The Origins of America’s First Tornado Prediction
In the backdrop of Getty Images, Air Force meteorologist Captain Robert Miller reported for duty at the Tinker Air Force Base near Oklahoma City. After examining the latest weather data sent from Washington, D.C., Miller anticipated a calm evening at the base, except for occasional gusts of wind. However, his prediction would soon be proven drastically wrong.
Around 9 p.m., neighboring weather stations detected lightning to the west of Tinker. Shortly after, Miller’s radar picked up fast-moving thunderstorms that appeared fierce. Concerned for the safety of the parked aircraft around the airfield, Miller, with a sense of urgency, issued a thunderstorm warning despite knowing it might already be too late to secure the planes.
Moments later, a report from Will Rogers Airport revealed a tornado heading towards the base. By 10 p.m., illuminated by flashes of lightning, the tornado wreaked havoc at Tinker Air Force Base, causing destruction and chaos. Miller and his colleague found themselves in a state of panic as they witnessed the devastation unfolding around them.
The tornado on March 20th became one of the most destructive storms in Oklahoma’s history, resulting in significant damage and injuries on the base. The aftermath prompted Air Force officials to investigate the failed forecast, leading to Miller and Major Ernest J. Fawbush explaining the unpredictable nature of tornadoes to a panel of generals.
The generals’ visit marked the beginning of a new era in meteorology, as Major General Fred Borum instructed Fawbush and Miller to devise a more effective solution for predicting tornadoes. This directive was extraordinary, considering the longstanding ban on using the word “tornado” in public forecasts, which had hindered tornado-related research for decades.
Despite the challenges, Miller and Fawbush, both soldiers and meteorologists, embarked on a journey to revolutionize tornado forecasting and mitigate the risks posed by these invisible threats.
They dutifully followed orders and immediately began investigating a longstanding mystery that had puzzled scientists for ages. Their discoveries would spark a revolution in weather prediction, forever altering the way meteorologists anticipate tornadoes. Just four days later, the skies unleashed their wrath once more.
The distinctive geography of the central United States, where warm, moist air from the Gulf of Mexico mixes with cool, dry air from Canada, leads to an abundance of these powerful storms compared to any other country. Indigenous tribes native to the Great Plains were familiar with tornadoes, with the Cheyenne referring to them as ma’xehevovetaso, or big whirlwinds, and the Kiowa portraying them in their calendars as the long, violent tails of horses.
While tornadoes occasionally hit the East Coast, they were less frequent. Governor John Winthrop of Massachusetts noted a tornado in a diary entry dating back to July 5, 1643, but their rarity meant they didn’t attract much scientific interest until the mid-1800s when settlers ventured into the Midwest, where tornadoes loomed each spring. Even then, tornadoes were thought to be unpredictable, sporadic events.
John Park Finley, an early challenger to this belief, was part of the Signal Corps, which was responsible for establishing the nation’s first weather service in 1870. Trained in science at the University of Michigan, Finley began his work at the Signal Service’s school in Arlington, Virginia, in 1877. He was among many servicemen collecting meteorological data, which was used to create some of the country’s earliest forecasts.
By 1878, the Corps had amassed weather reports from 284 stations and issued 24-hour forecasts to rural post offices nationwide. These reports, while detailed, were often inaccurate due to their reliance on statistical forecasting based on past data. Finley, after two years, was sent to the Great Plains to investigate the aftermath of a tornado outbreak. Like a detective, he interviewed witnesses, studied fallen trees, and examined damaged structures to piece together the events leading up to the tornado. Through his observations, Finley identified the atmospheric conditions that often precede tornadoes, noting that significant temperature and moisture variations hinted at an impending violent weather event.
This marked a significant step forward in understanding tornadoes and improving forecasting accuracy.
