Mid-Air Collision Avoidance While Flying

by Dave Loveman. This excellent article is courtesy of http://www.ultralightnews.ca/, reprinted with permission.

Picture yourself flying along at about 500 ft over a 60-mi. long lake, after a nice long leisurely weekend of cross-country flying. There isn’t a cloud in the sky, the sun is setting, you have your helmet on and your CD player playing through the ear muffs.

Then, all of a sudden, you hear this God-awful roar and at the same time hear a screeching sound like metal on metal, your ultralight aircraft pitches forward and down, something hits your rudder, and you see a set of conventional-size floats pass by the top of your windshield. On landing, you find that your king post has been damaged, and your rudder fabric torn.

Now, picture yourself coming in for a landing in your little Buccaneer Amphibian after a nice relaxing flight over your local lake, and setting up to land back on your mile-long sod field. At about 300 ft in the air, you hear a scratching sound like something is moving along the bottom of your hull, then your plane pitches over to the left.

As it pitches, you apply full power and climb out, glancing over to see a Quicksilver MX in the process of taking off, the pilot unaware that his king post has just scratched the bottom of your hull!

The above are only two of a number of incidents that have happened in real life. Add to that, last year I lost a friend and fellow pilot due to a mid-air collision between his Kitfox and a conventional aircraft.

The skies we fly in are not nearly as dangerous as the roads we drive on, but unlike roads, our routes are not confined to the area between the lines in one direction or the other. There is no aspect of our day-to-day life that can prepare us for locating things in the sky as we fly.

Our eyes have been trained to take in the stop lights, brakes lights, turn signals, on-coming traffic, and if you are a motorcyclist, the car coming up to a stop sign from the side road, or the car just about to enter the passing lane. But in the air, our eyes have none of this to relate to, and instead of a path that is 35 ft wide with no depth or altitude, they now have to contend with a path 20-mi. wide and thousands of feet both above and below.

It is interesting to note that in studying ultralight accidents over the last 30 years, and cross-referencing them to my own experiences, I have found that most of our ultralight accidents involving mid-air collisions or close calls are in the area of our home bases, and happen during the day, in good VFR conditions.

Why? Because that is when we fly our little planes, and that is when and where we are most likely to encounter other aircraft taking off or flying in a defined circuit or pattern. Another problem is that most of us fly to relax, to get away from all the noise, traffic, and congestion, so our guard has been let down!

Another interesting fact is that most of our accidents happen when two speeds of aircraft are either landing or taking off in the same direction. That is, a slower plane is landing as a faster and higher plane is also landing.

In the case above, I was landing in my Buc while the Quicksilver was taking off. I was unaware he was taking off—he was unaware I was landing. In the case of the conventional floatplane, I was cruising along below him, unaware that he was preparing to land.

So, what can we do to help prevent these kinds of close calls and accidents? Well, the first step is to train our eyes on how to look for other aircraft when flying. In training, students encounter problems when they continually glance at their instruments as they land, and then back out at the runway—they lose “focus.”

That is, the pilot looks at something near to him, adjusts for light, depth (while the brain adjusts for familiarity, the pilot knows what he is looking for on the dash) and then he looks out into the distance where the eye has to refocus with really no depth perception—that is, there is no dash to focus on, just 20 mi. of sky.

If the student continually watches the instruments, he looses his “perception” of speed, altitude and distance. Studies show that it can take two seconds or longer for the eye to adjust every time this happens.

Training our eyes to look for other aircraft

Photo: Michael Wimmer

The first step is to train our eyes on how to look for other
aircraft when flying.

At 60 mph, you are covering a mile a minute, nearly 100 ft per second. Now add in the reaction time that would be needed to avoid a collision—say 10 or 12 seconds (to first recognize then react)—and you have traveled over 1 000 ft.

Let’s look at some of the problems our poor old eyes face when we fly.

We’ve talked about the problem the eyes and brain have adjusting to things near and far. But what else will affect our “collision avoidance judgment”? The following are some examples:

  • weather conditions—clear days versus hazy days;
  • windshield condition—clean versus dirty or scratched. Remember when you look at a plane in the distance, it will first appear as a “spot” on the windshield. But if you already have a number of “spots,” you may not catch the one that is closing in on you at 150 mph;
  • where the sun is—a sun glaring in the windshield makes seeing distances impossible;
  • pilot’s eye condition—wearing glasses or even sunglasses;
  • optical illusions—how many times have you been flying, and seen something shiny in the distance that looks like a large aircraft coming at you, only to have it turn out to be the sun reflecting off a building? How many times have you seen what looks to be another plane doing aerobatics in the distance, only to have it turn out to be a model plane just in front of you?
  • aircraft design—many aircraft have blind spots. They must be recognized and compensated for;
  • some pilots do not properly “fit” their aircraft, that is, they do not have a good view over the dash because their seats are too low, or the windshield area restricted;
  • stress, alcohol, fatigue;
  • distractions—an engine that is not running right, or an engine gauge that suddenly starts to move into the danger zone;
  • daydreaming;
  • colour of the planes—picture two white planes over a frozen, snow-covered lake, or two dark colored planes just before dusk;
  • a dark aircraft below you over a dark field, or set of buildings.

