Wing Commanders (and Wing SEs for whom I have contact information):
Please disseminate the following message throughout your organizations UNLESS you have already forwarded a similar message. I know several pilots throughout NER at various levels of membership and I will be conducting an unofficial survey in a couple weeks to see if this message has filtered down to at least some folks...
* * * * * * * * *
To CAP Pilots and Aircrew Operating in Northeast Region:
I wanted to take this opportunity to review a couple critical points about icing conditions. Unfortunately, the old adage is true about flight safety being written with the "blood" of those who found out the hard way. I am sending this to you in the immediate aftermath of the Colgan Air (dba Continental Connection) Flight 3407 mishap near Buffalo, NY which claimed 49 lives plus one on the ground. Although the cardinal rule of air accident investigation is to NEVER speculate on the cause (especially ROOT cause) of a flight mishap in the early stages of evidence gathering, there is some factual information released by the FAA and the NTSB that remind us of the dangers of airframe (and engine) icing...regardless of whether that turns out to be causal in the Colgan Air Flight 3407 accident report. So here are some points to ponder:
CAPR 60-1 does not mention any flight restrictions for operations in known or forecast icing conditions. So we are left with guidance found in FAR Part 91 and each pilot's operating handbook (POH) for the models of aircraft that we fly. We must also remember to know our own personal limitations, exercise sound aeronautical decision-making, and apply ORM in instances which are not specifically addressed in regulations, laws, or manufacturers' operating manuals.
Remember that there are two basic types of airframe icing...rime and clear. Rime ice is bumpy and cloudy, while clear ice is smooth and harder to see. They form under different atmospheric conditions, but the takeaway here is that both types of icing are harder to detect on the lifting surfaces of high-wing aircraft and empennage. Airframe icing generally forms on the leading edges of unprotected structural components...wing struts, landing gear, wings, vertical stab and horizontal stab. Ice adds weight, increases drag, and (most importantly) changes the shape of the airfoil in ways we can't measure or anticipate. In short, icing destroys lift and changes the precise calculations of wing camber.
Remember that an aircraft can stall at any combination of weight, airspeed, flap setting and bank angle. An aircraft encounters an aerodynamic stall because the airfoil exceeds the CRITICAL ANGLE OF ATTACK. All the complex aerospace engineering boils down this critical angle of attack to target airspeeds to fly at various phases of flight, center of gravity requirements, flap settings and bank angles. However, all of that is based on one constant KNOWN condition...the airfoil remaining in the same shape and dimension as when it left the factory. Any alterations in that airfoil, negates all those other pre-computed values and you enter the realm of TEST PILOT! Even ice buildup on non-airfoil surfaces (like landing gear and engine cowling) could add weight and drag that lead to insidious airspeed decay. All pilots need to be vigilant (as in "semper vigilans") to unanticipated variations between power settings and airspeed.
Finally, please read the following excerpt from an NTSB briefing (www.fsinfo.org, or www.ntsb.gov):
At 30 minutes prior to the end of the recording, the flight crew briefed on the ILS approach to runway 23 at BUF, and destination weather. The crew also discussed their current visibility conditions flying at 16,000 feet, reporting 3 miles visibility in light snow and mist;
Commenting on restricted ("hazy") visibility at FL160, the pilot requested and was granted permission to descend to 12,000 feet. ATC cleared the aircraft to 11,000 feet soon after;
The flight crew activated airframe deicing systems, and discussed "significant" ice buildup on the windscreen and leading edges of the wings on the descent;
The pilot put the landing gear down one minute prior to the end of the recording, and selected Flaps 15 degrees 20 seconds after deploying the gear;
The flight director recorded a series of "severe pitch and roll excursions" seconds after the flaps were lowered. The crew apparently attempted to retract landing gear and flaps before the aircraft impacted.
The Q400 is equipped with pneumatic deicing boots on the leading edges of the wings, horizontal stabilizer, and vertical stabilizer.
Chealander warned the investigation has just begun, and it's too soon to draw any conclusions about what brought down the aircraft with 49 people onboard. "We're not ruling anything in, or anything out," he told reporters.
Many people will wonder why the aircraft could fly "just fine" with ice buildup and then depart controlled flight right at the end, just prior to landing. Without ascribing cause to this particular mishap, the hypothetical scenario cries out for further explanation. An amount of ice forming on a clean airfoil inflight might go unnoticed because it is conceivable that the ice could conform to the wing surface enough to allow continued lift performance from the airfoil. However, even a small amount of buildup on the wing (or tail, for that matter) could change the critical angle of attack just enough so that it wouldn't be noticed until there was some variable (change) introduced to the configuration. Lowering flaps changes the shape of the airfoil; reducing airspeed in anticipation of approach/landing also changes the angle of attack. Such changes introduced to a precarious and artificial equilibrium can certainly lead to a departure from controlled flight.
At best, a crew in that situation is faced with a KNOWN UNKNOWN; that is, they know they have icing but the extent to which ice buildup has changed their aerodynamic stability is not known. But even so, there are mitigating steps that can hopefully avert disaster. Kurt Pricer, NER/DOV reminds us to consider the following steps in case we encounter icing: "Keep the air speed up (especially on approach to landing), carb heat on, if you flew into it, make a 180 and fly out of it, and then land as soon as possible if necessary." Use pitot heat; remember to turn on any and all heat (deicing systems on your aircraft) early before ice buildup overcomes the equipment's capability. Keep in mind whether you are flying a carbureted aircraft, or an injected aircraft. You can also climb or descend to avoid icing. The best outcome is to avoid ice altogether, but if you start to accumulate ice, make smooth and gradual control inputs, airspeed and configuration changes. Pay attention to control forces on the yoke and listen to changes in engine and airflow noise.
The absolute worst case is the UNKNOWN UNKNOWN, where the crew has no idea they've accumulated structural ice and therefore doesn't even realize they must take mitigating steps to prevent catastrophe.
So all this boils down to my main point: Be extra careful with regard to the weather when operating in the northeast United States during winter (of course icing isn't restricted to the northeast, but my comments are directed toward pilots/aircrew of Northeast Region). Pay attention to those weather reports, especially the freezing level, listen for PIREPS, avoid atmospheric conditions of visible moisture, and check your airframe surfaces as best you can inflight...use your observers and scanners to look around and BEHIND you for ice buildup. There is obviously a larger chance of icing for instrument-rated pilots operating under IFR (not always feasible to avoid clouds).
Finally, don't get complacent...just because the flight has progressed perfectly to this point, doesn't mean that a change in aircraft configuration won't lead to rapid challenges in the minutes ahead. Thank you for your time, and remember to always FLY SAFE!
Vr
Bruce Brown
NER/SE
bbrown@ner.cap.gov 
