Introduction

It is frequently quoted that El Niño reduces hurricane activity in the Atlantic. The best modern reference is Gray, 1984, in which physical processes accompanying El Niño are used as an explanation for decreased hurricane activity. In this short note we analyze the frequency of hurricanes making landfall on the United States between the years 1949-1992. Only hurricanes are considered; not tropical storms. If El Niño has occurred in the previous winter, about 1 hurricane a year will hit the U.S. The probability of two or more hurricanes striking the U.S. in an El Niño year is 21 percent. The probability of 2 or more hurricanes striking the U.S. in a regular year is 46 percent. Therefore, it is 2.2 times more likely to have 2 or more landfalling hurricanes on the U.S. in a regular year than in an El Niño year.

What is an El Niño Year

El Niño means the occurrence of warm water along the equator in the Pacific Ocean. We prefer the identification of an El Niño event developed by the Japanese Meteorological Agency (JMA Atlas, 1991) rather than the Southern Oscillation Index which we believe to be very noisy. The JMA Index defines El Niño events based on sea surface temperature anomalies in the region 4deg. N to 4deg. S and 150deg. W to 90deg. W. An El Niño event is observed when the five month running average of SST anomalies is greater than 0.5deg.C for at least six consecutive months. In addition, the series of six consecutive months must begin before September, and must include October, November, and December. An El Niño typically starts in late summer and lasts for about a year, therefore, we choose the El Niño year to be the months of October to September. Such a definition allows us to analyze the impact of a particular ENSO event from its fully developed stage through the following year. Thus, in a particular hurricane season, the period June to September may be in an El Niño year while the period September through November may be in a normal year or vice versa. Table 1 gives the data (year and number) for the period 1949 - 1993. We also recognize the opposite of El Niño which is called El Viejo (La Niña). The definition of El Viejo is symmetric with the El Niño definition. If the JMA Index is 0.5deg. colder for 6 consecutive months then it is an El Viejo year.

Histograms of Hurricane landfall occurrence

The sample size is quite small; only 11 El Niño years and 22 neutral years. We use the bootstrap technique (Diaconis and Efron, 1983) to increase the sample size. Samples of size 11 are drawn randomly with instant replacement. This permits one to have an infinite number of samples. The expected value of the mean is equal to the actual mean of the original sample. The variance is conservative.

The histogram (figure 1) for the El Niño data and the normal data clearly indicate that both the means and variances are different. A t-test at 95% confidence level demonstrated that the means are significantly different. As noted above, the U.S. gets about 1 hurricane per year when El Niño has occurred in the previous winter, and about 2 per year otherwise. If we ask what is the probability of 2 or more landfalling hurricanes, we can use normal probability tables to determine that the probability is 21% after an El Niño and 44% after a neutral year. Thus, through a simple statistical technique, we can quantitatively relate the impact of warm water along the equatorial Pacific Ocean to the number of hurricanes striking the United States. If El Niño is identified the winter before the hurricane season, it will significantly reduce the number of hurricanes. The winter of 1994-95 was not an El Niño winter.

Acknowledgments

The Center for Ocean-Atmospheric Prediction Studies (COAPS) receives its base support from the Secretary of the Navy Grant to James J. O'Brien. Additional support for this work was received from the Office of Global Programs, NOAA. Todd Richards is supported by a Department of Defense ASSERT fellowship.

References

Diaconis, P. and B. Efron, 1983. Computer-intensive methods in statistics.

Sci. Amer., 248, 116-130.

Gray, W. M., 1984. Atlantic Seasonal Hurricane Frequency. Part I: El Niño

and 30 mb Quasi-Biennial Oscillation Influences. Mon. Weath. Rev., 112,

1649-1668

Japan Meteorological Agency, Marine Department. 1991. Climate Charts of

Sea Surface Temperatures of the Western North Pacific and the Global

Ocean. 51 pp.

List of Tables

Table 1. List of ENSO years (based on JMA-SST Index) and number of

Atlantic hurricanes striking the United States.

List of Figures

Figure 1. Relative frequency histogram of Atlantic hurricanes striking

the United States.