Health Stream Literature Summary - Issue 56 - December 2009

Adaptive management for mitigating Cryptosporidium risk in source water: A case study in an agricultural catchment in South Australia.
Bryan BA, Kandulu J, Deere DA et al. (2009) Journal of Environmental Management, 90(10); 3122-3134.

This paper presents a case study of the management of the pathogen Cryptosporidium in source water entering a drinking water supply in the Myponga catchment in South Australia using an adaptive management framework. The adaptive management framework includes a phase of explicit learning and adaptation which is used to inform planning of the next round of management. This study evaluated a recently completed water quality management program in the Myponga catchment. This information was then used to inform analysis and planning for a second round of management actions providing a practical example of an adaptive loop for the management of water quality.

The Myponga River catchment covers around 123 sq km and is situated 50 km south of Adelaide. The dominant land use in the Myponga catchment (61%) is broad scale grazing (mainly beef cattle and sheep, with some horses and deer). Other significant land uses include native vegetation (13%) and dairying (13%). The effectiveness of past water quality management programs in relation to the adoption of practices by landholders was evaluated using a socio-economic survey of land use and management in the catchment. During January to April 2007, a face-to-face interview and survey of 36 landholders in the Myponga catchment was undertaken. The survey included over 50 questions on land use and management relevant to water quality. Dairy farmers were also asked about effluent and nutrient management practices. The impact of past management on the mitigation of Cryptosporidium risk in source water was also evaluated on the basis of the analysis of water quality monitoring data. Water quality testing for enteric protozoa has been conducted in the Myponga catchment by SA Water since 1998. Samples were taken near the gauging station which captures run-off from 62% of the reservoir catchment area. Sampling is triggered by rainfall events rather than peak run-off events and therefore may underestimate Cryptosporidium loads.

Quantitative risk assessment was used in analysing and planning the next round of water quality management in the Myponga catchment. This study used an established, numerical process-based pathogen budget model to estimate the source apportionment of Cryptosporidium risk and the effectiveness of catchment mitigation benefits of a number of alternative management strategies. A baseline model was developed for the Myponga catchment including all potential sources of human-infectious Cryptosporidium under current land use and management. The survey of land use and management was the main data source to specify parameter values in the pathogen budget model with the addition of a previous livestock survey, spatial information on land use and property boundaries and other data. Scenario analysis was used to assess the impact of a range of management actions in the catchment in mitigating Cryptosporidium export to the reservoir. There were a total of 30 management scenarios specified. In the scenario analysis the baseline pathogen budget model parameter values were varied to estimate the effect of management actions on reducing Cryptosporidium export. The mitigation impact of catchment management scenarios was described relative to the baseline model by log removal and a measure of effectiveness (% change from baseline).

The survey results of the past catchment management programs revealed that the implementation of source water quality management programs led to the widespread adoption of best practice water quality management by dairy farmers. Low rates of adoption however, were found amongst other landholders. This difference may be explained largely by the fact that past programs were targeted towards dairy farmers. The analysis of water quality sampling data shows that from 2000 to 2007, SA Water's Cryptosporidium target of 0 cells/10 L in source water has been exceeded 100% of the time as detected at the inlet of the Myponga reservoir. Between 1 July 2001 and 30 June 2004 there were 51 water quality related incidents reported, 40 of which were linked to the detection of Cryptosporidium at the inlet to the reservoir. The median Cryptosporidium concentrations in source water samples increased from 3 oocysts/10 L in 2000 and peaked in 2003/2004 at 8 oocysts/10 L and have declined since then to 5/oocysts/10 L in 2007.

Most dairy farmers were found to practice stock and effluent management to minimise the contribution of pathogens to waterways. Dairy farmers were found to limit the access of young stock to water courses whereas non-dairy landholders did not. The decline in annual median oocysts concentration since 2004 was found to coincide with and is consistent with catchment management activities since 2000. A direct causal relationship was not established, however the monitoring data may reflect an initial beneficial impact of past catchment management programs on Cryptosporidium export. There may also be a number of other factors that influence Cryptosporidium concentrations such as land use change, sampling uncertainty and variation in catchment hydrology. The sampling results are nevertheless a positive sign. Despite the recent reductions, oocyst level still exceed SA Water targets and put ongoing pressure on the treatment system as a reliable barrier to mitigation.

The pathogen budget model showed under baseline land use and management conditions that almost all of the total human-infectious Cryptosporidium stock in Myponga catchment is generated by non-dairy calves (78%), dairy calves (17%) and lambs (5%). The model estimates that only about 1.4% of the total Cryptosporidium stock is exported to Myponga reservoir with around 65% of the export occurring in wet weather events. Non-dairy calves accounted for 87% of human-infectious Cryptosporidium export in dry weather and 95% in wet weather. Assessment of management scenarios suggests that most of the possible reduction in Cryptosporidium oocysts exported to the Myponga reservoir (91.8% in wet weather and 89.7% in dry weather) could be achieved by restricting watercourse access of non-dairy calves to 5%. Only slightly greater mitigation benefit (91.9% in dry weather and 89.8% in wet weather) could be achieved through restricting watercourse access of all non-dairy cattle to 5%. This case study of the Myponga catchment provides a practical and successful example of a passive adaptive management framework. The information gained was used to refocus, refine and re-prioritise the next round of management of Cryptosporidium risk in source water.

© Copyright Water Quality Research Australia Limited http://www.wqra.com.au/
Health Stream articles may be reproduced and communicated to third parties provided WQRA is acknowledged as the source. Literature summaries are derived in part from copyright material by a range of publishers. Original sources should be consulted and acknowledged.