Speculation over the crash of a single-engine turboprop plane into a cemetery shifted to ice on the wings Monday after it became less likely that overloading was to blame, given that most of the 14 people on board were small children.
While descending Sunday in preparation for landing at the Bert Mooney Airport in Butte, Mont., the plane passed through a layer of air at about 1,500 feet that was conducive to icing because the temperatures were below freezing and the air "had 100 percent relative humidity or was saturated," according to AccuWeather.com, a forecasting service in State College, Pa.
Safety experts said similar icing condition existed when a Continental Airlines twin-engine turboprop crashed into a home near Buffalo Niagara International Airport last month, killing 50.
A possible engine stall created by ice, and the pilot's reaction to it, has been the focus of the Buffalo investigation.
"It's Buffalo all over again, or it could be," said John Goglia, a former member of the National Transportation Safety Board. "Icing, given those conditions, is certainly going to be high on the list of things to look at for the investigators."
Mark Rosenker, acting chairman of the National Transportation Safety Board, told reporters in Montana that investigators would look at icing on the wings as a factor.
"We will be looking at everything as it relates to the weather," he said.
The single-engine plane crashed 500 feet short of the Montana airport runway Sunday, nose-diving into a cemetery and killing seven adults and seven children aboard. Relatives said the victims were headed to an exclusive resort on a ski vacation.
Safety experts said finding the cause of the crash is likely to be significantly complicated by the absence of either a cockpit voice recorder or a flight data recorder, which isn't required for smaller aircraft that don't fly commercial passenger like airlines and charter services.
Former NTSB chairman Jim Hall pointed to similarities between the Montana crash and a March 26, 2005 crash near Bellefronte, Pa., in which a pilot and five passengers were killed.
The plane in both cases was the Pilatus PC 12/45 and was on approach to an airport. In both cases there were reports of conditions conducive to icing at lower elevations and witness reports that the plane appeared to dive into the ground.
"I'm certain they are also going to look at the weather conditions at the time and the pilot's training," Hall said. He pointed to a recommendation on NTSB's "most wanted list" of safety improvements that FAA test the ability of turboprop planes to withstand a particular type of icing condition called "super cooled liquid drops" before certifying the aircraft design for flight. FAA officials have said they're working on that recommendation.
"If you had some precipitation and the temperature was in the right range, that again is an area that investigators would look at," Hall said.
Hours after the crash, federal investigators had focused on overloading as a possible cause. The plane was designed to carry a total of 10 people, including two pilots. Of the children, a relative said there were two 4-year-olds and the other children were ages 1, 3, 5, 7 and 9.
"It will take us a while to understand," Rosenker said. "We have to get the weights of all the passengers, we have to get the weight of the fuel, all of the luggage."
Goglia said the Pilatus has a powerful engine for its size and is unlikely to be affected by the additional weight of a few children "unless they had an awful lot of baggage."
Standard flight procedures are for the pilot to file a report on the planes weight, including the weight of the passengers and the baggage and how that weight would be distributed around the plane, before taking off, safety experts said.
Peter Felsch, a meteorologist at the National Weather Service, said conditions measured on the ground not long after the 2:30 p.m. MDT crash were fair — winds of about 9 mph, 10 miles visibility, a temperature of 44 degrees Fahrenheit and a "broken cloud deck at 6,500 feet."
The Pilatus PC 12/45 is certified for flight into known icing conditions, according to the manufacturers' Web site and pilots who have flown the plane.
However, like all turboprop planes, it relies on deicing boots — strips of rubber-like material on the leading edge of the wings and the horizontal part of the tail — that inflate and contract to break up ice. That technology, which goes back decades, isn't as effective at eliminating ice as the heat that jetliners divert from their engines to their wings.
One key in the Butte crash will be whether the pilot had changed the position of the aircraft's wing flaps for landing because changing the configuration of the wings by moving the flaps is where icing problems often show up, said Peter Goelz, a former NTSB managing director.
There won't be any radar data of the plane's final moments for investigators to examine — like thousands of small airports, the Butte airport doesn't have a radar facility. The radar at the FAA's en route center in Salt Lake City, which handled the flight's last leg, doesn't extend as far as the Butte airport because of the mountains between.
The last radio communication from the turboprop's pilot was with the Salt Lake City center when the plane was about 12 miles from Butte, said Doug Church, a spokesman for the National Air Traffic Controllers Association. The pilot told controllers he intended to land at Butte using visual landing procedures rather than relying on instruments, which is not unusual, Church said.
Rosenker confirmed that the pilot said nothing to controllers to indicate he was having trouble, including during radio conversations earlier in the flight when the pilot notified controllers he intended to divert from the flight's original destination of Bozeman, Mont., to Butte.
"We don't know the reason for the requested change to the flight plan," Church said. "We don't know whether weather was a factor in Bozeman. There was no apparent reason given for the change in flight plan from Bozeman to Butte."
John Cox, an aviation safety consultant and former crash investigator for the Air Line Pilots Association, said the lack of data means investigators will have to go "back to the old, traditional way of investigating aircraft accidents, looking at impact angles, looking at damage done to the aircraft, whether the engine was producing power or not."
"Then they'll have to look for anything unusual — bird feathers, a piece missing off the engine — it will be a series of exclusions," Cox said. "It will be sketchy, and it will not be nearly as definitive as it would be if they had had one of the recorders."
The particular plane that crashed Sunday was registered to Eagle Cap Leasing Inc. in Enterprise, Ore. It wasn't listed on any air carrier's operating specifications and therefore couldn't carry passengers for hire, but that wouldn't preclude leasing, FAA spokesman Les Dorr said.