by Lee Sandlin
But the hearing itself turned out to be a mild and perfunctory affair. Major Fawbush gave the board a summary of the night’s events. He stressed how difficult it had always been to predict tornadoes, and he provided a brief history of, and rationale for, the civilian Weather Bureau’s ban on tornado warnings. He did not mention Miller’s complaint about the upper-air data, much to Miller’s fury. He explained to Miller after the hearing that he wanted to avoid picking a fight with the bureau that would be of no use to either side. Miller grudgingly acquiesced.
The board members considered the matter over lunch. In the afternoon they came back with a report exonerating the weather officers. The tornado was an act of God, they ruled, and “due to the nature of the storm it was not forecastable given the present state of the art.”
They did inevitably have recommendations. They wanted the air force’s weather officers to collaborate with the civilian Weather Bureau in order to devise procedures for predicting tornadoes and issuing tornado warnings. This idea was, as the saying went in government circles, a nonstarter, and the board probably knew it. The Weather Bureau really did look down on the military’s weather officers, just as Miller darkly suspected, and any spirit of cooperation they may have felt during the war had long since melted away.
More realistically, the board ordered Tinker, and all the other air force bases in the Midwest, to devise and implement emergency preparedness plans in the event of a tornado. Tinker’s base commander, General Fred S. Borum, immediately seized on this last idea. That same afternoon, almost as soon as the board’s plane was back in the air for Washington, he had operational teams at Tinker working out procedures for the rapid securing of aircraft and other base matériel. He also ordered Miller and Fawbush to come up with a practical method for making tornado forecasts.
Borum knew nothing about the sorry, centuries-long history of failed tornado prediction. He didn’t think there was any reason Miller and Fawbush couldn’t come up with a workable method in a matter of days. The two weather officers couldn’t think of a good way of telling him he was wrong.
So, the next morning, they began an immediate ad hoc double-overtime seminar on the nature of tornadoes.
They didn’t have much to go on. The only known authority on tornadoes, even then, was John Park Finley; his book from 1888 was still the standard reference. There were a handful of research papers in the scientific periodicals; the Weather Bureau had also been doing some statistical studies of tornado occurrences and severity. But not a single book had been published on tornadoes since the 1880s.
The most substantial work that Miller and Fawbush could draw upon had been done by the military. The Pentagon had commissioned a major study of midwestern weather during World War II. This had been a practical necessity: the vast industrial structure of the war had taken over much of the prairie. The military had deliberately scattered bases and factories and industrial sites throughout the midwestern states as a way of helping the regional economy recover from the Depression and the Dust Bowl years. But the appearance of so much military construction had led to a growing public anxiety: Was it dangerous? What would happen if a tornado struck a munitions plant? The workers in the plants themselves were particularly jumpy. The appearance of black clouds in the southwestern sky was sometimes enough to have a plant’s warning sirens sounded and for everyone to run for the emergency shelters.
The Pentagon study concluded that the real danger of a tornado setting off a munitions explosion anywhere was vanishingly small (though it still recommended the organization of a tornado spotters network, just in case). But in order to reach this conclusion, the authors had to assemble a substantial body of research on what was and what was not known about tornadoes, and it was this that Fawbush and Miller devoured. Most of what they read was not essentially different from the ideas John Finley had worked with in the previous century, when he’d first set out to predict tornadoes. The stress was still on the clash between warm humid air at the surface and an overrunning domain of cooler air aloft. Tornadoes appeared when this atmospheric tension became catastrophically unstable. But what triggered the formation of the tornado’s violently rotating funnel cloud? It was still a complete mystery.
