Functional Evaluation of Three Manure-Borne Indicator Bacteria Release Models with Multiyear Field Experiment Data

Area of Science:

• Environmental Microbiology
• Water Quality Modeling
• Agricultural Hydrology

Background:

• Escherichia coli (E. coli) is a key indicator of fecal contamination in water quality assessments.
• Accurate modeling of E. coli transport from agricultural fields is crucial for water quality management.
• Existing E. coli release submodels have primarily been validated using laboratory or small-plot data.

Purpose of the Study:

• To evaluate and compare the field-scale performance of three manure-borne bacteria release submodels for E. coli.
• To integrate and test exponential release (EM), Bradford and Schijven (B-S), and Vadas-Kleinman-Sharpley (VKS) submodels within the KINEROS2/STWIR export model.
• To assess the efficacy of different submodels in simulating E. coli export from land-applied dairy manure.

Main Methods:

• Dairy manure was surface-applied annually for six years on a 0.28 ha experimental field.
• Two irrigation events followed each manure application to simulate runoff conditions.
• E. coli concentrations in manure, soil, and runoff were analyzed; model performance was assessed using Akaike criterion, R², and RMSE.

Main Results:

• The percentage of exported manure-borne E. coli ranged from 0.1% to 10%, influenced by irrigation timing relative to runoff initiation.
• The Vadas-Kleinman-Sharpley (VKS) submodel demonstrated superior performance, being preferred in 55% of evaluations.
• Two-parametric submodels (VKS and B-S) generally outperformed the single-parameter exponential release (EM) submodel.

Conclusions:

• The Vadas-Kleinman-Sharpley (VKS) submodel is recommended for field-scale E. coli export modeling from manure.
• Two-parametric models offer improved accuracy over single-parameter models for simulating manure-borne E. coli release.
• Accurate submodel selection enhances the reliability of microbial water quality assessments in agricultural watersheds.