Date of Award

Spring 1-1-2014

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Aerospace Engineering Sciences

First Advisor

Robert R. Leben

Second Advisor

William J. Emery

Third Advisor

Karl E. Gustafson

Fourth Advisor

Benjamin D. Hamlington

Fifth Advisor

Jeffrey P. Thayer

Abstract

A total of 20 Loop Current eddy separation event dates were derived from Seasat and ERS-1 satellite altimetry, Coastal Zone Color Scanner chlorophyll-a images, Advanced Very High Resolution Radiometer sea surface temperature images, Horizon Marine, Inc. EddyWatch(TM) reports, and Climatology and Simulation of Eddies Eddy Joint Industry Project Gulf Eddy Model analyses spanning mid-1978 - 1992. There were many inconsistencies between the new "pre-altimetry" reanalysis dates derived from mostly non-altimeter data and dates published in past literature based on earlier versions of the pre-altimetry record. The reanalysis dates were derived from a larger compilation of data types and, consequently, were not as affected by intermittent and seasonal data outages common with past records. Therefore, the reanalysis dates are likely more accurate. About 30 Loop Current eddy separation dates were derived from altimetry data spanning 1993 - 2012. The pre-altimetry and altimetry reanalysis dates along with similar altimetry dates published in other literature exhibit statistically significant seasonality. Eddy separation events are more likely in the months March, August, and September, and less likely in December. Reanalysis event dates were objectively divided into "spring" and "fall" seasons using a k-means clustering algorithm. The estimated spring and fall season centers are March 2nd and August 23rd, respectively, with seasonal boundaries on May 22nd and December 3rd. The altimetry data suggest that Loop Current intrusion/retreat is dominantly an annual process. Loop Current metrics such as maximum northern boundary latitude and area are relatively high from January through about July and low in September and October. February metrics are statistically different than metrics in either October or November or both. This annual process is primarily driven by and dynamically linked to geostrophic currents seaward of the Campeche Bank shelf break forced by Kelvin waves generated on the southeast United States and Gulf of Mexico continental shelves. The dominant mode of an altimetric coastal empirical orthogonal function analysis of the southeast United States and Gulf of Mexico continental shelves indicates that the primary coastal signal covaries with the Loop Current and accounts for 65% and 85% of the annual variance in Loop Current northern latitude and area, respectively.

Share

COinS