
R827915
R825249
R827622E02
Grant Title"
Watershed influences on transport, fete, and
bioavailability of mercury in Lake Superior
Methylmercury sources to lakes in forested
watersheds: has enhanced methylation increased
mercury in fish relative to atmospheric deposition
Responses of methylmercury production and
accumulation to changes in mercury loading: a
whole-ecosystem mercury loading study
Cycling of mercury in Saginaw Bay Watershed
Chemical and biological control of mercury
cycling in upland, wetland, and lake ecosystems
in the northeastern United States
Processes controlling the chemical speciation and
distribution of mercury from contaminated mine
sites
Mercury and methylmercury cycling in the coastal/
estuarine waters of Long Island Sound and its
river-seawater mixing zones
Understanding the role of sulfur in the
production and fate of methylmercury in
watersheds
The redox cycle of mercury in natural waters
Light induced mercury volatilization from
substrate: Mechanisms responsible and in situ
occurrence
Determining the role of plants and soils in
the biogeochemical cycling of mercury on an
ecosystem level
Principal Investigator
Hurley, James E,
Univ. of Wisconsin
Swain, Edward B.,
Minnesota Pollution
Control Agency
Gilmour, Cynthia C,
Academy of Natural
Sciences and Univ. of
Maryland
Nriagu, Jerome,
Univ. of Michigan
Driscoll, Charles T.,
Syracuse University
Brown, Gordon Jr.,
Stanford University
Fitzgerald, William E,
Univ. of Connecticut
Mason, Robert P,
Univ. of Maryland
Morel, Francois M.,
Princeton Univ.
Gustin, Mae S.,
Univ. of Nevada-Reno
Gustin, Mae S.,
Univ. of Nevada-Reno
Co-investigatorsa
D. E. Armstrong, Richard C.
Back, M. M. Shafer, Helen
Manolopoulos
Jim Almendinger, Jim Cotner,
Daniel Engstrom, Jeff Jeremiason,
Edward Nater
Andrew Heyes, Robert P Mason,
John M. Rudd
Gerald J. Keeler, John Lehman,
Steve Lindberg, Xia-Qin Qang,
Hong Zhang
Ronald Munson, Robert Newton,
Joseph Yavitt
Daniel Grolimund, Mae Sexauer
Gustin, Trevor R. Ireland,
Christopher S. Kim, James J.
Rytuba, Greg Lowry, Samuel
Shaw, Stephen B. Johnson, Aaron
S. Slowey
Pieter T. Visscher, Prentiss
H. Balcom, Chad R.
Hammerschmidt ,
Carl H. Lamborg
Cynthia C. Gilmour
N/A
N/A
Ray Alden, James Coleman,
Dale W Johnson, Steve Lindberg
a The grant title and co-investigators listed in this table correspond to the information provided in the final summary report for each project and may differ from
the title and staff listed on the original grant proposal.
b The two grants lead by Gustin were funded under different RFAs (Exploratory Research - Air Chemistry & Physics (1996) and Experimental Program to Stimulate
Competitive Research (1998)).
(3) Sources and distribution of
mercury in terrestrial and
aquatic systems. Research on
sources of inorganic mercury
and methylmercury as well as
factors influencing their distri-
bution in the environment that
are not already covered under
the first or second themes (e.g.,
inputs via atmospheric deposi-
tion, inputs from runoff and
erosion).
(4) Bioavailability and bioac-
cumulation of mercury in
aquatic systems. Research on
the availability of mercury for
uptake by aquatic organisms and
the factors that influence this
availability and accumulation.
In addition, STAR grant research-
ers developed several new analyti-
cal methods for measuring total or
speciated mercury to facilitate their
major research objectives. These
new methods are discussed as a fifth
research theme.
Table 3 presents a crosswalk of the
11 STAR grant projects and these five
research themes. Project results as they
relate to the aforementioned themes
are described in the next section.
-------
TABLE 2. CROSSWALK OF STAR GRANT CONTRIBUTIONS TO
ORD's MERCURY RESEARCH GOALS AND MEASURES
Mercury Fate and Transport STAR Research Grants
Annual Performance Goals and Measures
under LTG 2 of the Mercury MYP (EPA 2003b;
target year, assigned lab/center for APMs)
APG : Provide an assessment of key fate and transport
issues for tracking the fate of mercury from sources to
concentrations in fish tissue (2004)
APM: Hold workshop/SOS on mercury with emphasis
on Fate and Transport in watershed(s) and ecosystem
impacts (2004)
APM: Evaluate mercury cycling in complex ecosystems;
including, air/water interface to accurately assess
concentrations in water and fish. Focus is on human
exposure as the ecological endpoint (2004, NCER)
APG: Develop a model for tracking mercury from
deposition to concentrations in fish tissues (2006)
APG: Develop information on sources of mercury
emissions including the regional/global atmospheric
fate and transport of such emissions (2008)
APG : Produce an integrated multimedia modeling
framework for understanding mercury fate from source
to fish concentrations (2010)
APM: Develop an integrated multimedia modeling
framework for the complete scientific understanding
of mercury fate/transport and atmospheric chemistry/
processes (2010, NERL, NRMRL, NCER)a
„
JH
ffi
•
0
•
O
O
a
X
0
0
O
O
o
en
§
o
•
0
•
o
o
„
I
•
0
o
0
o
o
vn
1
Q
•
0
•
O
o
„
1
m
0
0
0
o
o
"0
d
oj
5?
ft
•
0
O
0
o
o
, ,
a
S
0
0
O
O
o
ON
"S
0
S
0
0
O
o
o
•a
c
o
1
O
0
0
O
O
o
o
Key: • = significant contribution based on available results; O = results provide background knowledge that may be useful in meeting goal/measure.
a This APM pertains most directly to the 2001 RFA; although as stated the FY 99 RFA grants provide supporting materials.
TABLE 3. CROSSWALK OF RESEARCH THEMES AND MERCURY STAR GRANT RESEARCH PROJECTS
Mercury Fate and Transport STAR Research Grants
Research Themes
Biogeochemical controls of mercury transformations (see Section 3.1)
Biogeochemical controls of mercury cycling
in the environment (see Section 3.2)
Sources and distribution of mercury in
terrestrial and aquatic systems (see Section 3-3)
Bioavailability and bioaccumulation of mercury
in aquatic systems (see Section 3.4)
New methods for mercury analysis (see Section 3-5)
^7
jy
ffl
•
*
•
en
0
*
1
O
0
*
"
1
z
0
vn
~0
Q
*
-------
f
The important results of the 11 STAR
research projects on mercury fate
and transport are described in this
section. This discussion is organized
according to the five research themes
described in Section 2.4. In general,
this section focuses on the study
results, but details regarding the
study approach are also included
to facilitate interpretation of the
results. Detailed descriptions of
the research conducted under each
individual grant can be found in the
final summary reports prepared by
the principal investigators and in the
references listed in the Appendix.
Throughout this report, the principal
investigator's name and/or grant refer-
ence number presented in Table 1 are
used to identify research conducted
under the different grants, even when
co-investigators may have led the
specific part of research that contrib-
uted to the result. This format is used
to simplify the presentation of the
grant findings. The primary sources
used to compile this report were the
final reports prepared by researchers
and submitted to NCER. Presenta-
tion materials from the November
21, 2003, STAR Mercury Fate and
Transport Final Progress Review
meeting were also used.2 In addition,
Dr. Robert Mason of the University of
Maryland, one of the grant principal
investigators, provided an informal
summary of what he believes to be the
most important findings of the nine
grants, and his input is reflected in this
synthesis report as well.
3.1 Biogeochemical Controls of
Mercury Transformations
STAR grant research included several
studies that investigated the pro-
cesses by which mercury transforms
from one species to another. These
ranged from ecosystem-level stud-
ies, in which the overall speciation
of mercury was analyzed, to labora-
tory-scale experiments, in which the
mechanisms of transformation were
investigated. Results described here
include new information on factors
affecting the methylation/demeth-
ylation of mercury and inorganic
transformations between elemental
mercury (Hg(0)) and divalent mer-
cury(Hg(II)).
Methylation/Demethylation
Net MeHg production in aquatic eco-
systems is a complex process because
there is rapid recycling between
inorganic Hg (Hg(II)) and MeHg,
as Hg is effectively methylated in
low oxygen environments by sulfate
reducing bacteria (SRB) in sediments,
but MeHg is also efficiently demethyl-
ated in both anoxic and oxic envi-
ronments. The recent STAR research
suggest that the turnover time of
MeHg in aquatic sediments is on the
order of days to weeks and that Hg is
likely recycled many times by these
processes between the two forms
before being bioaccumulated into
fish as MeHg, or lost from the system
as Hg(II), elemental Hg (Hg(0)) and
MeHg by other processes.
STAR grant research demonstrated
that rates of mercury methylation
and demethylation are affected by a
variety of factors that influence both
the availability of environmental
mercury and the activity of SRB that
are known to methylate inorganic
mercury. Results from methylation-
related research performed under
STAR grants include information
on factors that increase methyl-
mercury production rates, the
pathway of reaction in SRB meth-
ylation, and demethylation rates of
methylmercury.
In general, there has been more
study of methylation than of demeth-
ylation, and much of the STAR
research conducted in this area also
focused on factors that affect meth-
ylation of inorganic mercury. Study
results showed that methylation rates
increase with:
• Increased bacterial metabolic
rates in summer [7];
• Higher concentrations of
inorganic mercury in sediment
pore water [7];
• Increased levels of bioturbation
by benthic infauna [7];
• Increased mercury load to the
ecosystem [3]; and
• Increased sulfate load to a
wetland [2, 8].
2 One exception is noted. Information on research conducted by Hines and Brezonik under Swain's STAR grant that was not included in the final
summary report was obtained from separate pre-publication drafts of journal articles.
-------
Additionally, research demonstrated
that methylmercury production rates
decrease with:
• Increased concentrations of or-
ganic matter in sediments (which
binds inorganic mercury, limiting
its bioavailability) [7];
• Higher sediment distribution coef-
ficient (KD) values for inorganic
mercury [7]; and
• Increased sulfide loads in the
ecosystem [2, 7, 8].
STAR grant research also showed that
methylmercury demethylation can
occur via photolytically-induced reac-
tions [2], and that demethylation is a
major removal process of methylmer-
cury, along with burial [1]. Addition-
ally, researchers measured biological
demethylation in sediments and
related it to the net methylation rate
[3, 8]. The research supporting these
findings are described below.
Fitzgerald et al. [7] performed
multiple experiments in Long Island
Sound using mercury isotope trac-
ers to identify factors that increase
methylation in sediments. Methyla-
tion rates were enhanced in August
relative to March and June, illustrat-
ing the importance of methylating
bacteria in utilizing available substrate,
Hg(II). In areas where organic matter
was lower, there were higher Hg(II)
concentrations in the porewater
and correspondingly higher rates of
methylation. Conversely, the pres-
ence of organic matter (i.e., dissolved
organic carbon, or DOC) resulted in
decreased methylation rates by bind-
ing available inorganic mercury and
making it less available to the bacteria.
Sediment-water partition coefficients
(KDs) for inorganic mercury also were
measured, and higher coefficients
were correlated with lower methyla-
tion rates. Additionally, increased lev-
els of bioturbation by benthic infauna
appear to extend the zone of active
mercury methylation in sediment and
might move buried mercury to the
active methylating zone.
Other STAR research demonstrated
that increasing mercury or sulfate
loads to an ecosystem can increase
methylmercury formation. Gilmour
et al. [3] report their study to be
the first direct set of whole ecosys-
tem-level experiments to show the
rapid and linear increase in meth-
ylmercury production in an aquatic
ecosystem in response to increasing
surface loading of inorganic mercury.
Response time (i.e., time between
increase in mercury load and a mea-
sured change in methylmercury pro-
duction) varied with environmental
conditions. For shallow, warm sites
in the Florida Everglades, response
time was days to weeks, depend-
ing on temperature. Response time
for the Experimental Lakes Area in
Canada ranged from weeks for small,
shallow enclosures, to about a year
for the whole lake. Time to reach a
new equilibrium for the increased
mercury loading rate (about four
times higher than normal mercury
loading) in Experimental Lakes Area
is expected to be at least three years.
Gilmour et al. [3] noted that the
magnitude of the change in methyl-
mercury production due to mercury
loading depended on when and
where the loading occurred. Mer-
cury added to the lake surface was
readily methylated in lake sediments.
Mercury loaded to uplands and
wetlands was slow to move to sites
of methylation (i.e., subsurface areas
near the water table). Mercury mass
budgets calculated from isotope
measurements suggest that most of
the methylmercury in the lake was
formed in the lake rather than being
transported to the lake from other
parts of the watershed.
