(See update below.) The first reaction to news of the crash yesterday over I-287 in New Jersey, in which a married couple in their 40s, their two children, and a friend in his 30s all died (plus a dog), is the overwhelming tragedy of it all. I am so sorry for everyone affected by this loss. Anyone who has been touched by such events knows that their effects fade but never really go away.
The second is of course to wonder how it could have happened -- in a Socata TBM 700 that, while it may sound small because it is referred to as "single-engine," is in fact a very robust and capable turbo-prop airplane. As always, it will take a while to know. But it will be very surprising if the answer does not involve in-flight icing. As a reminder, icing also played a crucial part in the Colgan commuter-plane crash in Buffalo nearly three years ago, and in the Air France 447 crash over the Atlantic a few months after that. (About which, yes, I have a ton of material to update and share.)
For now, an explanation and some resources.
Explanation: Why can icing conditions be so dangerous? There are two different but mutually worsening problems. The first is icing's aerodynamic effect. When ice covers the plane's wings and tail surfaces, they provide less and less lift (because their shape is changed -- see below). Eventually they provide none at all, and the plane simply falls out of the air. Ice over the plane's body makes it "draggier" too. Thus a vicious cycle: the change in wing-shape means that the plane has to maintain a higher airspeed to keep flying, even as drag and loss of power are slowing it down.
Ice build-up on a wing:
![Airframe Icing, Langley Flying School.jpg]()
![pitot.jpg]()
The other effect is on instrumentation. Reports of the Air France crash stressed the importance of possible icing of the "pitot tubes." These are the small elbow-shaped metal probes (right) that stick out from the wing or the front of the fuselage and help determine how fast the airplane is moving through the air. The airspeed calculations arise from comparisons of the "ram pressure" of the air as it comes into the small opening at the front of the pitot tube, versus the "static pressure" felt at small ports on the side of the plane.
An airplane's ground speed -- the combination of its speed through the air plus whatever head- or tail-winds it is flying through -- matters if you want to know arrival time. But airspeed -- the rate at which air is flowing over the wings -- is all that matters for controlling the plane. (In a 100-knot headwind, a small airplane with a 100-knot cruise speed could theoretically stay over the same point on the ground and yet be flying perfectly well.) Pitot tubes have heaters and other protections to keep them functioning in bad conditions. But as is so often true, natural forces can overwhelm the protective systems, and icing can be so sudden and severe that the tubes freeze over and airspeed indications are lost. That seems to have been the first step toward the Air France crash -- later steps involved the responses of the plane's autopilot systems, and its human pilots. It could well have played a role here. Explanation of how and why, and the effect on automated and human-directed flight-control processes, later.
Resources: A pilot and former National Weather Service expert named Scott Dennstaedt, with whom I have flown and whose weather-science seminars I have taken and online tutorials I have bought, has a blog with very detailed weather data, including now some about this crash. What he has put up this morning is preliminary, with more updates scheduled later today. It will be worth checking this evening and afterwards. Of the info available now, one significant point may be a PIREP, or "pilot report," by an airline pilot of "moderate to severe rime icing" two hours earlier in the area and at the altitudes the TBM 700 would have been flying though. "Severe" icing means conditions where the ice builds up faster than any anti-ice systems can deal with it.
Dennstaedt also has this brief and, as always in these circumstances, nearly unbearable audio of the last transmissions from the pilot just before whatever went wrong happened. You'll hear him at the start, with the call sign "731 Charlie Alpha." Shortly thereafter you will hear airline pilots on the same frequency. What struck me on listening is how focused and concerned* the airline pilots sound, with their larger and more powerful airplanes, about knowing exactly where icing conditions have been reported and how they can be prepared to avoid it.
Sympathies to all involved.
