Climatology of Tornadoes Associated with Gulf Coast-Landfalling Hurricanes
Moore, Todd W., Dixon, Richard W., The Geographical Review
Tornadoes associated with hurricane landfalls, known as "hurricane-tornadoes" (IITS), are a notable hazard. Their infrequent occurrence and spatial dispersion set them apart from oilier hurricane-related hazards, and they have an impact not only on coastal regions but also on areas beyond the immediate environment of the hurricane, even far inland (Hill, Malkin, and Schulz 1966; Novlan and Gray 1974; Gen--try 1983; McCaul 1991; Schultz and Cecil 2009). In addition, they have the potential lo exacerbate fatalities and economic losses associated with landfalling hurricanes. HTs have contributed up to 10 percent of the overall fatalities caused by their associated hurricane (Novlan and Gray 1974). In September 2004, tornadoes associated with Hurricane Ivan were responsible for al least eight fatalities and seventeen injuries (Stewart 2005). And in 1980, the tornadoes associated with Hurricane Allen produced more than $70 million in losses, and in 2005 the tornadoes associated with Hurricane Cindy caused more than $40 million in losses (Gentry 1983; Stewart 2006).
HTs have been the focus of diverse research. The majority of the publications were case studies (for example, Gray 1919; Barbour 1924; Hills 1929; Malkin and Galway 1953; Sadowski 1962; Rudd 1964; Orton 1970; McCaul 1987; Suzuki and others 2000) or themalic studies that focus on the physical mechanisms contributing to HAT, on their geophysical characteristics, and on techniques for detecting HTS (for example, Weiss 1985; McCaul 1991; McCaul and Weisman 1996; Spratt and others 1997; Bogner, Barnes, and Franklin 2000; Curtis 2004; McCaul and others 2004; Eastin and Link 2009). Only a few studies, particularly in recent years, have provided long-term climatological descriptions of his (Smith 1965; Hill, Malkin, and Schulz 1966; Novlan and Gray 1974; Gentry 1983; McCaul 1991; Verbout and others J007; Schultz and Cecil 2009).
Aside from providing normal characteristics, HT climatological studies have attempted to distinguish patterns and linkages within IIT datasets aggregated from multiple storm events spanning large areas and long periods. For instance, one study presented a twenty-five-year (1948-1972) climatology of 373 HTS associated with last Coast- and Gulf Coast-landfalling hurricanes and tropical storms (Novlan and Gray 1974). Another examined a longer and more current (T954-2004) temporal record, included 1,123 tornadoes, but focused specifically on IIT outbreaks and their association with existing meteorological conditions (Verbout and others 2007).The most recent study, a fifty eight-year (1950 -2007) climatology of his associated with East Coast and Gulf Coast-landfalling hurricanes, included 1,767 tornadoes (Schultz and Cecil 2009).
Climatological studies have revealed common IIT characteristics, including temporal and spatial distributions. However, some of the previous ht climatological studies are of limited utility because of their short and outdated temporal record (Smith 1965; Hill, Malkin, and Schulz 1966; Novlan and Gray 1974). Much of this is likely due to the availability of data and the tedious nature of developing large datasets. Recent studies have covered longer temporal records but have combined HTA associated with both East Coast and Gulf Coast-landfalling hurricanes (McCaul 1991; Verbout and others 2007; Schultz and Cecil 2009). Previous studies indicated that Gulf Coast-landfalling hurricanes are more likely to produce tornadoes than are East Coast-landfalling hurricanes, due to differences in orientation of the right front quadrant of the hurricane to the coastline at landfall (Novlan and Gray 1974; Gentry 1983}. Therefore, Gulf Coast- and East Coast-landfalling hurricanes may differ with respect to IIT climatology; and, thus, combining the two may lead to mixed statistical distributions. The concept of mixed distributions poses the idea dial observations sampled from a population that is assumed to be homogeneous can also be interpreted as a composite sample representing multiple processes or subpopulations, each with its own distribution (Hirschboeck 1987). …