Keywords

QMRA; modelling; Monte Carlo; ducks; non-point pollution

Location

Session H1: Environmental Modeling, Software, and Data to Support Quantitative Microbial Risk Assessments (QMRAs)

Start Date

16-6-2014 10:40 AM

End Date

16-6-2014 12:00 PM

Abstract

The development of QMRA for recreational waters often involves assessing the inputs from multiple sources, including human sewage, agricultural sources and wild animals. The concentrations of microorganisms in faeces from individual animals are highly variable leading to a broad spectrum of risk. When modelling the load of faecal microbes from a group of animals, based on data from individuals, the group distribution must be correctly accounted for. A further complication to QMRA is that pathogen concentrations in faeces have a high proportion of non-detects making them difficult to model using standard mathematical distributions. Moreover, when only limited concentration data are available, determining the appropriate mathematical distribution is difficult. These issues are explored in this study using the scenarios of E. coli and Campylobacter depositions into a stream from wild ducks. Using Monte Carlo methods, different models to describe the load of microorganisms deposited by ducks into a stream were compared. In addition, parametric and non-parametric methods were used to generate microbial concentrations in an individual ducks’ faeces. Our study demonstrated that calculating the load of faecal microorganisms from groups of animals for QMRA requires models that accurately reflect the loads of the whole group. Non- parametric methods can be used for pathogen concentrations that do not appear to fit a standard mathematical distribution or when determining the appropriate distribution is difficult.

 
Jun 16th, 10:40 AM Jun 16th, 12:00 PM

Accounting for Groups of Animals in QMRA of Recreational Waters

Session H1: Environmental Modeling, Software, and Data to Support Quantitative Microbial Risk Assessments (QMRAs)

The development of QMRA for recreational waters often involves assessing the inputs from multiple sources, including human sewage, agricultural sources and wild animals. The concentrations of microorganisms in faeces from individual animals are highly variable leading to a broad spectrum of risk. When modelling the load of faecal microbes from a group of animals, based on data from individuals, the group distribution must be correctly accounted for. A further complication to QMRA is that pathogen concentrations in faeces have a high proportion of non-detects making them difficult to model using standard mathematical distributions. Moreover, when only limited concentration data are available, determining the appropriate mathematical distribution is difficult. These issues are explored in this study using the scenarios of E. coli and Campylobacter depositions into a stream from wild ducks. Using Monte Carlo methods, different models to describe the load of microorganisms deposited by ducks into a stream were compared. In addition, parametric and non-parametric methods were used to generate microbial concentrations in an individual ducks’ faeces. Our study demonstrated that calculating the load of faecal microorganisms from groups of animals for QMRA requires models that accurately reflect the loads of the whole group. Non- parametric methods can be used for pathogen concentrations that do not appear to fit a standard mathematical distribution or when determining the appropriate distribution is difficult.