Associate professor, University of Florida, Gainesville, Fla.
Environmental engineer, Water and Air Research, Inc., Gainesville, Fla.
Aquatic biologist, U.S. Environmental Protection Agency, Environmental Monitoring Systems Laboratory, Quality Assurance Div., Las Vegas, Nev.
Pages: 18 Published: Jan 1981
This paper presents methodology for the analysis of chronic toxicity test data. Emphasis is given to the determination of the lowest tested concentration whose mean response differs significantly from the control. In particular, the analysis of variance (ANOVA) approach for analyzing chronic toxicity test data is discussed in some detail.
The paper presents four experimental designs that have different experimental units, that is, the units on which the observations or measurements are made. The first type of experimental design is a completely randomized design in which the aquaria are the experimental units. The second is a nested design, with aquaria nested in concentration and fish nested in aquaria. The third is also a nested design, in which aquaria are nested in concentrations, fry chambers are nested in aquaria, and fish are nested in chambers. The fourth is a two-factor design in which sex is crossed with concentration. The final section of the paper is a general discussion of the utility of the ANOVA approach, its limitations, and the identification of some areas that still need research.
statistics, analysis of variance (ANOVA), chronic toxicity tests, fathead minnows, channel catfish, elemental phosphorus, nitroglycerin, aquatic toxicology, hazard assessment
Paper ID: STP34168S