Sulfate concentrations influence
environmental methylation rates
because SRB require sulfate to break
down organic matter to yield sulfide,
carbon dioxide, and energy, with
concomitant methylation of Hg(II).
Swain et al. [2] applied sulfate to
the surface of a naturally sulfate-
poor wetland and to lake sediment
samples. Sulfate sprayed onto half of
a 2-ha peatland at the Marcell Experi-
mental Forest in northeastern Minne-
sota increased the annual sulfate load
by approximately four times relative
to the control half of the wetland.
Measurements of sulfate and methyl-
mercury in the wetland two weeks
after the first application indicated
that methylmercury porewater con-
centrations had increased three-fold.
Three subsequent sulfate additions
that year did not further increase
methylmercury porewater concentra-
tions; methylmercury concentrations
in the treated portion of the wetland
remained elevated relative to the
control. Methylmercury concentra-
tions in the outflow from the wetland
increased following each addition.
Based on these results, Swain et al.
[2] concluded that methylmercury
production in sulfate-poor wetlands
is as much a function of atmospheric
deposition of sulfate as of mercury.
The addition of sulfate to lake
sediment samples in the laboratory,
however, reduced mercury methyla-
tion rates [2]. Addition of glucose
and ammonium to those sediment
samples also reduced net mercury
methylation rates. Swain et al. [2]
suggested that these additions proba-
bly reduced the bioavailability of mer-
cury through the binding of sulfides,
reduced pH, increased formation of
mercury-Cl species, or a combination
of these factors. Pyruvate treatments
were the only additions to the lake
sediment samples that increased net
mercury methylation rates. Swain et
al. [2] suggested that this occurred
because the SRB were ferment-
ing pyruvate instead of sulfate and
were not producing the sulfides that
reduce mercury bioavailability. In a
set of sediment cores taken across a
transect in the lake, methylmercury
concentrations were negatively cor-
-------
related with total inorganic sulfides.
Swain et al. [2] proposed that
bioavailability of mercury, which can
be reduced by binding with sulfides
(see Section 3.2), may control the
distribution of methylmercury in the
lake sediments.
Mason et al. [8] designed their exper-
iments to evaluate how different
types of mercury-sulfide complexes
affect the rate of mercury methylation
by SRB. In the presence of dissolved
sulfide, mercury forms both neutral
complexes (e.g., HgS°, Hg(SH)2°) and
charged complexes (e.g., Hg(SH)+,
HgS22"). Mason et al. [8] hypothesized
that SRB bioconcentrate neutral mer-
cury-sulfide complexes via passive
diffusion, but do not bioconcentrate
charged mercury-sulfide complexes.
Thus, the neutral mercury-sulfide
complexes are believed to be the
bioavailable form by which inorganic
mercury can enter these bacteria cells
and undergo methylation. Mason
et al. [8] measured a relatively high
octanol-water partition coefficient
(KOW) for two neutral mercury sulfide
complexes. Based on estimated cell
membrane permeability for com-
pounds with similar KQW values,
Mason et al. [8] estimated that SRB
could uptake these complexes at
rates that are more than sufficient
to achieve observed methylation
rates in cultures, both pure and field
cultures.
The fraction of mercury-sulfide
complexes in sediments that are
neutral is a function of the sulfide
concentration in the medium.
Chemical complexation modeling by
others has shown that the specia-
tion of mercury tends to shift toward
charged complexes as sulfide levels
increase. Mason et al. [8] conducted
mesocosm experiments in various
locations in the Florida Everglades
and observed that as sulfide concen-
trations increased, methylation rates
decreased. The investigators also
compared modeled uptake rates,
assuming passive diffusion of neutral
mercury-sulfide complexes (HgS° and
HOHgSH0), and compared them to
observed methylation rates in pure
cultures of the SRB Desulfobulbus
propionicus. Uptake and methylation
rates both increased with increas-
ing HgS° concentration with similar
slopes. Mason et al. [8] believe that
these findings confirm that speciation
of mercury in the medium surround-
ing methylating bacteria is a key
factor affecting mercury methylation
rates in sediments.
Mason et al. [8] also observed that
sulfate levels affect mercury methyla-
tion rates in sediments. Mason et al.
studied sediment cores taken from
various regions of the Florida Ever-
glades that differ in sulfate concentra-
tions. When the investigators added
sulfate to cores with low sulfate
levels, they observed an induction of
SRB activity and increased mercury
methylation, even though that activity
increased sulfide levels. Confirm-
ing previous findings, they found
that adding sulfide alone inhibited
methylation. Overall, the results of
both the laboratory and field studies
indicated that the balance between
sulfate levels, which affect SRB
activity, and sulfide production and
accumulation, which reduce mer-
cury bioavailability (see Section 3.2),
affects net mercury methylation rates.
One STAR grant indicated that bacte-
rial community structure also can
affect mercury methylation rates.
Previous research had shown that the
acetyl-coenzyme A pathway is key for
producing methylmercury in several
SRB strains. Morel et al. [9] hypoth-
esized that some methylation can
also occur via different pathways in
other strains. In a series of laboratory
experiments on seven strains of SRB,
Morel et al. used carbon monoxide
dehydrogenase (CODH) activity to
indicate activity of the acetyl-CoA
pathway and chloroform to inhibit
the acetyl-CoA pathway. They found
four incomplete-oxidizing SRB strains
that clearly do not utilize the acetyl-
CoA pathway for mercury methyla-
tion and, therefore, might not require
vitamin B12 as a coenzyme.
Demethylation of methylmercury is
important in determining the net for-
mation of methylmercury in an eco-
system. In Spring Lake in northern
Minnesota, Hines and Brezonik [2]
measured methylmercury photodeg-
radation rate constants of 0.0025 hr1
in the field. They estimated that pho-
tolytically-induced demethylation
accounts for twice as much removal
of methylmercury from surface
waters as does methylmercury burial
in sediments. The depth profile of
methylmercury photodegradation
showed loss of methylmercury up
to 50 to 60 cm, which corresponded
to the depth of 90 percent attenu-
ation of PAR (photosynthetic active
radiation). Using a mass-balance
approach to estimate the distribu-
tion and cycling of mercury species
in Lake Superior, Hurley et al. [1]
concluded that both sediment burial
and photo-demethylation are major
removal processes for methylmercury
in surface waters.
In addition to photodemethylation,
biological demethylation is also an
important process, as demonstrated
by Mason et al. [8] and Gilmour et al.
[3]. Results showed that methylmer-
cury speciation and bacterial activity
influence the rate of demethylation
in sediments and thus also affect the
overall net production of methyl-
mercury in the system. Therefore,
demethylation rates vary across eco-
systems. The researchers concluded
that studies should measure both
methylation and demethylation rates
to properly understand the rate of
net methylmercury production.
Inorganic Transformations
STAR research also focused on trans-
formations between inorganic spe-
cies of mercury via redox reactions,
-------
with a particular focus on photo-
induced reactions as an important
pathway. These redox transforma-
tions can affect the fate and transport
of mercury by converting mercury
to chemical species that are more or
less mobile in the environment. This
section presents STAR grant findings
related to rates of transformation,
chemical mechanisms, and factors
affecting these transformations.
Research focused more on the overall
transfer of mercury between media
(including transfers that might occur
due to redox reactions involving inor-
ganic mercury transformation) are
covered in the Section 3.2.
STAR grants showed that oxidation of
Hg(0)toHg(II):
• Occurs via photo-oxidation,
perhaps indirectly via hydroxy
radicals and other reactive chemi-
cal intermediates [2, 9];
• Can at times account for the ma-
jority of Hg(0) lost from surface
waters [9]; and
STAR grants also demonstrated that
reduction of Hg(II) to Hg(0):
• Can occur via photoreduction [2];
• May require the Hg(II) to be
bound to organic matter [7]; and
• Can occur at a higher rate than
photooxidation [2].
In some experiments, however,
both oxidation and reduction were
observed under similar conditions
[4]; additional research is needed to
understand such reactions.
Both Morel et al. [9] and Hines and
Brezonik [2] found photooxidation
of Hg(0) to be a significant means of
Hg(0) removal from surface waters.
In laboratory and field studies on
the St. Lawrence River, Morel et al.
[9] found the rate of loss of Hg(0)
to photooxidation to be about two
orders of magnitude greater than
its loss via volatilization to the air.
Through treatment with heat, chloro-
form, and filtration to eliminate SRB
activity, Morel et al. [9] confirmed
that the oxidation is chemically
mediated chiefly by UV radiation
rather than biologically mediated.
Their data suggested that factors
other than photon flux, however,
are rate-limiting. They proposed that
hydroxy radicals and reactive halogen
and oxygen intermediates may favor
Hg(0) oxidation.
In water samples from Spring Lake
in northern Minnesota, Hines and
Brezonik [2] calculated a pseudo-
first order rate constant of 0.58 h"1
for the photooxidative loss of Hg(0).
Measurements of air and water Hg(0)
concentrations indicated that lake
water was supersaturated with Hg(0).
Fluxes of Hg(0) from the lake to the
air were greatest in wanner, sun-
nier months, but correlation with
solar radiation was weak. The loss of
Hg(0) by photooxidation may have
exceeded loss by volatilization to air
at times. Other losses of Hg(0) from
the lake water were reaction with
ozone, hydroxyl radical, and possi-
bly singlet oxygen. Burial was also a
major loss of total mercury.
In studies of the photoreduction of
Hg(II), Fitzgerald et al. [7] concluded
that it must be bound to organic
material to be reduced by abiotic
processes in natural waters. Their
research demonstrated that the addi-
tion of humic acid to synthetic seawa-
ter increases the rate of Hg(II) reduc-
tion to Hg(0), as has been reported
by others. However, Fitzgerald et al.
also observed that adding ethylene-
diaminetetraacetic acid (EDTA) and
chlorine (Cl") as competitive ligands
at concentrations that are high
enough to outcompete the humic
acid as a ligand resulted in reduction
rates falling to almost zero. Fizgerald
et al. [7] concluded that higher Hg(0)
concentrations in surface water in the
summer are supported by the higher
summer concentrations of DOC and
other ligands that enhance reduc-
tion of Hg(II). However, they also
concluded that DOC may sometimes
be high enough to prevent reduction
to Hg(0) and that mercury-organic
associations can explain the reduc-
tion that was measured below the
photic zone (i.e., dark reductions
of mercury may be abiological and
organically mediated). Their data
suggest that organic ligands and DOC
affect the speciation of dissolved inor-
ganic mercury complexes in anoxic
low-sulfide sediments such as those
in Long Island Sound. That conclu-
sion appears to contrast with existing
chemical speciation models that indi-
cate sulfide to be the major ligand of
both dissolved methylmercury and
Hg(II) in natural porewaters.
Data collected by Nriagu et al. [4]
also suggest that photoreduction
of Hg(II) to Hg(0) is important
in surface waters. They measured
diurnal cycles of dissolved gaseous
mercury (DGM; dissolved gaseous
Hg(0)) concentrations in Great Lakes
water to correlate with solar radiation
both temporally and spatially. Their
calculations indicated that the Hg(0)
was supersaturated at most locations,
a finding consistent with those of
Hines and Brezonik [2]. In addition,
Nriagu et al. [4] investigated the pho-
toreduction of Hg(II) by examining
the role of amino acids in the redox
cycle of mercury. Preliminary experi-
ments demonstrated that Hg(II) was
photo-reduced in the presence of
eight amino acids, but not photo-
reduced in the presence of at least
four other amino acids. Hines and
Brezonik [2] also measured photo-
reduction of Hg(II) in the lake and
found the rate constant to be 1.1 h"1,
which was twice the rate constant
for photooxidation. They concluded
that redox reactions between Hg(0)
and Hg(II) readily occur, providing a
source of ionic mercury to the system
that can be methylated.
-------
These STAR results indicate that
both oxidation and reduction of
inorganic mercury species appear
to occur readily in natural waters
under the right conditions. Addi-
tionally, in surface water samples
spiked with Fe(III) and exposed to
sunlight, Nriagu et al. [4] found that
DGM increased in some samples
and decreased in others. Nriagu et
al. estimated a rate constant for the
reduction and proposed chemical
mechanisms for both reduction and
oxidation. It is not entirely clear why
both oxidation and reduction were
observed under similar conditions.
Experimental conditions introduced
by the investigators were proposed
as one explanation; however, further
research is necessary.
3.2 Biogeochemical Controls
of Mercury Cycling in the
Environment
Mercury cycling (i.e., phase parti-
tioning and inter-media transport)
is key to the distribution of mercury
in the environment. Most of the
STAR research in this area focused
on processes controlling mercury
transfer between surface water and
the atmosphere or surface water
and sediment as summarized in the
subsections below.
