A PA-32 Accident Report And The NTSB‘s Role

The airplane was en route from Reid-Hillview Airport in San Jose to Henderson Executive Airport in Nevada. Departure was at about 2:35 p.m. on December 19, 2015. The NTSB said the family was supposed to attend a surprise party that night in the Henderson area. A relative told investigators that the family had once lived in Henderson. The NTSB didn't provide details of what that party was all about, other than the relative saying it was for a family friend. A birthday? An anniversary? A Christmas party? What was going on to give the pilot a very strong motivation to complete the trip that day? The NTSB report doesn't provide the necessary background, in sharp contrast to its handling of major airline accidents in which an entire group of investigators delves deeply into human factors.

It's not as if the pilot was especially secretive. Reporters from a local newspaper published in San Jose were able to put together a profile within a day or so of the accident. The pilot and his wife regularly took their 9- and 10-year-old daughters and 14-year-old son on flying vacations. The son was taking flying lessons. The newspaper said the trip to the Las Vegas area was to visit and stay with friends. In addition to wanting to get to the party as the NTSB said, was the pilot motivated to impress his would-be-pilot son? Or the friends at the other end? Did he not want to break his record of successfully completing previous family vacation flights? Or show that he could fly the bigger PA-32RT with the same aplomb as he had flown a PA-28-236 he had owned for six months?    

The 42-year-old pilot was an engineer at a company in Gilroy, California. He previously served as a technician in the U.S. Coast Guard, and was a member of the Civil Air Patrol. A number of photos were posted on social media showing the pilot and his family posing with airplanes at the Reid-Hillview Airport.

The pilot received his private pilot certificate good for single-engine land airplanes in July 2012. He did not have an instrument rating. His third-class medical was issued in May 2012 and was good for five years.

As of the accident flight, he had just under 270 hours with 56.5 hours in the PA-32RT. His logbook showed 3.1 hours of simulated instrument time before he got his private pilot certificate and 0.8 hours in actual instrument conditions during two IFR training flights in February 2014. On January 30, 2015, 2.8 hours of dual instruction included two instrument approaches. He logged 28.3 hours of night experience. In April 2013, he purchased the PA-28-236 and logged 85 hours in it. He began flying the PA-32RT six months before the accident, and had 26 flights in it for 56.5 flight hours.

The airplane was owned by a corporation formed by a group of pilots. The accident pilot was not a member of the group, but he had been okayed to fly the airplane and was on the group's insurance policy.

The 1978 single-engine, T-tail airplane used a turbocharged Lycoming engine and had just over 3,840 hours total time. The airplane had an electrical backup vacuum/pressure system and two portable oxygen systems. One of the airplane's co-owners reported that he filled the oxygen tanks before the pilot took the airplane for the trip. The oxygen systems had only three cannulas among them, not enough for the five people onboard during the accident flight.

The airplane had a multifunction display in the panel and a WAAS-certified GPS/NAV/COMM. The airplane was approved for IFR but not equipped for icing conditions. It had its original autopilot which handled roll, pitch, heading, and altitude hold modes. On August 21, 2015, the pilot made a flight logbook entry that the "gyrocompass" and "autopilot" were inoperative. An entry in the airframe maintenance logbook, dated August 28, 2015, said "Installed overhauled directional gyro." One of the airplane's co-owners told investigators that he had used the autopilot regularly, including multiple times since the accident pilot's "inop" entry, and that the pilot probably didn't know how use it in all configurations.

The pilot seemed to plan for the flight seriously, even though he could have done more to stay updated on weather conditions. He signed on to DUATS the night before the flight for a weather briefing and to file a VFR flight plan. At 6:53 on the morning of the flight, he again got a DUATS weather briefing. The briefing included AIRMETs for IFR conditions, mountain obscuration, and moderate icing between the freezing level and FL180 for the area and time of the accident.

Had the pilot gotten a weather update, he likely would have been told about an Area Forecast issued at 12:45 p.m. by the Aviation Weather Center in Kansas City, Missouri, which covered the accident area. It called for broken ceilings at 1,000 feet, cloud tops to FL200, and visibility 3 to 5 miles in mist and scattered light rain showers.

During the initial climb, the pilot asked NorCal Approach for VFR flight following and said he wanted 15,500 feet MSL to stay above clouds. As the flight progressed, the controller began issuing advisories and vectors to aircraft in the San Jose area for approaching moderate-to-heavy precipitation in a 10- to 15-mile-wide band.

The NTSB did not provide a transcript of communications between the pilot and air traffic control, but it did summarize some of what was said.

At 3:02, the pilot contacted Oakland Center, and the airplane began to track south, leveling off at 15,500 feet. At about 3:12, it began to climb without the pilot advising the controller. SkyWest flight 2955 was about 4 miles east of the Piper and descending through 17,700 feet toward Monterey Regional Airport. The controller received a conflict alert, and SkyWest's collision avoidance system issued a resolution advisory requiring the crew to begin climbing.

The controller radioed the accident pilot that there was a traffic conflict, told him to maintain VFR, and asked if he was climbing. The pilot radioed that he was "going to climb over the top of this (unreadable)" and would level off at 16,500 feet.

At 3:16, the pilot asked for a climb to 17,500 feet and direct to Paso Robles, California. The airplane then abruptly turned east. It subsequently turned to the south toward Paso Robles. About that time, the controller issued an advisory to VFR traffic for moderate-to-heavy precipitation south of the Panoche VOR, about 40 miles north of the Piper.

