A STATISTICAL APPROACH FOR THE ASSESSMENT OF THE TOXIC INFLUENCES ON GAMMARUS-PULEX (AMPHIPODA) AND ASELLUS-AQUATICUS (ISOPODA) EXPOSED TO URBAN AQUATIC DISCHARGES

Citation
Rm. Mulliss et al., A STATISTICAL APPROACH FOR THE ASSESSMENT OF THE TOXIC INFLUENCES ON GAMMARUS-PULEX (AMPHIPODA) AND ASELLUS-AQUATICUS (ISOPODA) EXPOSED TO URBAN AQUATIC DISCHARGES, Water research, 30(5), 1996, pp. 1237-1243
Citations number
18
Categorie Soggetti
Engineering, Civil","Environmental Sciences","Water Resources
Journal title
ISSN journal
00431354
Volume
30
Issue
5
Year of publication
1996
Pages
1237 - 1243
Database
ISI
SICI code
0043-1354(1996)30:5<1237:ASAFTA>2.0.ZU;2-P
Abstract
A statistical procedure has been developed to quantify the mortality r esponse of caged macroinvertebrates in terms of measured environmental parameters. Populations of Gammarus pulex and Asellus aquaticus were caged in urban receiving waters downstream of a combined sewer overflo w (CSO) and a surface water outfall (SWO) for 36 d. During this period of exposure, the mortality and heavy metal bioaccumulation responses of populations of both species, as well as seventeen different chemica l and hydrological characteristics of the receiving watercourse were m onitored. Multivariate statistical techniques, comprising principal co mponent analysis (with VARIMAX rotation) and multiple regression analy sis were used to determine the measured variables that influenced mort ality. Certain parameters (BOD5, total aqueous copper concentration, f low rate and suspended solids) influenced the mortality responses of b oth species. Ammonia, total aqueous lead concentrations and dissolved aqueous concentrations of zinc and copper additionally affected the re sponse of Gammarus pulex. Body concentrations of zinc, lead, cadmium a nd copper and dissolved aqueous concentrations of cadmium were found t o be influential upon the mortality response of caged Asellus aquaticu s. The relationships between the mortality responses of each species a nd the selected independent variables are expressed in the form of sta ble quantitative equations by regressing the dependent variable (ie mo rtality) against the principal components of the independent parameter s. This statistical approach represents an important tool for interpre ting large intercorrelated sets of environmental data obtained in situ . Copyright (C) 1996 Elsevier Science Ltd.