You are here

Superensemble Forecasts of Hurricane Track and Intensity Using a Suite of Mesoscale Models

Title: Superensemble Forecasts of Hurricane Track and Intensity Using a Suite of Mesoscale Models.
Name(s): Kramer, Melanie A., author
Krishnamurti, T. N., professor directing thesis
Hart, Robert, committee member
Ruscher, Paul, committee member
Department of Earth, Ocean and Atmospheric Sciences, degree granting department
Florida State University, degree granting institution
Type of Resource: text
Genre: Text
Issuance: monographic
Date Issued: 2008
Publisher: Florida State University
Florida State University
Place of Publication: Tallahassee, Florida
Physical Form: computer
online resource
Extent: 1 online resource
Language(s): English
Abstract/Description: The tropical cyclone superensemble has provided skillful forecasts for Atlantic tropical cyclone activity for several years. Until very recently, the tropical cyclone superensemble had only been run using a suite of large scale models. Using large scale models within the Florida State Superensemble produced noticeable improvement over each of the respective member models with respect to track and intensity forecasting. The more recent development of operational tropical cyclone mesoscale models led to the development of the mesoscale tropical cyclone Florida State Superensemble, which is utilized for this study. This study uses a combination of four different mesoscale models. One model is currently being run operationally for the National Hurricane Center and the other three models are in-house models run at Florida State University. This research includes most of the tropical cyclones that occurred during the 2004, 2005, and 2006 Atlantic hurricane seasons. There are twenty six storms available for study during that period and most storms have a 0000 UTC and 1200 UTC initialization. This provides the Florida State Superensemble with fifty-seven forecast cases. For each storm the Florida State Superensemble issues track, minimum sea level pressure, and maximum wind forecasts out to seventy-two hours at six hourly intervals. These tracks and intensities are compared to the "best track" analysis as determined by the National Hurricane Center. The results show that the tropical cyclone Florida State Superensemble can provide accurate forecasts when using a suite of mesoscale models, though increased work needs to go into improving the accuracy of the member models, which on average have slightly higher errors than the similarly run Florida State Superensemble with large scale member models. This study quantifies the errors of the Florida State Superensemble, ensemble mean, and the respective mesoscale models, compares them to the large scale models, and examines several ways that the mesoscale tropical cyclone Florida State Superensemble can be enhanced for future work.
Identifier: FSU_migr_etd-2877 (IID)
Submitted Note: A Thesis submitted to the Department of Meteorology in partial fulfillment of the requirements for the degree of Master of Science.
Degree Awarded: Summer Semester, 2008.
Date of Defense: May 23, 2008.
Keywords: Tropical Cyclone Forecasting, Model Verification, Tropical Cyclone Intensity, Tropical Cyclone Track, Tropical Cyclone Model Limitations
Bibliography Note: Includes bibliographical references.
Advisory Committee: T. N. Krishnamurti, Professor Directing Thesis; Robert Hart, Committee Member; Paul Ruscher, Committee Member.
Subject(s): Meteorology
Persistent Link to This Record:
Use and Reproduction: This Item is protected by copyright and/or related rights. You are free to use this Item in any way that is permitted by the copyright and related rights legislation that applies to your use. For other uses you need to obtain permission from the rights-holder(s). The copyright in theses and dissertations completed at Florida State University is held by the students who author them.
Host Institution: FSU

Choose the citation style.
Kramer, M. A. (2008). Superensemble Forecasts of Hurricane Track and Intensity Using a Suite of Mesoscale Models. Retrieved from