1. Introduction
In recent years the contamination of New Zealand’s freshwaters by a range of indicator and pathogenic microorganisms has been studied under the Freshwater Microbiological Research Programme (McBride 2002). The results from this and earlier studies (for example, Smith et al. 1993) have confirmed that microbial contamination of lakes and rivers is widespread in New Zealand, with concentrations of the faecal indicator Escherchia coli (E. coli) often exceeding 1000 cfu per 100mL. Such findings, coupled with the high incidence of notified campylobacteriosis (Savill et al. 2001) and cryptosporidiosis (Duncanson et al. 2000) compared to other developed countries, has raised concerns over the public health risk from pathogens of faecal origin (including Campylobacter, Cryptosporidium oocysts, Giardia cysts, and Salmonellae) in New Zealand’s fresh waters. This risk to public health has substantial implications for land management practices, and for New Zealand’s international image with respect to trade and tourism. Furthermore, faecal contamination also restricts the recreational use of freshwaters, use for potable treatment, and shellfish aquaculture in estuaries receiving agricultural drainage.
The sources of faecal contamination of freshwaters are diverse and vary both spatially and temporally. Numerous studies (Wilcock 1986, Davies-Colley and Stroud 1995, and Gary et al. 1983) have demonstrated that grazing livestock cause faecal bacterial contamination of streams. This contamination arises through the delivery of faecal material in overland (Doran and Linn 1979) and subsurface (Collins 2002) flows to a watercourse and, where livestock have access to a stream, direct deposition of faecal material (Gary et al. 1983, Davies-Colley et al. 2002). Wild animals also contribute to faecal contamination of waterways (Niemi and Niemi 1991). Point source discharges of wastewater from sewage treatment and animal processing plants have also been shown to impair the bacterial quality of receiving waters (Smith 2001). Discharge of effluent to land, although considered preferable to discharge direct to a watercourse, leads to contamination of soil and soil water (Trevisan et al. 2002), which may, ultimately flow to surface waters.
The impact of each source of faecal contamination upon freshwater is influenced by a range of interacting environmental factors. These factors include the physical characteristics of a catchment, the land management practices within it, and dynamic processes such as rainfall, stream flow and recent climatic conditions. Understanding of the interactions between sources of contamination, and the watershed processes acting upon them remains, however, incomplete. This, in turn, may limit identification and application of land management strategies to minimise faecal contamination.
In an attempt to advance understanding and identify mitigating practices, stream water E. coli concentrations, collected from a regional monitoring program, were analysed in conjunction with a range of environmental variables. The microbial data, provided by Environment Waikato (EW), was collected from 73 stream sites throughout the Waikato region that encompass a diverse range of contaminant sources, catchment characteristics and land management practices. As part of the analysis a statistical model was developed to predict median E. coli concentrations across the region using basic environmental data. The modelling aids evaluation of the relative importance of sources and processes and provides a means of predicting likely faecal contamination at sites not currently monitored for E. coli.
This approach to improving prediction and management of faecal contamination through the use of environmental datasets has had recent, but limited, application elsewhere. Smith et al. (2001) used land cover and digital elevation data to analyse faecal contamination in South Carolina, and Crowther et al. (2001) used river discharge, tide, wind and sunshine data to provide insight into the factors affecting microbial water quality in coastal recreational waters in the U.K. The analysis of the EW dataset is, however, the most comprehensive to date in New Zealand
As part of this study AgResearch were subcontracted to undertake a field survey of selected watersheds (Longhurst and O’Connor, 2002). The objective of this survey was to identify and correct any major discrepancies in the data set, for example, in the delineation of catchment boundaries and/or descriptions of land use. The presence of riparian planting within a catchment was also noted, and an assessment made of the accessibility of streams to cattle. Direct deposition of faecal material into streams is likely to be an important process contributing to contamination, and riparian planting may attenuate microbes in overland flow and (if fenced) exclude stock from streams.
Contact for Enquiries
MAF Information Services
Pastoral House
25 The Terrace
PO Box 2526
Wellington, NEW ZEALAND
Fax: +64 4 894 0721
Contact this person
