Skip to main content
ARS Home » Plains Area » Clay Center, Nebraska » U.S. Meat Animal Research Center » Meat Safety & Quality Research » Research » Publications at this Location » Publication #370574

Research Project: Assessment of Genotypic and Phenotypic Factors for Foodborne Pathogen Transmission and Development of Intervention Strategies

Location: Meat Safety & Quality Research

Title: Predicting Escherichia coli loads in cascading dams with machine learning: An integration of hydrometeorology, animal density and grazing pattern

Author
item ABIMBOLA, OLUFEMI - University Of Nebraska
item MITTELSTET, AARON - University Of Nebraska
item MESSER, TIFFANY - University Of Nebraska
item Berry, Elaine
item BARTELT-HUNT, SHANNON - University Of Nebraska
item HANSEN, SAMUEL - University Of Nebraska

Submitted to: Science of the Total Environment
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 3/10/2020
Publication Date: 6/20/2020
Citation: Abimbola, O.P., Mittelstet, A.R., Messer, T.L., Berry, E.D., Bartelt-Hunt, S.L., Hansen, S.P. 2020. Predicting Escherichia coli loads in cascading dams with machine learning: An integration of hydrometeorology, animal density and grazing pattern. Science of the Total Environment. 722:137894. https://doi.org/10.1016/j.scitotenv.2020.137894.
DOI: https://doi.org/10.1016/j.scitotenv.2020.137894

Interpretive Summary: Surface water that is contaminated with fecal material is a common source of transmission of many pathogens that cause human and animal disease. The concentration of nonpathogenic total Escherichia coli in water is used as an indicator of the extent of fecal contamination, and simulation models for predicting E. coli loads in water are often used for the assessment and management of natural water systems. However, accurate prediction of E. coli loads is challenging due to the many physical, chemical and biological variables that can affect E. coli sources, occurrence, and numbers in surface water. This study used the novel approach of integrating hydrometeorological variables such as air temperature, water temperature, rainfall, water depth, and flowrate along with animal management variables such as animal density and pasture utilization in the E. coli prediction model to increase accuracy for a stream that flows through a cattle production area. Machine learning techniques were used to develop E. coli predictive models which were sufficiently accurate to be useful. Models with the fewest errors have the potential to be used to predict E. coli concentration for intervention plans and monitoring programs to implement effective best management practices to protect stream water quality in grazing cattle areas.

Technical Abstract: Accurate prediction of Escherichia coli contamination in surface waters is challenging due to considerable uncertainty in the physical, chemical and biological variables that control E. coli occurrence and sources in surface waters. This study proposes a novel approach by integrating hydro-climatic variables as well as animal density and grazing pattern in the feature selection modeling phase to increase E. coli prediction accuracy for two cascading dams at the US Meat Animal Research Center (USMARC), Nebraska. Predictive models were developed using regression techniques and an artificial neural network (ANN). Two adaptive neuro-fuzzy inference system (ANFIS) structures including subtractive clustering and fuzzy c-means (FCM) clustering were also used to develop models for predicting E. coli. The performances of the predictive models were evaluated and compared using root mean squared log error (RMSLE). Cross-validation and model performance results indicated that although the majority of models predicted E. coli accurately, ANFIS models resulted in fewer errors compared to the other models. The ANFIS models have the potential to be used to predict E. coli concentration for intervention plans and monitoring programs for cascading dams, and to implement effective best management practices for grazing and irrigation during the growing season.