A researcher's prediction is proven
The unfortunate pilots of Eastern 66 never knew what hit them, only that their jet rapidly lost lift. Nothing in their training and experience had taught them how to avoid the invisible weather hazard they flew into only seconds before they should have safely landed.
Weather events like the one that caused this accident are still occurring, of course. But you and other pilots flying today have the advantage of the scientific knowledge, new technologies, and practical flying advice that grew out of the Flight 66 crash and similar accidents over the following decade. This knowledge and practical flying advice should be a part of every pilot's weather education.
Showers were moving over the airport that afternoon, but the weather at JFK didn't appear threatening as Eastern 66 neared touchdown on Runway 22 Left at 4:05 p.m. Pilots of other airplanes approaching Kennedy had reported "wind shifts" and "wind shear," but these reports caused no alarm. Since aviation's earliest days pilots had known that gusty winds could complicate landings.
Eastern 66 was a little more than 2,400 feet from the end of the runway, about 200 feet above the ground, when it suddenly descended; the left wing clipped some approach lights, and the airplane crashed into Rockaway Boulevard, killing 113 of the 124 passengers and crew aboard.
The crash could have been blamed on bad luck and "pilot error," and would be all but forgotten today if the airline had not asked Theodore Fujita of the University of Chicago to take a detailed look at the weather at the time of the crash. Fujita, a renowned thunderstorm and tornado researcher, was famous for his ability to absorb reams of data and then visualize what was happening inside storms. (The F--for Fujita--scale used to rank tornadoes is named for him.)
He studied the flight data recorder from the crashed 727, reports from other pilots taking off and landing around the time of the crash, regional weather data, and radar and satellite images. He concluded that a strong, concentrated downdraft from a rain shower had caused the crash. Pilots and meteorologists had long known about downdrafts, but no one knew of any as strong and as concentrated as the one Fujita said had hit Eastern 66.
Fujita found that the 727 had plunged when a 16-knot headwind shifted to a 12.4-knot downdraft, causing airspeed to drop from 138 to 122 knots in seven seconds. He invented a new term, downburst, to describe the crash's cause. Fujita later sharpened the definition by inventing the term microburst for the especially concentrated--and dangerous--downbursts that were causing airplane crashes. By this definition, a microburst hits the ground in an oval pattern no more than four kilometers long.
The illustration at upper left is an idealized version by NASA of what's going on inside a microburst as envisioned by Fujita. If the airplane shown on final approach continued it would encounter winds much like those that caused the Eastern Flight 66 crash at JFK. As the jet flies into the center of the microburst, the headwind would die and the downdraft would push the airplane toward the ground. If it were still high enough after flying through the microburst center, the airplane would fly into tailwinds that would decrease lift.
Now, imagine that the landing airplane goes around and the microburst moves down the runway to the intersection with the cross runway. If the airliner that's holding short pulls onto the runway and tries to take off, its pilots would first encounter a headwind that would suddenly change into a tailwind, which could rob enough lift to cause the airplane to crash. The lower figure on p. 34 shows how this would happen.
Unfortunately, real microbursts aren't as visible as the one depicted. They are usually hidden by falling rain and low clouds and are moving and changing in strength.
Until Fujita advanced his theory, meteorologists had blamed thunderstorm gust fronts for the quick changes in wind speed and direction that sometimes occur in the vicinity of showers and thunderstorms. The gust front is the air that descends and spreads out as rain begins falling from a shower or thunderstorm, but not as a concentrated microburst.
Meteorologists knew that from time to time gust fronts cause rapid wind changes and sometimes bring high winds. But, no one had ever reported or suspected that the winds coming down from showers or thunderstorms were as concentrated as Fujita said. Researchers began using Doppler weather radar to look for Fujita's microbursts, and found them first in Illinois and then in Colorado, where research Doppler radars were available. By 1982 scientists had not only confirmed Fujita's microburst hypothesis, but had shown that microbursts were responsible for other aircraft accidents, including airliner crashes.
