A modelling framework for setting criteria of Microcystin violations in the Bay of Quinte

Session: 06. - Ecosystem Health and Recovery of the Bay of Quinte, Lake Ontario

Yuko Shimoda, University of Toronto Scarborough, [email protected]
Noreen Kelly, University of Toronto Scarborough, [email protected]
Susan Watson, University of Waterloo, [email protected]
Arthur Zastepa, Environment and Climate Change Canada, [email protected]
Aisha Javed, University of Toronto Scarborough, [email protected]
Dong-Kyun Kim, University of Toronto Scarborough, [email protected]
George Arhonditsis, University of Toronto Scarborough, [email protected]

Abstract

The Bay of Quinte experiences frequent exceedances of the total microcystin standard of 1.5 µg L-1 for drinking water. Our study introduces a Bayesian segmented modelling framework showing that chlorophyll a concentrations greater than 7-8 ?g L-1, TP concentrations greater than 25-30 ?g L-1, NH4 concentrations lower than 30 ?g L-1, water temperatures greater than 20oC, and wind speed  lower than 35 km hr-1 increase significantly the likelihood of microcystin standard violations. Probabilistic criteria (percentile standards) are increasingly used in environmental management, and rules for compliance with these standards are generally derived from classical one-sided hypothesis tests in which the p-value is used as a weight-of-evidence measure against the tested hypothesis. However, in developing such rules, we need to consider both risks of reaching false conclusions: the manager's risk (Type I error) and the consumer's risk (Type II error). Our objective is to develop a novel Bayesian framework that provides compliance assessment with microcystin criteria. We argue that it can give simpler and more direct information to a compliance assessor, even though it does call for a subjective element of information the assessor's belief regarding the probability of compliance before data become available.

1. Keyword
Bay of Quinte

2. Keyword
Microcystis

3. Keyword
drinking water

4. Additional Keyword
Criteria setting