But is there one good way to scan? If there is, I haven’t found it. Each plane I have flown is different. Each has its own little blind spots, which only seem to become evident the more the plane is flown. Thus, the pilot who is “married” to his plane will be able to “find that certain way of doing it” that is comfortable for him and his “partner.” Also, each phase of flight generally requires a different kind of scan.

The following are some techniques that all pilots should start from and then build on:

  • Don’t glance quickly!
  • Don’t stare at one area for long periods of time.
  • When you look at an area, look at the area up and down in a specific area. This area should be about 10 to 15° up-down-left-right looking for movement, then move over a little, and repeat. This may sound like it will take a long time—in fact it takes relatively little time when practiced.
  • Look first to the area that poses the most danger to you for the phase of flight you are in. If you are turning onto base, take a good look before the turn, during the turn, and ahead of the turn. As well, look over to the area that other planes might be coming in at you on if their circuit were to be further out than yours.
  • If you have radios, clear yourself before all turns.
  • While on approach, especially on the downwind leg, look for shadows of aircraft on the ground. By practicing this, you will get in the habit of looking down while landing—and a plane that is hard to see above or below you will usually cast a shadow to one side or the other.
  • A good practice to get into is doing gentle “S” turns while climbing out or landing.
  • Ultralight pilots should aim to touch down one-third of the way down the runway. Why? Because conventional pilots generally aim for the numbers at the end of the runway—thus, your higher approach will make you visible to the conventional plane below you, take you farther down the field with more time to react, and in the case of an engine out, you can still make the field.
  • If you are at altitude and flying cross-country, then a scan 60° to the right and left from a centre line and 10 to 15° up and down should allow the eyes to catch movement all the way over to the side windows.

Hopefully the above will help you fly safely, and so will the addition of strobe lights on your wing tips and fuselage, and a landing light or landing lights on the nose or wing struts. Aircraft colours such as yellow or other contrasting colours to the sky and clouds will also help in making your plane more visible in the sky (this from a man who is totally colour blind).

Here are a couple of questions you should be able to answer, if not—hit the books!

When two aircraft are converging head on, what should each pilot do?

How do you properly overtake a slower aircraft?

For more information on safe ultralight tips and news, visit http://www.ultralight.ca/.

During a three-year study of mid-air collisions involving civilian aircraft, the U.S. National Transportation Safety Board (NTSB) determined that:

  1. The occupants of most mid-air collisions were on a recreational flight with no flight plan filed.
  2. Nearly all mid-air collisions occurred in VFR conditions during weekend daylight hours.
  3. The majority of mid-air collisions were the result of a faster aircraft overtaking and hitting a slower aircraft.
  4. No pilot is immune. Experience levels in the study ranged from initial solo to the 15 000 hr veteran.
  5. The vast majority of mid-air collisions occurred at uncontrolled airports below 3 000 ft.
  6. En route mid-air collisions occurred below 8 000 ft and within 25 mi. of the airport.
  7. Flight instructors were onboard one of the aircraft in 37 percent of the mid-air collisions.

Almost 50 percent of mid-air collisions result in at least one death. Naturally, mid-air collision avoidance (MACA) is an important aviation safety topic. With the sky becoming more and more congested, the threat of a mid-air collision is increasing.

According to the NTSB, the most probable cause of mid-air collision is the “pilot-in-command failed to see and avoid other aircraft.” Aircraft speeds today challenge our ability to “see and avoid.”

Here are a few more facts about mid-air collision:

  1. Mid-air collisions generally occur during daylight hours; 56 percent of the accidents occurred in the afternoon; 32 percent of the accidents occurred in the morning; 12 percent of the accidents occurred at night, dusk, or dawn.
  2. Most mid-air collisions occur under good visibility.
  3. Flight fatigue (fatigue resulting directly from flight-related operations) was not a major factor in mid-air collisions. The average flight time prior to the collision is 45 min. This time varies from takeoff to over seven hours; 60 percent of the pilots on the mishap flight had been airborne 30 min or less; only 6 percent had been flying longer than two hours.

Sequential Operation and RNAV (GNSS) Approaches to Intersecting Runways in an Uncontrolled Environment

by Patrick Kessler, Civil Aviation Safety Inspector, System Safety, Transport Canada, Civil Aviation, Quebec Region

The Kuujjuaq airport, in northern Quebec, is located in uncontrolled airspace. It has a mandatory frequency (MF) area with a radius of 5 NM and a vertical limit of 3 200 ft above sea level (ASL). The airport has a flight service station (FSS) that provides advisory service. Pilots must follow the reporting procedures for IFR (Canadian Aviation Regulation [CAR] 602.104) or VFR (CAR 602.101) aircraft, as applicable.

