'Moderate' peril

By David Jack Kenny

Even in areas far removed from the weather hazards characterizing Alaska or the Rocky Mountain West airmets for moderate turbulence may be commonplace. Airmets cover large areas, and pilots flying in those areas often encounter nothing especially alarming. On the other hand, unless you spend a lot of time in it, “moderate turbulence” feels more like an emergency, and it’s easy to forget that forecasts can be wrong in either direction. Actual conditions can be worse than predicted, too.

Early in the evening of Dec. 14, 2012, a Beech King Air E90 broke up in flight about 15 minutes after takeoff from the Rick Husband International Airport in Amarillo, Texas. The pilot and passenger were both killed. Pieces of the airframe were scattered over a debris field about a mile in diameter, and all fracture signatures and cable separation points were consistent with overloads.

The flight took off from Amarillo about 10 minutes before 6 p.m., and then was handed off from Amarillo Tower to the Albuquerque Air Route Traffic Control Center (ARTCC). Albuquerque Center cleared the pilot to climb to FL 210 with approval for eastward deviations as needed; the area forecast predicted scattered thunderstorms, possibly intense, and a convective sigmet was in effect. Shortly thereafter, while climbing through 14,800 feet msl, the airplane began a right turn from a southerly heading to one roughly northbound. The controller asked the pilot about the change in direction, but got no response before the King Air disappeared from radar. The last data points downloaded from the onboard electronic ground proximity warning system indicated a final descent rate in excess of 32,000 feet per minute. Radar data indicated high reflectivity east of the accident site in western Oklahoma, but only light showers in the immediate vicinity.

The weather information available before the flight suggested a good chance of a rough ride, but nothing beyond a King Air’s design limits. An Airmet Tango warned of moderate turbulence below FL 180, and surface winds were strong and forecast to increase; over the course of the afternoon they had gusted up to 30 knots out of the west. About half an hour before takeoff, a sounding balloon measured winds aloft out of the south-southwest increasing from 51 knots at 11,000 feet to 63 knots at FL 200. The freezing level was around 8,400 feet, and moderate clear icing was forecast at 12,700 feet, with the potential for light icing between 10,700 and 13,500 feet and again between 15,600 and 17,300 feet. Three pilot reports recorded within 50 miles of the accident site in the three hours prior to the crash described moderate turbulence and mountain wave activity.

The NTSB concluded that the King Air was broken apart by “heavy to extreme” turbulence, possibly in icing conditions, that exceeded its structural limits. Examination of the wreckage found no evidence of any anomaly prior to the break-up. Beyond the general instability in the local atmosphere, there was no particularly obvious reason to expect conditions worse than any that had been forecast or reported. Still, forecasts are estimates, often wrong in some details—and if they are wrong, there’s no particularly good reason to expect the weather to be better than predicted rather than worse.