One recent piece of research particularly fascinated Fawbush and Miller: there appeared to be a correlation between tornadoes and upper-air wind shear. This made sense to them. The old idea, dating back to William Ferrel in the nineteenth century, was that the rotation of tornadic storms was caused by the Coriolis force; this had been disproven by Rossby. But there was no doubt that the interior of a severe thunderstorm was a chaotic environment where winds were moving violently in contrary directions. If a rising convection column were continuously deformed by wind shear—no matter where or how the shear originated—it would begin to rotate, and this could be the genesis of the tornado funnel. Fawbush and Miller then set out to collect every surface and upper-air weather chart they could find that recorded the conditions when tornadoes had touched down to see if they could locate consistent evidence of wind shear.
By the end of their three-day immersion, they took stock of what they’d come up with. Fawbush didn’t think much of it. They had some promising leads for research projects, particularly on conditions in the upper altitudes. They had a number of rules of thumb they thought could be used as the basis for a general guide to forecasting tornadoes, based on the observation of vertical instability and upper-air wind shear. But they hadn’t cracked the mystery.
Except that Miller thought he had. At least he’d had a kind of intuition. He couldn’t quantify it. He couldn’t even describe it clearly—then or later. Fawbush, reviewing the same data, was never able to work out exactly what Miller thought he saw. But looking at all the charts, Miller had gradually evolved a sense that the forces at play in the atmosphere took on a characteristic shape just before a tornado descended. And when, on the morning of March 25, the two men prepared to show their results to General Borum, Miller suddenly became aware that the map for that day’s weather was uncomfortably close to the pattern in his mind.
General Borum was something of a weather buff himself. He wasn’t an expert meteorologist, but he did know the basics. He was also good with the station’s old radar unit; he was often able to tweak it back to life after the weather officers had given up. Whenever bad storms came over the base, he liked to hang around the weather station and watch the melting glows of rain drift across the monitor screen. When he showed up at the weather station on the morning of the twenty-fifth, he took a look through the rules of thumb they had worked out over the last few days and had no trouble understanding them.
But what about Miller’s intuition about the day’s data? Borum then inspected the weather map. If he fully comprehended what Miller was talking about, he didn’t say. Instead, he simply asked whether they were thinking of issuing a tornado forecast.
That was a bad moment for Fawbush. He took refuge in vagueness. He would only say that Miller was right about one thing: the day’s map did look like the map from five days earlier.
Borum didn’t press the point. He let Fawbush off the hook by asking when they would be sure.
There was a silence. An honest answer from Fawbush would probably have been “Never.” As far as he knew, nobody had ever tried to predict a tornado for a specific place at a specific time, and he was positive that he did not want to be the first. Miller, meanwhile, was in fact quite sure of the forecast already, but he didn’t have the nerve to speak up.
Borum tried again: When was the tornado likely to arrive?
That much Fawbush was willing to commit to. If it happened at all, it would come in with a storm front that was then beginning to take shape to the southwest. He said he could predict the movement of the front easily enough: the crucial hour for the airbase would be between 5:00 and 6:00 p.m.
Borum told them to make the late-afternoon forecast for severe thunderstorms. It would at least flag the possibility of violent weather and trigger the
first phase of the base’s new disaster preparedness plan.
Miller and Fawbush spent all that afternoon reading the reports coming off the Teletype. The line of severe storms had duly formed and was moving to the northeast. Conditions across the prairie were deteriorating in advance of the front. The winds were rising, cumulonimbus clouds were sprouting, a squall line was gathering strength. General Borum stopped by the station again around 2:00 p.m. He asked when the squall line would reach Tinker. They told him that it was right on schedule and would hit around 6:00 p.m. Then Borum asked the same question again: Were they going to issue a tornado forecast?
Fawbush tried the same answer: “It sure does look like the last one.”
Borum told Fawbush he was sick of his sounding like a broken record. If they really believed they knew what they were doing, he said, then they had no choice but to issue the forecast.
Then Miller spoke up, but unexpectedly he urged caution. He pointed out that the chance of two tornadoes hitting the base inside of a week was beyond astronomical. Besides, he reminded the general, nobody had ever issued an operational tornado forecast before.
“You are about to set a precedent,” Borum answered.
