Find Post Investigations On:
Facebook Scribd Twitter
Friendfeed RSS Google Reader
» About This Blog | Meet the Investigative Team | Subscribe
Ongoing Investigation

Top Secret America

The Post explores the top secret world the government created in response to the attacks of Sept. 11.

Ongoing Investigation

The Hidden Life of Guns

How guns move through American society, from store counter to crime scene.

Have a Tip?

Talk to Us

If you have solid tips, news or documents on potential ethical violations or abuses of power, we want to know. Send us your suggestions.
• E-mail Us


Post Investigations
In-depth investigative news
and multimedia from The Washington Post.
• Special Reports
• The Blog

Reporters' Notebook
An insider's guide to investigative news: reporters offer insights on their stories.

The Daily Read
A daily look at investigative news of note across the Web.

Top Picks
A weekly review of the best
in-depth and investigative reports from across the nation.

Hot Documents
Court filings, letters, audits and other documents of interest.

D.C. Region
Post coverage of investigative news in Maryland, Virginia and the District.

Washington Watchdogs
A periodic look into official government investigations.

Help! What Is RSS?
Find out how to follow Post Investigations in your favorite RSS reader.

Hot Comments

Unfortunately I believe that we are limited in what we can focus on. I think that if we proceed with the partisan sideshow of prosecuting Bush admin. officials, healthcare will get lost in the brouhaha.
— Posted by denamom, Obama's Quandary...

Recent Posts
Bob Woodward

The Washington Post's permanent investigative unit was set up in 1982 under Bob Woodward.

See what you missed, find what you're looking for.
Blog Archive »
Investigations Archive »

Have a Tip?
Send us information on ethics violations or abuses of power.
E-Mail Us »

Notable investigative projects from other news outlets.
On the Web »
Top Picks »

'Black Box' Casts Doubt on Buffalo Crash Theory

POSTED: 05:59 PM ET, 03/25/2009 by The Editors
TAGS: aviation

What caused the worst air crash in the United States in more than seven years? The thinking has changed a bit today.

The initial theory about the Feb. 12 crash of Continental Connection Flight 3407 in Buffalo was that icing on the airframe sent the plane out of control. The Bombardier Dash 8 Q400 crashed into a house, killing all 49 people aboard and one man in the house.

Icing also has been cited as a possible cause of Sunday's crash of a chartered single-engine plane in Montana, which killed 14 people heading for a skiing vacation.

But information from the Buffalo plane's data recorders casts doubt on the icing theory in that case. The National Transportation Safety Board said today that the ice didn't appear to hamper the functioning of the aircraft. Moreoever, the data box shows that someone pulled back on the control stick after the plane's stall warning system activated -- which officials said could have increased the chance of a stall.

"The circumstances of the crash have raised several issues that go well beyond the widely discussed matter of airframe icing," NTSB Chairman Mark Rosenker says in a statement.

The NTSB has scheduled an unusual three-day public hearing on the crash beginning May 12. All five board members are scheduled to attend.

After a banner two years of safe air travel -- with no U.S. air fatalities in 2007 or 2008, the best years since the advent of jet travel -- there have now been three major plane crashes in the United States in 2009. (Besides Buffalo and Montana, there was the US Airways jetliner that landed in the Hudson River in New York City after a flock of geese hit the engines.)

According to the British insurance company Lloyd's, there have been five major air crashes worldwide with 92 fatalities so far this year. [See list from the Aviation Safety Network.] That still falls below the average rate of the last 10 years.

Lloyd's says that some authorities believe that safety risks could grow internationally, particularly in developing countries, due to increasing congestion, the economic downturn and the prospect that airlines may cut back on investments.

By The Editors |  March 25, 2009; 5:59 PM ET In the News
Previous: AIG Execs' Bonus Hold-Out, The Pitch for Expanded Powers, DOJ Invokes 'State Secrets | Next: More Departures at AIG?, DEA's Surveillance Plane Problems, Madoff Spotlight Swings to Brother


Please email us to report offensive comments.

So...will Jim Hall and Mary Schiavo now admit their histrionics about turboprops and icing right after the accident were inappropriate? Um...

Posted by: umtutsut | March 26, 2009 7:07 AM

The crew acted appropriately if they thought what they were experiencing was a tailplane stall due to icing, however.

The pilot may have been countertrained recently about tailplane icing, and worried about icing under those conditions, and made the wrong decision when the stick shaker came on.

Posted by: rhhardin | March 26, 2009 8:12 AM

I'm not an expert (nor do I play one on TV) but it still sounds like an icing problem to me. I recently read an article from a well-experienced pilot that explained the problem very well. In summary it goes as follows:

[1] plane is under autopilot.
[2] as the plane's wings are icing up the autopilot makes corrections to compensate for the icing
[3] the autopilot also compensates for the possibility of more ice accumulating on one wing than the other which is creating a balance problem
[4] as the plane approaches the runaway the pilot turns off the autopilot.
[5] with the autopilot turned off, the plane turns quickly due to the unbalance in icing on the wings and stalls as well due to the extra weight of the ice and the lack of lift.
[6] the pilot tries to pull up which also decreases the speed of the plane making the stall worse
[7] plane spirals straight down.

