Meeting Report: Key Outcomes from a Collaborative Summit on Agricultural Water Standards for Fresh Produce

Scale of analysis drives the observed ratio of spatial to non-spatial variance in microbial water quality: insights from two decades of citizen science data

AIMS

While fecal indicator bacteria (FIB) testing is used to monitor surface water for potential health hazards, observed variation in FIB levels may depend on the scale of analysis (SOA). Two decades of citizen science data, coupled with random effects models, were used to quantify the variance in FIB levels attributable to spatial versus temporal factors.

METHODS AND RESULTS

Separately, Bayesian models were used to quantify the ratio of spatial to non-spatial variance in FIB levels and identify associations between environmental factors and FIB levels. Separate analyses were performed for three SOA: waterway, watershed, and statewide. As SOA increased (from waterway to watershed to statewide models), variance attributable to spatial sources generally increased and variance attributable to temporal sources generally decreased. While relationships between FIB levels and environmental factors, such as flow conditions (base versus stormflow), were constant across SOA, the effect of land cover was highly dependent on SOA and consistently smaller than the effect of stormwater infrastructure (e.g. outfalls).

CONCLUSIONS

This study demonstrates the importance of SOA when developing water quality monitoring programs or designing future studies to inform water management.

Microbiological quality of irrigation water for cultivation of fruits and vegetables: An overview of available guidelines, water testing strategies and some factors that influence compliance

Contaminated irrigation water is among many potential vehicles of human pathogens to food plants, constituting significant public health risks especially for the fresh produce category. This review discusses some available guidelines or regulations for microbiological safety of irrigation water, and provides a summary of some common methods used for characterizing microbial contamination. The goal of such exploration is to understand some of the considerations that influence formulation of water testing guidelines, describe priority microbial parameters particularly with respect to food safety risks, and attempt to determine what methods are most suitable for their screening. Furthermore, the review discusses factors that influence the potential for microbiologically polluted irrigation water to pose substantial risks of pathogenic contamination to produce items. Some of these factors include type of water source exploited, irrigation methods, other agro ecosystem features/practices, as well as pathogen traits such as die-off rates. Additionally, the review examines factors such as food safety knowledge, other farmer attitudes or inclinations, level of social exposure and financial circumstances that influence adherence to water testing guidelines and other safe water application practices. A thorough understanding of relevant risk metrics for the application and management of irrigation water is necessary for the development of water testing criteria. To determine sampling and analytical approach for water testing, factors such as agricultural practices (which differ among farms and regionally), as well as environmental factors that modulate how water quality may affect the microbiological safety of produce should be considered. Research and technological advancements that can improve testing approach and the determination of target levels for hazard characterization or description for the many different pollution contexts as well as farmer adherence to testing requirements, are desirable.

Bio-based antimicrobial compositions and sensing technologies to improve food safety

Microbial contamination of food products is a significant challenge that impacts food safety and quality. This review focuses on bio-based technologies for enhancing the decontamination of raw foods during postharvest processing, preventing cross-contamination, and rapidly detecting microbial risks. The bio-based antimicrobial compositions include bio-based antimicrobial delivery systems and coatings. The antimicrobial delivery systems are developed using cell-based carriers, microbubbles, and lipid-based colloidal particles. The antimicrobial coatings are engineered by incorporating biopolymers with conventional antimicrobials or cell-based antimicrobial carriers. The bio-based sensing approaches focus on replacing antibodies with more stable and cost-effective bio-receptors, including antimicrobial peptides, bacteriophages, DNAzymes, and engineered liposomes. Together, these approaches can reduce microbial contamination risks and enhance the in-situ detection of microbes.

