ScholarWorks at UMass Boston Identification and Use of Indicator Data to Develop Models for Marine-Sourced Risks in Massachusetts Bay IDENTIFICATION AND USE OF INDICATOR DATA TO DEVELOP MODELS FOR MARINE-SOURCED RISKS IN MASSACHUSETTS BAY A Dissertation Presented
Directed by Professor Robert Bowen The coastal watersheds around Massachusetts Bay are home to millions of people, many of whom recreate in coastal waters and consume locally harvested shellfish. Epidemiological data on food-borne illness and illnesses associated with recreational water exposure are known to be incomplete. Of major food categories, seafood has the highest recorded rate of associated foodborne illness. In total, the health impacts from these marine-sourced risks are estimated to
... ks are estimated to cost millions of dollars each year in medical expenses or lost productivity. When recorded epidemiological data is incomplete it may be possible to estimate abundance or prevalence of specific pathogens or toxins in the source environment, but such environmental health challenges require an interdisciplinary approach. v This dissertation is divided into four sections: (1) a presentation of two frameworks for organizing research and responses to environmental health issues; (2) an exploration of human population dynamics in Massachusetts Bay coastal watersheds from 2000 to 2010 followed by a review of, and identification of potential indicators for, five marine-sourced risks: Enterococcus bacteria, Vibrio parahaemolyticus bacteria, Hepatitis A Virus, potentially toxigenic Pseudo-nitzschia genus diatoms, and anthropogenic antibiotics; (3) an introduction to environmental health research in the context of a changing data landscape, presentation of a generalized workflow for such research with a description of data sources relevant to marine environmental health for Massachusetts Bay; and (4) generation of models for the presence/absence of Enterococcus bacteria and Pseudo-nitzschia delicatissima complex diatoms and model selection using an information-theoretic approach. This dissertation produced estimates of coastal watershed demographics and usage levels for anthropogenic antibiotics, it also demonstrated that Pseudo-nitzschia delicatissima complex diatoms may be present in any season of the year. Of the modeling generation and selection, the Enterococcus model performed poorly overall, but the Pseudo-nitzschia delicatissima complex model performed adequately, demonstrating high sensitivity with a low rate of false negatives. This dissertation concludes that monitoring data collected for other purposes can be used to estimate marine-sourced risks in Massachusetts Bay, and such work would be improved by data from purpose-designed studies. vi ACKNOWLEDGEMENTS This dissertation would not have been possible without the guidance, support, thoughtful discussions, and intellectual sparkle provided by friends and family over the years. The list below is an incomplete record of all those who have my sincere thanks. First and foremost to my parents and family for support. Also first is my 'academic parent' and advisor Dr. Bob Bowen for everything since we met at a UMB Open House, and of course to my amazing committee members Dr. Helen Poynton, Dr. Jarrett Byrnes, and Dr. Todd Swannack for their guidance and feedback. The inter-dependent nature of scientific work has never been more apparent to me than during this process, thanks to specialists in multiple fields who shared their time and expertise with me. Special thanks to Wendy Leo of the MWRA for data that was essential to this work, to Dave Borkman of the University of Rhode Island for information about phytoplankton monitoring in Massachusetts Bay, Mingshun Jiang for discussions about physical processes in Massachusetts Bay and cogent answers to my out-of-the-blue email questions, and J. Duff for. Thanks to the wonderful faculty and staff in the UMB EEOS Dept/SfE, and the Urban Harbors Institute, you are inspirational in the way you make interdisciplinary work happen. Thanks to all my classmates for being great people to learn with, especially K. for their good humorvii especially during long hours in the GIS lab. Special thanks to Tom Angus for years of collegial support, inspiration, and many many conversations about Pseudo-nitzschia. For friends who receive me with good humor despite my faults, the Library of Congress would be too small to catalog everything I should thank you for, this is for you Ali, Anne, Adrienne, Kay, Grace, Jenn, and Michelle for being my core support and inspiration. Thanks to A. Jones for biology conversations and measured thought; to JW for infinite patience and kind assistance; to B. Lapointe and family for friendship and support; to the inspirational Jurneys and Alaina, Britta, and Hardysimply the best teachers and friends underwater and on land -I would not be on this path without you.