Traditionally, communities have often been caught unprepared for storms, as illustrated by John Finney’s photography. Over a period of five years, Finley traveled through the Plains states, recruiting a group of “tornado reporters” to collect observations during severe storms. Analyzing numerous accounts and weather charts, he developed a set of six rules for forecasting tornadoes, which he initially published in Science magazine in 1884 and later simplified for the American Meteorological Journal. Finley correctly deduced that tornadoes require high temperature gradients and the presence of a well-defined low-pressure area.
In 1884, Finley began experimenting with tornado predictions, dividing the eastern two-thirds of the country into districts and indicating whether conditions were favorable for tornado formation. However, his colleagues doubted the accuracy of these forecasts, with some critics accusing him of misidentifying derechos as tornadoes. Despite this, Finley believed his forecasts should be used to issue official tornado warnings. In 1885, the chief signal officer permitted special warnings for “violent storms” but prohibited the use of the word “tornado” to prevent unnecessary panic due to the imprecise nature of the forecasts.
Despite his efforts, Finley faced setbacks as Congress debated transferring the weather service from the Signal Corps to the War Department, which displayed little interest in his work. The 1887 Report of the Chief Signal Officer concluded that predicting tornadoes could potentially cause more harm than the tornadoes themselves, leading to the demise of Finley’s tornado-forecasting project.
Tornadoes are characterized by swirling funnel clouds that typically form within large thunderstorms when pockets of unstable cool, dry air and warm, moist air collide. These powerful twisters can cause extensive damage as they move across the landscape, sometimes covering long distances. Within supercell thunderstorms, updrafts draw warm, moist air upward while wind shear, which involves winds blowing at varying speeds and directions, contributes to the rotation necessary for tornado formation. As the warm updraft encounters cold sinking air, it triggers the creation of a mesocyclone, leading to the formation of tornadoes.
The temperature difference is believed to cause the funnel to narrow, spin faster, and draw in more air from the ground. A tornado typically weakens and dissipates when its surface supply of warm, moist air is exhausted.
Miller and the 33-year-old Fawbush, both experienced meteorologists who had served in WWII, took on the challenge of tornado prediction despite the political and scientific obstacles. They diligently worked to analyze atmospheric conditions that had led to previous tornadoes, including the devastating Woodward, Oklahoma tornado of 1947 that claimed 116 lives.
Studying a wealth of data and storm accounts, they identified specific weather parameters that seemed to indicate tornadic activity when present together in a geographic area. They highlighted the importance of wind shear—how winds change direction and speed with altitude—and the significance of atmospheric buoyancy, where warm air rises rapidly.
Their findings were put to the test shortly after they were documented, proving their theories correct. This stroke of luck allowed them to further explore and refine their predictions.
In the late 1800s, after a turbulent period of tornado studies and controversies, the U.S. Weather Bureau was established under civilian control within the Agriculture Department. This transition ushered in a new era of meteorological understanding, with experts like Willis L. Moore challenging past misconceptions about tornado occurrences and predictions. The evolving discourse on tornado forecasting highlighted the complexities and uncertainties surrounding severe weather events.
In Mark Harrington’s popular 1899 book, “About the Weather,” it was noted that exposure to dampness led to more deaths than all recorded tornadoes combined. The chance of being injured by a tornado was estimated at about one in a million. On March 20th, a Douglas C-54 Skymaster and numerous other aircraft were destroyed. Consequently, the 1905 regulations of the National Weather Bureau prohibited tornado forecasts, and research on tornadoes essentially halted. Despite this, tornadoes persisted.
On March 25, 1948, meteorologists Miller and Fawbush, after analyzing weather charts, predicted a significant tornado threat in central Oklahoma. Major General Borum, working on a new “Tornado Safety Plan,” sought their guidance on issuing a tornado forecast for the base. Faced with reluctance due to the rarity of such events, the meteorologists initially suggested a forecast of heavy thunderstorms instead. However, as the storms intensified and approached the base, Borum pressed for a tornado forecast. Despite the historical reluctance to issue such warnings, Fawbush was instructed to prepare an alert, marking a significant departure from previous practices.
