Modelling of Taste-Taint in Fish ... 49be non-ionic and not metabolised. In addtion,in the fugacity models, the bio-
transformation parameter was ignored. This means the output tends to over-
predict, leading to some errors (Weisbrod et al., 2007).
Food-web models have been synthesized to predict chemicals
concentrations in aquatic organism based on BCF, BAF or BSAF (Bio-
sediment accumulation) (Arnot and Gobas, 2004; Hendriks et al., 2001;
Morrison et al., 1997; Sharpe and Mackay, 2000). They are increasingly used
for chemical management programs (Weisbrod et al., 2007). However, the
limitations include restricted suitability for chemicals such as Persistent
Organic Pollutants (POPs) which have greater KOW, and slow metabolic
potential and dietary intake (Weisbrod et al., 2007). The reliability of the food-
web models is yet to be confirmed for all chemical classes, including ionizing
substances because of the lack of field data. There has to be specific data, such
as sediment water column disequilibria of the chemicals, dissolved and
particulate organic carbon concentration in water from the food-web, for
model operation (Weisbrod et al., 2007).
However, the weaknesses of this type of model are contaminant
concentrations can vary widely among individual organism in the
environment; and, site-specific data cause some discrepancies. For an
example, the use of population mean for highly variable elements can result in
unreliable predictions. It is generally accepted that food-web models are
difficult to use as these require site-specific ecosystem conditions and input
data for specific food-webs (Weisbrod et al., 2007).
Physiologically-based pharmacokinetic models have some disadvantages:
they are (reportedly) highly complex due to the number of equations being
used, require a large number of data for model validation, and; have data-
acquisition problems (Balant and Gex-fabry, 1990).
Therefore, it requires a great deal of knowledge to identify the specific
compartment kinetically (Wen et al., 1999). Understandably, two-
compartment and physiologically-based pharmacological models appear to be
more suitable models to describe the uptake and elimination of taste-taint
chemicals than bio-concentration or bio-magnification models. However the
mathematically derived physiological models have more parameters than
single compartment models.
This lack of reliable data to model the uptake and elimination kinetics of
taste-taint chemicals has led to a disregard of the physiologically-based
pharmacological models, and instead use of the simpler one-compartment
models. In addition, the unavailability of more descriptive data sets has