You are here

On the Obscured Relationship between Size and Intensity of Tropical Cyclones

Title: On the Obscured Relationship between Size and Intensity of Tropical Cyclones: A Preliminary Study.
Name(s): Hathaway, Nikki Marie, author
Cai, Ming, 1957-, professor directing thesis
Hart, Robert E. (Robert Edward), committee member
Liu, Guosheng (Professor of Earth, Ocean and Atmospheric Science), committee member
Florida State University, degree granting institution
College of Arts and Sciences, degree granting college
Department of Earth, Ocean, and Atmospheric Science, degree granting department
Type of Resource: text
Genre: Text
Master Thesis
Issuance: monographic
Date Issued: 2017
Publisher: Florida State University
Place of Publication: Tallahassee, Florida
Physical Form: computer
online resource
Extent: 1 online resource (65 pages)
Language(s): English
Abstract/Description: Both intuition and previous statistical analysis suggests that in general, hurricane size tends to increase with intensity. However, such statistical correlation, including the statistical significance and even the sign of the correlation, between hurricane size and intensity strongly depends on the sample hurricanes in the data pool for the correlation analysis. For example, there are ample instances when a hurricane at different times can have a similar size but differ in terms of intensity and vice versa. Therefore, predictions based on intuition or statistics often fail when considering an individual hurricane. In this thesis research, we attempt to apply a theoretical model in conjunction with observational case studies to gain insight on the main factors that make the relationship between both hurricane size and intensity, obscured. This theoretical model will apply an analytical analysis of the inertial instability neutral radial profile of an isolated gradient-wind balanced circular vortex in an f-plane shallow water equation model, which shows that the relationship between the size and the maximum tangential wind speed is not unique, because the size also depends on the radius of maximum wind. The radial profile of wind under neutral conditions of inertial instability reveals that hurricane size and intensity can have either a positive, near-zero, or negative correlation depending on the sample of hurricanes in the dataset from which such correlation is obtained. The main conclusion derived from the theoretical model is that the relationship between hurricane size and intensity can be obscured due to only one specific factor (i.e., the radius of maximum wind) that also influences the size. The theoretical model also predicts that the latitudinal position only weakly obscures the relationship, as long as the hurricane is not too close to the equator. We have examined whether the size inferred from the radial profile of inertially neutral wind would be able to capture its observational counterparts. Specifically, we examined five selected hurricanes derived from the Extended Best Track (EBT) Data, namely Katrina (2005), Ike (2008), Gustav (2008), Sandy (2012), and Joaquin (2015). We have performed a correlation analysis on the observed size and the size predicted by the simple theoretical model by using the information of maximum wind speed and its radius of each of the five hurricanes throughout the phases of each tropical cyclone’s (TC) life cycle. We found that the size obtained from the barotropic inertially neutral radial profile underestimates the size of observed hurricane by a factor of 2-2.5. This suggests that the observed hurricane wind’s radial profile does not follow angular momentum conservation or an air parcel would lose angular momentum as it converges towards the eyewall, mainly due to surface drag and eddy-mixing processes. This finding also implies that there are other parameters besides these three factors (intensity, radius of maximum wind, and latitude) that influence an individual hurricane size. This implies that the relationship between size and intensity is more complex than that predicted by the simple theoretical model. Our analysis suggests that about 1/3 (48 out of 174) of the observed cases show that other factors may strongly affect hurricane size. By removing these 48 data points that are indicative of possible strong impacts from the external factors, the R-squared value of the linear regression line between the observed size and the size predicted by the theoretical model increases substantially (from R2 = 46.3% to 71.5% on average). The inspection of the timing and location of these “external” data points indicate that they often occur in situations when (i) encountering big islands or land mass (e.g., Cuba for Ike) and (ii) undergo a very rapid weakening/intensifying transition (Joaquin). Therefore, the size information predicted by the simple theoretical model does capture the size record for most the track records (126 out of 174), suggesting that the most important factors that influence hurricane size are both maximum wind speed and the radius of maximum wind speed.
Identifier: FSU_SUMMER2017_Hathaway_fsu_0071N_13990 (IID)
Submitted Note: A Thesis submitted to the Department of Earth, Oceanic and Atmospheric Science in partial fulfillment of the requirements for the degree of Master of Science.
Degree Awarded: Summer Semester 2017.
Date of Defense: June 22, 2017.
Keywords: Intensity, Size, Tropical, Tropical Cyclone
Bibliography Note: Includes bibliographical references.
Advisory Committee: Ming Cai, Professor Directing Thesis; Robert Hart, Committee Member; Guosheng Liu, Committee Member.
Subject(s): Meteorology
Persistent Link to This Record:
Owner Institution: FSU

Choose the citation style.
Hathaway, N. M. (2017). On the Obscured Relationship between Size and Intensity of Tropical Cyclones: A Preliminary Study. Retrieved from