Survival: Ditching at sea

Martinique, the hard way

August 1, 2012

ditching at sea

We were delivering a recently converted PA–46T JetProp to Brazil. The airplane was a 2007 Malibu Mirage with 200 hours total time, equipped with an Avidyne Entegra EX5000 and all-glass panel. The modified airframe had only seven hours on a new Pratt & Whitney PT6A-35 when we departed California.

The airplane, like the four I had delivered before it, had been sold to a Brazilian businessman. After delivery, I was scheduled to train his new pilot.

My co-pilot Marcelo and I departed Clinton, Arkansas, before dawn, and after a stop in Fort Lauderdale, Florida, continued on to Borinquen, Puerto Rico. We covered more than 2,000 miles that day, and the airplane performed flawlessly. We fueled after landing, preparing for an early departure the following day.

After a full weather briefing—Puerto Rico would be the last stop where this would be available—we filed a flight plan, did a preflight, made sure everything was secure, and departed at 6:20 a.m. The sun was just rising behind a stack of cumulus clouds on the horizon. We climbed out, and before encountering the morning buildup, we were cleared by San Juan Center to go direct to Grenada, our next stop. We headed south.

It would be a short flight, two hours and 20 minutes, at FL270. We had more than four hours of fuel on board—enough to reach Boa Vista, Brazil—but that was risky. This was a route I had taken many times before. I knew that daily weather buildups caused frequent flight deviations, and a surplus of fuel meant we had more options.

The weather was as forecast: CAVU, no turbulence, and we had a slight tailwind. My co-pilot and I were watching the sun come up, enjoying being where we were and what we were doing. All I had to do was switch tanks when necessary and periodically scan the engine instruments.

And then it happened

We were 40 minutes out of Grenada, an hour and 40 minutes into the flight. I was beginning to plan our descent when things changed in an instant.

Four things happened at once: The airplane began to slow, the torque gauge quickly fell toward zero, the engine began to vibrate and shake, and dark smoke began pouring out the exhaust stacks.

My first concern was that the turbine, spinning at 48,000 rpm, might come apart. I quickly pulled the condition lever back to Flight Idle. At the same time I was concerned about fire. The header tank with 20 gallons of fuel was located only inches from the engine, and I didn’t know what might have come apart. I pulled back on the fuel control knob and feathered the prop. Then I closed the fuel shutoff lever, which closed a valve located between the header tank and the engine.

We were now a glider, with a range of approximately 60 nm. The nearest land was 148 nm to the east. We were going to have to ditch.

For the first time in more than 40 years of flying I declared an emergency. I was talking to Piarco Control, located in Trinidad and Tobago. The controller asked my intentions, and then went immediately to handling routine communications with another airplane. We were on our own.

The nearest land was Martinique, so we turned that direction, and were transferred to the Martinique controller. We were at the extreme edge of their radio range, but fortunately, the pilot of another airplane was able to relay our communications. He didn’t identify himself, but I’d like to buy him a cup of coffee sometime.

Preparing to hit

I was very fortunate in my choice of co-pilots. Marcelo is an experienced pilot who had traveled this route many times before. He took the controls long enough for me to don a life vest, and then he went to the rear of the airplane to prepare the airplane for ditching. Our gear was well secured, so I asked him to isolate what we would need: the life raft, first-aid kit, portable ELT/GPS, and any personal items. His most important job was to open the top half of the air-stair door prior to impact to insure we had a way out. He went one better by opening the emergency door as well. Good man.

He told me later that he had time to say a prayer. I’m glad he did.

I was busy up front: I shut down unnecessary equipment; activated the panel-mounted 406 MHz ELT; maintained contact with Martinique control as long as I could; stuffed my passport, wallet, cellphone, and hat into my shirt; and started to think about how best to hit the water.

Fortunately, I had a seaplane rating and had a rough idea of what we were facing. We would end up around 100 miles from land, in open water. I could see that the seas weren’t calm. Even as we descended through 15,000 feet, I could see they were rough, with occasional whitecaps.

I realized I wouldn’t be able to land into the wind, which would have helped slow our groundspeed, and the POH directed that water landings be made without flaps. With a low-wing airplane in rough water it would be easy to drag one wing and cartwheel. Again, that meant more airspeed at impact.

The waves were running close together and the surface was rough. Our best option was to land parallel to the swells.

That took care of my heading. The next problem was depth perception. I knew it was almost impossible to determine height above water without a reference point. I’d heard of parachutists landing in open water cutting their chutes away while hundreds of feet over the surface because they thought they were about to touch down. That kind of mistake can be fatal.

I lost my primary avionics at around 5,000 feet. I hadn’t thought about it, but our glide time exceeded the life of the twin batteries, and my backup gauges didn’t include a vertical speed indicator. I could have established a constant-rate descent at a given airspeed, but I wanted to hit tail-first at the slowest possible speed. I planned to stall it in.

