Date of Award

6-1-2014

Document Type

Campus Access Thesis

Degree Name

Master of Science (MS)

Department

Marine Sciences and Technology

First Advisor

Robert F. Chen

Second Advisor

George B. Gardner

Third Advisor

Crystal B. Schaaf

Abstract

The EPA has provided guidelines for the safety of recreational contact with river and beach water based on concentrations of the fecal indicator bacteria Escherichia coli and Enterococcus. However, it generally takes 24 hours to analyze water samples for these bacteria, and concentrations of the bacteria are known to fluctuate dramatically at a single location in timeframes of hours. Therefore, water quality warnings based on water sampling are inherently associated with timing mismatches, and are not as effective as they could be. Some monitoring organizations have tried to predict bacterial concentrations by modeling the correlation of bacterial levels with easily measurable hydro-meteorological parameters. This practice often results in reasonable forecasts that are 80-90% accurate, but this method is still plagued by timing issues because it depends on personnel to measure and input the variables daily. In this study, we tested how predictions and public warning systems can be improved by networking real-time continuous hydro-meteorological sensors with online automated water quality reports. Two sites have been studied, the Charles River basin in Boston, MA, and Wollaston Beach in Quincy Bay, MA, both of which are located in populous urban centers and are used heavily for recreation. Hydro-meteorological sensors were deployed at both locations. Data from these sensors were combined with data from other real-time sensors, such as USGS flow gauges and NOAA tide gauges, to feed an automated online model and warning system. The results of this study indicate that real-time data can dramatically improve the statistical sensitivity of predictive models, and that automation delivers powerful improvements in temporal coverage and significance to the public.

Comments

Free and open access to this Campus Access Thesis is made available to the UMass Boston community by ScholarWorks at UMass Boston. Those not on campus and those without a UMass Boston campus username and password may gain access to this thesis through resources like Proquest Dissertations & Theses Global or through Interlibrary Loan. If you have a UMass Boston campus username and password and would like to download this work from off-campus, click on the "Off-Campus UMass Boston Users" link above.

Download
Off-Campus UMass Boston Users

Share

COinS