Defense Date

12-4-2019

Document Type

Capstone

Degree Name

M.S. Marine Environmental Sciences

First Advisor

Alexander Soloviev, Ph.D.

Second Advisor

Matthew Johnston, Ph.D.

Third Advisor

Bernard Riegl, Ph.D.

Abstract

Hurricane prediction is an evolving challenge that has seen much improvement over the years. While hurricane models have improved in predicting the path of storms, forecasts of hurricane intensity are unreliable due to the complexity of environmental data, lack of understanding of how relative humidity, vertical wind shear, hurricane structure and other possible factors affect intensity. Rapid Intensification (RI), which is a wind speed increase of +30 kts over a 24-hr period, can contribute to major destruction and loss of life to coastal communities affected by hurricanes, and is especially difficult to predict. Given the continued development of coastal regions and the threat of RI occurring without warning, it is imperative to better examine all possible factors that might influence hurricane RI to better understand and predict RI. The need for more research into RI was underscored by the devastation caused by rapidly intensifying hurricanes in the Caribbean and the east and gulf coast regions of the U.S. during the 2017 and 2018 hurricane seasons, which included the first landfall of a Category 5 hurricane (Hurricane Michael, 2018) since Hurricane Andrew in 1992. Recent studies examining the barrier layer (BL) of the water column and its relationship to hurricane intensification have shown that BLs favor RI, and that barrier layer thickness (BLT) may influence the storm’s intensity. To determine if BLs might improve the prediction of hurricane intensity, this study examined all hurricanes in the Atlantic Ocean, Caribbean, and Gulf of Mexico spanning the years 2000-2018. Using relevant HYCOM data, daily temperature (T), salinity (S), mixed layer thickness (mlp), isothermal layer thickness (mld), and BLT were examined to determine each factor’s horizontal distribution and possible influence on RI events occurring during the 139 hurricanes. Additional analysis was conducted on 12 randomly selected hurricanes (six of which had RI events and six of which did not) to determine if these factors, specifically BLT, act as significant predictors for RI events. Although no known link has been shown in previous research, this study also sought to determine if there is a correlation between RI and the horizontal variability of BLT and other key factors near the center of a hurricane (within 1 degree lat/lon).Though BLs can exist in any ocean, they are constantly changing and not always present. In this study, however, it was observed that BLs were present during all hurricanes in the Atlantic (2000-2018), whether they experienced a period of RI or not. Using an untested horizontal statistical analysis, this study shows that barrier layer thickness (BLT) does not appear to be a significant predictor of the probability of an RI event to occur, with no clear relationship shown between BLT and the magnitude of intensification, but these results cannot be taken as definitive. Given the limitations of this study, future research on hurricane RI should incorporate all known factors that impact hurricane intensity, testing each using multiple intensity models across all ocean basins.

ORCID ID

0000-0001-8109-9943

Download

Share

COinS