TideTrack
Click the Poster to View Full Screen, Right click to save image
Liam Healy
CoPIs:
Tzu-Han Lin, Joao Borges, Max Caverly
College:
The Dorothy and George Hennings College of Science, Mathematics, and Technology
Major:
Computer Information Systems
Faculty Research Advisor(s):
Jing-chiou Liou
Abstract:
The goal of this project is to utilize machine learning algorithms to develop an AI model
for providing predictions of Harmful Algae Blooms (HABs) severity in the West Florida Shelf
area of the Gulf of Mexico. This information will be displayed via a software system with an
interactive map showing both the predictions and historical data on blooms in the area. HABs
due to the algae Karenia Brevis, happen seasonally in the Gulf under similar circumstances
every year from roughly August to December. Official information from the National Oceanic and
Atmospheric Administration (NOAA), including wind speed/direction, historical cell counts, water
salinity, and water temperature, will be fed into our system. Predictions of the severity will be
made based on this data, a similar system exists from the NOAA but only looking ahead by one
day. The TideTrack system will predict into a longer time frame, ten days at the minimum.
Karenia Brevis releases a neurotoxin into the water it inhabits, poisoning the ocean life in
the area making them unfit for human consumption. This causes these blooms to be particularly
damaging, both to people’s health and the economic viability of the area. Fisheries operating in
the Gulf have an incentive to avoid these blooms in order to not lose revenue due to wasted
catches, labor, and wear on fishing equipment. In addition to that, scientists studying such
blooms in the area can save on labor costs by seeing if HABs are likely in an area before
committing to field work. The severity predictions displayed on this system will help these
groups avoid these potential economic losses.