Sediment—Surface Water Transfers
As MeHg formation appears to occur
primarily in sediments, the processes
responsible for its transport from the
sediment into the water column are
very important in determining its rate
of bioaccumulation into fish. Parti-
tioning of mercury between dissolved
and solid phases in surface water
determines the mechanisms by which
it can be transferred to other media
(e.g., mercury bound to particles in
the water column will be subject to
deposition and to resuspension from
the benthic sediment, whereas dis-
solved mercury complexes will not;
some dissolved complexes may be
bioavailable whereas other are not).
Fitzgerald et al. [7] found that most
of the ionic inorganic mercury dis-
solved in the Long Island Sound
water column is complexed with
dissolved organic matter with a high
affinity, given that the conditional sta-
bility constants were very high (log K'
of about 21 to 24). Principal sources
of organic ligands in Long Island
Sound appear to be influx from river
water as terrestrial organic matter
(47 percent) and phytoplankton
exudation (31 percent). The primary
loss of organic ligands is via tidal
exchange with low DOC/low ligand
waters of the continental shelf. The
Connecticut River is the primary
source of organic matter to Long
Island Sound, which is reflected in
the seasonal variation in the ligand
abundance (highest in summer/
spring, lowest in winter).
Fitzgerald et al. [7] studied factors
affecting the fraction of mercury in
more labile forms which can poten-
tially increase the rates of methyl-
mercury formation. In the mixing
zone of the Connecticut River and
the Long Island Sound, Fitzgerald et
al. found that the fraction of mercury
present as reactive mercury increased
from levels present in river water
upstream of the mixing zone. The
reactive fraction was primarily in
the particulate phase. On the basis
of field measurements and a simple
competitive ligand model, Fitzgerald
et al. concluded that the enhanced
reactive mercury in the estuary is a
result of dilution of the dominant
organic ligand class (characteristic of
fresh water mercury) and competi-
tion with chloride (from Long Island
Sound) in the mixing zone, followed
by coagulation/adsorption onto sus-
pended particles.
Fitzgerald et al. [7] found both sedi-
mentary organic matter and acid-vola-
tile sulfide to affect partitioning of
methylmercury and Hg(II) between
particulate and dissolved phases
in waters of Long Island Sound. As
expected on the basis of bioavailabil-
ity of Hg(II) to SRB, Fitzgerald et al.
[7] found mercury methylation rates
to vary inversely with the KD of Hg(II)
and positively with the concentration
of Hg(II) in porewater (which exists
mostly as HgS°).
Mason et al. [8] also explored how
mercury partitioning to sediment
particles impacts mercury concentra-
tion and bioavailability in sediment
pore waters. They developed a model
accounting for adsorption and sulfide
concentration to predict Hg(II) con-
centrations in sediment pore water
and sorbed to particles. The model
was applied to two ecosystems, the
Patuxent River and the Florida Ever-
glades, and the results were validated
with measurements. The model's
predictions of neutral mercury-sul-
fide complexes correlated well with
the methylmercury concentrations
for the samples.
Binding of mercury to dissolved
organic carbon (DOC) in surface
waters has been found to be impor-
tant, but the findings of Mason et al.
[8] suggest that the influence of DOC
on mercury partitioning in sediments
is more complex. Their research
suggested that binding of mercury to
DOC is less important in sediment
pore waters under typical DOC levels
and when there is greater than 0.01
micromolar sulfide. Mercury bind-
ing with DOC in the solid phase is
important, and these studies [8]
suggest that in aerobic sediments,
binding with the organic fraction of
sediments is more important than
binding with methyl oxide phases.
Surface Water—Atmospheric Transfers
Deposition of mercury from the
atmosphere to surface waters (and
soils and plants) occurs via both wet
and dry deposition. Dry deposition
is less well studied than wet, and
includes deposition of atmospheric
paniculate Hg and gas phase Hg. Gas
phase Hg consists of both Hg(II),
often called reactive gaseous mercury
(RGHg), and Hg(0). Dry deposition
-------
of RGHg depends on its chemical
composition (e.g., HgBr2, HgCl2),
wind speed, and other factors. Less
well studied is the loss of mercury
from surface waters via volatilization
of Hg(0) to the atmosphere. Several
STAR grants investigated the relative
importance of Hg(0) volatilization
from surface water in the total mer-
cury mass balance.
Estimating net dry deposition of
Hg(0) to surface water (and soils) is
complex because Hg(0) is constantly
degassing (i.e., volatilizing) from the
surface. As described above, Hines
and Brezonik [2] and Nriagu et al. [4]
found Hg(0) concentrations in lakes
typically to be above saturation with
respect to the atmosphere, indicating
that dry deposition of Hg(0) will not
occur; rather, Hg(0) will escape from
the surface water into the atmo-
sphere. Both Hurley et al. [1] and
Fitzgerald et al. [7] found volatiliza-
tion of Hg(0) from surface water to
be a major process by which Hg(0)
leaves surface water. Based on a mass
balance of mercury derived for Lake
Superior, Hurley et al. suggested that
volatilization of Hg(0) is one of the
main removal processes for total mer-
cury from that ecosystem, followed
by sediment burial. Fitzgerald et al.
found volatilization of Hg(0) to be
important in removing Hg(0) from
Long Island Sound as well, and noted
that this transfer is a potentially
important source of atmospheric
Hg(0). Fitzgerald et al. estimated
annual emissions of Hg(0) from Long
Island Sound to be 80 kg in total.
This corresponds to 35 percent of
the annual input of mercury to Long
Island Sound being remobilized into
the atmosphere as Hg(0). Fitzgerald
et al. concluded that volatilization
competes with sediment methyla-
tion of mercury as a mechanism of
removal of Hg(0) from surface water.
Results from a short-term study of a
Tobico marsh at the coast of Saginaw
Bay suggested to Nriagu et al. [4]
that loss of Hg(0) from that marsh
resulted primarily from volatilization
of Hg(0). They estimated a "lifetime"
of about one hour for DGM in the
marshlands. Volatilization of Hg(0)
may be more or less important than
other processes that remove Hg(0)
from surface waters in other types
of ecosystems. As stated earlier,
results from Nriagu et al. indicate
that mercury cycling between surface
water and air varies diurnally and
seasonally, with rather large daily
variations in DGM in surface water of
Saginaw Bay. Bay waters were gener-
ally supersaturated with Hg(0) rela-
tive to the partial pressure of Hg(0)
in the atmosphere. Distinct diel
variations in DGM concentrations
were observed that indicate a high
dependence on the intensity of solar
radiation. In their research, Nriagu
observed a maximum DGM in surface
water samples around noon, with
DGM levels generally highest in June.
These variations seem to indicate that
the dissolved mercury pool is being
actively recycled in the water column.
Soil—Atmospheric Transfers
Deposition of mercury from the
atmosphere to surface soils, which
occurs via both wet and dry deposi-
tion, has been extensively studied.
Less well studied is the transfer of
mercury from soils back into the
atmosphere. Several STAR grants
investigated factors affecting the
emission of mercury from soils to the
atmosphere.
Gustin et al. [10] performed labora-
tory, field, and mesocosm experi-
ments to investigate light-enhanced
mercury volatilization. In laboratory
experiments, a well-mixed single pass
gas exchange chamber was used to
test various hypotheses about mer-
cury volatilization from soil. All mer-
cury species and mercury-containing
substrates constantly emitted mer-
cury to the air. However, of the pure
phase mercury species, only mercury
sulfide exhibited light-enhanced
emissions. In a test of whether
mercury adsorbed to iron oxides
and organic matter would exhibit
light-enhanced emissions, iron oxide
amended with mercury chloride, and
organic material amended with Hg(0)
and mercury chloride exhibited light-
enhanced emissions.
At various mining, Superfund, and
geothermal sites in Nevada and
California, Gustin et al. [10] used
chamber and micrometeorological
flux measurements to assess light-
enhanced mercury emissions. At all
sites, light-enhanced emissions of
mercury were observed with little
increase in soil temperature. A rapid
increase in emissions was observed
in the morning (e.g., at New Idria,
CA, emissions from soils doubled as
morning light irradiated the soil),
and on cloudy days the flux would
mimic the shadowing and lighting
of the sites. A 100-fold increase in
emissions was observed by artificially
lighting tailings at night at the Carson
River Superfund site. Gustin et al.
[10] also calculated the activation
energy for the mercury flux in dark
and light conditions. The activation
energy was lower for the nighttime
flux, indicating that the light energy
in the day is being used for a process
in addition to volatilization, perhaps
to generate elemental mercury by
photoreduction.
In follow-up field work, Gustin et al.
(under the Brown et al. [6] grant)
compiled data on mercury emis-
sions from mine wastes and sur-
rounding mercury-enriched terrains
by measuring fluxes at 18 areas of
mining activity. They studied the dif-
ferent emissions under light versus
dark conditions and determined the
speciation of the mercury sources
at each site. Overall, the researchers
found that mercury-contaminated
mining sites with higher concentra-
tions of soluble mercury-containing
species emit more mercury to the
atmosphere than those containing
less soluble mercury-containing spe-
cies. Specifically: (1) as particle size
decreases, mercury concentration,
-------
mercury emissions, and light-to-dark
emission ratio increase; (2) metacin-
nabar-bearing samples exhibit higher
light-to-dark emission ratios than
those containing cinnabar; and (3)
the presence of more soluble mercury
species also correlates positively with
higher light-to-dark emission ratios.
Gustin et al. [11] used large meso-
cosms (Ecologically Controlled
Lysimeter Laboratories (EcoCELLs)),
smaller plant-exposure chambers
(ecopods), and a single-plant gas-
exchange system to further study
the role of soils (and plants, see next
section) in controlling the fate and
transport of mercury in the environ-
ment at the ecosystem level. Mea-
surements of mercury emissions from
the soil were made in the EcoCELLs.
Dominant factors driving mercury
emissions from soil were incident
light, precipitation, and the presence
of vegetation. Soil temperature and
moisture also influenced flux. Data
indicated that soil gas efflux is not
a diffusion-driven process, except
during certain periods of the day
(particularly midday). Therefore, soil
gas concentrations are not an effec-
tive predictor of soil mercury flux.
Plant Transfers
In the EcoCELL experiments, Gustin
et al. [11] found that nearly all the
mercury in plant foliage was derived
from the atmosphere, as opposed
to the soil. Mercury concentrations
in the foliage of trees grown in
mercury-enriched soil (concentra-
tion of 12.3 jtig/g) was similar to the
foliage of trees grown in soil with a
background mercury concentration
(0.03 jtig/g). These results indicate
that foliage may act as a significant
sink for atmospheric mercury, and
that the mercury content in litter-
fall would represent a significant
mercury input to terrestrial ecosys-
tems, consistent with the findings of
Driscoll et al. [5] described below.
Additionally, experiments (using
leaf washing) showed that less than
three percent of mercury would be
removed with precipitation. Gustin
et al. concluded that litterfall may be
an effective way to estimate mercury
deposition in deciduous forests.
Ecopod and gas exchange chamber
experiments supported these results
by showing that mercury concentra-
tion in foliage increased with increas-
ing air concentrations of mercury,
and that mercury soil concentrations
either did not significantly influence
foliar concentration or influenced
it much less than the air concen-
trations. Experiments in the gas
exchange chamber indicated that
some mercury is transported from
the soil via the transpiration stream,
because there was a pulse in water
vapor and mercury emitted when the
chamber was first illuminated.
In their study of two upland forest
sites (one deciduous, one conifer-
ous) in the Sunday Lake watershed,
Driscoll et al. [5] also found forest
vegetation to be very important in
mediating the inputs of mercury
to the forest floor. They found that
the flux of mercury to the forest
ecosystem was dominated by dry
deposition (70 percent of total
deposition, including both litter fall
and throughfall). Ten-fold higher
levels of total mercury were reported
for the throughfall (i.e., deposition
passing through tree canopy) at the
coniferous plot, however, than at the
deciduous plot. The major pathway
for dry deposition in the coniferous
plot was throughfall, while the major
pathway in the deciduous plot was
litterfall, which is consistent with the
conclusions from Gustin et al. [11]
regarding litterfall being an effective
measure of deposition in deciduous
forests.
Studies by Nriagu et al. [4] indicate
that plants are contributing in a
minor way to the release of mer-
cury from soils to the air. Mercury
flux measurements over soybean
plants (3 ng/m2-hr) were higher
than mercury flux measurements
over soil with no plants that had
previously been sown with corn
seed dressed with mercury (from
< 0.1 to 1.7 ng/m2-hr). Results of
these measurements indicated that
the mercury in the soil from the
seeds was a minor source of mercury
to the atmosphere, but that vegeta-
tion could play a role in transferring
mercury from soil to the atmosphere.
3.3 Sources and Distribution of
Mercury in Terrestrial and
Aquatic Systems
Several STAR grants identified the
major sources and means of distribu-
tion of mercury in a given study area.
