The Effects of Age on Pimephales Promelas Survival in Acute Whole Effluent Toxicity Tests

Abstract

The effects of age on P. promelas survival in acute whole effluent toxicity (WET) tests were studied. Definitive 48-hour, static, non-renewal WET tests were conducted at a constant temperature of 25°C using hard synthetic fresh water as the dilution and control water with organisms 1 to 2, 3 to 4, 6 to 7, 11 to 12, and 13 to 14-days of age using the reference toxicants potassium chloride (KCl), sodium dodecyl sulfate (SDS), and a mixture of KCl and SDS. Standard analysis of variance (ANOVA) indicated, 1) that the mortality rates of 1 to 2-day old fish exposed to either of the toxicants KCl or SDS are significantly lower than the mortality rates of fish of 3 to 14-days of age exposed to these toxicants (p < 0.05), and 2) that there are no statistically significant differences in the mortality rates of, 1 to 2, 3 to 4, 7 to 8, 11 to 12, or 13 to 14-day age groups of fish exposed to the KCl plus SDS toxicant combination. Mean LC 50 values for 1 to 2-day age groups of fish exposed to the toxicants KCl, SDS, and the KCl plus SDS combination were all numerically higher than the mean LC 50 values for 3 to 4, 7 to 8, 11 to 12, or 13 to 14-day age groups of fish exposed to these toxicants. Standard ANOVA indicated that these differences were not statistically significant. More sophisticated statistical analyses using a cross validation scheme based on a regression model indicated that LC 50 values for 1 to 2-day age groups of fish exposed to either of the toxicants KCl or SDS were statistically higher than the LC 50 values for 3 to 14-day age groups of fish exposed to these toxicants. Power analysis also indicated that the data set may be too small to identify statistically significant differences in the LC 50 values across the different age groups of fish used in this study using only standard ANOVA. Average coefficients of variation (CVs) for tests conducted with 1 to 14-day old fish were numerically greater than CVs for tests conducted with 3 to 14-day old fish for the toxicants KCl, SDS, and the KCl plus SDS combination. These results suggest that LC 50 data generated using the 1 to 2-day old fish are more variable than the LC 50 data generated using 3 to 14-day old fish. The results of this study are consistent with those reported by Markle et al. (2000), who concluded that the age of organisms used for testing needs to be selected and/or specified to the laboratory conducting the bioassay in order to ensure uniform sensitivity and maximize precision. The results of this study indicate that fish age as a cause of inter-and intra-laboratory variability has not been sufficiently addressed by the USEPA in the publication of standard methods for conducting WET tests with P. promelas.