Aftermath – I Hope We Have Enough Fuel: An Aviation Fuel Emergency Turns Tragic

Here's a heartening statistic: To judge from an unscientific search of NTSB accident records, fewer than one in eight fuel-exhaustion accidents during the period 2006 through 2010 were fatal. Notice that these are the accidents that come to the attention of the NTSB because they involve injury or property damage of a certain magnitude. We do not know how many pilots run out of fuel and land without damage or injury; I suspect, however, that their number must fall somewhere between the 246 who escaped with their lives from a banged-up airplane in that five-year period and the 34 who died along with most of their passengers.

Perhaps it is an intuitive sense of these rather hopeful survival statistics that encourages so many pilots to flirt with Empty. Perhaps it is just irrational optimism.

A retired businessman, 83, and an old friend of his had for several years been in the habit of making a summer trip in a B55 Baron from their home in Georgia to Alaska. In August 2008 they repeated the trip, traveling by way of Kansas, Wyoming, Idaho and Washington before departing the Lower 48 for Skagway, their first planned destination in Alaska. According to an Internet posting by a friend of the passenger, the pilot had said that this would be the last time they would make the trip; he intended to sell the airplane.

They left Bellingham, Washington, late in the afternoon. Four hours and 40 minutes later the Baron landed for fuel at Gustavus, in the middle of the fractured coastline of southeastern Alaska. Finding no one at the airport, they took off again and contacted Anchorage Center to report that they were en route to Sitka.

“I hope we have enough fuel,” the pilot remarked.

The controller inquired how much fuel they had.

“About an hour,” the pilot replied.

Sitka, not counting maneuvering for the IFR approach, was half an hour away to the south-southeast — the direction from which they had just come. A couple of minutes later, the controller asked the pilot whether he would rather make the LDA approach at Juneau, only 37 nm from Gustavus, instead. The pilot replied, “I don’t think so, uh ... we haven’t done one, but I think Sitka would probably be OK, wouldn’t it?”

After the controller observed that the current Juneau weather was better than that at Sitka, which was reporting 2,500 overcast and 10 miles, the pilot changed his mind. The controller requested his altitude and instructed him to remain VFR. The pilot responded with, “We’re at 6,500 and we’re pretty well socked in.” The controller then instructed the pilot to climb to 10,000 feet while maintaining VFR. The pilot’s partly unintelligible reply seemed to suggest that he was unsure that he could maintain VFR up to or at 10,000 feet.

“Are you sure you don’t want to return to Gustavus with the weather like it is?” the controller asked, perhaps sensing that things were not going quite by the book.

“I told you,” the pilot replied, “there’s no one there at Gustavus; the place is locked, and we can’t go, no phone, nothing.”

The controller cleared him for the approach, which comes in from the west on a 68-degree heading with a slight right jog for landing on Runway 08. Minimums are 2,100 feet and four miles, with the missed approach point 3.2 miles from the runway threshold; the last segment is to be flown visually.

The pilot failed to intercept the localizer, and the controller instructed him to discontinue the approach and climb. The pilot then declined the offer of another try at Juneau, saying, “No, why don’t we just go to Sitka.” The controller pointed out that the weather was worse at Sitka and the approach there was another LDA.

“Can you do that?” he asked. “At Sitka?” the pilot replied. “Yeah, we ought to be able to do that.”

The minimums at Sitka are 400 and one.

About 22 minutes after their first radio contact, the controller asked the pilot for his fuel remaining and souls on board. The pilot replied that there were two on board and he had “about an hour and 10 minutes of fuel left.”

Eleven minutes later, the pilot called the controller and said, “ ... These LDAs are just like an ILS, aren’t they?”

“Affirmative,” the controller said. “It just doesn’t have a glideslope.”

Thirty-six minutes after the first contact, and perhaps 40 minutes after the Baron departed from Gustavus, the controller called to ask for a better estimate of remaining fuel. At first there was no response; then came a partially garbled one: “Looks like we’re having trouble with our left engine.”

