Premium
Assessing the efficacy of environmental restoration via noninferiority: A Bayesian perspective
Author(s) -
Robinson Timothy J.,
Wulff Shaun S.,
Webb Michael,
Banda Jo Ann,
Nair Sreejayn
Publication year - 2019
Publication title -
quality and reliability engineering international
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.913
H-Index - 62
eISSN - 1099-1638
pISSN - 0748-8017
DOI - 10.1002/qre.2535
Subject(s) - environmental science , watershed , tributary , pollution , hydrology (agriculture) , water quality , environmental resource management , geography , computer science , ecology , cartography , engineering , geotechnical engineering , machine learning , biology
The Ashtabula River covers an area that lies in northeast Ohio, USA, flowing into the central basin of Lake Erie at the city of Ashtabula where its drainage covers an area of 355 km 2 . In the mid‐1900s, several chemical production companies began operation along tributaries of the Ashtabula, and over time, discharges from these facilities left the lower Ashtabula River heavily contaminated with polychlorinated biphenyls (PCBs), heavy metals, and a suite of other agents. This pollution lead to the designation of the lower Ashtabula River as one of the most contaminated sites in the Great Lakes watershed, as defined in the 1987 Great Lakes Water Quality Agreement. This manuscript is a case study used to demonstrate the advantages of noninferiority testing over the classical hypothesis testing framework to demonstrate when one or more biological variables at the impacted site are no worse or inferior to the distribution of the same biological variable(s) at a reference site. In order to implement this approach, it is essential to define a tolerance level that represents a meaningful difference in parameter values. We demonstrate the use of Bayesian logistic regression for developing a data‐driven approach for tolerance level determination when consensus on the tolerance level value cannot be reached.