Structural Equation Models Suggest That On-Farm Noncrop Vegetation Removal Is Not Associated with Improved Food Safety Outcomes but Is Linked to Impaired Water Quality

While growers have reported pressures to minimize wildlife intrusion into produce fields through noncrop vegetation (NCV) removal, NCV provides key ecosystem services. To model food safety and environmental tradeoffs associated with NCV removal, published and publicly available food safety and water quality data from the Northeastern United States were obtained. Because data on NCV removal are not widely available, forest-wetland cover was used as a proxy, consistent with previous studies. Structural equation models (SEMs) were used to quantify the effect of forest-wetland cover on (i) food safety outcomes (e.g., detecting pathogens in soil) and (ii) water quality (e.g., nutrient levels). Based on the SEMs, NCV was not associated with or had a protective effect on food safety outcomes (more NCV was associated with a reduced likelihood of pathogen detection). The probabilities of detecting Listeria spp. in soil (effect estimate [EE] = -0.17; P = 0.005) and enterohemorrhagic Escherichia coli in stream samples (EE = -0.27; P < 0.001) were negatively associated with the amount of NCV surrounding the sampling site. Larger amounts of NCV were also associated with lower nutrient, salinity, and sediment levels, and higher dissolved oxygen levels. Total phosphorous levels were negatively associated with the amount of NCV in the upstream watershed (EE = -0.27; P < 0.001). Similar negative associations (P < 0.05) were observed for other physicochemical parameters, such as nitrate (EE = -0.38). Our findings suggest that NCV should not be considered an inherent produce safety risk or result in farm audit demerits. This study also provides a framework for evaluating environmental tradeoffs associated with using specific preharvest food safety strategies. IMPORTANCE Currently, on-farm food safety decisions are typically made independently of conservation considerations, often with detrimental impacts on agroecosystems. Comanaging agricultural environments to simultaneously meet conservation and food safety aims is complicated because farms are closely linked to surrounding environments, and management decisions can have unexpected environmental, economic, and food safety consequences. Thus, there is a need for research on the conservation and food safety tradeoffs associated with implementing specific preharvest food safety practices. Understanding these tradeoffs is critical for developing adaptive comanagement strategies and ensuring the short- and long-term safety, sustainability, and profitability of agricultural systems. This study quantifies tradeoffs and synergies between food safety and environmental aims, and outlines a framework for modeling tradeoffs and synergies between management aims that can be used to support future comanagement research.

Determination of Phylloplane Associated Bacteria of Lettuce from a Small-Scale Aquaponic System via 16S rRNA Gene Amplicon Sequence Analysis

Fresh vegetables harbour diverse bacterial populations on their surfaces which are important for plant health and growth. Information on epiphytic bacteria is limited to only a few types of vegetables and it is unknown how the lettuce epiphytic bacterial community structure may respond when a probiotic product is added to an aquaponic system. In this study, we evaluated lettuce growth and analysed epiphytic bacterial communities of lettuce based on metabarcoding analysis of the V3-V4 region of the 16S rRNA gene obtained from paired-end Illumina MiSeq reads. The addition of Bacillus probiotics resulted in a significant increase of nitrate and phosphate in the deep-water culture solution, as well as increased vegetative growth of lettuce. Metabarcoding analysis revealed that the most abundant phyla on lettuce leaf surfaces were Proteobacteria, Bacteroidetes, Firmicutes, and Actinobacteria. The in-depth bacterial composition analysis indicated that genera Chryseobacterium, Bacillus, Pantoea, Pseudoduganella, Flavobacterium, Paludibacter, and Cloacibacterium were dominant in leaf samples obtained from Bacillus-treated systems. Analysis of lettuce epiphytic bacterial communities of the fresh lettuce leaf surfaces also indicated the presence of food-borne pathogens belonging to the Shigella and Aeromonas genera, which were less abundant in the probiotic treated systems. This study provides the first characterization of the epiphytic bacterial community structure and how it can be modulated by the addition of a probiotic mixture to the nutrient solution of aquaponic systems.

