National Weather Service United States Department of Commerce

NOAA Technical Memorandum NWS SR-206 (Updated Through 2002)


Donovan Landreneau
National Weather Service Office
Lake Charles, Louisiana

1. Introduction

 Tropical storms and hurricanes have affected every coastal state along the Atlantic and Gulf of Mexico from Texas to Maine. Even some inland states, such as Arkansas and Tennessee, have adversely experienced the effects of such storms. Anyone living in the eastern half of the U.S. should be aware of the effects of tropical storms and hurricanes, and how they could affect their lives and businesses.

This study was initiated for the purpose of addressing such concerns and others like them. By using a large part of the work done by Neumann, et al. (1993), several statistics are revealed, including frequency and return period of tropical storms and/or hurricanes which have affected the various coastal and inland states. The distance between landfalls for hurricanes is introduced to show which coastal state has the most concentration of landfalls over time, rather than just looking at which state has the highest number of landfalls.

2. Data Collection

The majority of the information used for this study came from Neumann, et al. (1999). The Atlantic track file (Jarvinen, et al. 1984) was used to complement this publication. Additional data for the years 1999-2002, as well as an updated Atlantic track file through 2002, were obtained from the National Hurricane Center Web site. Tropical depressions were excluded from this study due to the absence of data for these weak tropical systems.

Information on coastline length was obtained from Famighetti (1996). During the process of measuring the coastline, Connecticut was eliminated and had no measurement of a coastline. A CD-ROM mapping program, DeLorme (1997), was used to estimate a coastline length for Connecticut.

3. Analysis and Results

Hurricanes are ranked according to strength and by the amount of damage they cause. Table 1 is a brief description of the Saffir/Simpson hurricane intensity scale. The weakest hurricane is designated a Category One with a maximum sustained wind from 74 to 95 mph and an average storm surge of 4 to 5 ft above sea level. In contrast, a Category Five hurricane has a maximum sustained wind greater than 155 mph and a storm surge of greater than 18 ft. Storm depends on many factors such as the shape of the continental shelf just offshore, whether the hurricane makes landfall at high or low tide, and the location of the onshore and offshore winds relative to the eye of the hurricane.

Appendix A is a chronological list of hurricanes of various intensities which have struck from Texas to Maine for the years 1899-2002. In this study, a storm affects a state only once. For example, Hurricane Erin of 1995 made landfall on the east coast of Florida, moved over the peninsula, and struck the Florida panhandle two days later. Such situations are counted once for simplicity since it was the same storm. Table 2 further divides these direct hits according to category using the Saffir/Simpson scale. As one would expect, Category One hurricanes have struck most frequently with 63 landfalls, and Category Five storms are rare with only three landfalls. Notice the secondary maximum of Category Three landfalls.

The Glossary of Meteorology (Huschke 1959) and Elsner and Kara (1999) define frequency as the number of times a specified event occurs in a given series of observations, or period of time. In Table 2, the landfall frequency is represented by dividing the number of storms which made a landfall or direct hit by time. In this case, the time is 104 years. With the exception of the total for the United States, the results for each state are smaller than one, since no state averages a hurricane landfall or direct hit every year. The frequency of 1.62 for the United States signifies an average of one to two hurricane landfalls per year, somewhere along the Gulf or Atlantic coastline.

The above references define return period as the average time interval between the occurrence of a given quantity and that of an equal or greater quantity. This would represent the reciprocal of frequency, or the average number of years between each hurricane landfall. Table 2 shows, for example, that the average number of years between a hurricane landfall for Louisiana is 3.9. In contrast, the average number of years between landfalls in Georgia is 20.0. This gives an idea of the climatological average.