That was the last communication from the Baron. When it did not arrive at Sitka and an ELT signal was reported along its route, a Coast Guard helicopter went out to search for it. The wreckage was spotted, less than two hours after the last radio transmission, in mountainous, forested terrain at the north end of Baranof Island, about 28 miles from Sitka. A rescuer, lowered to the accident site, reported that both occupants were dead.

Four members of a mountain rescue team and a pilot-rated state trooper went to the site the next morning. The trooper reported that there was no fuel in the airplane’s tanks nor any smell of fuel in the vicinity. The airplane was recovered several months later; no evidence of any mechanical problem was found, and both engines ran on a test stand.

The Baron, manufactured in 1978, had Continental IO-470s rated at 260 horsepower. Its usable fuel capacity was 136 gallons. Beech estimated that its cruise fuel consumption was “approximately 24.2 gallons per hour.” Cruise means different things to different people, but the pilot had bought 66.5 gallons of fuel after arriving at Bellingham from Boise, Idaho, suggesting that he was getting a block mileage of around 6.2 nm per gallon. At that rate, the 4.7-hour, 750 nm flight from Bellingham to Gustavus, with an allowance for taxi, takeoff and climb, would have used around 120 gallons, unless the pilot, mindful of the trip’s length, had leaned a little more aggressively than usual. In any case, his initial estimate of an hour’s fuel remaining when leaving Gustavus seems barely reasonable. Even assuming a direct flight and a successful instrument approach, he would likely have arrived at Sitka on fumes. By the pilot’s own estimate, the Baron cannot have met the IFR requirement for a 45-minute fuel reserve.

The pilot’s statement to the controller that there was “no phone, nothing” at Gustavus was puzzling. Although the airport was attended only until 1500, it had, as you would expect, a phone for pilots to call for someone to come out and pump 100LL or jet-A, and a sign explaining the procedure. There were hotels and motels in town. So the reason for the pilot’s decision to leave Gustavus for Sitka late in the day, in IMC weather and with absolutely marginal fuel, is unclear. It was a choice for which it is difficult to imagine any rational justification.

The pilot held single-engine and multiengine land, single-engine sea, instrument and glider ratings. He had reported 7,500 hours on his most recent application for a medical, which had been rejected by the FAA because of his history of “atrial fibrillation with excessive pauses.” The autopsy revealed that his heart was enlarged, weighing about twice the average, and his brain showed evidence of “several previous small strokes.” There was no suggestion, however, that his heart disease was related to the accident, which the NTSB blamed on “inadequate fuel planning and navigation.”

Considering the pilot’s experience and ratings and his multiple trips to Alaska, where LDA approaches — localizers that because of terrain are not aligned with the runway centerline — are common, it is surprising that he was unfamiliar with them. It may be that, by the same token that he had continued to fly despite losing his medical certification, he had allowed his instrument skills to become rusty in the belief that he would always avoid hard IFR and in a pinch he could still manage.

If the pilot did in fact intend to sell his airplane, he must have felt that his flying days were coming to an end. (As far as FAA Aeromedical was concerned, they had already ended.) That must be a difficult and painful curtailment to accept, but knowing when to stop is as important late in life as it is late in a long day's flying.
This article is based on the NTSB's report of the accident and is intended to bring the issues raised to our readers' attention. It is not intended to judge or to reach any definitive conclusions about the ability or capacity of any person, living or dead, or any aircraft or accessory._

Peter Garrison taught himself to use a slide rule and tin snips, built an airplane in his backyard, and flew it to Japan. He began contributing to FLYING in 1968, and he continues to share his columns, "Technicalities" and "Aftermath," with FLYING readers.

Your email address will not be published. Required fields are marked *

Subscribe to Our Newsletter

Get the latest FLYING stories delivered directly to your inbox

Subscribe to our newsletter