R. Costanza and D. Mauriello
The group examined the case studies with regard to their use or nonuse of models. Each was evaluated according to the answers to these questions:
Were models used?
If models were not used, would they have improved the assessment?
If models were used, could their use have been improved?
The group also considered some general issues regarding the use of mathematical models in risk assessment and risk management.
Use of Models in the Case Studies
Models were not used in the hazard identification phase. They were used to predict the rate of leaching of TBT into the water from ships painted with antifouling paint. The decision to ban the use of the paints in Virginia was based only on hazard assessment. Such a decision might not have been made if the vulnerable organisms had not included commercially valued species.
This case study described a rigorous approach to hazard identification and exposure-response assessment. Models were extensively used in determinations of the sensitivity of end- point species to pesticide exposure. The case study paper pointed out that little basic knowledge is available on the overall ecology of agroecosystems and that this would be a fertile subject for future modeling efforts. The discussion group agreed that larger-scale models are required to deal with geographic variability and to guide future research in pesticide ecotoxicology.