As we descended through 5,000 feet I began calling out our altitude in thousands to warn Marcelo. I had taken my headset off, and we were able to communicate without an intercom. At 1,000 feet I turned parallel to the swells and told Marcelo to open the top half of the door and prepare for impact. I wanted to hit tail-first in a full stall, and we did, but I think I flared a bit high. We hit hard.

I was aware of only one impact. Marcelo later told me there were three. I had a pretty good size lump on the side of my head. I think the initial impact knocked me out momentarily.

When I turned to the rear of the airplane I saw the life raft was partially inflated, inside the airplane, and Marcelo was trapped in his seat. Working together we got the raft out of the airplane. By the time I got out, Marcelo was drifting rapidly away. He’d told me the day before that he wasn’t a strong swimmer, and while he had inflated his life preserver and was holding on to the side of the raft, he was unable to get in.

I wanted to grab some gear and maybe secure the door; I think the airplane might have floated a long time. It appeared to be intact, with the exception of the elevator, which was torn off. It had hit tail-first.

Instead I jumped in and swam to my co-pilot and the raft. I got in, and went to pull him in. We immediately capsized. I tried again, more cautiously, and we were in.

The three-man raft barely held the two of us and with the waves occasionally breaking over us, the raft was more like a bathtub full of water. I estimated the height of the waves to be running around six feet.

We were down. We were alive and had minimal injuries. We congratulated ourselves and said a silent thank you for any unseen help we were given, and we talked about a lot of things.

Forty-five minutes after hitting the water a Cessna 404 appeared and closed on our position. It dropped a smoke canister and set up a racetrack pattern around us. On three of his circuits he dropped down almost to the water to let us know he was watching over us. He was quite a pilot.

Shortly after that a military aircraft appeared. About the size of a C–130, it dropped marker dye and set up in a higher orbit. I believed it was coordinating the eventual rescue.

Ninety minutes after we hit the water a helicopter arrived. It dropped two swimmers and then retrieved the four of us. On the way to Martinique the Cessna 404 pulled into a tight formation with us. The helicopter had a large opening on the side, and we were able to silently communicate our thanks. Waving to the pilot brought tears to my eyes.

An hour later we were on a hospital helipad where we had a chance to thank the pilot and well-trained, professional crew that had collected us. I asked the pilot how many pilots he had rescued; he said we were the first. We had a few cuts and bruises, and some broken ribs. Given the circumstances we were extremely fortunate.


I’ve replayed the “movie” in my mind countless times. What could I have done differently? Should I have chosen to follow the chain of the Caribbean islands rather than go direct? It’s a fair question. When I started flying turboprops, I asked about their reliability and learned that more than 135,000 PT6s had been produced—and that the number of catastrophic failures could be counted on the fingers of one hand.

Since then I’ve flown much of North and South America, day and night. The route I was on was one I’ve flown many times before, as have other pilots. Many similar airplanes have been delivered to Europe, with much longer water crossings, with no reports of engine failures. I’m a believer in turbine reliability.

There is no question I could have followed the islands and might have been able to find an airport when the engine died. But our next three legs would have been over the Amazon, and had the engine died over that endless expanse, I doubt I would be here to write this account.

Was there anything that could have been done to restart the engine and nurse the airplane to an airport? I don’t think so. The engine was unable to produce any power, and any attempt to run it under those conditions would have increased the very real risk of fire.

What else did I learn? First, bring along a knowledgeable, experienced co-pilot, and the right gear. A bigger life raft would have been better. They don’t take up much room, and the slight increase in cost for a larger one is money well spent. Also, we had lap belts with a shoulder strap running diagonally across our torsos. I have no doubt that had the airplane been equipped with four-point harnesses (two shoulder straps), neither of us would have broken ribs, and I might have avoided a concussion. Other waterproof gear should include a first-aid kit, handheld radio, and portable ELT with GPS locator capability—plus food and fresh water to suit the circumstances.

Many experienced pilots advocate putting more hours on a new engine before taking it offshore. It’s said that there are two times in an engine’s life when it is vulnerable: when it’s very young, and very old. Despite the statistical reliability of turbines, I think that’s good advice.

Flights over water don’t require seaplane ratings, but I believe that the experience of having made water landings before was central to our survival.

We were lucky. We ditched early in the day, in good weather. We had onboard and portable 406 MHz ELTs with GPS capabilities. We were in constant contact with ATC, and we survived the impact with few injuries. We were quickly spotted from the air, and we were close to an island with professional air/sea rescue capabilities.

I couldn’t have picked a better companion than Marcelo. He was cool, calm, and collected, and without him in the back of the airplane to handle the door and lifesaving equipment, I don’t know that I would have survived.

Peter K. Boeck has been flying for more than 43 years and has 2,200 hours total time.