Fact Sheet - El Nino/La Niata and Kentucky Tornadoes
Joshua D. Durkee Research Assistant and Glen Conner
State Climatologist Emeritus for Kentucky
El Niato refers to a prolonged period of unusually high sea surface temperature over the eastern tropical and equatorial Pacific Ocean. This warm weather phase causes the trade winds to slacken or even reverse in direction. This drives the warm water eastward toward South America. This weather pattern can persist from longer than a year, and occurs about every three to seven years (1).

La Niata is found in the central and eastern tropical zones of the Pacific Ocean. This cold weather phase consists of a period of strong trade winds and low sea surface temperatures and occurs about half as frequent as El Niato (2). Its causes are the opposite of El Niato (1). A normal weather phase exists when neither El Niato or La Niata are occurring.

Each El Niato, La Niata, and normal phase that occurred from 1951-1998 is shown on achart prepared by NOAA's Climate Prediction Center (3). Each year is separated into four three-month cells (e.g., AMJ= April, May, June). A W represents El Niato for warmer than normal ocean waters. A C represents La Niata for colder than normal ocean waters. The Nspaces have been added to the normal periods not affected by either El Niato or La Niata. The plus or minus signs following a W or aC, symbolizes either strong or weak events.

From the 485 recorded tornadoes in Kentucky from 1951-1998, many initial touchdown locations were not available in this research. Therefore, for the actual analysis in this study, tornado paths were not considered. For example, two counties that shared the same tornado experience, were recorded as occurrences in each county.

Since each tornado was recorded by date, the next step was to transpose the basic data into the format of the chart. This shows the total tornadoes that occurred in each cell from 1951-1998. Data by county for each of the three weather phases (El Niato, La Niata,Normal) were then placed into map form.
Ratio Method
Each of the three weather phases has occurred at different rates. For example, the data showed that the total for W is greater than the total for C. One might conclude that El Niato produces more tornadoes than La Niata, but this may not be true; El Niato may have occurred more times than La Niata, producing a higher total. Therefore, the total number of tornadoes was compared to the total occurrence of the weather phase.

These ratios were generated through a standardized T-Testing method. The T-Test was used to compare values to determine if there were any significant differences between the two data sets. The objective was to compare tornado values in each phase to try to detect any differences between outcomes. For example, the values for the normal phase were compared to the values in the El Niato and the La Niata phases. Then the La Niata phase was compared to the El Niato phase. This way each phase was compared to each of the other two. The T-Test decided if there was a significant difference in outcomes between each of the three phases.
Climatic Division Method
The next step in this research was to check the spatial distribution of tornadoes by county for any patterns. The objective in this part of the study was to compare tornado sums in each division.

Kentucky has four climate divisions: Western, Central, Bluegrass, and Eastern. County Totals were assessed in each division using the same ratio method. For example, the Western division during La Niata was compared its totals during the El Niato and Normalphases. Next, El Niato was compared to the normal phase. Again, this way each phase was compared to the other two. The T-Test was applied to check for differences between the Western division, in all three phases. Any differences between the Western division and all three phases would suggest a shift. No differences suggests similar conditions and no controlling factors in touchdown sums and locations.
After comparing the values between each weather phase by using the T-Testing method, the data sets for El Niato, La Niata, and normal phases were found to be similar. In fact, the data sets showed no significant difference among them. Therefore, the T-Test suggested that the difference among El Niato, La Niata, and normal year tornado production was not significant. This suggested that the phases were alike in the production of tornadoes in Kentucky.

For the climatic division method, the statistics showed each division, in each phase had no significant difference. For example, the Western division's tornado production during El Niato, La Niata, and normal years was about the same. The significance levels varied somewhat, however the limit for no significant difference was still obtained.

Similar results were found for the other three divisions, as well. Therefore, the T-Tests suggests that the tornado production in each division in each phase was not significantly different. This suggested that the change of phase had no significant effect on the distribution of touchdowns by county.
Westwell, Ivan. Encyclopedia of the WEATHER, JD Press, 1997, North Dighton, MA., pp. 71, 109
Bove, Mark. Impacts of ENSO on the United States Tornadic Activity, Florida State University, June 3, 1997, Tallahassee, Florida.
Conner, Glen. Tornadoes in Kentucky 1950-1998, Kentucky Climate Center Publication number 27, June 1991, Bowling Green, Ky., pp. 1-2.
Greco, Stephen. Storm Data, National Climatic Data Center, Noel Risnychok Publication, 1951-1998, Asheville, North Carolina.
McGrew, Jr., J., Monroe, Charles. Statistical Problem Solving in Geography, Wm. C. Brown Publishers, 1993, Dubuque, Iowa, pp. 163-165.