Interpretability Versus Accuracy: A Comparison of Machine Learning Models Built Using Different Algorithms, Performance Measures, and Features to Predict E. coli Levels in Agricultural Water

Since E. coli is considered a fecal indicator in surface water, government water quality standards and industry guidance often rely on E. coli monitoring to identify when there is an increased risk of pathogen contamination of water used for produce production (e.g., for irrigation). However, studies have indicated that E. coli testing can present an economic burden to growers and that time lags between sampling and obtaining results may reduce the utility of these data. Models that predict E. coli levels in agricultural water may provide a mechanism for overcoming these obstacles. Thus, this proof-of-concept study uses previously published datasets to train, test, and compare E. coli predictive models using multiple algorithms and performance measures. Since the collection of different feature data carries specific costs for growers, predictive performance was compared for models built using different feature types [geospatial, water quality, stream traits, and/or weather features]. Model performance was assessed against baseline regression models. Model performance varied considerably with root-mean-squared errors and Kendall's Tau ranging between 0.37 and 1.03, and 0.07 and 0.55, respectively. Overall, models that included turbidity, rain, and temperature outperformed all other models regardless of the algorithm used. Turbidity and weather factors were also found to drive model accuracy even when other feature types were included in the model. These findings confirm previous conclusions that machine learning models may be useful for predicting when, where, and at what level E. coli (and associated hazards) are likely to be present in preharvest agricultural water sources. This study also identifies specific algorithm-predictor combinations that should be the foci of future efforts to develop deployable models (i.e., models that can be used to guide on-farm decision-making and risk mitigation). When deploying E. coli predictive models in the field, it is important to note that past research indicates an inconsistent relationship between E. coli levels and foodborne pathogen presence. Thus, models that predict E. coli levels in agricultural water may be useful for assessing fecal contamination status and ensuring compliance with regulations but should not be used to assess the risk that specific pathogens of concern (e.g., Salmonella, Listeria) are present.

Predictive Models May Complement or Provide an Alternative to Existing Strategies for Assessing the Enteric Pathogen Contamination Status of Northeastern Streams Used to Provide Water for Produce Production

While the Food Safety Modernization Act established standards for the use of surface water for produce production, water quality is known to vary over space and time. Targeted approaches for identifying hazards in water that account for this variation may improve growers' ability to address pre-harvest food safety risks. Models that utilize publicly-available data (e.g., land-use, real-time weather) may be useful for developing these approaches. The objective of this study was to use pre-existing datasets collected in 2017 (N = 181 samples) and 2018 (N = 191 samples) to train and test models that predict the likelihood of detecting Salmonella and pathogenic E. coli markers (eaeA, stx) in agricultural water. Four types of features were used to train the models: microbial, physicochemical, spatial and weather. "Full models" were built using all four features types, while "nested models" were built using between one and three types. Twenty learners were used to develop separate full models for each pathogen. Separately, to assess information gain associated with using different feature types, six learners were randomly selected and used to develop nine, nested models each. Performance measures for each model were then calculated and compared against baseline models where E. coli concentration was the sole covariate. In the methods, we outline the advantages and disadvantages of each learner. Overall, full models built using ensemble (e.g., Node Harvest) and "black-box" (e.g., SVMs) learners out-performed full models built using more interpretable learners (e.g., tree- and rule-based learners) for both outcomes. However, nested eaeA-stx models built using interpretable learners and microbial data performed almost as well as these full models. While none of the nested Salmonella models performed as well as the full models, nested models built using spatial data consistently out-performed models that excluded spatial data. These findings demonstrate that machine learning approaches can be used to predict when and where pathogens are likely to be present in agricultural water. This study serves as a proof-of-concept that can be built upon once larger datasets become available and provides guidance on the learner-data combinations that should be the foci of future efforts (e.g., tree-based microbial models for pathogenic E. coli).

