Evaluation of Grower-Friendly, Science-Based Sampling Approaches for the Detection ofSalmonellain Ponds Used for Irrigation of Fresh Produce

Abstract

The recognition that irrigation water sources contribute to preharvest contamination of produce has led to new regulations on testing microbial water quality. To best identify contamination problems, growers who depend on irrigation ponds need guidance on how and where to collect water samples for testing. In this study, we evaluated several sampling strategies to identify Salmonella and Escherichia coli contamination in five ponds used for irrigation on produce farms in southern Georgia. Both Salmonella and E. coli were detected regularly in all the ponds over the 19-month study period, with overall prevalence and concentrations increasing in late summer and early fall. Of 507 water samples, 217 (42.8%) were positive for Salmonella, with a very low geometric mean (GM) concentration of 0.06 most probable number (MPN)/100 mL, and 442 (87.1%) tested positive for E. coli, with a GM of 6.40 MPN/100 mL. We found no significant differences in Salmonella or E. coli detection rates or concentrations between sampling at the bank closest to the pump intake versus sampling from the bank around the pond perimeter, when comparing with results from the pump intake, which we considered our gold standard. However, samples collected from the bank closest to the intake had a greater level of agreement with the intake (Cohen's kappa statistic = 0.53; p < 0.001) than the samples collected around the pond perimeter (kappa = 0.34; p = 0.009). E. coli concentrations were associated with increased odds of Salmonella detection (odds ratio = 1.31; 95% confidence interval = 1.10-1.56). All the ponds would have met the Produce Safety Rule standards for E. coli, although Salmonella was also detected. Results from this study provide important information to growers and regulators about pathogen detection in irrigation ponds and inform best practices for surface water sampling.

Prevalence of Salmonella and Listeria monocytogenes in non-traditional irrigation waters in the Mid-Atlantic United States is affected by water type, season, and recovery method

“…Water samples collected for filtration in ponds were collected from banks of the pond. Other research has shown that location of water collection from within pond sites can affect the quantitative recovery of Salmonella and E. coli [30,31].…”

The Persistence of Bacterial Pathogens in Surface Water and Its Impact on Global Food Safety

“…For biological indicators, generic E. coli, fecal coliforms, total coliforms, enterococci, and aerobic/heterotrophic bacterial plate counts have been compared to Salmonella presence/absence and concentrations. A few studies have identified a predictive relationship between the levels of generic E. coli and the probability of detecting Salmonella [81,82,87,100], while others have not seen this relationship [78,80,88]. The same dichotomy was seen with the other biological indicators [78,80,83,87,91,92,98,100].…”

“…Most studies in the U.S. have focused on the East and West Coasts in areas where agricultural production is high. Along the East Coast, Salmonella prevalence ranges between 3% and 100% [4,5,8,40,[77][78][79][80][81][82][83][84][85][86][87]. Most studies on the West Coast have been performed in CA, where prevalence ranges from 6% to 65% [6,[88][89][90][91].…”

“…Chemical indicators such as pH, conductivity, redox potential, dissolved oxygen and nitrogen/nitrate concentrations have also been studied for their ability to predict or correlate to the presence of Salmonella. The majority of the work has concluded that most of these indicators do not work well [80,81,83,91]. However, some have found correlations, such as Manan et al (2020), who saw a positive correlation between nitrate-nitrogen levels and Salmonella presence but a negative correlation with dissolved oxygen and pH in waters along the mid-Atlantic region of the U.S. [8].…”

“…Like these biological, chemical and physical indicators, weather patterns, such as seasonality, rainfall, and air temperature have been investigated. Several research studies have found seasonality to play a role in the likelihood of detecting Salmonella, where more Salmonella positive water samples were found in warmer months typically of the summer and fall seasons [80][81][82]97,101]. Other studies did not necessarily see a seasonal effect in the ability to detect Salmonella but did see a seasonal effect on the serovar diversity [77,80,99].…”