ASK A SCIENTIST
Question: How likely is it to have a tornado in New York?
Answer: The quick answer to your question is 'very likely,' but let's look at how we can come to such a conclusion and what it means. Reasonably reliable records of tornado occurrences have been kept since 1950, so we have a relatively short period of time from which to determine likelihood. These records are available from the National Climatic Data Center (www.ncdc.noaa.gov). Since 1950, 359 tornadoes have been recorded for New York State. The traditional method of determining the likelihood or probability of a tornado occurrence is to divide the number of tornadoes in the record (359) by the period of record (57 years) and multiply that by 100. The result is the percent chance that a tornado will occur somewhere in New York State in any given year. Our result is 630%. Because this is greater than 100%, what this means is that New York will very likely experience a tornado, somewhere, every year. However, we have to interpret this carefully because the record of tornadoes does not show at least one occurrence every year, and in some years, there were numerous tornadoes. In fact, a severe thunderstorm can generate multiple tornadoes in a short period of time, which can affect different counties as the storm moves along its path. One problem is that we have only 57 years of record, and tornadoes occurred long before 1950. So, is the past 57 years representative of all tornado occurrences? We don't know. Another problem is that, during this time period, our ability to spot tornadoes has improved with Doppler radar and other technologies. As a result, the number of tornado occurrences in the United States shows an increase over time, though not consistently for every year. Part of this may be due to an increase in severe storms that generate tornadoes; part could be due to better technologies. We also are much better at relating characteristics of wind damage after an event to tornado activity, so sometimes events are classified as tornadoes after the fact. My best guess is that the increase in occurrences is a function of all of these and not merely of more tornado activity.
We must also look at where tornadoes have occurred in New York. With 62 counties in the state, we might consider an average, which would be about 6 tornadoes per county over the 57 years of record. Using the formula above, this gives a 10.5% probability per county per year. However, every county has not had similar experiences. Broome County has had seven tornadoes since 1950, one above the average, giving an annual probability of 12%. Other counties have had none. Chautauqua County has had the most, 24 tornadoes, resulting in a 42% annual probability. So the likelihood of a tornado occurrence is not equal throughout the state. Various aspects of a location can contribute to tornado occurrence. For instance, Chautauqua County has Lake Erie as its western border. The lake effect can contribute to more thunderstorms and to more severe thunderstorms, and therefore more tornadoes. Erie County, just north of Chautauqua, has a record of 18 tornadoes, the third highest for the state. Second is Suffolk County on Long Island, with 19 tornadoes.
Another important aspect of tornadoes is their magnitude, measured on the Fujita Scale, which runs from F0 to F5. Most of the tornadoes in New York, and indeed most tornadoes everywhere are low magnitude, less than F3. In New York, 28 out of the 359 tornadoes, or 8%, were F3 or F4. The probability of an F3 or greater is about 5% in any given year. There have been no F5 tornadoes in the state. Interestingly, eight of the F3 tornadoes occurred on the same day, May 5, 1983, within a five-hour span. Four of the five F4 tornadoes recorded in New York occurred on the same day, as well, this time July 10, 1989, affecting four different counties all within one hour. It is these high intensity tornadoes that cause the most loss and fortunately, New York does not experience many.
Ask a Scientist appears Thursdays. Questions are answered by faculty at Binghamton University. Teachers in the greater Binghamton area who wish to participate in the program are asked to write to Ask A Scientist, c/o Binghamton University, Office of Communications and Marketing, PO Box 6000, Binghamton, NY 13902-6000 or e-mail email@example.com. Check out the Ask a Scientist Web site at askascientist.binghamton.edu. To submit a question, download the submission form(.pdf, 460kb).