48 Priyantha I. Hathurusingha and Kenneth R. Davey
The model simply explains how the uptake and elimination kinetics
impact on accumulation of the taste-taint chemicals. Further, theoretical
predictions were not validated with any empirical or independent data.
Martin et al. (1990) studied the distribution and elimination of MIB using
a two-compartment model. In their research, a known concentration of MIB
was injected into channel cat fish, and its concentration then measured in the
plasma and selected organs at various time intervals. The decrease of MIB
concentration in plasma was found to be consistent with the projected results
of the two-compartment model. They showed that the distribution of MIB in
different tissues depended on lipid concentrations. Similar findings with regard
to this kinetic two-compartment model have been reported by Schlenk et al.
(1999).
One of the major problems encountered in the literature for reporting of
predictive models is the lack of data for kinetics of GSM and MIB. However,
there are some reported kinetic studies for similar chemical contaminants
(Howgate, 2004). Therefore, it is thought these data can be used as a guide to
develop an adequate model for both GSM and MIB.
INADEQUACIES OF EXISTING MODELS
With any model, the model predictions need to agree with observed data
(Wen et al., 1999). It is important nevertheless to emphasise that there can be
uncertainties in any model.
A number of bio-magnification (BCF = bio-concentration or BAF = bio-
accumulation) models have been derived (Clark et al., 1990; Neely et al.,
1974), but these do not adequately predict the chemical concentration in fish-
flesh.
Possible reasons for this include: the lack of kinetic data, biological
variability, too few factors considered, errors in experimental and field
samplings, and; variation in applied methodologies (Sijm et al., 1992).
Although a life-cycle bio-magnification model is more descriptive and
comprehensive, a major drawback is the processes have produced inconsistent
results leading to inaccurate predictions (Sijm et al., 1992). That BCF models
need to be integrated with bio-magnification and bio-transformation processes
have been ignored in present models.
More research with BCF models needs to be done with existing data to
evaluate the reliability of these processes (Weisbrod et al., 2007). Other
shortcomings of BCF models are the taste-taint chemicals of interest need to