Mercury in the Marine Environment - Mercury in the Environment - page 182

this the case, the historical increase in mercury emissions
due to anthropogenic activity would be unlikely to have
resulted in a substantial increase in MMHg levels in marine
fi sh during the past 200 years, although such an asser-
tion is also debatable (Fitzgerald et al., 2007; Sunderland
and Mason, 2007). Alternatively, if MMHg in the oceans
is produced primarily in intermediate waters or coastal
sediments, then the increased atmospheric load of mercury
from anthropogenic emissions would be much more likely
to have increased MMHg production and bioaccumulation
in marine organisms, as observed in freshwater systems.

2002b). The vertical and lateral movement of the oxic-
suboxic boundary caused by tidal pumping and the alter-
nating patterns of wetting and drying may also contribute
to the high MMHg production rates in coastal and salt-
marsh sediments compared to nearshore or shelf sediments
(Catallo, 1999; Tseng et al., 2001; Taillefert et al., 2007).
There are concerns of increased mercury methylation
and bioaccumulation related to efforts to restore wetlands
following their degradation over the past two centuries.
Because wetlands are hot spots for MMHg production, wet-
land restoration may exacerbate existing mercury problems
of biomagnifi cation and toxicity in aquatic food chains.
However, not all wetlands produce or release copious
amounts of MMHg. MMHg concentrations in sediments of
a mangrove in Brazil were only slightly greater than those
in an adjacent estuary (Kehrig et al., 2003), and a compari-
son found that MMHg levels were lower in Venice Lagoon
in Italy than in Lavaca Bay in Texas, despite roughly 30%
of Venice Lagoon's margin consisting of various types of
intertidal wetlands, as compared with less than 5% for
Lavaca Bay (Bloom et al., 2004b).

COASTAL WETLANDS, SALT MARSHES, AND
MANGROVE SWAMPS

The high rates of nutrient and organic matter deposition and
cycling in wetlands drive high microbial respiration rates
in wetland sediments, including sulfate reduction linked
to MMHg production (Benoit et al., 2003; Compeau and
Bartha, 1985; Gilmour et al., 1992). It is, therefore, not sur-
prising that salt marshes and coastal wetlands are sites of
elevated MMHg production (Marvin-DiPasquale et al.,
2003; Choe et al., 2004; Canário et al., 2007b; Heim et al.,
2007; Hall et al., 2008; Mitchell and Gilmour, 2008).
Microbial methylation of mercury and MMHg concen-
trations in sediments are generally highest at or near the
oxic/anoxic interface (Choe et al., 2004; Covelli et al.,
2008; Hammerschmidt et al., 2004; Han et al., 2007; Hines
et al., 2000; King et al., 2001, Marvin-DiPasquale and Agee,
2003; Ouddane et al., 2008). Because of the abundance of
rooted macrophytes in wetlands that facilitate and medi-
ate the exchange of oxygen and other compounds in the
rhizosphere-ectorhizosphere zone (Grosse et al., 1996), the
size of this oxic-suboxic boundary layer in salt marshes and
other wetlands is much greater than in pelagic sediments.
These plants increase the volume of sediment potentially
involved in MMHg production per unit area relative to other
areas, and they may be responsible for the higher MMHg con-
centrations and Hg methylation rates commonly reported
in marsh sediments colonized by plants as compared with
nonvegetated areas (Choe et al., 2004; Marvin-DiPasquale et
al., 2003; Canário et al., 2007b; Heim et al., 2007; Windham-
Myers et al., 2009). Seasonal changes in MMHg production
and degradation in wetlands likely refl ect temporal changes
in temperature and inputs of organic carbon and nutrients
(Marvin-DiPasquale and Agee, 2003; Stoichev et al., 2004),
which can infl uence Hg(II) speciation and microbial respi-
ration rates, and changes in plant growth and physiology,
which infl uence microbial community composition and
activity (Hines et al., 1999).
Bioturbation and bioirrigation by microbenthos and
macrobenthos abundant in coastal wetlands are also
responsible for increasing the size of the oxic-anoxic
boundary layer and the continual reoxidation of surfi cial
sediments, resulting in heterogeneity in microbial popula-
tions and redox zonation, in addition to increasing rates of
iron reduction (Koretsky et al., 2005; Kostka et al., 2002a,

NEARSHORE SEDIMENTS

In situ production of MMHg in nearshore sediments is the
greatest source of MMHg to coastal waters, and studies of the
production, export, and biogeochemical cycling of MMHg
in surfi cial estuary and coastal sediments make nearshore
sediments the most widely studied source of MMHg in the
marine environment. Sediment concentrations of MMHg in
the solid phase, pore waters, and estimates of benthic fl uxes
of MMHg for various estuary and coastal areas are listed
in Table 10.3. Concentrations of MMHg in the majority of
coastal sediments range from 0.01 to 50 ng g 1 dry weight
(dw). MMHg in pore waters of coastal sediments range from
0.02 to 25 ng L 1 . MMHg typically constitutes less than 1.5%
of the total mercury pool in the solid phase of sediments,
but comprises a much larger portion in pore waters, where it
can be the dominant form of mercury present. Solid-phase
MMHg concentrations in sediments generally correlate to
mercury methylation potential (Benoit et al., 2003; Gilmour
et al., 1998 ; Heyes et al., 2004, 2006; Hines et al., 2006;
Sunderland et al, 2004), although the nature of this relation-
ship varies by site and season.
Measurements of MMHg benthic fl uxes generally quan-
tify either fl uxes due only to diffusion and bioirrigation
(using concentration gradients between pore waters and
overlying waters or with laboratory-based fl ux chambers
using sediment cores) or fl uxes due to the combination
of diffusive and advective processes (using in situ benthic
fl ux chambers). Because the diffusive fl uxes of MMHg out
of sediments that have been measured (0-34 ng m 2 d 1 )
are generally much smaller than advective fl uxes (-330
to 2370 ng m -2 d 1 ), diffusive fl uxes usually represent a

relatively minor portion of MMHg benthic fl uxes from
coastal sediments (Choe et al., 2004; Covelli et al., 1999,
2008; Gill et al., 1999). However, this is unlikely to be true