Researchers identified and investi-
gated important processes influenc-
ing movement of mercury through-
out a watershed (e.g., production of
methylmercury, predominant mer-
cury fate and transport processes). In
addition, several researchers per-
formed general mass balance studies
to quantify sources and inputs within
a given watershed. Relevant results
are summarized here.
Mass Balance Studies
Using data from open water cruises
and GIS-based studies, Hurley et al.
[2] developed a mass balance model
of mercury in Lake Superior. They
reported that total mercury inputs
were dominated by atmospheric
deposition (58 percent), followed
by tributary inputs (21 percent)
and contributions from particulate
remineralization (19 percent), while
sources of methylmercury were
somewhat evenly divided between
the atmosphere, tributaries, ground-
water, remineralization, and in-situ
methylation. Elevated concentra-
tions of mercury in nearshore waters
appeared to be the result of tributary
inputs. Hurley et al. [2] estimated
volatilization of Hg(0) to be the
principal removal process of total
mercury from the surface water,
followed by sedimentation, while
they estimated methylmercury to be
removed primarily by sedimentation
and photo-demethylation. Based
on trace amounts of methylmercury
detected in open surface waters
-------
(about 1 percent of total mercury in
the lake), Hurley et al. [2] concluded
that wet deposition may be a small
but measurable source of methyl-
mercury to surface water in the lake.
They found the proportion of total
mercury as methylmercury to be
maximum at the thermocline in open
surface waters of the lake, probably
because plankton, with their accumu-
lated methylmercury, aggregate at the
thermocline. Using their data, Hurley
et al. [2] developed a simple model
to relate annual tributary loading
from a watershed to the lake based
on land use/type in the watershed.
Driscoll et al. [5] performed mass
balance calculations on mercury
cycling in the Sunday Lake watershed
of the Adirondacks. Results from
this study indicate that overall, the
watershed and lake are sinks for total
mercury, while the wetlands and the
lake are sources of methylmercury.
At both coniferous and deciduous
upland plots, mass balance calcula-
tions indicated that the soil acts as a
net sink for inputs of total mercury
and methylmercury. Discrepancies
in these calculations suggest that
mercury is either accumulating in
the forest floor or is lost by vola-
tilization or both. As a part of this
study, total mercury concentrations
were measured in soil horizons.
The concentrations were highest in
the surface layer, lowest in the next
layer, peaked slightly in the third
layer, and then decreased in the
lowest layer analyzed. In measure-
ments taken in soil water, fluxes of
total mercury decreased with depth,
which coincided with decreases in
DOC. Driscoll et al. [5] concluded
that mercury is immobilized in the
mineral soil by the deposition of
organic matter.
Swain et al. [2] performed a mass bal-
ance study of mercury in the Spring
Lake watershed in northern Min-
nesota. They concluded that atmo-
spheric deposition is the main source
of total mercury to the lake, methyla-
tion in the lake is the main source of
methylmercury, demethylation and
photodegradation consume most of
the methylmercury in the lake, and
Hg(0) is lost from the lake mostly by
photooxidation and volatilization.
In Situ Methylmercury Production
As studies described thus far indicate,
although atmospheric deposition
and watershed inputs are important
sources of methylmercury (MeHg),
in situ production within the aquatic
system is the dominant source of
MeHg, at least for freshwater lakes
(except for those that have a large
watershed or extensive abridging
wetlands) and for estuarine sys-
tems (e.g., Long Island Sound and
the Chesapeake Bay). Hines and
Brezonik [2] performed mercury
mass balance calculations for Spring
Lake in northern Minnesota (a lake
with little wetland coverage in the
watershed) and found that in situ
methylation is the main source of
methylmercury, accounting for 76
percent of input). In Long Island
Sound, Fitzgerald et al. [7] concluded
that in situ production is also the
major source of methylmercury. The
calculated diffusive sediment-water
flux of methylmercury in Long Island
Sound (11 kg/yr) is higher than
methylmercury inputs to the Sound
from external sources (estimated at
5.2 kg/yr). Results from both a model
developed by Driscoll et al. [5] (i.e.,
the Mercury Cycling Model for Head-
water Drainage Lake Systems) and
the mercury mass balance calcula-
tions for the Sunday Lake watershed
in the Adirondacks indicate that
both the surrounding wetlands and
Sunday Lake itself are the dominant
sources of methylmercury to the lake.
Gilmour et al. [3] studied methylmer-
cury sources in both the Everglades
and the Experimental Lake Area in
Canada. Calculations of mercury
mass balance based on mercury
isotope measurements in the systems
suggested that most of the methyl-
mercury found in both systems is
formed in situ. Mercury added to sur-
face water was readily methylated in
the sediments, while mercury loaded
to uplands and wetlands was slow
to move to sites of methylation. The
major locations for new methylation
in both the shallow, warm sites in the
Everglades and the deeper, cooler
Experimental Lakes Area were satu-
rated surface peats and sediments,
and for the Experimental Lakes Area,
also anoxic bottom waters. The newly
deposited mercury was more readily
methylated than existing mercury
at those sites (i.e., sediments and
soils), suggesting that older mercury
deposits are less bioavailable for
methylation.
In Lake Superior, Hurley et al. [1]
measured sediment-water partition
coefficients (log KD) for total mercury
and methylmercury in sediment that
were relatively low compared to
other lakes, suggesting that mercury
is more mobile in Lake Superior sedi-
ments than in other lakes. The high-
est concentrations in sediment were
observed in spring and near tributary
inputs. They proposed several factors
as likely to control methylmercury
concentrations in sediments, but
did not acquire quantitative data on
those factors during this study.
Morel et al. [9] conducted a statistical
analysis of methylmercury concen-
trations in marine fish tissues and a
simple modeling exercise of the Pacific
Ocean. Statistical analyses of methyl-
mercury concentrations in yellowfin
tuna caught off of Hawaii in 1971
and 1998 indicated no statistically
significant change in methylmercury
concentrations over this time period.
By contrast, a simple, three-layer box
model of the ocean over this time span
indicated an increase in methylmer-
cury concentrations in surface waters
if methylmercury were formed in the
thermocline or the mixed layer (and
-------
therefore were susceptible to changes
in mercury deposition from the atmo-
sphere). Based on this contrast, Morel
et al. [9] suggested that methylation is
occurring in deep waters, sediments,
or hydrothermal vents where mercury
concentrations are not greatly affected
by human activities.
Inputs via Atmospheric Deposition
As mass balance studies indicate
(e.g., Hurley et al. [1] and Swain
et al. [2] results described above),
atmospheric deposition can be a
major source of total mercury to
an ecosystem and can have a direct
effect on surface water mercury
concentrations. Based on studies in
Spring Lake in northern Minnesota,
Hines, and Brezonik [2] reported
that atmospheric deposition is the
primary input of total mercury to the
lake. Likewise, Hurley et al. [1] found
that atmospheric deposition accounts
for 58 percent of the total mercury
input to Lake Superior. In a direct
ecosystem-level experiment (see
Section 3.1 and previous section),
Gilmour et al. [3] observed a rapid
and linear increase in methylmer-
cury production within the aquatic
ecosystem in response to changes
in mercury load by surface applica-
tion designed to simulate elevated
atmospheric deposition. Findings
from that study also indicate that the
newly deposited mercury was more
bioavailable than existing mercury at
these sites and contributes substan-
tially to methylmercury production
and bioaccumulation (Section 3.4).
As described in Section 3.2 under
Plant Transfers, Driscoll et al. [5]
analyzed mercury deposition in two
upland forest sites in the Sunday
Lake watershed and found 70 per-
cent of the total mercury deposition
to be via dry deposition.
Driscoll et al. [5] also sampled sec-
tions of sediment cores from eight
lakes in the Adirondacks for total
mercury to quantify the increase in
mercury deposition in that area over
the last 200 years. Sediment samples
were dated using 210Pb to measure
changes in sediment deposition
throughout this period. On average,
sites showed a 5.8-fold increase in
sediment mercury deposition from
background values (before 1900) to
peak values (which occurred from
1973 to 1995). Current mercury sedi-
ment deposition has decreased from
the peak to a level approximately 3.5
times the background values. Driscoll
et al. [5] also estimated rates for cur-
rent and pre-anthropogenic mercury
deposition in a perched seepage
lake. Using the sediment deposition
data, Driscoll et al. concluded that
retention of mercury in Adirondack
lakes has decreased over the past 200
years, but the mechanism for this
decrease is unclear.
Inputs from Runoff/Erosion
As described in Section 3.3, mass
balance calculations by Hurley et al.
[1] indicate that tributary loading is
a significant input of total mercury to
Lake Superior, suggesting that runoff
and erosion can be major sources
of mercury to a water body. In their
watershed study of Lake Superior
involving primarily groundwater
sampling in the basin, Hurley et al.
[1] found (among other results) that
forested areas in a watershed basin
can be a significant source of methyl-
mercury, apparently generated by in
situ methylation. Measurements of
methylmercury in groundwater and
surface water samples of the Sunday
Lake watershed indicated to Driscoll
et al. [5] that the riparian wetland
was a net source of methylmercury
to the lake and presumably a site for
methylation. Furthermore, Driscoll
et al.'s mass balance calculations
indicated that fluxes of methylmer-
cury were elevated in waters draining
riparian wetlands, and that the entire
watershed and the lake were net
sources of methylmercury to down-
stream surface waters. However, the
mass balance calculations did suggest
that the upland forest floor may be
an active zone of demethylation.
On the other hand, mass balance
calculations by Gilmour et al. [3]
using measured mercury isotope
concentrations from their whole-
ecosystem study in the Experimental
Lakes Area of Canada show that most
of the mercury loaded to uplands
has been slow to move to sites of
methylation in the lake and wetlands.
Only a small fraction of the mercury
loaded to upland areas had migrated
to the lake after two years; much of
the deposited mercury remained on
surface vegetation. Mercury loaded
to the wetland surface has been slow
to move to subsurface areas near
the water table where methylation
occurs. The differences between
these observations and those of
Hurley et al. [1] and Driscoll et al. [5]
may indicate the site-specific nature
of inorganic mercury mobility in the
environment.
Subsurface Transport
As noted above, Hurley et al. [1]
sampled ground water along a
stream passing through wetlands and
forested area in the Lake Superior
basin. A strong link between methyl-
mercury concentrations in subsurface
samples and the hydrologic cycle was
observed, indicating that subsurface
transport and/or production of meth-
ylmercury can be an important source
of methylmercury to surface waters.
Hurley et al. [1] proposed that the
source of the methylmercury from the
forested areas may be the hyporheic
zone of streams (i.e., the subsurface
interface occurring between ground-
water and stream water).
Brown et al. [6] studied how mercury
from mining activities has entered
watersheds. The researchers per-
formed column experiments to evalu-
ate the fate of Hg(0) introduced via
dredge tailings during gold mining
and subsequent restorations (such as
the Clear Creek tributary to the Sacra-
mento River in Redding, CA). Varying
influent ionic strengths and low
molecular weight organic acids were
applied to columns filled with tailings
-------
containing mercury in an attempt to
simulate infiltration through the root-
ing zone of re vegetated mine waste.
Significant concentrations of mercury
leached from the tailings in dissolved
and particle-associated forms. Chemi-
cal extractions indicated that Hg(0)
was transformed into readily soluble
species such as mercury oxides and
chlorides (3-4 percent), intermedi-
ately extractable phases that likely
include sorption complexes and
amalgams (75-87 percent), and frac-
tions of highly insoluble forms such
as mercury sulfides (6-20 percent;
e.g., cinnabar and metacinnabar).
These results imply that Hg(0) is
transformed into other forms of mer-
cury (soluble mercury and insoluble
mercury sulfides) during its transport
from placer gold mining regions
to major wetlands such as the San
Francisco Bay Delta. Also the results
indicate that colloid-associated
mercury from revegetated mercury
mine tailings sites may be of potential
environmental concern, and that the
gradual infiltration of even very low
concentrations of organic acids into
mine tailings will eventually instigate
conditions where colloid mobiliza-
tion can occur.
Brown et al [6] also attempted to
measure the variation of different
mercury isotopes from a variety of
natural mining samples to determine
if any natural variations in the ratios
may be useful as a mercury source
indicator. No significant variation
was found (above the experimental
error), but further studies were sug-
gested.
3.4 Bioavailabilityand
Bioaccumulation of Mercury
in Aquatic Systems
Bioavailability
Several STAR grant findings related to
factors that affect the bioavailability of
mercury to SRB have been discussed
in the preceding sections on mercury
transformations and mercury cycling
in the environment. Additional find-
ings on the topic are described here.