How could microbursts have been missed until 1975? Fortunately, they are rare and the odds are small that one will happen to hit a weather station. Even if one did, the station's single anemometer would show only the wind in one place, not the starburst pattern of winds hitting the ground that Fujita visualized as he studied the data from the JFK crash.
By the early 1980s researchers, pilots who were interested in weather, and others were developing programs to train pilots how to recognize and avoid microbursts, and designing systems such as Doppler radars that could help controllers alert pilots to a microburst danger. Until August 1985, however, neither the FAA nor many others in the aviation industry saw the need for a major effort to reduce microburst crashes. The August 2, 1985, microburst-caused crash of a Delta Air Lines Lockheed L-1011 at Dallas-Fort Worth International Airport, which killed 165 people, ended the apathy. The FAA and other aviation interests began implementing the many recommendations that scientists, aviators, and others had been making, including a proposal to install special Terminal Doppler Weather Radars (TDWR) at air carrier airports east of the Rockies--where microbursts are most likely. These radars give a detailed picture of air movements over the runways and on approach and takeoff paths at nearby airports, unlike ordinary Doppler weather radars that track all kinds of weather over wide regions.
Until the DFW crash, microbursts had been causing an airline accident in the United States approximately every 18 months. Since 1985, microbursts have caused only one air carrier accident: the July 2, 1994, crash of a USAir DC-9 at Charlotte, North Carolina, that killed 37 people. At the time, Charlotte's TDWR had not yet been installed.
Microbursts, however, continue to cause general aviation accidents. A microburst causes low-level wind shear, and pilot weather training materials often describe microbursts as "low-level wind shear." The term wind shear can be confusing because meteorologists use it to describe phenomena that seem to be quite different. All kinds of wind shear have one thing in common: the wind's speed, its direction, or both change by a large amount over a relatively small area.
Low-level shear, as the name says, occurs near the ground and can be the most dangerous because pilots have less time to recover from the loss of airspeed and lift that such shear can cause. This brings us to the microburst, a phenomenon that creates the most dangerous kind of wind shear--the kind that caused the JFK and DFW crashes and many others. But microbursts don't cause all low-level wind shear. A gust front can cause low-level wind shear, often miles from the thunderstorm that produced the gust front. Microbursts, however, cause the most violent low-level wind shear. Unfortunately the size and obvious menace of a thunderstorm isn't a good guide to the microburst danger.
Avoiding the obvious hazards of thunderstorms will go a long way toward ensuring you won't encounter a microburst. But this isn't a 100-percent guarantee. The thunderstorm that caused the DFW microburst in 1985 was small, with its top only 23,000 feet high; it had produced only a couple of lightning bolts. A gust front flowing from a 50,000-foot thunderstorm to the northeast had triggered the much smaller, but deadly, storm.
This is why it's not a good idea to wait to take off or land when a shower, or a cumulus cloud that looks like it could soon begin producing rain, is overhead. You can't go wrong by assuming that any rain shower could be hiding a microburst.
In the western United States where the humidity is usually much lower than in the East, cloud bases are usually higher above the ground, and you often see rain evaporating before it reaches the ground--this is called virga. You should consider virga to be a microburst alert. As the falling rain evaporates it cools the air, making it denser, which means the air falls to the ground, maybe as what meteorologists call a dry microburst. At times you can see western microbursts when they hit ground, kicking up a swirl of dust.
Wet microbursts are much more common in the East. Causes include dry air that intrudes into the storm cloud. Rain falling into this dry air evaporates to cool the air and send it hurtling toward the ground. Clouds and rain hide all of this and the only way to really know what's going on is to have Terminal Doppler Weather Radar keep an eye on all clouds over the airport or along the approach and departure paths.
Since general aviation pilots are more likely to take off and land at airports without TDWR, they should be cautious about flying into or under showers while taking off and landing.
Jack Williams is coordinator of public outreach for the American Meteorological Society. An instrument-rated private pilot, he is the author of The USA Today Weather Book and The Complete Idiot's Guide to the Arctic and Antarctic, and co-author with Bob Sheets of Hurricane Watch: Forecasting the Deadliest Storms on Earth.
Related Articles