Fawbush reluctantly wrote out the forecast. Miller typed it and sent it on to base operations. It was formally issued to the base just before 3:00 p.m. After the general left the weather station, Miller remembered, he and Fawbush remained where they were, sunk in gloom, certain they were about to become laughingstocks. Miller was sure the blame was going to fall on them rather than on General Borum. “We couldn’t win,” he later wrote, “and the General couldn’t lose.” After all, what would happen when the forecast turned out to be a bust? “I figured General Borum wasn’t about to say, ‘I made them do it.’ More likely it would be, ‘Major Fawbush and Captain Miller thought it looked a great deal like the 20th—ask them.’ I wondered how I would manage as a civilian, perhaps as an elevator operator. It seemed improbable that anyone would employ, as a weather forecaster, an idiot who issued a tornado forecast for a precise location.”
Meanwhile, all over the base, in response to the forecast, work crews were guiding the aircraft into hangars, moving cars and trucks into garages, policing the grounds for stray objects. The air controllers were directing incoming planes to divert to other bases. As 6:00 p.m. approached, and the western sky was flickering with lightning, the controllers shut down the tower and moved to shelter.
Fawbush and Miller were still reading reports from weather stations. The line of storms was continuing its stately progress across the prairie toward Oklahoma City. But these still appeared to be ordinary spring thunderstorms. There were no reports of hail or high winds—much less any sightings of funnel clouds. The mood in the station grew funereal. The two men decided that they had only one chance to save their reputations: the national commander of the Air Weather Service might take mercy on them if only the thunderstorms were bad enough. “At this point,” Miller wrote, “we would settle gratefully for a brilliant lightning display and a wind gust to 30 or 40 mph.” At 5:00 p.m. the squall line passed over the Oklahoma City municipal airport: the report there was light rain, a scattering of pea-size hail, and wind gusts of a paltry 26 miles per hour.
It was the end of Miller’s shift; he couldn’t bear to stay on any longer, and he went home.
Miller lived in Midwest City, a suburb that had sprung up between Tinker and Oklahoma City. It was a low sprawl of shopping centers, ranch houses, and light-industrial factories. From his back patio Miller had an unbroken view of the southwestern skies. When he got home, he stood there for a long while watching the storm approaching. He told his wife, Beverly, what had happened. “She was reasonably sympathetic,” he wrote, but he didn’t feel even remotely consoled. Instead, “I sat down to aggravate my depression systematically.” (By which he appears to mean that he started drinking.) Shortly after 6:00 p.m., the squall line passed over the base and neared Miller’s house. Miller walked back out onto the patio. The sky was darker than he’d expected. Some areas of the cloud deck were agitated, and there were small white clouds darting beneath the base of the thunderhead. Then the rain started falling. He went back inside. His wife had turned on the radio. “I was in another part of the house,” he recalled, “but I caught the words destructive tornado and Tinker Field.”
His first thought was that the newscast was some kind of garbled replay from earlier in the week. “Good grief,” he thought, “they’re still talking about last week’s tornado—but why break into the news?” He picked up the phone to call the base weather station. The line was dead. He felt “a strange unbelieving excitement.” He told his wife he was going back to the base.
It turned out to be a tough expedition. The streetlights were out, and there was storm debris strewn all over the highway. He came to the base entrance to find a scene of chaos. Power lines were down everywhere; emergency floodlights had been brought in, and they were sweeping erratically across the endless pavements, revealing vistas heaped with wreckage.
Miller threaded his car slowly through the blocked streets of the base until he arrived at the weather station. The Operations Building was dark. Fawbush was standing outside the front door, smoking a cigarette. His mood, as Miller later described it, was “jubilant.” He eagerly recounted what had taken place after Miller had left for home. When the main force of the squall line had come over the base around 6:00 p.m., he had gone outside to watch. Just above the southwest fences of the base, he saw what looked like two thunderclouds coming together and rotating. The sky around them turned greenish black. An enormous wing of a B-29 went sailing up languidly toward the clouds and disintegrated in midair. Then the funnel cloud descended.