STAR grant researchers found that
newly deposited mercury is more
bioavailable than "older" mercury
and that DOC levels in a water body
appear to affect bioavailability. In
both the Florida Everglades and the
Experimental Lakes Area in Canada,
Gilmour et al. [3] found that newly
deposited mercury was more readily
methylated than existing mercury. In
the Everglades, Gilmour et al. dem-
onstrated that sulfate and DOC have
a much larger effect on the meth-
ylation of newly deposited mercury
than on the methylation of mercury
already present in soils. Additionally,
new methylmercury production from
mercury isotope spikes added during
the experiments was a good predic-
tor of total methylmercury concentra-
tions in surface sediments. After two
years of mercury loading, Gilmour et
al. [3] found that about 25 percent of
the mercury in perch in the Experi-
mental Lakes Area lake originated
from the isotopically-labeled spike.
Partitioning of methylmercury to
biota appeared to decrease with time
in Hurley's study. Hurley recorded
the uptake of methylmercury by
Selenastrum over time; a partition
factor of 105 was measured through
24 hours. After this time, the factor
decreased to 104. The reason for this
change is unclear, but may reflect the
increased algal growth rate.
Hurley et al. [1] found that mercury
concentrations in biota in Lake Supe-
rior are generally higher in the spring
than summer, a trend that might result
from bio-dilution by growing phyto-
and zooplankton during the summer.
Their results also showed higher con-
centrations of methylmercury in zoo-
plankton from the epilimnion. These
observations suggest that mercury
in surface waters deposited from the
atmosphere may be more bioavailable
than mercury in deeper waters.
Hurley et al. [1] also reported that
samples of water from Lake Superior
with higher DOC had lower algal
concentrations of methylmercury, sug-
gesting that DOC may be complexing
mercury and making it unavailable for
uptake. That observation agrees with
those of Fitzgerald et al. [7] (discussed
in Section 3.1) that increasing DOC
reduces methylation rates.
Driscoll et al. [5] found that biocon-
centration factor (BCF) values offish
in Sunday Lake in the Adirondacks
increased with each trophic level.
Total mercury concentrations in
yellow perch were also higher than
values measured in many remote
regions (locations not specified).
However, BCF values were generally
lower for yellow perch in Sunday
Lake than values for this species in
other Adirondack lakes. Driscoll et
al. suggested that this may occur
because methylmercury in Sunday
Lake is less bioavailable due to bind-
ing with the high concentrations of
DOC.
Mason's team [8] evaluated the role
of pH in mercury-sulfide complex-
ation, given that a number of studies
have shown an inverse relationship
between lake water pH and mer-
cury contamination offish tissues.
Mason et al. modeled mercury-sulfide
speciation with changing pH and
found that at low pH, as sulfide levels
increase, the tendency for speciation
of mercury to shift toward charged
complexes is not as strong as it is
at higher pH. In other words, when
sulfide levels are high, neutral mer-
cury-sulfide complexes, which are the
bioavailable forms (see Section 3.1),
are more likely to form if the pH is
low than if the pH is high. This rela-
tionship may help explain why fish
tend to have higher tissue mercury
residues in waters of lower pH.
The bioavailability of mercury at
mining sites could vary based on the
mercury speciation at the site and
presence of ions such as chloride and
sulfate. Brown et al. [6] took samples
from multiple mining locations in
California Coast Range mercury
mineral belt, the placer gold mining
-------
belt in the Sierra Nevada foothills,
and Nevada and found significant
variations in mercury speciation from
site to site. Highly soluble forms
of mercury, such as the minerals
corderoite, montroydite, schuetteite,
egglestonite, and terlinguite, found at
some sites are potentially much more
bioavailable than highly insoluble
forms of mercury, such as cinnabar
and metacinnabar. Therefore, bio-
availability may vary from site-to-site
with mine waste contamination.
However, Brown et al. [6] also
found that sorption of Hg(II) onto
mineral particles may effectively
sequester mercury in mine tailing
and aquatic environments if the sorp-
tion complexes are strongly bonded
to the particle surfaces. An assess-
ment of Hg(II) sorption complexes
that form on mineral particles in
mercury-mine waste environments
showed that Hg(II) forms dominantly
inner-sphere sorption complexes
on these common mineral particles.
An important result of this study is
the finding that iron and aluminum
(hydr)oxides can play a significant
role in the uptake of Hg(II) through
direct inner-sphere sorption pro-
cesses. Additionally, the sorption of
Hg(II) on these particles was exam-
ined in the presence of chloride and
sulfate ions as a function of pH. The
presence of chloride reduced Hg(II)
sorption (through the formation of
HgCl2 complexes), and the presence
of sulfate enhanced Hg(II) sorption
(through the reduction in electro-
static repulsion at positively-charged
particle surfaces).
Bioaccumulation
Hurley et al. [1] and Fitzgerald et al.
[7] studied the river/lake interfaces
(i.e., the riverine mixing zones)
of Lake Superior and Long Island
Sound, respectively, and observed
that the concentration of mercury
bound to the particulate phase
increased in the mixing zones relative
to the river water, which can affect
bioavailability and bioaccumulation.
Hurley et al. [1] measured mercury
and methylmercury concentra-
tions at the riverine mixing zone in
Lake Superior. The methylmercury
concentration on particles increased
within the mixing plume with dis-
tance from the interface; in other
words, the methylmercury content
of particles is enriched at the outer
edge of the plume. Larger particles
(and the methylmercury associated
with these particles) settle out closer
to the mouth of the river; smaller
particles remain in solution and carry
their load further offshore. Therefore,
methylmercury in the particulate
phase enters the lake water column
from the river associated predomi-
nantly with small particles, which
can be ingested by zooplankton. The
result is enhanced bioaccumulation
of methylmercury in plankton in the
mixing zone.
3.5 New Methods for
Mercury Analysis
Several new analytic methods for
measuring total and speciated mer-
cury in the environment were devel-
oped under the STAR grants:
• A batch culture bioassay using Selen-
astrum capricorutum was devel-
oped to measure bioaccumulation
of methylmercury by algae [ 1 ].
• Refined methods to measure
methylation rates in sediments at
near ambient mercury levels were
developed using individual stable
mercury isotopes and ICP-MS
analysis [1, 8]. These methods
also allow tracking of both the in
situ mercury methylation and the
fate of the added mercury. Using
these techniques, methylation and
demethylation were measured in
the same sample concurrently [8].
Synchrotron-based x-ray absorp-
tion fine structure (XAFS) spec-
troscopy was used to determine
the molecular-level speciation
of Hg in many different mining
localities in California [6]. This
method was compared to sequen-
tial extraction methods and solid
phase mercury thermodesorption
in a collaborative study [10].
An in vitro reductible-mercury
titration approach was developed
to determine the concentration
and conditional stability constants
of dissolved organic matter to-
wards mercury [7].
A semi-automated dissolved el-
emental mercury analyzer (DEMA)
was developed for measuring
Hg(0) concentrations in labora-
tory samples of surface water [7].
A method was refined to estimate
the bioavailability of mercury at
environmental levels through
measurement of the octanol-water
partitioning of mercury [8]. Previ-
ous measures were effective only
for high concentrations of mer-
cury. Clean techniques and other
modifications allowed detection of
mercury at picomolar levels.
A new technique for measuring
low levels of sulfide in natural
waters was developed to detect at
nanomolar to micromolar levels of
mercury [8].
EcoCELLs were used for investigat-
ing the biogeochemical cycling of an
environmental contaminant [11].
-------
4. APPLICATIONS OF FINDINGS FROM STAR GRANTS
'•'**m
In this section, the findings presented
by research theme in Section 3 are
summarized in three ways according
to their potential application in future
mercury research and modeling. First,
results are grouped according to how
they may be used in the develop-
ment or improvement of mercury fate
and transport models. Second, key
variables studied here that affect the
fate and transport of mercury in the
environment are cross-walked with
mercury cycling processes that are
influenced by these variables. Finally,
information about how these research
results broaden our understanding
of variation across ecosystems and
regions is presented.
4.1 Potential Improvements to
Mercury Fate and Iransport
Models
This section summarizes new mod-
els developed as a part of the STAR
research and groups key findings that
will be useful in the refinement of
existing mercury fate and transport
models.
New Models
Work under two of the STAR grants
included developing models that
incorporated the observations made
by the investigators in their ecosys-
tem studies. Methods and algorithms
from these models could potentially
be used in the development of other
models as well. Driscoll et al. [5]
constructed the Mercury Cycling
Model for Headwater Drainage Lake
Systems (MCM-HD) to simulate
mercury interactions in headwater
drainage lakes and the adjacent
watersheds, including wetlands, and
found that results from the model
were generally in agreement with
mass balance calculations. Hurley
et al. [1] developed a simple model
that related annual tributary loading
from a watershed to the lake based
on watershed composition (i.e., land
use/type). The model was applied to
the entire lake watershed. Data from
that study [1] also are being incorpo-
rated into the Lake Superior dynamic
Mercury Cycling Model (D-MCM) to
elucidate the effects of mercury loads
in specific nearshore regions.
New Data
STAR grant investigators gained new
understanding of various processes
that affect mercury fate and cycling in
the environment that can be incor-
porated into existing or new mercury
models. Findings are summarized
according to their relevance to meth-
ylation, inorganic transformations,
and mercury transport.
Factors Affecting Methylation
Several investigators studied mer-
cury methylation and the factors
that enhance or inhibit it. Fitzgerald
et al. [7] demonstrated that meth-
ylation rates increase with elevated
bacterial metabolic rates and higher
concentrations of inorganic mercury.
Factors affecting methylation studied
by other researchers can be grouped
into these same two categories.
Bacteria methylation rates influence
the total amount of methylmercury
that is formed. Three important fac-
tors that influence bacteria methyla-
tion rates were studied, including:
• Time of year/temperature. Meth-
ylation rates are higher during the
warmer summer months [7].
• Sulfate levels. Increasing sulfate
levels can increase methylation in
sulfate-poor systems [2, 8].
• Availability of Hg(II). Increasing
inorganic mercury loading to a water-
shed increases methylation rates [3].
The STAR grant investigators demon-
strated the effect of several factors on
the availability of inorganic mercury
for methylation. Many of these factors
would be important in mercury fate
models.
• Organic Matter. Increasing
organic matter concentrations
decreases the concentration of
methylmercury, probably be-
cause inorganic mercury forms
complexes with DOC, making it
unavailable for uptake by methyl-
ating bacteria [1,7].
• Sulfide Concentration. The ad-
dition of sulfide can decrease the
rate of methylation [8]. The effect
of sulfide on methylation depends
on whether neutral or charged
sulfide mercury complexes form.
Neutral complexes enhance meth-
ylation, while charged complexes
inhibit it [8].
-------
• Particulate Binding. Inorganic
mercury binding to sediment and
paniculate matter will decrease
methylation. Higher sediment-
water distribution coefficients for
inorganic mercury are associated
with lower methylation rates [7].
• Mixing Zones. The fraction of
reactive mercury increases in the
mixing zones of rivers with lakes
or sounds, thereby increasing the
amount available for methylation
and bioaccumulation [1,7].
• Bioturbation. Increased bioturba-
tion of benthic infauna in the sedi-
ment can increase available mercury
and increase methylation [7].
• "Newness" of Mercury. In some
environments, newly deposited
mercury contributes more substan-
tially to production of methylmer-
cury than "older" mercury [3]. Sul-
fate and DOC have a much larger
effect on the methylation of newly
deposited mercury than of mercury
already present in soils [3].
Factors Affecting Inorganic
Mercury Transformations
Transformations between Hg(II) and
Hg(0) also were studied by STAR
grant investigators, and their results
include key information that could be
incorporated into mercury environ-
mental fate models. Reduction of
Hg(II), photooxidation of Hg(0), and
volatilization of Hg(0) are all impor-
tant processes in the mercury cycle
that can ultimately affect the availabil-
ity of mercury for methylation and
subsequent accumulation in the food
chain, as indicated below.
• Photochemical Reduction.
Hg(II) is readily reduced to Hg(0)
in aquatic surface waters by pho-
tochemical reduction; as a result,
Hg(0) formation and subsequent
volatilization display a diurnal
cycle [4].
• Organic Matter. Some data indi-
cate that Hg(II) must be bound
to organic material to be reduced
by abiotic processes in natural
waters. However, excess ligands
can result in significantly lower
reduction rates of Hg(II) [7].
• Photooxidation. Photooxidation
is responsible removing much
of the Hg(0) in surface waters.
In some cases, photooxidative
flux can exceed volatilization as
a loss mechanism for Hg(0) in an
aquatic system [2, 9]-
Factors Affecting Transport
Some data points collected on the
major transport processes for mer-
cury through a watershed could be
incorporated into fate and transport
models.