Longitudinal Assessment of the Dynamics of Escherichia coli, Total Coliforms, Enterococcus spp., and Aeromonas spp. in Alternative Irrigation Water Sources: a CONSERVE Study

As climate change continues to stress freshwater resources, we have a pressing need to identify alternative (nontraditional) sources of microbially safe water for irrigation of fresh produce. This study is part of the center CONSERVE, which aims to facilitate the adoption of adequate agricultural water sources. A 26-month longitudinal study was conducted at 11 sites to assess the prevalence of bacteria indicating water quality, fecal contamination, and crop contamination risk (Escherichia coli, total coliforms [TC], Enterococcus, and Aeromonas). Sites included nontidal freshwater rivers/creeks (NF), a tidal brackish river (TB), irrigation ponds (PW), and reclaimed water sites (RW). Water samples were filtered for bacterial quantification. E. coli, TC, enterococci (∼86%, 98%, and 90% positive, respectively; n = 333), and Aeromonas (∼98% positive; n = 133) were widespread in water samples tested. Highest E. coli counts were in rivers, TC counts in TB, and enterococci in rivers and ponds (P < 0.001 in all cases) compared to other water types. Aeromonas counts were consistent across sites. Seasonal dynamics were detected in NF and PW samples only. E. coli counts were higher in the vegetable crop-growing (May-October) than nongrowing (November-April) season in all water types (P < 0.05). Only one RW and both PW sites met the U.S. Food Safety Modernization Act water standards. However, implementation of recommended mitigation measures of allowing time for microbial die-off between irrigation and harvest would bring all other sites into compliance within 2 days. This study provides comprehensive microbial data on alternative irrigation water and serves as an important resource for food safety planning and policy setting.IMPORTANCE Increasing demands for fresh fruit and vegetables, a variable climate affecting agricultural water availability, and microbial food safety goals are pressing the need to identify new, safe, alternative sources of irrigation water. Our study generated microbial data collected over a 2-year period from potential sources of irrigation (rivers, ponds, and reclaimed water sites). Pond water was found to comply with Food Safety Modernization Act (FSMA) microbial standards for irrigation of fruit and vegetables. Bacterial counts in reclaimed water, a resource that is not universally allowed on fresh produce in the United States, generally met microbial standards or needed minimal mitigation. We detected the most seasonality and the highest microbial loads in river water, which emerged as the water type that would require the most mitigation to be compliant with established FSMA standards. This data set represents one of the most comprehensive, longitudinal analyses of alternative irrigation water sources in the United States.

Effect of Weather on the Die-Off of Escherichia coli and Attenuated Salmonella enterica Serovar Typhimurium on Preharvest Leafy Greens following Irrigation with Contaminated Water

The Food Safety Modernization Act (FSMA) includes a time-to-harvest interval following the application of noncompliant water to preharvest produce to allow for microbial die-off. However, additional scientific evidence is needed to support this rule. This study aimed to determine the impact of weather on the die-off rate of Escherichia coli and Salmonella on spinach and lettuce under field conditions. Standardized, replicated field trials were conducted in California, New York, and Spain over 2 years. Baby spinach and lettuce were grown and inoculated with an ∼104-CFU/ml cocktail of E. coli and attenuated Salmonella Leaf samples were collected at 7 time points (0 to 96 h) following inoculation; E. coli and Salmonella were enumerated. The associations of die-off with study design factors (location, produce type, and bacteria) and weather were assessed using log-linear and biphasic segmented log-linear regression. A segmented log-linear model best fit die-off on inoculated leaves in most cases, with a greater variation in the segment 1 die-off rate across trials (-0.46 [95% confidence interval {95% CI}, -0.52, -0.41] to -6.99 [95% CI, -7.38, -6.59] log10 die-off/day) than in the segment 2 die-off rate (0.28 [95% CI, -0.20, 0.77] to -1.00 [95% CI, -1.16, -0.85] log10 die-off/day). A lower relative humidity was associated with a faster segment 1 die-off and an earlier breakpoint (the time when segment 1 die-off rate switches to the segment 2 rate). Relative humidity was also found to be associated with whether die-off would comply with FSMA's specified die-off rate of -0.5 log10 die-off/day.IMPORTANCE The log-linear die-off rate proposed by FSMA is not always appropriate, as the die-off rates of foodborne bacterial pathogens and specified agricultural water quality indicator organisms appear to commonly follow a biphasic pattern with an initial rapid decline followed by a period of tailing. While we observed substantial variation in the net culturable population levels of Salmonella and E. coli at each time point, die-off rate and FSMA compliance (i.e., at least a 2 log10 die-off over 4 days) appear to be impacted by produce type, bacteria, and weather; die-off on lettuce tended to be faster than that on spinach, die-off of E. coli tended to be faster than that of attenuated Salmonella, and die-off tended to become faster as relative humidity decreased. Thus, the use of a single die-off rate for estimating time-to-harvest intervals across different weather conditions, produce types, and bacteria should be revised.

