A Rapidly Closing Weather Window

The instrument-rated private pilot had logged 2,258 hours with 188 in the Malibu conversion. His third class medical required glasses for distant and near vision. He told the FAA he wasn't using medications, but toxicological testing after the accident found he was using the erectile dysfunction drug Sildenafil, the sleep aid Zolpidem, the pain reliever Ibuprofen and the heartburn medication Ranitidine. He also was found to have used marijuana. The NTSB said that it's not likely the drugs contributed to the accident, but can anyone be sure there wasn't even a minor cumulative effect? Investigators couldn't determine from his flight records whether the pilot had a current flight review and was current for instruments.

The flight was from Aspen-Pitkin County Airport/Sardy Field (KASE) at Aspen, Colorado, to Brenham Municipal Airport (11R) in Texas, a direct distance of 851 statute miles. It was supposed to take 3 hours and 30 minutes. The pilot had filed a cruise speed of 250 knots at FL270 (approximately 27,000 feet). The pressurized six-seat PA46-310P Malibu typically can get up to FL250 where it cruises at about 200 knots. This airplane, which was built in 1985, received an STC conversion to a much more powerful 560 shp turbine powerplant in 1999. With the Pratt & Whitney PT6A-34 engine, it could fly higher, faster and get above more weather. The NTSB pointed out a minor complication, which the pilot ignored, and really didn't contribute to the accident. Although the Pilot Operating Handbook for the turbine conversion approved operation as high as FL270, the airplane's altimeters were only certified to 25,000 feet.

At 1:09 p.m., Mountain Daylight Time, about 17 minutes before departure, the pilot phoned flight service for a weather update and to file his instrument flight plan. He indicated that he already knew some of what was going on with the weather, and was given an abbreviated weather briefing. The briefer described the flow of moisture north from the Gulf of Mexico, and discussed a Convective SIGMET for a developing line of thunderstorms near the destination. If, indeed, the pilot had given himself a comprehensive weather briefing on the computer before telephoning flight service, he would have been ready for a mix of visual and instrument conditions en route. There would be some turbulence getting out of the Aspen area, but then a fairly good ride with the wind at FL270 from 240 degrees at about 32 knots. The airplane could be in and out of the clouds, especially once over Texas. With all the moisture and cloud buildups, and an outside air temperature of minus 25 degrees C, the pilot might have to give the airplane's de-ice boots a bit of a workout.

By 3:17 p.m., Central Daylight Time, after having been airborne for about 51 minutes, the pilot was being handled by Albuquerque Center at his cruise altitude of FL270. When the pilot checked in, he reported a smooth ride. The controller announced that a Fort Worth Center Weather Advisory could be heard on the flight watch frequency. A few minutes later, the Malibu was handed off to another Albuquerque Center controller. At about 3:49 p.m., this controller announced that a Convective SIGMET for Texas had been issued and could be heard on flight watch. Neither of the controllers broadcast details of the unfolding weather. Pilots tuning to flight watch would have heard that severe thunderstorms had developed and were forecast to continue over portions of Oklahoma and Texas. Hail was forecast to be up to 2 inches, there was extreme turbulence, surface winds could be expected to hit 60 knots, and the cloud tops could reach up to 55,000 feet. Along the route of the Malibu's flight, it might not be quite so bad, but it was still severe, with cloud tops to 45,000 feet, hail to 1½ inches in diameter and wind gusts on the ground to 50 knots.

At 3:54:17, the controller alerted the Malibu pilot that "...I've seen you as much as five hundred feet high and as much as four hundred feet low in the last minute." The pilot replied, "....we're having some autopilot issues here on the hold, uh, we're trying to get her..." The rest was unintelligible.

Whether the airplane was being bounced around in heavy updrafts and downdrafts, or really did have an autopilot problem, wasn't pursued at the time. After the accident, the autopilot was checked by the manufacturer and seemed to be okay. About two minutes later, the pilot radioed, "...we're going to do ten degrees left for build-up." The controller said, "...left deviation approved and, uh, advise on course." The pilot confirmed, "...we'll advise back on course." Instead of turning to the left to fly east of his course, the pilot turned to the right, deviating west of course.

Over the next few minutes, the controller's scope continued to show thunderstorms popping, and the Malibu pilot continued his maneuvering for weather avoidance. At about 4:06, the controller radioed, "...does it look like you're going to, ah, to the west of that cell that's off your, ah, twelve o'clock all the way down to your, ah, nine o'clock position, you going west of that..." The pilot responded, "Yeah, we're going west of that one and we might be able to go between those, we'll take a look and see, but otherwise we'll go around it." The controller radioed,
"...not a problem, uh, deviations right of course are approved and, uh, just advise when you can make it to your, ah, destination." The controller put the code "DR" in the data block adjacent to the Malibu's radar return indicating that the airplane was deviating to the right. The flight was then handed off to Fort Worth Center.

When the pilot checked in with the new controller, he advised being at FL260, but was going back up to FL270 while doing a turn for weather avoidance. The controller acknowledged, and didn't volunteer any information about the weather she was seeing on her scope. After all, the indication "DR" meant he was deviating and, therefore, had to know about the weather. She did ask if the pilot was going to go north of a cell. The pilot responded, "...ah, we're trying to go in this window here." The controller asked the pilot to "...say again...," and he told her, "We're trying to go through a window here." The controller replied "...right, thanks..." and, about 30 seconds later, asked whether the pilot needed FL290. She asked twice more whether he needed flight level 290. At about 4:32, she radioed, "...I show you're right in the middle of moderate extreme precipitation, did you need a different altitude?" About 20 seconds after that transmission, the accident pilot transmitted, "Mayday."

