Summary
The goal of this project is to develop a risk-based approach for sampling of irrigation waters used for produce production to minimize the risk of crop contamination by foodborne bacteria. Irrigation water has been implicated in a number of outbreaks associated with fresh produce. Currently, there are no scientific methods for determining where and how often water quality sampling should take place in constructed irrigation system’s typical of Arizona and Southern California. We propose to use a risk assessment considering factors which are known to influence contamination of surface waters including rainfall, watershed characteristics (landscape features, urban development, etc), the type of produce, and the irrigation method (e.g. spray vs. flood) to develop recommendations for risk-based sampling strategies for growers. Additionally, since rainfall plays a significant role in surface water quality a user friendly application, for use with mobile phones or other hardware, will be developed to aid in determining the need for risk-based sampling based on downloadable local weather information. This study will offer recommendations towards risk based sampling strategies (frequency, timing, location, volume) for E.coli indicator bacteria in irrigation waters that provide the greatest risk reduction to produce.
Technical Abstract
Irrigation water has been implicated in a number of outbreaks associated with fresh produce. General guidelines for water quality sampling for indicator bacteria (Escherichia coli) and sampling frequency have recently been proposed by the Food and Drug Administration (Federal Register, 2013), however, it is not apparent if they are based on site specific conditions with quantifiable benefits related to risk reduction. The goals of this research are to assess and quantify factors which 1) determine variability of generic (indicator) E.coli, pathogenic E.coli (Shiga toxin producing stains – STEC), and Salmonella occurrence in irrigation water over time, based on historic data and data collected as part of this study, at specific locations in Arizona/Southern California. This data will be used to assess the impact of risk events such as rainfall, water quality factors including temperature and turbidity, canal size, and watershed characteristics (potential sources of fecal contamination), on the occurrence of these organisms. 2) Assess the impact of occurrence, duration and intensity of rainfall events on E coli/Salmonella in irrigation waters with the goal to determine how long after a specific rainfall event the irrigation water quality will be impacted. 3) Use an exposure scenario risk based model for E.coli/Salmonella in irrigation waters to quantify the risks of infection with different sampling frequencies of irrigation waters based on environmental factors (e.g. rainfall), irrigation methods, and type of produce. 4) Develop a cell-phone/computer application (app) that can be used for guidance for frequency of sampling after high risk (rainfall) events.
Research Objectives
1) Determine the variability of E.coli/STEC/Salmonella occurrence in irrigation waters over time based on historic data at specific locations in Arizona/Southern California. This data will be used to initially assess the impact of rainfall events and water quality factors (e.g., temperature, turbidity), canal size, and watershed characteristics (e.g., bridges, drainage ways, urban development), on the occurrence of these organisms.
2) Assess the impact of occurrence, duration, and intensity of rainfall events on the presence of E.coli/Salmonella in irrigation waters and to determine the effect of sample volume on being able to detect E.coli and Salmonella in water in comparison to the traditional 100 ml IDEXX Colilert Quantitray® test.
3) Use an exposure scenario risk based model for E. coli/Salmonella in irrigation waters to quantify the risks of infection with different sampling frequencies of irrigation waters based on environmental factors (e.g., rainfall), irrigation methods, and the type of produce.
4) Develop a cell-phone/computer application (App) that can be used for guidance for frequency of sampling based on risk factors (e.g., after rainfall events).
Findings & Recommendations
Overall, the findings from this project support previous work by the project PIs to determine risk related factors that are likely to influence water quality. More specifically, this project has the following recommendations:
• Data assessment indicates that water quality is highly dependent on localized environmental conditions, and every effort should be made by industry to better understand their water sources through collection of water quality data and historical analysis.
• Scientific data collected and analyzed by our research team indicate that the main influential factors in the region evaluated were air temperature, solar radiation, rainfall and electrical conductivity (Appendix C). Surprisingly, the ability of a user to input electrical conductivity into developed models greatly increased risk assessment confidence. This lends itself towards the recommendation to industry to include EC in routine water quality monitoring plans to increase the likelihood of predicting coliform bacteria and E. coli in water sources.
• There is no “one-size fits all” model to predict water quality, however, the development of multiple models allows for a wider range of users based on available data and location. The complete model developed by our research team provides excellent predictions of water quality based on the data available and the region evaluated.
• Grower Apps can be useful tools that allow industry to make more informed decisions about their water sources from both a water use and sampling perspective.
• Future work should include testing of additional regional water sources and comparison of water quality data against development models (n=13) in order to validate their use in regions beyond the desert Southwest.