Methods for Deriving Pesticide Aquatic Life Criteria - Reviews of Environmental Contamination and Toxicology - page 89

TTM = Σ ( C / WQG )

(2.15)

i i

where
TTM = total toxicity of the mixture
C i = concentration of the i th component of the mixture
WQG i = water quality guideline for that component
If the TTM exceeds 1.0, then the water quality guideline has been exceeded.
For more complex mixtures (>5 components), the Australia/New Zealand guide-
lines prefer the technique of DTA of effluents and receiving waters (equivalent to
whole effluent or ambient toxicity testing in the US). DTA is a good tool to deter-
mine if waters are able to support aquatic communities, but without follow-up
toxicity identification evaluation, it does not provide information to define what
chemical or chemicals, in a mixture, may cause any observed toxicity. Again, this
is a monitoring and compliance tool, rather than a way to address derivation of
criteria for mixtures.
The Dutch, Danish, OECD, South African, Canadian, EU, Spanish, and French
methodologies do not directly address mixtures or multiple stressors (RIVM 2001;
Samsoe-Petersen and Pedersen 1995; Roux et al. 1996; CCME 1999; Bro-Rasmussen
et al. 1994; Lepper 2002). The EU risk assessment TGD, and the German guidelines
include mixture effects on lists of uncertainties that may be addressed by using AFs,
but offers no further guidance on mixtures (ECB 2003; Irmer et al. 1995). In the UK,
combined EQSs may be derived for structurally similar substances with similar
modes of action (Zabel and Cole 1999). Lepper (2002) proposes a method for deriva-
tion of quality standards that does not explicitly account for the toxicity of mixtures,
but does utilize the AF method, described in the EU risk assessment TGD (ECB
2003), which includes factors for mixture effects.
Another way to address mixtures is provided by Könemann (1981). He devel-
oped a maximum toxicity index (MTI), which can be used to quantify toxicity of
mixtures of two or more chemicals that have either similar or independent action.
Könemann's model and the others discussed above, only work when trying to deter-
mine if toxicity is additive, more than additive, or less than additive. Rider and
LeBlanc (2005) have recently proposed a model that incorporates toxicokinetic
chemical interactions, as well as concentration and response addition. This model,
called the integrated addition and interaction (IAI) model, correctly predicted joint
toxicity of 30 ternary mixtures containing known interactive chemicals. Models
that assumed no interaction did not accurately predict joint toxicity.
Finally, SSDs offer a means of assessing mixture toxicity. Traas et al. (2002)
discussed how to determine a multisubstance potentially affected fraction (msPAF;
essentially the same as the HC p ) for cases of concentration addition (i.e., similar
mode of action), and response addition (i.e., different modes of action). The calcu-
lations are somewhat complicated, and the reader is referred to Traas et al. (2002)
for details. To calculate an overall msPAF, SSDs are first calculated individually for
each chemical in the mixture. The concentration addition calculation method is
then applied to groups with similar modes of action, yielding an msPAF for each
mode of action in the mixture. The individual PAF values that did not fit into any