Cross-Validation Indicates Predictive Models May Provide an Alternative to Indicator Organism Monitoring for Evaluating Pathogen Presence in Southwestern US Agricultural Water

Abstract

Pathogen contamination of agricultural water has been identified as a probable cause of recalls and outbreaks. However, variability in pathogen presence and concentration complicates the reliable identification of agricultural water at elevated risk of pathogen presence. In this study, we collected data on the presence of Salmonella and genetic markers for enterohemorrhagic E. coli (EHEC; PCR-based detection of stx and eaeA) in southwestern US canal water, which is used as agricultural water for produce. We developed and assessed the accuracy of models to predict the likelihood of pathogen contamination of southwestern US canal water. Based on 169 samples from 60 surface water canals (each sampled 1–3 times), 36% (60/169) and 21% (36/169) of samples were positive for Salmonella presence and EHEC markers, respectively. Water quality parameters (e.g., generic E. coli level, turbidity), surrounding land-use (e.g., natural cover, cropland cover), weather conditions (e.g., temperature), and sampling site characteristics (e.g., canal type) data were collected as predictor variables. Separate conditional forest models were trained for Salmonella isolation and EHEC marker detection, and cross-validated to assess predictive performance. For Salmonella, turbidity, day of year, generic E. coli level, and % natural cover in a 500–1,000 ft (~150–300 m) buffer around the sampling site were the top 4 predictors identified by the conditional forest model. For EHEC markers, generic E. coli level, day of year, % natural cover in a 250–500 ft (~75–150 m) buffer, and % natural cover in a 500–1,000 ft (~150–300 m) buffer were the top 4 predictors. Predictive performance measures (e.g., area under the curve [AUC]) indicated predictive modeling shows potential as an alternative method for assessing the likelihood of pathogen presence in agricultural water. Secondary conditional forest models with generic E. coli level excluded as a predictor showed < 0.01 difference in AUC as compared to the AUC values for the original models (i.e., with generic E. coli level included as a predictor) for both Salmonella (AUC = 0.84) and EHEC markers (AUC = 0.92). Our data suggests models that do not require the inclusion of microbiological data (e.g., indicator organism) show promise for real-time prediction of pathogen contamination of agricultural water (e.g., in surface water canals).

Salmonella Prevalence Is Strongly Associated with Spatial Factors while Listeria monocytogenes Prevalence Is Strongly Associated with Temporal Factors on Virginia Produce Farms

“…These findings support future environmental studies on the Eastern Shore of Virginia with a temporal component to enhance sampling surveys in order to understand ecological dynamics of the environment that appear favorable for Salmonella. While there are some studies (12,25,44,45) examining associations between land use and Salmonella prevalence in environmental samples, universal trends are lacking, and the associations instead appear to be based on study or location. For example, a 2021 study, observed percent developed open space in a 1,000-to 5,000-ft buffer zone was negatively associated with the odds of agricultural water in the southwestern United States being positive for Salmonella (44).…”

“…While there are some studies (12,25,44,45) examining associations between land use and Salmonella prevalence in environmental samples, universal trends are lacking, and the associations instead appear to be based on study or location. For example, a 2021 study, observed percent developed open space in a 1,000-to 5,000-ft buffer zone was negatively associated with the odds of agricultural water in the southwestern United States being positive for Salmonella (44). Another survey of waterways in California found a significantly (P , 0.05) lower prevalence of Salmonella in human-impacted areas than animal-impacted areas (44).…”

“…For example, a 2021 study, observed percent developed open space in a 1,000-to 5,000-ft buffer zone was negatively associated with the odds of agricultural water in the southwestern United States being positive for Salmonella (44). Another survey of waterways in California found a significantly (P , 0.05) lower prevalence of Salmonella in human-impacted areas than animal-impacted areas (44). Previous work in Virginia showed levels of fecal indicator bacteria (FIB) (Escherichia coli and total coliform bacteria) in streams were positively correlated with proportions of total developed land (urban plus agriculture) within a 100-m-wide buffer (28).…”

Prevalence of STEC virulence markers and Salmonella as a function of abiotic factors in agricultural water in the southeastern United States

“…This is relatively potentiated with the establishment of food safety practices to protect the safety of fresh produce ( Truitt et al, 2018 ; Devarajan et al, 2023 ). Previous studies have investigated the prevalence of foodborne pathogens in irrigation water sources for vegetable crops ( McEgan et al, 2013 ; Strawn et al, 2013b ; Falardeau et al, 2017 ; Topalcengiz et al, 2017 ; Truitt et al, 2018 ; Weller et al, 2020a , 2020b ; Belias et al, 2021 ; Murphy et al, 2022 ). Although there is a substantial body of research available, the existing studies exhibit a notable degree of inconsistency, often contradicting each other, especially when evaluating abiotic factors and pathogen prevalence.…”

Prediction of E. coli Concentrations in Agricultural Pond Waters: Application and Comparison of Machine Learning Algorithms

“…By achieving these goals, researchers may be able to eliminate unimportant predictors from measurement programs which can potentially save a great deal of time and resources. The application of ML regressions in the field of microbial water quality is relatively new (Park et al, 2015 ; García-Alba et al, 2019 ; Stocker et al, 2019 ; Abimbola et al, 2020 ; Ballesté et al, 2020 ; Li et al, 2020 ; Belias et al, 2021 ; Wang et al, 2021 ). For agricultural waters, research into predicting E. coli concentrations using ML was done for streams (Weller et al, 2021 ), but so far no studies have utilized ML regressions to predict E. coli concentrations in agricultural irrigation ponds which serve as an important source of irrigation water across the United States and abroad.…”