|
(60-08) Accounting for hierachical variability in species sensitivity distributions. Verdonck, Frederik*,1, Janssen, Colin2, Jaworska, Joanna3, Thas, Olivier1, Vanrolleghem, Peter1, 1 Ghent University, Coupure Links 653, Gent, Belgium2 Ghent University, Jozef Plateaustraat 22, Gent, Belgium3 Procter & Gamble, Temselaan 100, Strombeek-Bever, Belgium ABSTRACT- One of the problems with the current use of species sensitivity distributions (SSD's) is that information available within the raw data on (1) intra-species sensitivity (variations in sensitivity between several individuals of the same species) and (2) inter-laboratory variations is discarded e.g. by averaging the data available for a particular species. These extra levels of variations are called here hierarchical variations. Two possible alternative methodologies to include hierarchical variations in SSD's and, thus, into the derivation of a hazardous concentration HC5, were proposed and tested. Accounting for this extra information results in a more realistic effect assessment. SSD's for atrazine, LAS, Cu and Zn are used as illustrative examples. Weighted characterisation of all available chronic toxicity data into one single distribution is found to be the best method. In this way, more information on intra-species or inter-laboratory variability is considered. Weighting avoids that a particular species with abundant data is overrepresented in the SSD. Key words: inter-laboratory variability, intra-species variability, uncertainty |