• Volatilization. Volatilization is an
important process for removing
Hg(0) from surface waters and a
potentially significant source of
Hg(0) to the atmosphere [1, 2, 4,
6, 7]. For example, Fitzgerald et
al. [7] estimated annual emissions
of Hg(0) from Long Island Sound
to be 35 percent of the input of
mercury to Long Island Sound.
Volatilization of mercury from
soils, which has been shown to
vary based on light, could also be
an important source of mercury to
the atmosphere [10, 11].
• Deposition and Plant Transfers.
Several studies confirmed that
atmospheric deposition is the key
pathway for emitted mercury to en-
ter watersheds [1, 2, 5], and some
provided quantitative data on this
relationship (e.g., dry deposition
was 70 percent of the total deposi-
tion of mercury to the forest floor
[5]). Additionally, plant foliage may
be a significant sink of mercury
from the atmosphere, and conse-
quentially may be a large source of
mercury to watershed via litterfall
and deposition [11].
• Runoff/Erosion. Results were
mixed regarding the extent to
which uplands and wetlands func-
tion as a source of methylmercury
to downstream water bodies. The
rate of transport and methylation
in a watershed depends on ecosys-
tem-specific properties [1, 3, 5]
as well as sorption onto mineral
particles and availability of com-
plexing ligands [6].
4.2 Understanding Key Variables
Affecting Mercury Fate and
Transport
Several specific variables that affect
mercury fate and transport in the
environment were identified and
studied in the STAR mercury research
projects. Key variables are summa-
rized in Table 4. The applicability of
each variable to mercury transport,
availability, or speciation (as deter-
mined by these studies) is indicated.
4.3 Understanding Variation
in Key Variables by Ecoregion
and Ecotype
The STAR grant research on mercury
was performed in a variety of ecosys-
tems across North America, includ-
ing an Adirondack lake, Long Island
Sound, the Experimental Lakes Area
in northwestern Ontario, Canada, the
Florida Everglades, Lake Superior,
the Patuxent River in Maryland, the
St. Lawrence River, the Saginaw Bay
watershed in Michigan, the Marcell
Experimental Forest in northeastern
Minnesota, and mining and other
mercury source locations in Cali-
fornia and Nevada. Some regional
differences were noted by research-
ers; however, no study attempted
to systematically identify regional
characteristics affecting mercury fate
and transport in the environment.
Regional differences are expected
for processes that are affected by
temperature and solar radiation.
For example, Gilmour et al. [3]
performed similar experiments
in the Florida Everglades and the
Experimental Lakes Area in Canada
and observed differences in methyl-
mercury response to mercury surface
-------
TABLE 4. VARIABLES AFFECTING MERCURY FATE AND TRANSPORT
Variable
Deposition rate / mercury load
Sulfate load
Sulfide concentrations and charge of
mercury-sulfide complexes
Dissolved organic carbon (DOC)
Suspended particulate matter
Sediment burial
Temperature
Bacterial activity
"Newness" of mercury
Phase partitioning of mercury between
solid and liquid in sediment and surface
water
Bioturbation of sediments
pH
Photo-demethylation of methylmercury
Oxidation of Hg(0)
Reduction of Hg(II)
Volatilization of Hg(0) (phase
partitioning between water and air)
Vegetation
Processes Affected by Variable
Transport
between media
•
•
•
•
•
'
•
•
Availability to
methylating bacteria
•
•
•
•
•
•
"
•
•
•
Transformation
between Hg species3
•
•>
•
•>
•>
•
-
•
•
•
1 For reasons other than availability to methylating bacteria, which is covered in the previous column.
b Principally due to effects on bacteria methylation rates.
-------
application at these two locations,
with shorter response times mea-
sured in the shallow, warm sites of
the Everglades. Both photooxida-
tion of Hg(0) and photoreduction of
Hg(II) will be more prevalent in areas
with less cloud cover and, during the
summer, at lower latitudes with more
intense solar radiation. The balance
between those two processes is likely
to be affected by more localized fac-
tors, however.
Ecosystem differences are expected
for processes that are affected by
DOC and particulate matter in sur-
face waters, which can vary substan-
tially depending on the type of water
body and characteristics of the water-
shed. The effects of these factors on
mercury bioavailability methylation,
and bioaccumulation are complex,
and will need to be considered in
conjunction with other factors. In
freshwater systems, high levels of
suspended sediment particles tend
to result in lower dissolved Hg(II)
concentrations as the mercury sorbs
to the particles and may become
buried in the benthic sediments.
Where rivers meet large lakes or the
ocean, on the other hand, the smaller
suspended sediment particles might
enhance transfer of mercury through
aquatic food chains as the mercury-
containing particles are consumed
by zooplankton in the mixing zone.
DOC can affect mercury cycling in
aquatic systems by binding dissolved
inorganic and organic mercury in
complexes limiting its bioavailabil-
ity to methylating bacteria and to
other aquatic organisms, by limiting
UV light penetration and photolytic
processes, and by generating reac-
tive molecules that assist in Hg(II)
reduction. The affinity of DOC for
mercury varies across ecosystems
as well. Given the importance of
these localized factors, differences
in mercury distribution can occur
within relatively small geographic
regions. For example, Driscoll et al.
[5] compared measured BCF values
in Sunday Lake in the Adirondacks to
values measured in other Adirondack
lakes and observed that BCF values
were lower in Sunday Lake. He sug-
gested that the methylmercury in
Sunday Lake may be less bioavailable
due to binding associated with high
concentrations of DOC.
-------
5. FUTURE RESEARCH NEEDS
Several overarching future research
needs related to mercury fate and
transport were identified by compar-
ing the STAR research results for these
eight projects to the goals outlined
in the ORD Research Plan and the
original STAR Mercury RFA (see Sec-
tion 2.2). Additionally, areas of further
research that were specifically identi-
fied by investigators as follow-up to
the work completed under these STAR
grants are summarized below;
Topics Included in ORD
Research Plan
The research summarized here
contributes new data and analyses
that improve general understanding
of processes and factors that affect
the fate and transport of mercury
in the environment. Thus, these
research projects have contributed
significantly to the first two ORD
goals/measures listed in Table 2.
Specific ORD research goals and per-
formance measures that are related
to the grants described here (see
Section 2.3), but may require further
work to fully accomplish, include
development of a model for mercury
in fish, identification of sources of
mercury emissions, and the eventual
creation of an integrated multimedia
modeling framework for mercury
in the environment. The research
covered by the eight STAR grants
described here will help EPA accom-
plish these broader, long-term goals;
however, it is expected that other
ORD-related research will also play a
role in ultimately meeting those goals
(especially the long-term goal of an
integrated multimedia framework).
Topics Of STAR RFAs
The specific research questions
posed in the original RFA are listed
in Section 2.2. The majority of
the STAR grant research projects
described in this report focused on
Question 2 (i.e., variables controlling
mercury transformations). Specific
variables were identified that control
the transformation of mercury to
methylmercury as well as transforma-
tions between inorganic species of
mercury.
Question 1 (i.e., bioavailability and
fish concentrations of mercury) was
addressed by some investigators
(e.g., Hurley et al. [1] and Driscoll
et al. [5]). Further research into the
relationship between mercury in
the watershed to mercury in biota,
however, may be necessary to close
information gaps regarding uptake by
fish and other organisms.
The STAR research described here
was performed in a range of loca-
tions; however, follow-up research
systematically analyzing the differ-
ences between results for these dif-
ferent ecosystems would be required
to comprehensively answer Question
3 (i.e., how mercury cycling varies
in different geographic regions). In
addition to further analyzing the
existing results across regions, stud-
ies that conduct research in parallel
in more than one region (similar
to the Gilmour et al. [3] study in
Canada and Florida) may be useful
in clearly identifying how mercury
cycling varies by region. The focus
for most of the research presented
here was on freshwater ecosystems;
therefore, further investigation into
the differences between freshwater
and estuarine or coastal ecosystems
would be beneficial, especially
because research presented here
(Fitzgerald [7]) indicates that pro-
cesses may be different in the estuary
environment than in freshwater.
Topics Specified by Researchers
Many of the STAR grant final reports
list areas where further research
is warranted. Investigators have
reported both general subjects wor-
thy of future consideration and more
specific research topics that follow
directly from work completed to
date. Both types of recommendations
are summarized here.
General areas for continued or future
research identified by the STAR grant
investigators include the following.
• Effects of sulfate on methyla-
tion. Driscoll et al. [5] suggested
developing a new formulation for
the response of methylation to
changes in sulfate concentrations
in a wetland environment. Other
research mentioned by Driscoll et
al. has shown a greater-than-linear
response in methylation when
sulfate is added; however, the
current model, based on data for
lakes, does not predict that type of
response.
• Intracellular sequestration of
mercury. Mason et al. [8] noted
that not all of the mercury enter-
ing a sulfate-reducing bacterium
-------
in neutral sulfide complexes is
methylated. Methylating bacteria
carry out both methylation pro-
cesses and competing processes
that sequester mercury and render
it unavailable for methylation. It
has been hypothesized that differ-
ences in cell physiology (e.g., size
and membrane composition) and
in mercury partitioning within the
cell cause the variation in methyla-
tion rates observed among bacteri-
al strains. Further work is needed
to understand the mechanisms
and kinetics of these intracellular
sequestration pathways.
Demethylation. Mason et al. [8]
also observed that the factors in-
fluencing demethylation processes
are not well understood and
deserve further study.
"Newness" of mercury. Mason et
al. [8] further suggested that more
work is needed to elucidate the
differences in methylation rates of
newly added mercury compared
to in situ mercury. It is difficult
to assess the pool of mercury
available for methylation, and this
information is also necessary to
accurately estimate methylation
rates.
Photooxidation. Morel et al. [9]
recommended further research
in different types of water bodies
to determine if more of the Hg(0)
in surface waters is converted to
Hg(II) via photooxidation or is
lost to the atmosphere via volatil-
ization of Hg(0).
SRB studies. Morel et al. [9]
suggested additional research
on bacterial methylation by SRB,
including: the role of vitamin B12
and other methyl-transferring
coenzymes in methylmercury pro-
duction by SRB; the role internal
mercury speciation plays in meth-
ylation; the pathways responsible
for methylation; and the biome-
chanical mechanism of bacterial
methylation.
• Mercury in seafood. Morel et al.
[9] and Nriagu et al. [4] both not-
ed that more research is needed
to determine the ultimate sources
of mercury in seafood.
Specific research needs noted by
investigators that relate to sites or
topics studied in STAR grants include
the following.
• Gilmour et al. [3] suggested
follow-up research in the Experi-
mental Lakes Area in Canada to
determine time to equilibrium
for increased mercury loading at
Experimental Lakes Area; to deter-
mine the time course for mercury
movement through uplands and
wetlands; to investigate bioavail-
ability of mercury from uplands
and wetlands for methylation
and bioaccumulation; to investi-
gate biogeochemical parameters
that affect bioavailability; and to
more fully investigate "aging" of
dosed mercury (including ligand
exchange of mercury).
• Hurley et al. [1] listed several
specific next steps for research in
Lake Superior, including the inves-
tigation of atmospheric deposition
(i.e., effects of local urban sources
on nearshore regions and direct
deposition of methylmercury in
the basin), mixing zones/water-
shed influences (i.e., mechanisms
for algal and zooplankton uptake
and fate of colloidal methylmer-
cury), and in-lake processes (i.e.,
fate of changing anoxia on meth-
ylmercury dynamics in Lake Erie,
mechanisms of bioaccumulation
in offshore zones, and methyla-
tion in deep sediments).
• Nriagu et al. [4] observed associa-
tions between photosynthetically
active radiation and indicators of
biological activity (e.g., chloro-
phyll-a). They plan to use those
data to further explore the interac-
tion of biological processes with
the mercury cycle in the Saginaw
Bay.
• Brown et al [6] saw in laboratory
experiments that organic acids
may affect mercury transport from
mining sites, and suggested fur-
ther field studies into the poten-
tial impact of revegetation on the
mobilization of colloidal materials
from mine tailings. Addition-
ally, Brown et al. [6] suggested
further studies into the question
of whether ratios of mercury iso-
topes could be used to determine
the source of mercury. Specifically,
measurements of the isotope
ratios need to be made with more
precision (using more sensitive
multi-collector, inductively cou-
pled plasma mass spectrometer).
• Gustin et al. [11] follow-up
research has already begun and
includes using the EcoCELLs to
study mercury cycling associated
with tall grass prairie vegetation
and soils from Oklahoma.
In addition, these STAR grant results
have provided a wealth of data that
merits further analysis and compara-
tive review. For example, the research
of Nriagu et al. [4] could be analyzed
for additional patterns and reasons
for reduction of mercury and other
transformations. Measured mercury
concentrations in environmental
media at the different locations stud-
ied by the STAR grant investigators
could be tabulated, and ratios of total
mercury, Hg(0), Hg(II), and meth-
ylmercury, could be compared to
identify regional and temporal trends
or patterns of mercury distribution in
the environment.
