Abstract
To support Regulatory Impact Analysis of the United States Environmental Protection Agency (EPA), there is a need to characterize the occurrence of Cryptosporidium in the nation's source waters. The Information Collection Rule (ICR) and two supplemental surveys were designed by EPA to provide such occurrence data. A Poisson hierarchical model in Bayesian scheme, which allows for several covariates, is applied to the data collected in the two supplemental surveys. We employ an MCMC computing strategy. Our results show that flowing streams have higher mean Cryptosporidium concentrations than other water types, and medium systems have higher mean Cryptosporidium concentrations than large systems; the mean Cryptosporidium concentration does not change over time significantly; and the Cryptosporidium concentration is positively associated with sample turbidity.
Get full access to this article
View all access options for this article.