Escherichia coli concentrations and loads in an urbanised catchment: The Yarra River, Australia

The Yarra River flows through the city of Melbourne in southeast Australia. The estuary of the river is in the centre of the city, and so is an area of recreational activities with strict water quality requirements to limit human health risks. These requirements are based on ensuring that the measured E. colt concentrations are below certain values. The objectives of the work presented in this paper were (i) to identify possible patterns in spatial and temporal variability of E. colt concentrations along the Yarra River, (ii) to understand the relationship between E. colt concentrations and other pollutants along the Yarra River, and (iii) to quantify the importance of stormwater drains in determining the magnitude of E. colt levels in the estuary. Two long-term water quality data sets, including E. colt concentrations, along the estuary were analysed: one data set comprises approximately monthly data collected in the period 1994-2010 at six sites, and the other is composed by weekly samples in 12 sites collected in 2005-2010. Additionally, data from two major stormwater drains were analysed. Median values of E. coli concentrations were observed to increase further downstream with concentrations showing larger variability in the estuary. Inter-annual and seasonal patterns were evident in the upper catchment, although differences were sometimes not statistically significant; in the more urbanised part of the catchment, including the estuary, inter-annual and seasonal variations were no longer present. A multiple linear regression analysis showed that the E. colt concentrations along the river were positively correlated with phosphorus. Estimates of loads from the streams and some of the drains supplying the estuary suggest that stormwater drains might be a considerable source of E. colt contamination especially during wet-weather flow; however, the loads from the drains appear to be at least one order of magnitude lower than the inputs of E. colt from upstream of the estuary. This result suggests that mitigation efforts for the Yarra estuary may need to focus also on areas upstream, as well as those impervious areas contributing directly to the estuary. (C) 2013 Elsevier B.V. All rights reserved.