Spatial-temporal variations of microbial water quality in surface reservoirs and canals used for irrigation

Guidelines for microbial irrigation water quality have been proposed by regulatory agencies and organizations to reduce potential risks of waterborne contamination of produce. Most recommendations emphasize threshold values of fecal indicators and typically rely on only a limited number of water tests over the course of irrigation season. In this study, two irrigation canals and four surface reservoirs located in Ohio, USA were repeatedly sampled (n=227) to investigate fluctuations of fecal indicators concentration over an irrigation season (2010). Bootstrap analysis was applied to determine the sensitivity of the measured parameters as a function of sampling frequency. Escherichia coli counts in water collected from irrigation canals were approximately one order of magnitude higher (2.48±0.79 log MPN per 100ml) than that in reservoirs (1.54±0.04 log MPN per 100ml) and increased following heavy rainfall events (>20mm) (P<0.01). Sampling frequency was estimated in terms of accuracy and precision. The 95% bootstrap confidence interval width surrounding coliform and E. coli estimates based on a single sample were broad, including or exceeding the upper limit for acceptable use standards recommended by several organizations (126CFU/100ml). In conclusion, a single water sample imprecisely reflected the quality of water over the course of the irrigation period. Caution should be used when drawing conclusions about the microbial acceptability of water used for irrigation purposes based on a limited number of water quality measurements. Environmental factors influencing the spatiotemporal variation in the water quality (i.e. the type of water source and recent heavy precipitation events) and the expected interval between testing and the time of harvest should be considered in developing irrigation water testing frequency guidelines.