Table 2 also gives information on the coastline length of each state and the distance between each hurricane landfall. We assume storms are, on average, distributed randomly along the coast, and we obtain this figure by dividing the state's coastline length by the number of hurricanes which have affected that state. This value is introduced to show the concentration of landfalling hurricanes for each state. The smaller the number, the smaller the distance between each landfall resulting in a greater concentration of landfalls over time. For example, the total coastline length for Texas is 367 mi. Dividing this figure by the total number of hurricane landfalls, in this case 37, gives the distance between landfalls of 9.9 mi. Alabama, on the other hand, only had 11 hurricane landfalls during this time period. Since its coastline is only 53 mi, the resulting distance between landfalls is 4.8 mi, a higher concentration than Texas. The relative numbers for states should be used with some caution, however, because the assumption of random distribution may not be valid. Portions of the Atlantic and Gulf coasts of Florida, for instance, may have significantly different landfall frequencies.

A tropical storm has maximum sustained winds of 39 to 73 mph. In this analysis, a tropical storm is considered to have affected a state if the center of the storm intersected any portion of the state while the storm was at tropical storm intensity. This does not include periphery effects from storms affecting adjacent states or countries. To obtain these data, a very detailed analysis of each track was performed using the yearly track charts (Neumann, et al. 1999) in conjunction with the Atlantic track file (Jarvinen, et al. 1984). The results are shown in Appendix B.

Using the results in Appendix B, Table 3 was constructed to represent the total number of tropical storms which have affected each state, along with the frequency and return period. The methodology for computing frequency and return period is the same as Table 2, except each hit represents a tropical storm passing through any part of the state, and not just a coastal landfall. Table 4 shows the number of landfalling hurricanes and tropical storms which have affected each coastal state, along with the frequency and return period. Once again, the methodology for computing frequency and return period is the same as Tables 2 and 3.

4. Summary

The main purpose for this study is to show which states are more susceptible to tropical storms and hurricanes by using frequency of occurrence and the return period. The distance between landfalling hurricanes was introduced to compare which states have the most concentration of landfalls, or in other words, the smallest average distance between landfalls over the 104-year data span. I emphasize again, however, that the length of a state's coastline plays a significant role in the likelihood of a land-falling storm (i.e., exposure), and a longer coastline decreases the possibility that land-falling storms will strike with equal likelihood along all parts of the coast. The latter is especially true for Florida and Texas (Elsner and Kara 1999).

Over the years, certain cycles and patterns of tropical cyclones affecting coastal states can be observed. Such patterns include periods when most of the storms made landfall along the east coast of the U.S. verses the Gulf coast, during El Niño/La Niña events, etc. Such distinctions were not attempted in this paper, as many NWS offices within the studied area have performed local studies to address these issues. The subject is also well covered in many published papers and texts, see for example Elsner and Kara (1999) and its references.

Results presented in this study represent averages, and are not intended for use as a forecast of when the next tropical storm or hurricane will affect a state. Instead, they may serve as a general tropical cyclone climatology for coastal states as well as some inland states as noted in this analysis.

5. Acknowledgments

The author would like to thank the many authors and editors of the various publications used in this study, especially Neumann, et al. (1993), without which this analysis would have been close to impossible to complete. Many thanks also go to Lee Harrison (MIC) and Ken Falk (SOO) of WFO Shreveport, as well as Steve Rinard (MIC) of WFO Lake Charles for allowing time to complete this project and reviewing the format and content of this paper.

6. References

DeLorme, 1997: Street Atlas USA. Vers. 3.0. CD-ROM Computer Software.

Elsner, James B. and A. Birol Kara, 1999: Hurricanes of the North Atlantic - Climate and Society. Oxford University Press.

Famighetti, Robert, 1996: The World Almanac and Book of Facts. 1997 ed.

Huschke, Ralph E., 1959: Glossary of Meteorology. American Meteorological Society, Boston, MA.

Jarvinen, Brian R., Charles J. Neumann, and Mary A. S. Davis, 1984: A Tropical Cyclone Data Tape for the North Atlantic Basin, 1886-1983: Contents, Limitations, and Uses. NOAA Technical Memorandum NWS-NHC-22.

Neumann, Charles J., Brian R. Jarvinen, Colin J. McAdie, and Joe D. Elms, 1999: Tropical Cyclones of the North Atlantic Ocean, 1871-1998. Historical Climatology Series 6-2, Asheville, North Carolina, National Climatic Data Center.