The ban on tornado forecasts by the Weather Bureau had previously led to a high death toll from tornadoes, with residents relying on personal observations and the Farmers’ Almanac to predict storms. However, with the evolving weather patterns and increasing risks, the traditional approaches were proving inadequate.
In 1925, a devastating tornado outbreak swept through the country, claiming 794 lives in what would become known as the Tri-State Tornado. The tornado carved a path of destruction across Missouri, Illinois, and Indiana, leaving whole towns in ruins and hundreds dead. This tragic event, along with advancements in meteorological tools, inspired increased research into tornado prediction during the 1920s.
During World War II, the National Weather Bureau lifted its ban on using the word “tornado” in weather reports for emergency personnel only. Volunteer storm spotters were established near military installations to warn of approaching storms, but after the war, the urgency to predict severe weather waned. In 1948, meteorologists Fawbush and Miller took a chance and issued a tornado warning, fearing it would end their careers.
Their forecast covered a large area, but as they anxiously awaited the storm, only mild thunderstorms were reported nearby. However, a subsequent urgent news bulletin revealed a destructive tornado near Tinker Field. Rushing back to the base, Miller found the aftermath of another tornado strike, but thanks to their warning, damage was minimized, and there were no injuries.
Their prediction was remarkably accurate, turning the men into instant heroes. Two years after the renowned forecast by Miller and Fawbush, the Weather Bureau lifted its ban on using the term “tornado.” By 1951, after further research and successful predictions, Fawbush and Miller were appointed to lead the Severe Weather Warning Center, a new unit tasked with forecasting hazardous conditions at all Air Force sites in the Continental U.S. That year, the center issued 156 tornado warnings, with a 65 percent accuracy rate. The National Weather Bureau later adopted the military’s forecasting methods for public safety. The following year, the agency established its own severe weather unit, which eventually evolved into the Storm Prediction Center in Norman, Oklahoma.
Russell Schneider, now heading the center, describes it as the national tornado forecasting hub, crediting Fawbush and Miller’s pioneering forecasts in 1948 as the foundation for their work today. The devastating twister on March 20th, which was the costliest storm in Oklahoma’s history at the time, caused damages equivalent to $132 million, according to the Oklahoma Historical Society.
Advancements in technology over the past 75 years have enhanced tornado prediction reliability, notes University of Oklahoma professor Howard Bluestein. Contemporary scientists utilize a blend of satellite data, radar, weather balloons, and supercomputers to develop forecasts. Despite improved forecasting models, predicting the exact location of a tornado remains elusive; the focus now is on estimating the likelihood within certain areas.
The Storm Prediction Center employs probabilistic forecasting, generating a range of potential outcomes based on various simulations with slightly different data inputs. Bluestein characterizes tornado forecasting as a blend of science and art, combining meteorological knowledge, model predictions, and visual observations. Tornado watches indicate favorable conditions for tornadoes, while warnings are only issued when tornadoes are spotted.
For forecasters like Schneider, predicting severe storms, with their potential for significant damage and threats to lives, remains a challenging responsibility. The legacy of Miller and Fawbush endures at the Storm Prediction Center, where a signed map by Miller serves as a reminder of their groundbreaking forecast. Schneider emphasizes the ongoing commitment to building upon the legacy of those early meteorologists.
The United States experiences approximately 1,200 tornadoes annually, making it a hotspot for these destructive weather events.
The unique combination of weather conditions in Tornado Alley, stretching from Texas to South Dakota, can lead to the formation of destructive tornadoes. Cool, dry air from Canada, driven by the jet stream, moves over the Rocky Mountains towards areas of low pressure in the Great Plains. This air collides with warm, moist air from the Gulf Coast and warm, dry air from the Southwest, creating an unstable atmosphere that gives rise to powerful thunderstorms and tornadoes. Recent studies indicate that climate change may be causing stronger storms to move from Tornado Alley towards southeast regions, impacting communities that may not be adequately prepared for such events.
