Environmental Engineering Reference
In-Depth Information
Thus, benchmark doses need to incorporate some level of
exposure uncertainty, which is sometimes accomplished by
adding a safety factor of 10 (Gochfeld and Burger, 2007).
Several methods have been used to evaluate potential
harm, including benchmark doses, no observed effect or
no adverse effect levels (NOELs and NOAELs), and lowest
observed adverse effect levels (LOAELs). These in turn can
be used to calculate RfDs, MRLs, or ADIs used for human
health guidelines to calculate HQs. NOELs, NOAELs, and
LOAELs, mainly rely on animal models, whereby different
doses of a chemical, such as mercury, are given to labora-
tory animals to determine the highest dose that produces
no detectable effect of any type (NOEL) or no adverse effect
(NOAEL). These are the highest nonzero (or noncontrol)
dose at which there is no measurable or detectable effect.
In many studies, even the lowest dose tried may produce
an effect, thus there is no NOEL or NOAEL, and the lowest
dose becomes the LOAEL, the lowest level at which adverse
effects were detected. Although the NOAEL may be a “safe”
dose, the LOAEL obviously is not. If an adverse response is
related directly to dose, it is sometimes diffi cult to deter-
mine the dose at which the chemical is “safe”. A threshold
is presumed to exist between the NOAEL and the LOAEL.
Thus, toxicologists use safety factors (also called uncer-
tainty factors [UFs]) to account for uncertainties in the tests
species differences, and individual susceptibilities.
concentrations of contaminants in tissues (e.g. liver,
muscle, heart, feathers, hair) that are associated with
adverse effects in laboratory studies and have been found
to result in similar adverse effects in wild animals (see
Burger, 1997a; Burger and Gochfeld, 1997b). Developing
these screening levels requires controlled laboratory exper-
iments with a particular toxicant, such as mercury, and val-
idation under fi eld conditions. Such validation can include
either dosing animals in the wild to achieve certain tissue
levels and observing adverse effects, or observing adverse
effects and collecting tissues to determine levels.
Total Maximum Daily Load
Another important determination that affects risk of a con-
taminant is the total maximum daily load (TMDL) that is
allowed by the U.S. Clean Water Act. A TMDL is a calcula-
tion of the maximum amount of a pollutant that a body
of water can receive and still meet water-quality standards.
The USEPA has set water quality criteria based on methyl-
mercury concentrations in fi sh at 0.3 µg/g (0.3 ppm; USEPA,
2001c), which has been adopted as a mercury freshwater
quality standard of 0.3 ppm methylmercury in fi sh for most
states. This was set to recognize the important fact that lev-
els of mercury in water do not relate directly to the levels of
mercury in fi sh. And, it is the levels of mercury in fi sh that
are the important end point in terms of mercury risk. Some
states have more stringent standards. Maine and Minnesota
have water-quality standards of 0.2 µg/g. When fi sh mer-
cury levels exceed the water-quality standard, the Clean
Water Act requires the calculation of how much mercury
loading must be reduced.
Hazard Quotient Risk Characterization
One of the tools for ERA, which is used either as part of for-
mal ERA or alone with risk management, is the chemical-
specifi c HQ risk characterization. Chemical-specifi c HQ risk
characterization is a ratio of an exposure estimate to a toxic-
ity reference value or benchmark. An HQ less than a value of
1 indicates that adverse impacts to ecologic receptors are con-
sidered unlikely (USEPA, 1997, 2001b). Managers and others
fi nd HQs useful because they provide a risk number, which
allows comparisons between and among sites, between and
among species, and between and among chemicals. How-
ever, there are a number of uncertainties with HQs that can
be masked by the apparent precision of having a number.
Another diffi culty with chemical-specifi c HQs is that they
deal with species, and not with the many effects that subtle
changes may produce at the population, community, and
ecosystem level (Tannenbaum et al., 2003; Sorensen et al.,
2004). Other methods have also been used, including com-
pensatory restoration, performance-based ecologic monitor-
ing, ecologic signifi cance criteria, and net environmental
benefi t analysis (Sorensen et al., 2004). These methods, how-
ever, are more applicable to communities and ecosystems
than to the analysis of a single chemical, such as mercury.
Risk Evaluations and Mercury
The general risk-evaluation methods discussed above are
used to evaluate the potential risk from mercury to humans
and ecosystems, including their component species, popu-
lations, and communities. Before addressing specifi c risk-
evaluation methods for mercury, there are two differences
between HRA and ERA that impact risk evaluations with
mercury: (1) HRA focuses on the health of individual people,
while ERA focuses on the health of populations (except in
the case of endangered species), and (2) HRA mainly focuses
on chemical toxicity and exposure, while ERA also focuses
on the ecologic characteristics of a site. The focus on chemical
toxicity resulted from emphasis on the Resource Conser-
vation and Recovery Act (RCRA) and the Comprehensive
Environmental Response, Compensation, and Liability
Act (CERCLA) and other environmental protection laws
that require chemical measurement of site conditions and
not ecologic conditions, and was a function of focusing on
humans (Sorensen et al., 2004). These two differences affect
how we evaluate the risk from mercury because it creates a
dichotomy between how risk is evaluated for humans and
ecosystems. Many people have advocated that this barrier
Screening Levels
Another, simpler, method of evaluating risk is to com-
pare levels of contaminants in biota with fi eld-validated
risk-based screening levels. Such screening levels are the
