WeatherThis Week in Weather History


WX HISTORY SPECIAL | The incredible story of the first tornado warning

The coincidence that changed weather forecasting
Posted at 3:11 PM, Mar 25, 2024
and last updated 2024-03-25 16:11:24-04

History is filled with coincidences. Seemingly random events can alter the course of the world. For example, if Archduke Franz Ferdinand's car hadn't taken a wrong turn in June 1914, he may not have been assassinated and thus no World War I. Coincidences can alter the course of world history and our lives.

One example of a coincidence changing history is what happened in late March 1948 at Tinker Air Force Base in Oklahoma City. On March 25, a tornado tore through the base just five days after another tornado hit the same base. For the second tornado, two meteorologists stationed at the base did something no one else had done before, they issued a tornado warning for the storm. This coincidence spawned the modern era of tornado warning and research, and the story behind it is fascinating. In this special installment of This Week in Weather History, we look at the improbable week of March 1948 and the birth of the Tornado Warning.


Before the 1880s, tornadoes were a phenomenon that wasn't studied well. Europe rarely had tornadoes, so it was not until people began moving to the US that tornadoes became a concern. Research into tornadoes was conducted in the 1880s by Michigan-born John Park Finley, who studied hundreds of tornadoes in his research to attempt to define what they were and how they formed. He even tried to issue "tornado forecasts" for a given region, with varying success.

John Park Finley.jpg
John Park Finley is often known as the "first tornado researcher" in US history for his work on tornadoes in the 1880s. Although his work was not picked up at the time, his valuable contributions to meteorology have lived on to the present day. This photograph of him was taken in 1913.

Although Finley's work was received well by some, others were more skeptical. In 1890, the US Weather Bureau (precursor to the National Weather Service) was formed under the direction of the Department of Commerce. Through the 1890s, several weather bureau meteorologists came out against the need to issue tornado forecasts and warnings. There was a variety of reasons for this: 1.) tornadoes were rare, and most destruction in storms was due to other phenomenon; 2.) The chances of a tornado impacting a region were so small that it would be useless to issue warnings; 3.) Warnings would cause unnecessary panic; and 4.) Some politicians in the Midwest did not want to associate their state with tornadoes, so tried to keep research down.

First Tornado Photo.jpg
Despite the belief that this is the first tornado photograph in Howard, South Dakota in August 1884, it is the 2nd earliest tornado photograph.

Based on those reasonings, by the turn of the century, the Weather Bureau formally banned the word "tornado" from being used in official forecasts. Instead, the term "severe local storms" could be used if conditions were favorable for tornadoes. This ban would last several decades until 1938. The ban on tornado warnings was consequential, between 1920 to 1939 over 4,000 people were killed by tornadoes. Periodically, the public would call for a tornado warning system, but their pleas were largely ignored.

First Actual Tornado Photo.jpg
The first known photograph of a tornado was taken in Gannet, Kansas in April 1884.


The story of the first tornado warning centers around two men: Captain Robert C. Miller, and Major Ernest J. Fawbush. If it was not for these two men at the helm in March 1948, there is a chance the tornado warning was never issued.

Fawbush and Miller.jpg
Major Ernest J. Fawbush (left) and Captain Robert C. Miller (right) were the two men responsible for issuing the first tornado warning in March 1948. This photograph was taken in the early 1950s.

Robert Miller did not want to be a meteorologist growing up in sunny Los Angeles, California. He had aspirations to become a math or science teacher and thus went to college to study mathematics. He enlisted in the army when the US entered World War II, and got himself assigned to Grand Rapids, Michigan to train to become a meteorologist for the Air Force. After the war, he continued his studies in weather, being assigned to Tinker Air Force Base in Oklahoma City in early 1948, just a few weeks before the tornadoes hit.

When Miller arrived at Tinker AFB, he met Major Ernest J. Fawbush. Fawbush was born in 1915 in Virginia but spent most of his childhood in New Jersey. He entered the Air Force after his childhood and became a meteorologist at Tinker AFB during the 1940s. It would be Miller and Fawbush that would be in the hot seat come March 1948.


March 20 began as an ordinary day for Miller and Fawbush. They looked over the various weather maps and concluded that besides some gusty winds, no thunderstorms were expected. What the two did not know at the time was one of the upper air maps had an error printed on it, and the situation was ripe for thunderstorms to develop.

March 20 Weather Map.png
One of the many weather maps that Fawbush and Miller looked over on March 20, 1948.

By the evening, a line of thunderstorms marched toward Oklahoma City. At Will Rogers Airport, 7 miles southwest of Tinker AFB, the line of storms brought wind gusts to 92 mph and a confirmed sighting of a tornado from Air Traffic Control. The tornado hit the base a few minutes later, shattering the windows in Air Traffic Control tower and damaging multiple hangars on the base. 54 aircraft were destroyed, and 50 other planes were damaged. No one was killed, but three people were injured in the control tower by flying glass.

Tinker Tornado March 20.jpg
Some of the many planes that were destroyed during the Tinker AFB tornado of March 20, 1948.

