A Poisson random field model of pathogen transport in surface water

Abstract

To address the uncertainty associated with microbial transport and surface water contamination events, we developed a new comprehensive stochastic framework that combines processes on the microscopic (single microorganism) and macroscopic (ensembles of microorganisms) scales. The spatial and temporal population behavior is modeled as a nonhomogeneous Poisson random field with Markovian field dynamics. The model parameters are based on the actual physical and biological characteristics of the Cryptosporidium parvum transport process and can be extended to cover a variety of other pathogens. Since soil particles have been shown to be a major vehicle in microbial transport, a U.S. Department of Agriculture approved erosion model (Water Erosion Prediction Project) is incorporated into the model. Risk assessment is an integral part of the stochastic model and is conducted using a set of simple calculations. Poisson intensity functions and correlations are computed. The results consistently indicate that surface water contamination events are transient, with traveling high peaks of microorganism concentrations. Correlations between microorganism populations at different points in time and space reach relatively significant levels even at large distances from one another. This information is aimed to assist water resources management teams in the decision‐making process to identify the likely timing and locations of high‐risk areas and thus to avoid collection of contaminated water.