Environmental Engineering Reference
In-Depth Information
Ecosystem Responses to Mercury
systems. While other benchmarks can and have been devel-
oped, their usefulness depends on the appropriate regulators
accepting them for the risk assessment for mercury.
Whereas HRA concentrates on only one species (humans),
ERA focuses not only on individuals and their popula-
tions, but also on ecosystems, since effects at the species
level can have cascading effects on the structure and func-
tion of the system. Ecosystem approaches to evaluating the
risk of mercury involve assessing mercury deposition in
airsheds and watersheds (both terrestrial and aquatic com-
ponents), monitoring and evaluating trends in sediment
and water indicators, monitoring and evaluating trends in
aquatic biota, and examining wildlife indicators (Harris et al.,
2006). All these require evaluating the fate and effects of
mercury within components of the ecosystem.
Hazard Quotients
HQs have been used to evaluate the ecologic risk from
exposure to mercury (see Table 12.3). One problem, how-
ever, is that data on toxicity levels for mercury are avail-
able for only a very small percentage of the species, even for
birds and mammals in which much of the work has been
done. Thus, assessors are forced to extrapolate not only
from laboratory studies, but from one species to another,
not to mention adjusting for potential differences due to
subspecies differences. Very rarely are such toxicity data for
mercury validated in the fi eld.
HQs are a type of analysis used for risk characterization
in screening assessments. The HQ is derived by dividing the
ambient exposure concentration by a toxicologically effec-
tive concentration. For wildlife, doses are used in place of
concentrations. If the HQ is greater than 1, then the chemi-
cal is worthy of concern and further risk evaluations. Large
quotient values suggest large effects or greater uncertainty
concerning the end point. When more chemicals than just
mercury are involved, the sum of the toxic units are added
together to produce an index of toxicity (Suter et al., 2000),
sometimes referred to as a hazard index. However, there is lit-
tle empirical evidence to support the additivity of HQs. HQ's
less than 1 do not allow a substance to be ignored completely.
Additional Issues for Risk Evaluations
for Mercury
Monitoring, Biomonitoring, and Surveillance
Environmental monitoring examines the status and
trends in indicators to determine whether the environ-
ment is improving or degrading, and as such, have been
used extensively by managers (Suter, 2001). Both human
and ecological risk assessment for mercury depend on
biologic monitoring of species or tissues. To assess the
impact of elevated levels of mercury in media (usually
inorganic mercury) or biota (such as methylmercury in
fi sh), biologic monitoring is conducted, usually with reg-
ular frequency. For humans, the tissue of choice varies
with the type of exposure. Elemental and inorganic mer-
cury exposure is usually monitored with urine analysis for
total mercury. Since methylmercury is excreted mainly in
the feces, urine monitoring does not provide consistent
results, and hair or blood levels are used. The EPA con-
siders hair levels below 10 ppm and blood levels below
5 µg/L to be unlikely to be associated with adverse effects
(USEPA, 2006). Such monitoring for mercury is recom-
mended for people who have high levels of consumption
of fi sh or shellfi sh (Hightower and Moore, 2002), in com-
munities of Native Americans who eat high levels of sub-
sistence foods (Harris and Harper, 1997; Harnly et al.,
1997; Wheatley and Wheatley, 2000; Burger and Goch-
feld, 2007), for anyone consuming large quantities of fi sh
(e.g. recreationists, high-end commercial fi sh consumers),
and for people living near gold-mining operations (Rojas
et al., 2006) and chlor-alkali plants (Ullrich et al., 2007).
Optimally, monitoring plans should be developed to
show spatial and temporal trends in mercury levels in
media (e.g. soil, sediment, water), biota and the food chain,
and humans (and their tissues). Increasingly organizations
and governments are designing monitoring plans to assess
mercury in the food chain, including humans (Muir et al.,
2005). Monitoring data can be used for ecological risk
assessment models at the local, regional, and global levels
(Barnthouse, 1992), as well as for constructing food-web
and population-based models. Burger et al. (2001b) used
Toxicity Reference Value
Another risk assessment approach is the use of the toxicity
reference value (TRV), which is based on terrestrial mam-
mal data. Since the TRV for mammals does not incorporate
all the uncertainty values of the RfD, it is less conservative
(Hung et al., 2007). As with HRA, ERA uncertainty factors
include interspecifi c to intraspecifi c comparisons, short to
long term, lowest (LOAEL) to NOAEL, and laboratory to
fi eld extrapolation. There is little standardization in how
these uncertainty factors are dealt with in ecological risk
assessment (Chapman et al., 1998).
Adverse Effects Levels and Screening Levels
In a practical sense, health professionals interested in risk
often use established human health guidelines to deter-
mine whether, for example, people should eat fi sh. Similarly
ecotoxicologists use a combination of NOAEL and LOAEL
to determine what levels of mercury in animals result in
adverse effects. From controlled laboratory experiments,
ecotoxicologists and risk assessors can determine screening
levels, which are effects threshold values based on toxic-
ity databases (Naito et al., 2006). There is a wide range of
benchmarks that can be used in the screening of mercury
and other chemicals (Suter, 1996).