_____
* Update. Reader Steve Polychronopoulos writes to say:
![Email this Article]( "Email this Article")
![Add to digg]( "Add to digg")
![Add to Reddit]( "Add to Reddit")
![Add to Twitter]( "Add to Twitter")
![Add to del.icio.us]( "Add to del.icio.us")
![Add to StumbleUpon]( "Add to StumbleUpon")
![Add to Facebook]( "Add to Facebook")
![]()
![]()
![]()
The second is of course to wonder how it could have happened -- in a Socata TBM 700 that, while it may sound small because it is referred to as "single-engine," is in fact a very robust and capable turbo-prop airplane. As always, it will take a while to know. But it will be very surprising if the answer does not involve in-flight icing. As a reminder, icing also played a crucial part in the Colgan commuter-plane crash in Buffalo nearly three years ago, and in the Air France 447 crash over the Atlantic a few months after that. (About which, yes, I have a ton of material to update and share.)
For now, an explanation and some resources.
Explanation: Why can icing conditions be so dangerous? There are two different but mutually worsening problems. The first is icing's aerodynamic effect. When ice covers the plane's wings and tail surfaces, they provide less and less lift (because their shape is changed -- see below). Eventually they provide none at all, and the plane simply falls out of the air. Ice over the plane's body makes it "draggier" too. Thus a vicious cycle: the change in wing-shape means that the plane has to maintain a higher airspeed to keep flying, even as drag and loss of power are slowing it down.
Ice build-up on a wing:
The other effect is on instrumentation. Reports of the Air France crash stressed the importance of possible icing of the "pitot tubes." These are the small elbow-shaped metal probes (right) that stick out from the wing or the front of the fuselage and help determine how fast the airplane is moving through the air. The airspeed calculations arise from comparisons of the "ram pressure" of the air as it comes into the small opening at the front of the pitot tube, versus the "static pressure" felt at small ports on the side of the plane. An airplane's ground speed -- the combination of its speed through the air plus whatever head- or tail-winds it is flying through -- matters if you want to know arrival time. But airspeed -- the rate at which air is flowing over the wings -- is all that matters for controlling the plane. (In a 100-knot headwind, a small airplane with a 100-knot cruise speed could theoretically stay over the same point on the ground and yet be flying perfectly well.) Pitot tubes have heaters and other protections to keep them functioning in bad conditions. But as is so often true, natural forces can overwhelm the protective systems, and icing can be so sudden and severe that the tubes freeze over and airspeed indications are lost. That seems to have been the first step toward the Air France crash -- later steps involved the responses of the plane's autopilot systems, and its human pilots. It could well have played a role here. Explanation of how and why, and the effect on automated and human-directed flight-control processes, later.
Resources: A pilot and former National Weather Service expert named Scott Dennstaedt, with whom I have flown and whose weather-science seminars I have taken and online tutorials I have bought, has a blog with very detailed weather data, including now some about this crash. What he has put up this morning is preliminary, with more updates scheduled later today. It will be worth checking this evening and afterwards. Of the info available now, one significant point may be a PIREP, or "pilot report," by an airline pilot of "moderate to severe rime icing" two hours earlier in the area and at the altitudes the TBM 700 would have been flying though. "Severe" icing means conditions where the ice builds up faster than any anti-ice systems can deal with it.
Dennstaedt also has this brief and, as always in these circumstances, nearly unbearable audio of the last transmissions from the pilot just before whatever went wrong happened. You'll hear him at the start, with the call sign "731 Charlie Alpha." Shortly thereafter you will hear airline pilots on the same frequency. What struck me on listening is how focused and concerned* the airline pilots sound, with their larger and more powerful airplanes, about knowing exactly where icing conditions have been reported and how they can be prepared to avoid it.
Sympathies to all involved.
_____
* Update. Reader Steve Polychronopoulos writes to say:
Professional pilots know that planes are really cool and fun to fly, but that bad things can happen while flying them, ranging from scrubbed tires to having your brains spread evenly over the windshield. They also realize that their extensive training and experience means nothing without constant vigilance. You describe them as sounding "focused and concerned", but they sound closer to "scared" to me. Severe icing isn't a concern or a problem, it's an emergency.