Total Coliform and Generic E. coli Levels, and Salmonella Presence in Eight Experimental Aquaponics and Hydroponics Systems: A Brief Report Highlighting Exploratory Data

Although many studies have investigated foodborne pathogen prevalence in conventional produce production environments, relatively few have investigated prevalence in aquaponics and hydroponics systems. This study sought to address this knowledge gap by enumerating total coliform and generic E. coli levels, and testing for Salmonella presence in circulating water samples collected from five hydroponic systems and three aquaponic systems (No. of samples = 79). While total coliform levels ranged between 6.3 Most Probable Number (MPN)/100-mL and the upper limit of detection (2496 MPN/100-mL), only three samples had detectable levels of E. coli and no samples had detectable levels of Salmonella. Of the three E. coli positive samples, two samples had just one MPN of E. coli/100-mL while the third had 53.9 MPN of E. coli/100-mL. While the sample size reported here was small and site selection was not randomized, this study adds key data on the microbial quality of aquaponics and hydroponics systems to the literature. Moreover, these data suggest that contamination in these systems occurs at relatively low-levels, and that future studies are needed to more fully explore when and how microbial contamination of aquaponics and hydroponic systems is likely to occur.

Complex Interactions Between Weather, and Microbial and Physicochemical Water Quality Impact the Likelihood of Detecting Foodborne Pathogens in Agricultural Water

Agricultural water is an important source of foodborne pathogens on produce farms. Managing water-associated risks does not lend itself to one-size-fits-all approaches due to the heterogeneous nature of freshwater environments. To improve our ability to develop location-specific risk management practices, a study was conducted in two produce-growing regions to (i) characterize the relationship between Escherichia coli levels and pathogen presence in agricultural water, and (ii) identify environmental factors associated with pathogen detection. Three AZ and six NY waterways were sampled longitudinally using 10-L grab samples (GS) and 24-h Moore swabs (MS). Regression showed that the likelihood of Salmonella detection (Odds Ratio [OR] = 2.18), and eaeA-stx codetection (OR = 6.49) was significantly greater for MS compared to GS, while the likelihood of detecting L. monocytogenes was not. Regression also showed that eaeA-stx codetection in AZ (OR = 50.2) and NY (OR = 18.4), and Salmonella detection in AZ (OR = 4.4) were significantly associated with E. coli levels, while Salmonella detection in NY was not. Random forest analysis indicated that interactions between environmental factors (e.g., rainfall, temperature, turbidity) (i) were associated with likelihood of pathogen detection and (ii) mediated the relationship between E. coli levels and likelihood of pathogen detection. Our findings suggest that (i) environmental heterogeneity, including interactions between factors, affects microbial water quality, and (ii) E. coli levels alone may not be a suitable indicator of food safety risks. Instead, targeted methods that utilize environmental and microbial data (e.g., models that use turbidity and E. coli levels to predict when there is a high or low risk of surface water being contaminated by pathogens) are needed to assess and mitigate the food safety risks associated with preharvest water use. By identifying environmental factors associated with an increased likelihood of detecting pathogens in agricultural water, this study provides information that (i) can be used to assess when pathogen contamination of agricultural water is likely to occur, and (ii) facilitate development of targeted interventions for individual water sources, providing an alternative to existing one-size-fits-all approaches.