Health Stream Literature Summary - Issue 50 - June 2008

Microbial groundwater quality and its health implications for a border-strip irrigated diary farm catchment, South Island, New Zealand
Close, M., Dann, R., Ball, A., Pirie, R., Savill, M. and Smith, Z. (2008) Journal of Water & Health, 6(1); 83-98.

The objectives of this study were to assess the microbial quality of groundwater in a border strip irrigation dairy catchment on the South Island of New Zealand; to determine any relationship between timing of stocking, border strip irrigation and contamination of the groundwater using an indicator bacteria (E. coli) and the pathogen Campylobacter; and to conduct a preliminary assessment of the health risk related to drinking groundwater sourced from shallow aquifers in this area. Border strip irrigation involves the controlled flooding of contoured, bordered paddocks near river with around 100 mm of water. There is concern that in such situations macropore flow could lead to significant contamination where pathogens are being released to the ground surface from animal faeces and in parts of the farm, effluent irrigation. The study included five wells in the Waikakihi catchment which is a largely dairying catchment in South Canterbury with border strip irrigation. Well depths ranged from 4.6 to 11 m below ground level and wells were selected to avoid possible contamination from septic tanks and to ensure good wellhead protection. Information on stocking and irrigation practices was collected from the well owners and climatic data was collated. During the three year study 135 groundwater samples were collected with sampling regimes varying throughout the three years. E.coli was detected in 75% of samples, and Campylobacter in 12%.

A health risk assessment was undertaken to determine the risk to consumers associated with the Campylobacter found in the groundwater. The risk was calculated using a Monte Carlo approach with the @RISK modelling program. An epidemiological survey of the effects of dairy and irrigation in the greater Canterbury region was undertaken using the ESR EpiSurv notifiable disease database along with dairy farm location and major irrigation scheme data. An analysis of dairy and irrigation data was undertaken to produce three subset groups. The three subsets investigated were ‘Subject Group’ (SG) encompassing dairy with major irrigation schemes; ‘Control Group 1’ (CG1) all other dairy farms (i.e. not with major irrigation schemes) and ‘Control Group 2’ (CG2) the remainder of Canterbury region.

There was a general trend of high concentrations of E. coli being detected in groundwater approximately 20-30 days after a coincidence of stocking and irrigation (or a large rainfall event) in the nearby paddocks. There appeared to be a trend of Campylobacter contamination of wells after periods of major rainfall events generally accompanied by increases in E. coli counts. The lag period between ‘high risk’ irrigation events and elevated numbers of E. coli in groundwater samples showed a variable time delay which mostly ranged from 20-40 days.

The probability of infection on any given day from drinking 1 litre of well water in the study area was estimated at 0.50% to 0.76%, with the overall probability of infection during the 6 month irrigation season estimated at 60% to 75%. Infection does not necessarily result in illness, and the probability of illness was estimated between 20% and 40%. The epidemiological assessment in which the subject group was compared to the control groups indicated that there was a statistically significant increase in age-standardised rates of campylobacteriosis (CG1 Relative Risk (RR) = 1.15 (95% CI = 1.31-1.75); CG2 RR = 1.15 (1.33-1.72)); cryptosporidiosis (CG1 RR = 2.08 (1.55-2.79); CG2 RR = 5.33 (4.12-6.90)); and salmonellosis (only significant for CG2 RR = 2.05 (1.55-2.71)).

The results show that both E. coli and Campylobacter were consistently detected in groundwater samples from wells in the catchment of border strip irrigation with dairying. Other concurrent studies of microbial contamination of dairy farms using different irrigation methods have only identified very low levels of contamination compared to this area with border strip irrigation. The health risk assessment undertaken in this study suggests a substantial probability of infection from Campylobacter during the irrigation season among people drinking the well water. Epidemiological analysis showed significantly higher notifications of illness for campylobacteriosis, salmonellosis and cryptosporidiosis in dairying areas with major irrigation schemes.