When the controller didn't immediately respond to the mayday call, the pilot of an Eclipse jet operating at FL410 relayed to the controller that there had been a mayday call. He also heard the pilot say that he was "spinning." The controller radioed the Malibu, asking for the pilot's altitude. The pilot replied, "Nineteen." A United Airlines pilot reported hearing the Malibu pilot say he was spinning and had lost sight of the ground or horizon. The controller asked American Airlines and Alaska Airlines pilots to try to make contact.

FAA radar data, as captured by a commercial service, showed that just before the mayday call, the airplane had entered a climb of more than 1,000 feet per minute. About a minute later, it went into a descent of almost 3,000 feet per minute. Less than a minute after that, the descent reached a rate of 6,774 feet per minute, then slowed to 4,500 feet per minute. The airplane was descending at a bit under 3,000 feet per minute just before radar contact was lost.

The NTSB said the crash occurred at 4:35 p.m., in an open field about one-half mile east of Lehman, Texas. The wreckage was pretty much intact, and showed extensive vertical crushing. Investigators concluded that the airplane had come down in a flat spin.

An NTSB meteorologist overlaid the Malibu's flight track on weather radar images and lightning strike data. Beginning at about 4:00, the pilot would have flown between cells reaching 27,500 and 34,000 feet. By 4:12 p.m., cells to the east and south had grown to 44,000 feet. Within the next four minutes, storms to the south had grown to 47,000 feet. Some cells within 10 miles of the airplane contained hail greater than 2 inches in diameter. At 4:33 p.m., instead of continuing to avoid the storm cells, the airplane turned to the east-northeast into a growing area of intense activity. An intense core was only seven miles east of the airplane's location. The cell was 20 miles wide and reached about 46,000 feet. The pilot flew directly into it.

The NTSB's report and published backup material don't document the avionics carried onboard the Malibu. One might expect to find airborne radar on a Malibu like this one, especially when the pilot plans to operate in the flight levels during thunderstorm season, but not all have it. Some Malibus have a radar set that uses a small antenna located in a wing. Others use an aftermarket higher-powered set with a larger, more effective antenna in a pod hanging from the right wing.

My NTSB source was told by an investigator that the airplane was equipped to receive and display weather data. The investigator said they couldn't determine whether the pilot had a current subscription to a weather service, nor what the pilot may have been viewing during the accident flight. If he was still in visual conditions, the pilot should have been able to see that he was headed into some towering giants. If he was looking at a NEXRAD mosaic, any window he saw between cells could already have closed. There are processing and transmission delays when NEXRAD images are sent by commercial services. Even though there are time hacks associated with the images, sometimes pilots will fail to make the connection. The NTSB, for one, has warned about this potential hazard. We don't know the level of the pilot's training in interpreting weather radar images. If he was looking at a radar image, old or real-time, was his window through the weather something he misinterpreted, or really in the image as a result of signal attenuation? One thing we can do to help avoid repeating what happened here is to learn more about reading radar.

In looking at wreckage photos, I could see that the airplane's panel did have an Argus lightning detector. If it was working, the pilot might have seen some of the 36 cloud-to-ground and 2,117 in-cloud lightning strikes recorded within 25 miles of the accident site at around the time of the crash. A lot of the lightning was in the direction he had turned, and should have been taken as a warning to stay away.

The NTSB report doesn't provide much enlightenment about how the pilot wound up in a spin. We might speculate that the pilot entered clouds and there was some ice accumulation, resulting in decreased aerodynamic capability. Couple that with the airplane at its maximum approved altitude, perhaps forced even higher by updrafts, throw in varying angles of attack due to turbulence, and you have an aircraft perilously close to its operational threshold. A military pilot near the Malibu, but lower, reported picking up ice. Testing after the accident found the Malibu's de-ice boots were functional.

The Eclipse pilot I mentioned earlier was bothered by what sounded to him like the controller's laid-back or laissez-faire reaction to the unfolding emergency. He subsequently telephoned ATC to discuss it, and was assured that everything was handled according to current directives. He reported that his Eclipse was equipped with X-band airborne weather radar, a StormScope and XM Satellite Weather. He said at FL410 he had to deviate around the line of weather that was over the accident location. He estimated the cloud tops were at 48,000 feet and there was a lenticular-shaped cap cloud on top. He said he briefly found himself in instrument conditions while maneuvering around the weather. He said that the weather was very deceptive around the time of the accident, with cells growing rapidly and merging. He believed that the accident airplane, down lower, would have been in IMC.

While the NTSB faults the controller for failing to volunteer weather avoidance information, we have to remember that the pilot didn't ask for any kind of help until it was likely too late. A simple request to the controller a bit earlier for vectors or a request from the pilot for her to describe the weather returns she was seeing might have resulted in a different outcome. PP

Peter Katz is editor and publisher of NTSB Reporter, an independent monthly update on aircraft
accident investigations and other news concerning the National Transportation Safety Board.
To subscribe, visit www.ntsbreporter.us or write to: NTSB Reporter, Subscription Dept., P.O. Box 831, White Plains, NY 10602-0831.