According to the experienced pilot who wrote the article the proper procedure would have been to nose the plane down to build up speed to increase lift. It's quite counter intuitive but it is the correct procedure to get out of the situation.

Regardless of how the NTSB will spin their results, icing was the primary cause of the crash. Secondary causes will likely be pilot error (or more likely inexperience with this specific situation), overuse of autopilot, and an inability of the planes de-icing mechanisms to keep up with the ice accumulating on the wings.

Posted by: ssejhill | March 26, 2009 12:54 PM

rhhardin describes an analysis I agree with so far. The pilot had most of his hours in a plane that was more prone to tailplane stall than the Q400 is reported to be, hence his reaction to low power and stick back, which made the actual wing stall worse. The NASA film on tailplane icing sure looked like the answer initially, especially with the flaps out sequence in the NASA film exactly matching the flaps out and stall sequence of Colgan 3407. That NTSB mentions sterile cockpit rules is curious as the first officer's voice to tower sounds giddy in her last transmission, suggesting perhaps that non-flight discussions will appear in the CVR transcript as they did for the Kentucky crash where distracted pilots attempted takeoff on the wrong (too-short) runway. Colgan 3407 was at 2,300 feet for many minutes before the crash -- well under the 10,000 foot ceiling where only flight conduct and flight discussion are allowed by the crew. As to icing, pilots have to watch for that in all seasons and at all locations (even warm climates), so Buffalo snow likely is not unsafe for flying.

Posted by: RJW_NY | March 26, 2009 3:16 PM

Unfortunately it appears icing, most likely, had no contributing factor in this event. Conditions at the time of the accident were reported by other pilots to be no worse than moderate icing. A Dash-8 can hold incredible amounts of ice. It's a daily occurrence for Dash pilots such as myself. The Colgan pilots had their deice equipment operating the entire flight; so unless it malfunctioned, the icing that night should have been a non-factor. You are right to blame the auto-pilot though. What most likely happened is the crew went to flight idle in a descent in order to slow down to configure for landing. They got distracted as the plane leveled off, allowing the airspeed bleed off. As the plane approaches stall the auto-pilot will disconnect and a stick-pusher (NOT shaker like the captain's old plane) will attempt to drop the nose. Apparently whoever was flying reacted improperly to the situation and fought the airplane, pitching up extremely, not adding power, thus creating a severely aggravated stall (identified by the extreme rolling). The stick pusher does not activate with a tail stall. Hate to break it to you folks but pilots talk all the time, women often sound cheerful on the radio, and the NTSB has no reason to "SPIN" an investigation... RIP all involved

Posted by: nyplt | March 27, 2009 9:47 PM

The stick pusher, though, exactly imitates the symptoms of a tailplane stall, the recovery from which is to raise the nose.

The problem is to account raising the nose in a stall, which is counter to every pilot's drilled-in instincts.

Posted by: rhhardin | March 28, 2009 1:33 AM

Kind of weard that all of a sudden planes start falling out of the sky and all the FCC can say is (maybe ice) well think for a moment ..when did this all start ? ANSWER.
When Iran sent up that satalite are they not telling us that they have a problem that maybe that satalite is messing with the planes ? When has anyone ever really told us american's what really is going on ...

Posted by: AMERICANDREAMER | March 28, 2009 7:45 AM

The nose drops in a wing stall too.... A tailplane stall generally occurs at higher air speeds, just as the flaps are lowered to an approach setting. The recovery from a tailplane stall almost always includes reversing the flap down(which instantly initiated the stall) selection and reducing power. The fact that Colgan had it's approach flaps down for 30 seconds and power at flight idle, prior to the upset, should have been a clue. My guess is the unfamiliarity with the EFIS scrolling tape airspeed indicator vs. a SAAB's steam gauge, contributed to their lack of situational awareness.

To recover from a wing stall, the nose will drop, and within a second you have recovered and are PULLING back on the controls. At the same time you are adding power to regain airspeed in level/climbing flight. During a tailplane stall the nose will drop but you are aggravating the stall, as the airspeed increases. The recovery has you pull back and raise the flaps, using power to maintain a lower speed than which the stall occurred, while in level/climbing flight. NO where does it say to pitch up to 31 degrees and let the airspeed fall off to nothing. These pilots were surprised and were unlucky.

Posted by: nyplt | March 28, 2009 10:29 AM

Post a Comment

We encourage users to analyze, comment on and even challenge's articles, blogs, reviews and multimedia features.

User reviews and comments that include profanity or personal attacks or other inappropriate comments or material will be removed from the site. Additionally, entries that are unsigned or contain "signatures" by someone other than the actual author will be removed. Finally, we will take steps to block users who violate any of our posting standards, terms of use or privacy policies or any other policies governing this site. Please review the full rules governing commentaries and discussions.

characters remaining


© 2010 The Washington Post Company