Methods for Deriving Pesticide Aquatic Life Criteria - Reviews of Environmental Contamination and Toxicology - page 30

The OECD methodology (1995) recognizes three levels of aquatic effects
assessment and derives MTCs for each level. An initial, or primary, assessment
is based on laboratory toxicity data from only one or two representatives of
primary producers, primary consumers and predators. An intermediate or refined
assessment is based on results of chronic or semi-chronic laboratory tests. Field
or semi-field studies are used for comprehensive assessments. MTCs derived by
the OECD methodology (1995) are used to set environmental quality objectives.
However, MTCs have different levels of reliability, depending on how they
are derived. An MTC derived from quantitative structure activity relationships
(QSARs) has lower status than one derived from acute toxicity tests; one derived
from acute toxicity tests has lower status than one derived from chronic tests; an
MTC derived from a reliable, representative field test has the highest status.
Lower status MTCs are used for setting priorities, rather than for setting enforce-
able objectives.
In Australia and New Zealand, TVs of low, medium, and high reliability are
derived (ANZECC and ARMCANZ 2000). The reliability rating is dependent on
how much data support the value. Only medium and high reliability values are used
as final guideline TVs. Low reliability values, which are similar to USEPA advisory
concentrations, are interim figures, which, if exceeded, indicate the need for further
data collection. High and medium reliability TVs are not pass/fail levels. If
exceeded, a TV is reevaluated and refined in a site-specific assessment. Further
regulatory action usually occurs only if the site-specific value is exceeded (although
risk managers have the option of using the more conservative, national TVs as
enforceable values).
By whatever name, all of the values discussed (including those not currently
used in setting water quality standards or objectives) represent efforts to esti-
mate concentrations of chemicals below which beneficial uses can reliably be
regarded as protected. When data are limited, numeric criteria of low site-specif-
icity and high uncertainty can be derived. As more data become available, crite-
ria can be refined for better site-specificity and greater certainty (Di Toro 2003;
La Point et al. 2003).

5 Protection and Confidence

Aquatic life water quality criteria are intended to protect aquatic life from exposure
to toxic substances. But what really is the goal? Is it overall ecosystem protection,
or protection of each individual in the ecosystem? And, how certain can one be of
meeting that goal? This section discusses how aquatic life protection goals are stated
in various derivation methodologies, and how those goals have to be approached,
considering the need to extrapolate ecosystem effects from single-species toxicity
data. It is also important to know that, with a quantified level of certainty, criteria
are achieving the intended level of protection.