Applying species‐sensitivity distributions in ecological risk assessment: Assumptions of distribution type and sufficient numbers of species

Abstract

Species-sensitivity distribution methods assemble single-species toxicity data to predict hazardous concentrations (HCps) affecting a certain percentage (p) of species in a community. The fit of the lognormal model and required number of individual species values were evaluated with 30 published data sets. The increasingly common assumption that a lognormal model best fits these data was not supported. Fifteen data sets failed a formal test of conformity to a lognormal distribution; other distributions often provided better fit to the data than the lognormal distribution. An alternate bootstrap method provided accurate estimates of HCp without the assumption of a specific distribution. Approximate sample sizes producing HC5 estimates with minimal variance ranged from 15 to 55, and had a median of 30 species-sensitivity values. These sample sizes are higher than those suggested in recent regulatory documents. A bootstrap method is recommended that predicts with 95% confidence the concentration affecting 5% or fewer species.

Revisions to the derivation of the Australian and New Zealand guidelines for toxicants in fresh and marine waters

“…This is a considerable increase on the current minimum data requirements (by 60%), but still not as large as the USA (USEPA, 1999) or EU requirements (EC, 2011). However, the minimum data requirements of all these jurisdictions fall considerably short of the minimum data requirements recommended by Newman et al (2000) of between 15 and 55 with a median of 30 and by Wheeler et al (2002) of 10 to 15 species. The reason for not increasing the minimum data requirements to at least 10 (as per Wheeler et al, 2002) was a matter of balancing competing factors.…”

Including Mixtures in the Determination of Water Quality Criteria for Herbicides in Surface Water

“…Wheeler et al (37) recommend that a minimum of 10 data points be used, as they showed that results of SSD analysis appear to stabilize with 10-15 data points. Newman et al (36) wrote that optimal sample sizes range from 15 to 55 data points depending on the chemical and data quality. The Technical Guidance Document on Risk Assessment (38) recommends the use of at least 10 NOEC values (preferably more than 15) for different species covering at least 8 taxonomic groups.…”

Toxicity Reference Values and Tissue Residue Criteria for Protecting Avian Wildlife Exposed to Methylmercury in China

“…If the duration of exposure was deemed to be insuf fi cient, or if only an unbounded LOAEL was produced in some studies, the data were corrected before fi tting the SSD function, by using UF S or UF L (US EPA 2005 ) . The basic assumption of the SSD approach is that sensitivity among species can be described by using a speci fi ed statistical distribution, such as the normal (Wagner and Løkke 1991 ;Aldenberg and Jaworska 2000 ) , logistic (Kooijman 1987 ;Aldenberg and Slob 1993 ) , triangular (Stephan et al 1985 ) , or Weibull ( Caldwell et al 2008 ) probability functions, or by using distribution-free, nonparametric methods (Ling 2004 ;Newman et al 2000 ) . It was assumed that the toxicity data selected for MeHg are skewed and can be described by using a log-normal distribution (Aldenberg and Jaworska 2000 ) .…”