-------
Environmental Protection Agency
(EPA). (1997a). Mercury Study
Report to Congress. EPA-452/R-
97-003 to -010. U.S. EPA, Office
of Air Quality Planning and Stan-
dards, and Office of Research
and Development, Washington,
DC.
EPA. (1997b). Great Waters Second
Report to Congress. EPA-453/R-
97-011. U.S. EPA, Office of Air
Quality Planning and Standards,
Research Triangle Park, NC.
EPA. (1998). Utility Air Toxics Report
to Congress. EPA-453/R-98-004.
U.S. EPA, Office of Air Qual-
ity Planning and Standards,
Research Triangle Park, NC.
EPA. (2001). Office of Research and
Development Strategic Plan.
EPA/600/R-01/003. U.S. EPA,
Office of Research and Develop-
ment, Washington, DC. January.
EPA. (2003a). 2003-2008 EPA Stra-
tegic Plan: Direction for the
Future. EPA-190-R-03-003. U.S.
EPA, Office of the Chief Financial
Officer, Office of Planning, Analy-
sis, and Accountability, Washing-
ton, DC. September.
EPA. (2003b). Mercury Research
Multi-Year Plan. FY2005 Plan-
ning - Final Version, May 9, 2003.
U.S. EPA, Office of Research and
Development, Washington, DC.
Available at: http://www.epa.
gov/osp/myp/mercurypdf.
EPA. (2003c). Air Toxics Multi-
Year Plan. U.S. EPA, Office of
Research and Development,
Washington, DC. April 2003
Update.
Mason RP, Abbott ML, Bodaly RA,
Bullock Jr. OR, Driscoll CT, Evers
D, Lindberg SE, Murray M, Swain
EB. (2005). Monitoring the
response to changing mercury
deposition. Environmental
Science and Technology A-Pages
39(1):14A-22A
-------
Babiarz CL, Hurley JP, Hoffmann SR,
Andren AW, Shafer MM, Armstrong
DE. (2001). Partitioning of total
mercury and methylmercury to the
colloidal phase in fresh waters. Envi-
ronmental Science and Technology,
35(24):4773-4782.
Babiarz CL, Hoffmann SR, Shafer
MM, Hurley JP, Andren AW, Armstrong
DE. (2000). A critical evaluation of
tangential-flow ultrafiltration for trace
metal investigations in fresh water
systems: Part II Total and Methylmer-
cury. Environmental Science and
Technology, 34(16):3428-3434.
Back RC, Hurley JP, Rolfhus KR.
(2002). Watershed influences on the
transport, fate and bioavailability of
mercury in Lake Superior: Field mea-
surements and modeling approaches.
Lakes and Reservoirs: Research and
Management, 7:201-206.
Back RC, Gorski PR, Cleckner LB,
Hurley JR (2003). Mercury content
and speciation in the plankton and
benthos of Lake Superior. Science of
the Total Environment, 304:327-348.
Cleckner LB, Back RC, Gorski PR,
Hurley JP, Byler S. (2003). Seasonal
and size-specific distribution of meth-
ylmercury in seston and zooplank-
ton of two contrasting Great Lakes
embayments./owm
-------
Hines NA, Brezonik PL. (2004).
Mercury dynamics in a small north-
ern Minnesota lake: Water to air
exchange and photoreactions of
mercury. Marine Chemistry, 90(1-
4): 137-139.
Johnson BM. (2004). Sulfate reducing
bacteria and the role of nutrients in
mercury methylation in sediments of
Spring Lake, Minnesota. M.S. Thesis,
University of Minnesota, Minneapolis,
Minnesota.
Hines NA, Brezonik PA, Engstrom DR.
Sediment and porewater profiles and
fluxes of mercury and methylmercury
in a small seepage lake in northern
Minnesota. (Submitted).
Babiarz CL, Hurley JP, Krabbenhoft
DP, Gilmour CC, Branfireun B.
(2003). Application of ultrafiltration
and stable isotopic amendments to
field studies of mercury partitioning
to filterable carbon in lake water and
overland runoff. Science of the Total
Environment, 304:295-303.
Benoit JM, Gilmour CC, Mason RP
(2001). Aspects of the bioavailability
of mercury for methylation in pure
cultures of Desulfobulbouspropioni-
cus (Ipr3). Applied and Environmen-
tal Microbiology, 67:51-58.
Benoit JM, Gilmour CC, Mason RR
(2001). The influence of sulfide on
solid-phase mercury bioavailability
for methylation by pure cultures of
Desulfobulbouspropionicus (Ipr3).
Environmental Science and Technol-
ogy, 35:127-132.
Benoit JM, Mason RP, Gilmour CC,
Aiken GR. (2001). Mercury bind-
ing constants for dissolved organic
carbon isolates from the Florida Ever-
glades. Geochimica et Cosmochimica
Acta, 65:4445-4451.
Benoit J, Gilmour C, Heyes A, Mason
RP, Miller C. (2003). Geochemical and
biological controls over methylmer-
cury production and degradation in
aquatic ecosystems. In: Biogeochem-
istry of Environmentally Important
Trace Elements, ACS Symposium
Series #835, Y Chai and O.C. Braids,
Eds. American Chemical Society,
Washington, DC. pp. 262-297.
Jay JA, Murray KJ, Gilmour CC, Mason
RP, Morel FMM, Roberts AL, Hemond
HE (2002). Mercury methylation by
Desulfovibrio desulfuricans ND132
in the presence of polysulfides.
Applied and Environmental Microbi-
ology, 68:5741-5745.
Zhang H, Lindberg S. (2001). Sun-
light and iron(III)-induced photo-
chemical production of dissolved
gaseous mercury in freshwater.
Environmental Science and Technol-
ogy, 35:928-935.
Zhang H, Lindberg S, Gustin M, Xu
X. Towards a better understanding of
mercury emissions from soils. In: Cai
Y, Braids O [eds.] Biogeochemistry of
Environmentally Important Elements.
American Chemical Society Sympo-
sium Series Book, American Chemical
Society, Oxford University Press.
Driscoll CT, Munson RK, Yavitt J,
Newton RM, Demers J, Kalicin M,
McLaughlin E. Chemical and bio-
logical control of mercury cycling in
upland, wetland and lake ecosystems
in the Adirondack Region of New
York, USA. Environmental Pollution.
Kalicin M, Driscoll C, Yavitt J, Newton
R, Munson R. The dynamics of mer-
cury in upland forests of the Adiron-
dack region of New York. Environ-
mental Pollution.
McLaughlin E, Driscoll C, Yavitt J,
Newton R, Munson R. Mercury in
upland and riparian wetland vegeta-
tion. Environmental Pollution.
McLaughlin E, Driscoll C, Suther-
land J, Yavitt J, Newton R, Munson
R. Trophic transfer of mercury in an
Adirondack lake ecosystem. Environ-
mental Pollution.
Munson RK, Harris RC, Driscoll CT,
Yavitt J, Newton RM. The mercury
cycling model for headwater drainage
lakes (MCM-HD): Model theory and
processes. Environmental Pollution.
Perry ER, Norton SA, Kamman KG,
Lorey PM, Haines T, Driscoll CT. Mer-
cury accumulation in lake sediments
in the northeastern United States
during the last 150 years. Ecotoxicol-
ogy (In press).
Yavitt JB, Demers J, Driscoll CT, Kali-
cin M, Newton R, Munson R. Wetland
characteristics and mercury behavior
within an Adirondack (New York
State) watershed. Environmental
Pollution.
Coolbaugh ME, Gustin MS, Rytuba JJ.
(2002). Annual emissions of mer-
cury to the atmosphere from three
natural source areas in Nevada and
California, Environmental Geology,
42:338-349.
Engle MA, Gustin MS, Zhang H.
(2001). Quantifying natural source
mercury emissions from the Ivanhoe
Mining District, north-central Nevada,
USA. Atmospheric Environment,
35:3987-3997.
Engle MA, Gustin MS. (2002). Scaling
of atmospheric mercury emissions
from three naturally enriched areas:
Flowery Peak, Nevada, Peavine Peak,
Nevada and Long Valley Caldera, Cali-
fornia. Science of the Total Environ-
ment, 290(1-3) :91-104.
-------
GigliniT. (2003). Measurement of
Total and Reactive Mercury above a
Naturally Enriched, an Anthropogeni-
cally Contaminated and a Pristine
Site, Nevada. M.S. Thesis, University
of Nevada, Reno.
Gustin MS. (2003). Are mercury
emissions from geologic sources
significant? A status report submitted
to Science of the Total Environment,
304:153-167.
Gustin MS, Lindberg SE, Austin K,
Coolbaugh M, Vette A, Zhang H.
(2000). Assessing the contribution of
natural sources to regional atmo-
spheric mercury budgets. Science of
the Total Environment 259(1) :6l-72.v
Gustin MS, Biester H, Kim CS. (2002).
Investigation of light enhanced
emissions of mercury from naturally
enriched substrate. Atmospheric Envi-
ronment, 36(20) :324l-3254.
Gustin MS, Coolbaugh M, Engle M,
Fitzgerald B, Keislar R, Lindberg
S, Nacht D, Quashnick J, Rytuba
J, Sladek C, Zhang H, Zehner RE.
(2003). Atmospheric mercury
emissions from mine wastes and
surrounding geologically enriched
terrains. Environmental Geology,
43:339-351.
Johnson SB, Yoon TH, Slowey AJ,
Brown GE Jr. (2004). Adsorption
of organic matter at mineral/water
interfaces: 3. Implications of surface
dissolution for adsorption of oxalate.
Langmuir, 20(26): 11480-11492.
Kim CS. (2002). Mercury Speciation
and Sorption Processes in Mining
Environments. Ph.D. Thesis, Depart-
ment of Geological & Environmental
Sciences, Stanford University, Stan-
ford, CA.
Kim CS, Rytuba JJ, Brown GE Jr.
(1999). Utility of EXAFS in character-
ization and speciation of mercury-
bearing mine wastes. Journal of
Synchrotron Radiation, 6:648-650.
Kim CS, Brown GE Jr., Rytuba JJ.
(2000). Characterization and specia-
tion of mercury-bearing mine wastes
using X-ray absorption spectroscopy
(XAS). Science of the Total Environ-
ment, 261:157-168.
Kim CS, Catalano JG, Grolimund D,
Warner JA, Morin G, Juillot F, Galas
GC, Ildefonse P, Rytuba JJ, Parks GA,
Brown GE Jr. (2000). EXAFS determi-
nation of the chemical speciation and
sorption processes of Hg(II), Sr(II),
and Zn(II) in natural and model sys-
tems. 1999 Activity Report, Stanford
Synchrotron Radiation Laboratory
Report, Stanford, CA.
Kim CS, Bloom NS, Rytuba JJ, Brown
GE Jr. (2003). Mercury speciation
by extended x-ray absorption fine
structure (EXAFS) spectroscopy and
sequential chemical extractions: A
comparison of speciation methods.
Environmental Science and Technol-
ogy, 37:5102-5108.
Kim CS, Rytuba JJ, Brown GE Jr.
(2004). Geological and anthropo-
genic factors influencing mercury
speciation in mine wastes. Applied
Geochemistry, 19579-393.
Kim CS, Rytuba JJ, Brown GE Jr.
(2004). Mercury(II) sorption to
Fe- andAl-(hydr)oxides: II. Effects of
chloride and sulfate. Journal of Col-
loid and Interface Science, 270:9-20.
Kim CS, Rytuba JJ, Brown GE Jr.
(2004). Mercury(II) sorption to
Fe- andAl-(hydr)oxides: I. Effects of
pH. Journal of Colloid and Interface
Science, 271:1-15.
Lowry G\ Shaw S, Kim CS, Rytuba
JJ, Brown GE Jr. (2004). Particle-
facilitated mercury transport from
New Idria and Sulphur Bank mercury
mine tailings. 1. Column experiments
and macroscopic analysis. Environ-
mental Science and Technology,
38(19):5101-5111.
Nacht DM. (2002). Measurement
of Reactive Gaseous Mercury and
Mercury Flux from Substrates in
California and Nevada. M.S. Thesis,
University of Nevada, Reno.
Nacht DM, Gustin MS. (2004). Mer-
cury emissions from background and
altered geologic units throughout
Nevada. Water, Air, and Soil Pollu-
tion, 151:179-193.
Sladek C, Gustin MS, Kim CS, Biester
H. (2002). Application of three meth-
ods for determining mercury specia-
tion in mine waste. Geochemistry:
Exploration, Environment, Analysis,
2:369-376.
Slowey AJ, Rytuba JJ, Brown GE Jr.
