ABSTRACT
The purposes of mesocosm studies have been identified as to negate a presumption of unreasonable risk, and to define the intensity and duration of adverse effects of pesticides in aquatic systems. A variety of approaches were presented and have been historically utilized to design mesocosm studies for pesticide risk assessments. In order to develop information for a risk assessment that will stand the test of time, both type I errors and type II errors must be minimized in mesocosm studies. Based on the information presented in this series of six papers, it was apparent that there is little probability that mesocosm studies as currently designed with ANOVA approaches, t-test approaches, or regression approaches will find effects that may be manifested in the field. Perhaps the guidance document can be treated as merely guidance, and more scientifically defensible and optimal experimental designs will produce more accurate data, conclusions, and regulatory decisions.
