Methods for Deriving Pesticide Aquatic Life Criteria - Reviews of Environmental Contamination and Toxicology - page 45

quality targets, in terms of toxicant concentration in suspended particulate matter.
To protect German fisheries, BCF values are also used to derive water quality targets
for fish from pesticide MPR (maximum permissible residue) values.
The USEPA guidelines (1985) require collection of bioaccumulation data only
when residues are known to be of toxicological concern. Physical-chemical data,
such as volatility, solubility, and degradability, are required for evaluation of toxicity
data. BCFs and BAFs are used to derive the FRV.
For development of a full guideline, Canada (CCME 1999) requires collection
of environmental fate data. Specifically, information must be available on the
mobility of the substance and its final disposition, abiotic and biotic transforma-
tions that occur during transport and after deposition, the final chemical form of the
substance, and on the persistence of the substance in water, sediment, and biota.
The Danish methodology (Samsoe-Petersen and Pedersen 1995) does not clearly
specify what kinds of physical-chemical data must be collected, but criteria deriva-
tion documents indicate consideration of a wide-range data. Such data include CAS
number, empirical formula, molecular weight, water solubility, K H , BCF, and K ow , as
well as biodegradability data. Bioaccumulation data are used to determine the size
of the AF to be applied. Biodegradation data are used to determine whether criteria
ought to be derived for the parent chemical or for a stable, toxic metabolite. If little
is known about degradation products, then AFs will reflect this uncertainty.
According to EU guidance (Bro-Rasmussen et al. 1994), physical-chemical data
requirements are very general, and simply state that “a summary of the main chemi-
cal and physicochemical characteristics” must be collected. For criteria derivation,
bioaccumulative potential and persistence can affect the size of the applied AF.
Also, K ow values may be used to derive QSAR estimates of toxicity, when toxicity
data are lacking. For assessment of secondary poisoning potential, the EU risk
assessment TGD (ECB 2003) utilizes K ow values, adsorption data, hydrolysis and
other degradation data, and molecular weight.
Spanish guidelines (Lepper 2002) require collection of physical-chemical data
that may have some bearing on the toxicity of the substance. These include specia-
tion, toxicokinetic properties, and relationships between toxicity and water quality
parameters. The UK (Zabel and Cole 1999) and South African (Roux et al. 1996)
guidelines do not specify particular uses for physical-chemical data in criteria
derivation.
Physical-chemical data are used by various methodologies to improve interpre-
tation of ecotoxicity data and to determine whether water quality criteria are set at
levels that could potentially harm nonaquatic species (including humans). Without
adequate physical-chemical data, it would not be possible to adequately assess
potential effects of chemicals. If explicit details, regarding the collection of physi-
cal-chemical data, are available, they are important contributors to criteria deriva-
tion methodology.
QSARs describe and model mathematical relationships between a chemical's
structure and its toxicity. According to Jaworska et al. (2003) QSARs are simplified
mathematic representations of complex chemical-biological interactions. They
are most commonly developed by regression analysis, neural nets, or classification