At 3:24, after reaching Paso Robles, the Piper began to turn east, and the pilot radioed that he was turning toward Bakersfield. He was handed off to Los Angeles Center. A 3:37, about 55 miles west-northwest of Bakersfield, the controller advised, "moderate precipitation from one to two o'clock to your 9 o'clock position, first group of cells begins in about 5 miles, and then there is a secondary line from north to south that begins in about two zero miles and extends one five miles."

The pilot radioed that he could see the clouds. At about 3:40, he advised he would be descending to 15,500 feet. The pilot of a Cessna 414 reported the cloud tops in his area were at 18,000 feet. The accident pilot asked where the Cessna was, and the controller advised it was about 30 miles to the east. The pilot replied, "Roger, just wondering when I can get over to their altitude and clear the clouds." The controller repeated the alert about moderate-to-heavy precipitation along his flight route. The pilot asked about the cloud bottoms, and the controller said he didn't know and added that he had information about light rime icing up to 19,000 feet to the southeast. The pilot replied, "OK, we're going to deviate to the south and try and go through Barstow." Another pilot then reported cloud tops in the Palmdale area of about 21,000 feet, and the controller relayed the information to the accident pilot because that was the way he was headed.

About 3:50, the controller asked the pilot if he would like to file for an IFR clearance to Henderson. The pilot responded that he would, requesting an altitude of 15,000 feet. About 90 seconds later, the controller advised he was ready with the IFR clearance and while the controller issued it and the pilot did the readback, the airplane was turning left to the northeast. The controller asked the pilot about the turn and the pilot replied, "...I just took a heading off of Bakersfield and I'm going to change it to the current assigned IFR."

The controller then issued a 10-degree heading change to an airplane which was approaching the accident airplane at the same altitude, 15,000 feet. About 45 seconds later, the controller told the accident pilot to fly a heading of 095 degrees. The pilot responded to the heading request; however, rather than turning right, the airplane continued the same northeast heading and then began a climbing left turn to about 350 degrees, reaching 15,600 feet 40 seconds later. The controller then told the Piper pilot to make an "immediate right turn to 095," followed by directing the conflicting traffic to descend to 14,000 feet. By this time, the accident airplane had dropped to 13,800 feet and was going eastbound.

A few seconds later, the pilot transmitted, "Air traffic control Lance 402 mayday mayday mayday." Twenty seconds later, the pilot made another mayday call. I listened to a recording of those transmissions. The pilot sounded excited, but not panicked. His words were clear; there was no slurring of the kind you might expect if he was suffering from hypoxia. In the background of the second mayday call, I could hear something which was chilling: girls screaming in the background, at a low level compared with the pilot's voice.

Ten seconds later, at 3:56:10, the last radar target was recorded and was showing an altitude of 11,200 feet. The controller was unable to make contact with the Piper pilot. Another pilot was asked to listen to the emergency frequency 121.5 but heard nothing. Other pilots were asked if they could see or contact the accident airplane. One pilot reported clouds enveloped the area.

Most of the wreckage was found in an almond orchard, directly below the last radar target, about 9 miles southwest of Bakersfield. Analysis determined that both wings and the stabilator separated after being overstressed by excessive air loads. All five occupants were ejected from the airplane during the breakup sequence. No problems were found with the engine or flight control continuity prior to the breakup. There was no evidence of vacuum pump or instrument failure.

Investigators said the pilot likely flew into clouds, became spatially disoriented, lost control, tried to recover control and overstressed the airplane with excessive air loads. They said airframe icing or pitot-static system icing could not be ruled out. Using radar data, they calculated that for most of the last 22 minutes of flight, the airplane's speed exceeded the maximum maneuvering speed of 132 knots calibrated airspeed (KCAS). During the time the controller was issuing the instrument clearance, the airspeed reached a calculated peak of 168 KCAS. During the mayday calls, the airplane was descending while tracking generally east and "zig-zagging" to the north and south.   

Investigators determined that, as prescribed by regulations and good aeromedical judgment, everyone onboard should have been on oxygen for at least half of the flight. But, that wasn't possible because there were only three cannulas. An oxygen mask was found entangled with the pilot's jacket, suggesting that he was on oxygen, which also is supported by the clarity of his speech during radio transmissions. But, the Safety Board's report does not explore just what might motivate a pilot to risk the health and well-being of his family by exposing them to a situation in which they could not be provided with needed oxygen.

The NTSB determined that the probable cause of this accident was the non-instrument-rated pilot's decision to conduct and continue the flight despite forecast and en route instrument meteorological conditions which were not conducive to safe operation under visual flight rules. Also causal to the accident was the pilot's decision to accept an instrument flight rules clearance and fly into IMC during cruise flight, which led to his spatial disorientation and a resultant loss of control and in-flight breakup. Contributing to the accident was the pilot's self-induced pressure to arrive at the destination for a party that night.

While the probable cause statement reflects the tidy package put together by the Safety Board's investigators, it also reflects the minimal attention given to what we really need to know about the pilot's behavior and thought processes if we are to prevent future accidents like this. The throwaway contributing factor statement does little to enhance our understanding of the pilot's decision-making process and what motivated him to take one risky action after another. How can we avoid "self-induced pressure to arrive at the destination for a party" so we're not taking risks with our own life and the lives of our passengers? The NTSB isn't much help this time. We'll have to use this accident as a reference point and be smart enough to figure it out for ourselves.

That's not to condemn the NTSB for its handling of the investigation and report. It fulfilled its obligation to be factual and create a record.

And, I certainly recognize that it has limited resources which must be utilized wisely so as to have the greatest impact. But, in an ideal aviation safety world, I'd like to be able to pick up a report like this and learn about the factors which caused the pilot to make the decisions he did from the beginning, and the processes which kept getting him in deeper and deeper until there was no way out. My selfish objective: to recognize instantly if it's happening to me, so I can do something about it.

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.