Environmental Engineering Reference
In-Depth Information
have taken mercury from its long-term storage in geologic
reservoirs and transferred it to the atmosphere. While the
lifetime of mercury in the atmosphere is about a year, cycling
between the atmosphere and the land and ocean surface
effectively lengthens the amount of time mercury circulates
in the environment (Selin et al., 2008). Mason and Sheu
(2002) estimate that it will take about 10,000 years for mer-
cury to return to long-term sedimentary storage. Until then,
this historical mercury continues to be released again to the
atmosphere from land and ocean surfaces. The magnitude
of these fl uxes and the processes controlling them are not
well constrained, but measurements have shown that fl uxes
can depend on temperature (Kim et al., 1995; Lindberg
et al., 1995), solar radiation (Carpi and Lindberg, 1998; Gustin
et al., 2002), or soil moisture (Gustin and Stamenkovic,
2005). Isotopic fi eld studies have also shown that mercury
recently deposited to ecosystems is more available for emis-
sion (Graydon et al., 2006; Hintelmann et al., 2002).
from the Southern Hemisphere at Cape Point (Baker
et al., 2002), from Antarctica (Ebinghaus et al., 2002) ,and
from ocean cruises (Lamborg et al., 1999; Laurier et al.,
2003; Temme et al., 2003) have generally reported lower
concentrations than in the Northern Hemisphere, which
indicates that most mercury sources are in the Northern
Hemisphere. The interhemispheric gradient of Hg(0), in
combination with the balance of sources between the
Northern and Southern hemispheres, provides constraints
on the atmospheric lifetime of Hg(0), as the interhemi-
spheric exchange time for air is about a year (Jacob et al.,
1987). The longer the atmospheric lifetime of mercury, the
smaller the interhemispheric gradient is expected to be,
since mercury would have an opportunity to mix between
the hemispheres before it is removed from the atmosphere.
Seasonal variation of Hg(0) is consistent at most sites in
the Northern Hemisphere (Kellerhals et al., 2003; Selin et al.,
2007), with a maximum in winter and minimum in late
summer. This behavior has been measured, for example,
at a network of stations in Canada (CAMNet) (Kellerhals
et al., 2003), and the seasonal variation is statistically sig-
nifi cant for available sites in the northern midlatitudes
(Selin et al., 2007). This suggests a photochemical sink of
Hg(0), which is oxidation to Hg(II). However, the dominant
atmospheric oxidant of Hg(0) is at present uncertain, as dis-
cussed below. Seasonal variation of Hg(0) in the Southern
Hemisphere is more puzzling. Hg(0) measurements at Cape
Point observatory in South Africa (Slemr et al., 2008) are
maximum in summer and minimum in winter, opposite
what would be expected from photochemical oxidation in
this hemisphere. Slemr et al. suggest, based on the Cape
Point data, that the seasonal behavior of mercury is driven
by its sources rather than its sinks. Obrist (2007) reviewed
the seasonal data from mercury measurements and sug-
gested that the spring and summer declines in atmospheric
mercury could be driven by the uptake of mercury by
vegetation rather than its oxidation sink. This is a subject
of continuing scientifi c investigation and discussion.
Measurements of Hg(II) and Hg(P) are fewer, though the
number of measurements of these species are increasing.
Hg(II) is measured in the atmosphere as reactive gaseous
mercury (RGM) using an operationally defi ned method.
Typically, Hg(II) measurements are made by collecting the
species on KCl-coated denuders and reducing it to Hg(0)
before measurement (Landis et al., 2002).
RGM has been shown to vary diurnally in the atmo-
sphere, with a peak around midday and at a minimum at
night. Jaffe et al. (2005) measured RGM at Okinawa, Japan,
and found that levels peaked in the afternoon and were
at a minimum at night. RGM at this site did not correlate
with Hg(0), indicating that RGM results here from oxida-
tion of Hg(0) and is not directly emitted from anthropo-
genic sources. It is thought that this refl ects production of
Hg(II) via oxidation of Hg(0). Laurier and Mason (2007)
measured RGM at two sites in Maryland and on an Atlantic
cruise and reported diurnal variation in RGM consistent
Forms and Distribution of Mercury
in the Atmosphere
In the atmosphere, mercury exists in three major forms.
The majority of mercury in the atmosphere is in the form
of gaseous, elemental mercury, which is termed Hg(0)
(Mason and Sheu, 2002; Schroeder and Munthe, 1998).
Typical concentrations of Hg(0) in the atmosphere are about
1.6 ng m
-3
at the surface. Hg(0) has a Henry's law constant
of 0.11 M atm
-1
at 298 K (Lin and Pehkonen, 1999), which
makes it less soluble than other forms of atmospheric Hg,
and therefore more likely to be present in the gas phase in
the atmosphere. Hg(0) has a lifetime of between 0.5 and
2 years in the global atmosphere, which means that it has
the ability to transport globally. The two other forms of
mercury are both shorter-lived. Divalent mercury [Hg(II)] is
more soluble than Hg(0), which means that it is more likely
to deposit to the surface through wet deposition and also
dry deposition (which is enhanced for more soluble spe-
cies). Because it deposits so readily, its lifetime in the atmo-
sphere is shorter than that of Hg(0)—on the order of days
to weeks. Typical concentrations of atmospheric Hg(II) vary
between 1 and 100 pg m
-3
. It is thought that most divalent
mercury in the atmosphere is in the form of HgCl
2
(Lin
et al., 2006). Mercury can also be associated with atmo-
spheric particulate matter, termed Hg(P). Atmospheric con-
centrations of Hg(P) are of the same order of magnitude as
Hg(II). Depending on particle size, it will also be deposited
to the surface through wet and dry deposition on timescales
of days to weeks. Mercury can also exist in organic forms
in the atmosphere (e.g., methylmercury), though concen-
trations are more than an order of magnitude smaller than
inorganic forms (Hammerschmidt et al., 2007).
Measurements of Hg(0) in the atmosphere are available
at a number of land-based stations and from some ocean
cruises and aircraft missions. Most land-based measure-
ments are from the northern midlatitudes. Measurements