(2005). Speciation of mercury and
mode of transport from placer gold
mine tailings. Environmental Science
and Technology, 39(6): 1547-1554.
Zehner RE, Gustin MS. (2002). Esti-
mation of mercury vapor flux from
natural substrate in Nevada. Envi-
ronmental Science and Technology,
36(19) :4039-4045.
Zhang H, Lindberg SE, Gustin MS, Xu
X (2003). Toward a better under-
standing of mercury emissions from
soils. In: Biogeochemistry of Environ-
mentally Important Trace Elements,
ACS Symposium Series #835, Y
Chai and O.C. Braids, Eds. American
Chemical Society, Washington, DC.
pp. 246-261.
Kim CS, Gustin MS, Rytuba JJ, Brown
GE Jr. Associations between Hg
speciation and Hg vapor flux rates in
mine wastes. Environmental Science
and Technology.
Nacht DM, Gustin MS, Engle MA,
Zehner RE, Giglini AD. Quantifying
total and reactive gaseous mercury
at the Sulphur Bank Mercury Mine
Superfund Site, Northern California.
Environmental Science and Technol-
ogy. (Submitted).
-------
Slowey AJ, Johnson SB, Rytuba JJ,
Brown GE Jr. Role of Organic Acids
in Promoting Colloid Transport of
Mercury from Mine Tailings. Envi-
ronmental Science and Technology.
(Submitted).
Fitzgerald WF, Vandal GM, Rolfhus
KR, Lamborg CH, Langer CS. (2000).
Mercury emissions and cycling in the
coastal zone. Journal of Environmen-
tal Science, 12(1):92-101.
Fitzgerald WF, Lamborg CH. (2003).
Geochemistry of mercury in the
environment. In: Sherwood-Lollar B,
ed. Treatise on Geochemistry, Vol. 9:
Environmental Geochemistry. Else-
vier: St. Louis, MO.
Hammerschmidt CR, Fitzgerald WF.
(2001). Formation of artifact methyl
mercury during extraction from a
sediment reference material. Analyti-
cal Chemistry, 73(24):5930-5936.
Hammerschmidt CR, Fitzgerald WF.
(2004). Geochemical controls of
the production and distribution of
methylmercury in near-shore marine
sediments. Environmental Science
and Technology, 38(5): 1480-1486.
Lamborg CH, Tseng CM, Fitzgerald
WF, Balcom PH, Hammerschmidt CR.
(2003). Determination of mercury
complexation characteristics of
dissolved organic matter in natural
waters through "reducible Hg" titra-
tions. Environmental Science and
Technology, 37(15):3315-3322.
Langer CS, Fitzgerald WF, Visscher
PT, Vandal GM. (2001). Biogeochemi-
cal cycling of methylmercury at Barn
Island Salt Marsh, Stonington, CT,
USA. Wetlands Ecology and Manage-
ment, 9(4):295-310.
Rolfhus KR, Lamborg CH, Fitzgerald
WF, Balcom PH. (2003). Evidence
for enhanced mercury reactivity in
response to estuarine mixing./owr-
nal of Geophysical Research-Oceans,
108(C11):3353.
Rolfhus KR, Fitzgerald WF. (2001).
The evasion and spatial/temporal dis-
tribution of mercury species in Long
Island Sound, CT-NY Geochimica et
Cosmochimica Acta, 65 (3): 407-417.
Tseng CM, Balcom PH, Lamborg CH,
Fitzgerald WF. (2003). Dissolved
elemental mercury investigations in
Long Island Sound using on-line Au
amalgamation-flow injection analysis.
Environmental Science and Technol-
ogy, 37(6): 1183-1188.
Vandal GM, Fitzgerald WF, Rolfhus
KR, Lamborg CH, Langer CS, Balcom
PH. (2002). Sources and cycling of
mercury and methylmercury in Long
Island Sound (Project No. CWF-326-
R). Final Report to the Connecticut
Department of Environmental Protec-
tion, Long Island Sound Program.
Balcom PH, Fitzgerald WF, Vandal
GM, Lamborg CH, Rolfhus KR, Langer
CS, Hammerschmidt CH. Mercury
sources and cycling in the Connecti-
cut River and Long Island Sound.
Marine Chemistry (Submitted).
Hammerschmidt CR, Fitzgerald WF,
Lamborg CH, Balcom PH, Visscher PT.
Biogeochemistry of methylmercury
in sediments of Long Island Sound.
Marine Chemistry (Submitted).
Lamborg CH, Fitzgerald WF, Skoog
A, Visscher PT. The abundance and
source of mercury-binding organic
ligands in Long Island Sound. Marine
Chemistry (Submitted).
mental Microbiology, 67:51-58.
Benoit JM, Gilmour CC, Mason RP
(2001). The influence of sulfide on
solid-phase mercury bioavailability
for methylation by pure cultures of
Desulfobulbous proprionicus (1PR3).
Environmental Science and Technol-
ogy, 35:127-132.
Benoit JM, Mason RP, Gilmour CC,
Aiken GR. (2001). Constants for
mercury binding by dissolved organic
matter isolates from the Florida Ever-
glades. Geochimica et Cosmochimica
Acta, 65:4445-4451.
Benoit JM, Gilmour CC, Heyes A,
Mason RP, Miller CL. (2003). Geo-
chemical and biological controls over
mercury production and degradation
in aquatic systems, pp. 262-297. In:
Y Cai and OC Brouds [eds.], Biogeo-
chemistry of Environmentally Impor-
tant Trace Elements, ACS Symposium
Series 835, ACS, Washington, DC.
Heyes A, Miller C, Mason RP (2004).
Mercury and methylmercury in the
Hudson River sediment: impact of
resuspension on partitioning and
methylation. Marine Chemistry,
90:75-89.
Mason RP, Benoit JM. (2003). Organo-
mercury compounds in the environ-
ment, pp. 57-99. In: P Craig [ed.],
Organometallics in the Environ-
ment, John Wiley & Sons, New York.
Benoit JM, Mason RP, Gilmour CC.
(2001). Aspects of the bioavailability
of mercury for methylation in pure
cultures of Desulfobulbous propri-
onicus (1PR3). Applied and Environ-
Amyot M, AuClair JC, Poissant L.
(2001). In situ high temporal resolu-
tion analysis of elemental mercury
in natural water. Analytica Chimica
Acta, 447:153-159.
Ekstrom EB, Morel FMM, Benoit JM.
(2003). Mercury Methylation Inde-
pendent of the Acetyl-Coenzyme A
Pathway in Sulfate-Reducing Bacteria.
Applied and Environmental Microbi-
ology, 69(9):54l4-5422.
-------
Kraepiel AL, KeUer K, Chin HB,
Malcolm EG, Morel FMM. (2003).
Sources and Variations of Mercury in
Tuna. Environmental Science and
Technology, 37:5551-5558.
LaLonde JD, Amyot M, Kraepiel AML,
and Morel FMM. (2001). Photooxida-
tion of Hg(0) in Artificial and Natural
Waters. Environmental Science and
Technology, 35:1367-1372.
LaLonde JD, Amyot M, Orvoine J,
Morel FMM, AuClair JC, Ariya PA.
(2004). Photoinduced Oxidation of
HgO(aq) in the Waters from the St.
Lawrence Estuary. Environmental
Science and Technology, 38:508-514.
Engle ME. (2003). The mobility of
mercury in epithermal mercury
deposits in an arid environment.
Masters Thesis, University of Nevada,
Reno, 155 p.
Engle MA, Gustin MS, Zhang H.
(2001). Quantifying natural source
mercury emissions from the Ivanhoe
Mining District, north-central Nevada,
USA. Atmospheric Environment,
35:3987-3997.
Engle MA, Gustin MS. (2002). Scaling
up atmospheric mercury emissions
from three naturally enriched areas:
Flowery Peak, Nevada, Peavine Peak,
Nevada and Long Valley Caldera, Cali-
fornia. Science of the Total Environ-
ment, 290(1-3) =91-104.
Gustin MS. (2003). Are mercury emis-
sions from geologic sources signifi-
cant?: A status report. Science of the
Total Environment, 304:153-167.
Gustin MS, Rasmussen P, Edwards
G, SchroederW, KempJ. (1999).
Application of a laboratory gas
exchange chamber for assessment of
in situ mercury emissions. Journal of
Geophysical Research - Atmospheres,
104(D17):21, 873-78.
Gustin MS, Lindberg SE, Austin K,
Coolbaugh M, Vette A, Zhang H.
(2000). Assessing the contribution of
natural sources to regional atmo-
spheric mercury budgets. The Science
of the Total Environment, 259:61-72.
Gustin MS, Biester H, Kim C. (2002).
Investigation of light enhanced
emission of mercury from naturally
enriched substrate. Atmospheric
Environment, 36:3241-3254.
Gustin MS, Coolbaugh M, Engle M,
Fitzgerald B, Keislar R, Lindberg
S, Nacht D, Quashnick J, Rytuba J,
Sladek C, Zhang H, Zehner R. (2003).
Atmospheric mercury emissions from
mine wastes and surrounding geo-
logically enriched terranes. Environ-
mental Geology, 43:339-351.
Lindberg SE, Zhang H, Vette AF,
Gustin MS, Barnette MO, Kuiken
T (2002). Dynamic flux chamber
measurement of gaseous mercury
emission fluxes over soils: Effect of
flushing flow rate and verification of
a two-resistance exchange interface
simulation model. Atmospheric Envi-
ronment, 36:847-859-
Sladek C. (2001). Investigation of
methods for determining mercury
speciation and mobility in substrate.
Masters Thesis, University of Nevada,
Reno.
Sladek C, Gustin MS, Biester H, Kim
C. (2002). Application of three meth-
ods for determining mercury specia-
tion in mine waste. Geochemistry:
Exploration, Environment, Analysis,
2 (4): 369-3 75.
Sladek C, Gustin MS. (2003). Evalua-
tion of sequential extraction methods
for determination of mercury spe-
ciation and mobility in mine waste.
Applied Geochemistry, 18(4):567-576.
Zehner RE, Gustin MS. (2002). Esti-
mation of mercury vapor flux from
natural substrate in Nevada. Envi-
ronmental Science and Technology
36:4039-4045.
Zhang H, Lindberg SE, Barnette MO,
Vette AF, Gustin MS. (2002). Simula-
tion of gaseous mercury emissions
from soils measured with a dynamic
flux chambers using a two-resistance
model. Atmospheric Environment,
36:835-846.
Zhang H, Lindberg SE, Gustin MS, Xu
X. (2002). Towards a better under-
standing of mercury emissions from
soils, in Biogeochemistry of Environ-
mentally Important Trace Elements.
Cai Y and Braids OC, eds. American
Chemical Society, Washington DC,
246-261.
Benesch, JA. (2002). Assessing the
role of deciduous forests in the
biogeochemical cycling of mercury.
Masters Thesis, University of Nevada,
Reno, 90p.
Ericksen JA, Gustin MS, Schor-
ran DE, Johnson DW, Lindberg SE,
ColemanJS. (2003). Accumulation
of atmospheric mercury in forest
foliage. Atmospheric Environment,
37(12):1613-1622.
Ericksen JA, Gustin MS. (2004). Foliar
exchange of mercury as function
of soil and air concentration. The
Science of the Total Environment,
324:271-279.
Frescholtz TF. (2002). Assessing the
role of vegetation as sources and
sinks of atmospheric mercury using
Quaking Aspen. Masters Thesis, Uni-
versity of Nevada, Reno, 67p.
Frescholtz TF, Gustin MS, Schorran
DE, Fernandez GC. (2003). Assess-
ing the source of mercury in foliar
tissue of quaking aspen. Environ-
mental Toxicology and Chemistry,
22(9):2114-2119.
Frescholtz TF Gustin MS. (2004).
Soil and foliar mercury emission as a
function of soil concentration. Water,
Air, Soil Pollution, 155:223-237.
-------
Gustin MS, Ericksen JA, Schorran DE,
Johnson D^ Lindberg SE, Coleman
JS. (2004). Application of controlled
mesocosm for understanding mer-
cury plant-soil-air exchange. Environ-
mental Science and Technology, 38:
6044-6050.
Gustin MS, Stamenkovic J. (2005).
Effect of watering and soil moisture
on mercury emissions from soils.
Biogeochemistry, 76(2):215-232.
Johnson D, Benesch JA, Gustin MS,
Schorran DS, Lindberg SE, Coleman
JS. (2003). Experimental evidence
against diffusion control of Hg eva-
sion from soils. Science of the Total
Environment, 304(1-3)=175-184.
-------
-------
vvEPA
United States
Environmental Protection
Agency
Office of Research and Development (8101R)
Washington, DC 20460
EPA/600/S-06/013
January 2006
www.epa.gov
/ XV Recycled/Recyclable
7~\
------- |