| Abstract: | Abstract Quantitative electric fishing and multiple mesh-size gillnetting of fish communities in the Hawkesbury-Nepean River system, near Sydney, were carried out upstream of, within and well below each of three major points of treated sewage effluent discharge. Approximately half of the fish sampled were caught by electric fishing, comprising 515 individuals in 3 h of electric fishing (172 fish per hour). The remaining 439 fish were caught in 72 h of gillnetting (six fish per hour). In total, 16 fish species were collected by electric fishing and only seven species by the gill nets. Differences were found in the total abundances and numbers of species collected by each method in relation to points of treated sewage effluent discharge, but neither community variable showed consistent differences between the areas above, within and well below points of treated sewage effluent discharge in all three river regions. However, classification and ordination of electric fishing catch data produced distinct fish community groups in relation to sewage discharge, while similar analyses of the gillnetting data failed to reveal such differences in community structure. It was suggested that the increased sampling efficiency of electric fishing, both in terms of the number of fish sampled per unit effort and number of species collected, allowed for significant increases in the power of univariate and multivariate analysis methods in detecting such environmental disturbances in large rivers. |