The funnel was there and gone inside of a minute. Many planes that hadn’t been securely braced were destroyed, but the total damage was barely half of what it had been a week earlier, and there were no injuries at all among the base personnel. That alone made Borum’s tornado disaster plan a spectacular success in the eyes of the Pentagon. Fawbush and Miller were “instant heroes,” Miller remembered. “In my case,” he wrote, “the rest of my life would be intimately associated with tornadoes and severe thunderstorms.”
On March 25, 1949, exactly one year after the operational forecast, Fawbush and Miller put out a second tornado warning. This one was for a large area of eastern Oklahoma south of Tulsa and east of McAlester. They issued the warning at three in the afternoon. At nine in the evening, two tornadoes were spotted on the ground northeast of McAlester. A month later, on April 30, Fawbush and Miller issued a forecast for tornadoes in an area within thirty miles on either side of a line from the town of Altus to south of Tulsa. Thirteen tornadoes came down in eastern Oklahoma, and ten of them were within or close to the warning area. A week later, another forecast, for an area between Amarillo and Lubbock, Texas: two tornadoes touched down there. And another nine days later, for an area between Amarillo and Gage, Oklahoma: a very large and violent tornado touched down and destroyed a veterans’ housing development with more than 170 units. Seven people died and eighty-two were injured.
These forecasts were all kept confidential. They were classified documents issued through the military’s own communication network for the benefit of base personnel only. Fawbush and Miller didn’t attempt to make them public. Public forecasting was the exclusive business of the civilian Weather Bureau. But they were quickly growing uncomfortable with their position. On the one hand, they didn’t want to be court-martialed for leaking classified documents; on the other, they didn’t want to be held responsible for civilian deaths because they’d failed to warn people of imminent danger.
Besides, the forecasts were beginning to leak out anyway. Military people in the know were calling their relatives and warning them if they were in a threat area. Discreetly, and only with the tacit sanction of their superiors, Fawbush and Miller began notifying certain civilian authorities like the Red Cross and the Oklahoma Highway Patrol whenever they issued a new fo
recast.
The results were only to be expected. A rumor began to spread through Oklahoma and its neighboring states that the air force had a secret method of predicting tornadoes. There was an immediate public outcry. As it happened, there was already a rumor circulating that tornadoes in the Great Plains were increasing in number and severity due to the nuclear tests in the southwestern desert. The idea that the government might be withholding knowledge of when and where these sinister new tornadoes would strike led to newspaper editorial campaigns demanding full disclosure and urgent congressional inquiries to the Weather Bureau about why this secret method wasn’t being made public in order to protect the people of the heartland.
The Weather Bureau’s response was halting and confused. It’d had little to say in public concerning the rumors about nuclear testing, other than to issue rote dismissals of the whole issue as nonsense. (The Pentagon hadn’t commented at all; it maintained a solemn silence about everything involving America’s nuclear arsenal.) Now, faced with this new flurry of criticism, it made some gestures at dealing with the question of tornado forecasting. In July 1950, the head of the bureau, Francis Reichelderfer, sent a memo to all its offices concerning the long-standing forecasting ban. The memo didn’t so much lift the ban as wish away its whole existence. Reichelderfer urged the bureau’s regional offices to remember that they had the authority to issue tornado forecasts whenever they deemed it appropriate. Also, he wrote, “Weather Bureau employees should avoid statements that can be interpreted as a negation of the Bureau’s willingness or ability to make tornado forecasts.”
Meanwhile, a meeting was arranged between senior Weather Bureau meteorologists and the team of weather officers from Tinker AFB. There Fawbush and Miller laid out their provisional tornado-forecasting guidelines in detail and explained their success with their operational forecast. The delegation from the Weather Bureau listened politely and then rejected the whole presentation out of hand.