Table 2. The Ratio of the 90 to the 10 Percentile of Bioaccumulation Factors (BAF) Predicted by the Gobas and Thomann Models for a Piscivorous Fish for a Log n-Octanol/Water Partition Coefficient Kow of 6.5 Using the Uncertainties of the Individual Input Parameters

Sensitivities ranged up to about 20 (indicating a decrease in BAF)

for the feeding preference of a benthic invertebrate on

phytoplankton in the Thomann model. The models were less

sensitive to changes in organism weight, temperature (input to

Gobas model only), and sediment organic carbon (input to Gobas

model only).

As described (Burkhard 1998), the uncertainty associated with the

model output depends not only upon its sensitivity to the input

parameters described but also on the uncertainty associated with the

individual input parameters. For example, a low sensitivity input

parameter can add considerable uncertainty to estimates of output if

the measurement uncertainty of the input is relatively large.

Uncertainty associated with the input parameters may result from

analytical errors in the measurement of the parameter, sampling that

is not representative of the population, or lack of sufficient

information about the parameter.

The dual influences of sensitivity and uncertainty in the input

parameters must be considered when evaluating the overall model

uncertainty. Monte Carlo simulations were performed for each input

parameter, using a plausible range of values for each input

parameter that reflect uncertainty, not variability (Burkhard 1988).

The ratios of the 90th and the 10th percentiles of the model output

derived from the simulations were compared among input

parameters. For both models, ΠSOCW, Kow, and feeding preferences

resulted in the largest range of simulated output values. Table 2

summarizes results from Burkhard (1998).

Table 2

The Ratio of the 90 to the 10 Percentile of Bioaccumulation

Factors (BAF) Predicted by the Gobas and Thomann Models for a

Piscivorous Fish for a Log n-Octanol/Water Partition Coefficient

Kow of 6.5 Using the Uncertainties of the Individual Input

Parameters

Ratio of 90th to 10th

Input Parameter

Percentile Predicted BAFlfd

Uncertainty

(C.V.), %

(assumed

Parameter

distribution)

Gobas Model

Thomann Model

Kow

0.2 (lognormal)

1.41

2.88

Temperature

10 (normal)

1.15

Not used

Sediment organic carbon

63 (normal)

1.00

Not used

ΠSOCW

15 (lognormal)

3.09

2.19

Weight of piscivorous fish

50 (normal)

1.05

1.00

Lipid content of piscivorous

fish

5 (normal)

1.12

1.10

Feeding preference of smelt

(fish)

40 (normal)

1.58

1.05

Chapter 5 Uncertainty in Tier IV Risk Assessments