Modeling coastal pathogen populations: Advancements in predictive modeling of Vibrio vulnificus in a tropical urban estuary in Honolulu, Hawaiʻi

dc.contributor.advisor Nelson, Craig E. Bullington, Jessica
dc.contributor.department Oceanography 2021-09-30T18:19:06Z 2021-09-30T18:19:06Z 2021 M.S.
dc.subject Biological oceanography
dc.subject Ecology
dc.subject Microbiology
dc.subject Climate Change
dc.subject Estuary
dc.subject Hawaii
dc.subject Pathogen
dc.subject Predictive Modeling
dc.subject Vibrio vulnificus
dc.title Modeling coastal pathogen populations: Advancements in predictive modeling of Vibrio vulnificus in a tropical urban estuary in Honolulu, Hawaiʻi
dc.type Thesis
dcterms.abstract The south shore of Oʻahu, Hawaiʻi is one of the most visited coastal tourism areas in the United States with some of the highest instances of recreational waterborne disease. The pathogenic bacterium, Vibrio vulnificus, lives in the warm temperature and intermediate salinity waters of the estuarine Ala Wai Canal in Honolulu which surrounds the heavily populated tourism center of Waikīkī. Heavy rainfall has been shown to transport water from the canal into the nearshore coastal system threatening human and ecological health as well as the state's tourism-based economy. I developed a biological-physical statistical model to predict V. vulnificus dynamics using environmental measurements that can be captured with moored oceanographic sensors in real time. I parameterized and validated the model using a novel approach that integrates data from three sources: 1) monthly profile measurements at each of 18 sites spanning an annual cycle which included in situ discrete sampling at 3 depths (surface, pycnocline, and bottom water) and continuous sensor profiling, 2) moored instrument continuous time series from the Pacific Islands Ocean Observing System (PacIOOS) and 3) the Waikīkī Regional Ocean Modeling System (ROMS). Discrete samples (n = 213) were analyzed for 36 biogeochemical variables and V. vulnificus density using quantitative PCR of the hemolysin A toxicity gene (vvhA). I found significant monthly variation in V. vulnificus density ranging over 4 orders of magnitude largely predictable by seasonality in biogeochemistry. The best multiple regression model of V. vulnificus concentration, explaining 70% of variation, included salinity, five-day average rainfall, daily maximum air temperature, silicate, dissolved oxygen and two metrics of dissolved organic matter fluorescence (visible humic-like fluorescent dissolved organic matter and the Humification Index). This result suggests a consistent response of V. vulnificus to freshwater inputs, particularly groundwater, and changes in organic matter resources. These linear models, paired with the PacIOOS time series and ROMS products, were then implemented to predict V. vulnificus density in the canal and coastal waters in real time across the nearshore Waikīkī region and were validated against measured values. Long-term climate model projections of local rainfall and air temperature were used to predict an overall increase in V. vulnificus density of approximately 20 to 40 % by 2100. Improving these predictive models of microbial populations is critical for management of waterborne pathogen risk exposure, particularly in the wake of a changing global climate.
dcterms.extent 86 pages
dcterms.language en
dcterms.publisher University of Hawai'i at Manoa
dcterms.rights All UHM dissertations and theses are protected by copyright. They may be viewed from this source for any purpose, but reproduction or distribution in any format is prohibited without written permission from the copyright owner.
dcterms.type Text
Original bundle
Now showing 1 - 1 of 1
No Thumbnail Available
3.98 MB
Adobe Portable Document Format