For the next several days, the two men were tasked alongside other meteorologists to analyze the weather conditions that brought about the March 20 tornado. For several days, the weather maps were poured over alongside other maps from past tornado outbreaks. Patterns began to emerge, and the men soon began to gain a broad understanding of how tornadoes can form using a broader lens. They knew that the information would be important for tornado forecasting down the road, but they never expected how soon it would come.


On the morning of March 25, Fawbush and Miller began their routine of analyzing weather maps for the day, but something unsettled them. The weather map for March 25 looked eerily like the weather map from March 20. Based on their recent knowledge, the men became concerned that severe thunderstorms including tornadoes were possible at Tinker AFB that evening.

By early afternoon, the two spoke to their commanding officer, General Fred Borum, who told the men that if they believed that the conditions were similar to March 20 they should issue a tornado forecast for Tinker between 5-6 pm. At 2:30 pm, Fawbush and Miller reluctantly issued the tornado forecast for Tinker that evening, a decision they knew would be monumental. The chances of a tornado hitting the same place twice were 1 in 200,000 they surmised, and the chances of it hitting within a week were almost unheard of. Still, they followed their officer and issued the alert, and watched developments unfold.

As the squall line approached, all their hopes dimmed as the line was weak. Instead of the 92 mph wind report with a tornado from Will Rogers Airport back on March 20, the observation for the March 25 storm was 26 mph winds and pea size hail. Dejected, Miller went home around 5:30 pm and left Fawbush to finish watching the line of storms.

As he sat eating dinner with his life, the unbelievable happened:

During the evening the radio broadcast we were listening to was interrupted for a major news bulletin. I was in another part of the house but caught the words destructive tornado and Tinker Field. "Good grief," I thought, "they're still talking about last week's tornado—but why break into the news?" I tried to call the weather station but the lines were dead. I felt a strange unbelieving excitement rising, told my wife I was going to the station and drove away.
The First Operational Tornado Forecast Twenty Million to One by Robert Miller and Charlie Crisp

When he arrived at the base, he met with Fawbush who described what happened:

At six o'clock thunder began at the base as the squall line moved in from the southwest. [Fawbush] and my friend, the Sargent, were outside, observing the motion of the clouds. As the line approached the southwest corner of the field, two thunderstorms seemed to join and quickly took on a greenish black hue. They could observe a slow counterclockwise cloud rotation around the point at which the storms merged. Suddenly a large cone shaped funnel bulged down rotating counterclockwise at great speed. At the same time they saw a wing from one of the moth-balled World War II B-29's float lazily upward toward the visible part of the funnel. A second or two later the wing disintegrated...
The First Operational Tornado Forecast Twenty Million to One by Robert Miller and Charlie Crisp

The second tornado moved less than 100 yards away from where the previous tornado hit. Due to the advanced tornado warning, many planes were in hangars and no one was on the field when the tornado hit. Still, 35 planes were destroyed as well as several hangars.

Tinker Tornado March 25.jpg
Damage to a plane from the tornado of March 25, 1948, at Tinker AFB. This was the second tornado to strike the base within five days.


The unbelievable series of events in late March 1948 brought attention from the meteorological community. It was the first successful tornado warning, and it was responsible for saving money and lives that day. Fawbush and Miller were now in charge of propelling their tornado research further. In 1951, the Severe Weather Warning Center (SWWC) was established. The SWWC was in charge of issuing tornado forecasts and warnings to various Air Force stations in the US, it was not a public warning system. Eventually, the public became aware of the tornado warning system and put pressure on the US Weather Bureau to respond.

In early 1952, the Weather Bureau created its storms forecasting unit, and issued its first tornado forecast on March 17, 1952, for parts of the south. Four days later, it issued another tornado forecast for parts of the Deep South. Unfortunately, many of the tornadoes on March 21 occurred outside the tornado forecast. Just north of the tornado forecast zone, a significant tornado outbreak in Arkansas into Tennessee killed over 200 people.

First Tornado Forecast.PNG
One of the first public tornado forecasts from the Weather Bureau was issued on March 21, 1952. Many of the tornadoes occurred just outside the region. Later, another tornado forecast was issued for western Tennessee as the storms moved in.

Following one of the most devasting tornado outbreak sequences in June 1953, the Weather Bureau formed the Severe Local Storm Warning Weather Service (SELS) to oversee the issuing of tornado forecasts to the public. The SELS is the precursor to the modern-day Storm Prediction Center (SPC), which is based in Norman, Oklahoma. In 1965, these "tornado forecasts" have morphed into what we know of as "tornado watches". That same year, tornado warnings were issued during the Palm Sunday Tornado Outbreak of April 11, 1965, for specific storms that were producing tornadoes.

Jumping ahead to 2024, the apparatus of severe weather watches and warnings can be traced back to two men who had to make a momentous decision in 1948. Fawbush and Miller continued their severe weather research well into their golden years. Fawbush died in Texas in 1982, and Miller passed away in 1998. These two were pioneers in tornado forecasting and research, and we have much to thank them for.