Runways 07/25 and 13/31 have nine different approaches, four of which are RNAV (GNSS) approaches. These include waypoints, flight paths, a final approach fix (FAF) and a missed approach path, which are shown only on the landing chart of the runway being used.

The radio navigation aids—non-directional beacon (NDB), VHF omnidirectional range (VOR) and instrument landing system (ILS)—are indicated on approach charts, and most are also indicated on enroute IFR charts and VFR navigation charts (VNC). Most pilots using this airport are familiar with the names and locations of these aids.

A recent incident highlighted the complexity of this environment when a number of aircraft with estimated times of arrival (ETA) within a short period headed to intersecting runways using different approaches.

The crossing altitudes can be equal to or higher than the sector altitude and this decision is left to the discretion of the pilot. The track of one of the aircraft heading toward a waypoint in order to initiate the approach can intersect the path of another aircraft.

The RNAV (GNSS) approaches for runways 25 and 31 at Kuujjuaq are shown on the same illustration, below, which demonstrates the proximity of waypoints “EPMIB” and “IMUVA”.

Superimposition of RNAV (GNSS) approaches for runways 25 and 31 at Kuujjuaq

Superimposition of RNAV (GNSS) approaches
for runways 25 and 31 at Kuujjuaq

Most of the time, pilots do not have equipment that could provide this kind of composite illustration. They have to be able to situate themselves mentally in the air in order to ensure separation from other aircraft.

Proper, effective and concise communication should reduce conflicting traffic situations. The use of a traffic alert and collision avoidance system (TCAS) and transponders contributes to improved safety.

Whiteout Claims Life of Inexperienced Helicopter Pilot

On March 19, 2008, a Bell 206B III helicopter was departing Réservoir Gouin, Que., on a private visual flight rules (VFR) flight to the pilot’s cottage located 42 NM to the east-southeast. Shortly after takeoff, at 08:37 Eastern Daylight Time (EDT), the aircraft struck the frozen, snow-covered surface of the lake. The pilot, the sole occupant on board, was fatally injured. The helicopter was destroyed. This article is based on the Transportation Safety Board of Canada (TSB) Final Report A08Q0054.

The aircraft was owned by a commercial helicopter company based in Alma, Que. The pilot was a friend of the company’s co-owner and would occasionally borrow the Bell 206 for private use when it was available. The occurrence flight was a private flight. The  company’s co-owner, himself a fixed-wing and rotary-wing private pilot, also privately operates a Cessna 206 fixed-wing aircraft.

At 07:00 EDT, the Cessna pilot and the Bell 206 pilot called the company operations in Alma from the Bell 206 pilot’s cottage via satellite telephone to get the weather conditions and forecast. It was partly sunny in Alma, 67 NM to the east; however, snow was expected by mid-morning. The weather at the cottage at that time was estimated to be one and a half to three miles visibility in light snow showers, with a ceiling at approximately 800 ft above ground level (AGL). The Bell 206 was preflighted and the two pilots took off for Réservoir Gouin at approximately 07:42 EDT to retrieve the Cessna, which had been stuck in the soft snow and slush-covered surface of the reservoir for over a week. They landed and shutdown behind the parked Cessna at approximately 08:07 EDT.

The Cessna, flown by the company’s co-owner, took off for Alma at 08:25 EDT in weather conditions considered to be instrument meteorological conditions (IMC). Under IMC, pilots are required to operate in accordance with instrument flight rules (IFR). The Cessna arrived in Alma at 09:37 EDT. At 10:00 EDT, when the Bell 206 pilot did not return to his cottage as planned, the operator was notified.

The operator uses a Guardian SkyTrax (SkyTrax) flight-following system to track its helicopter fleet, and the helicopter accident site was found at 14:09 EDT, 1.2 NM east of its take-off point on the flat, frozen, snow-covered surface of Réservoir Gouin. The pilot was fatally injured and the helicopter was destroyed. The weather at the time of the search was as follows: estimated ceiling at 1 500 ft AGL, vertical visibility approximately 800 ft, and horizontal visibility approximately 1 mi., at times one-half mile in constant moderate snow showers.

Whiteout claims life of inexperienced helicopter pilot

The helicopter struck the snow-covered surface of Réservoir Gouin on a northerly heading, in a 45° nose-down, left-side-low attitude. The helicopter struck the lake surface while in a high rate of descent. The main rotor blades struck the lake surface and the front cabin. The helicopter then tumbled, destroying the cabin sections and rupturing the fuel cell. The engine compressor and turbine casing deformation revealed signs of power at the time of impact.

Examination of the helicopter did not reveal any pre-existing mechanical abnormalities that could have contributed to the occurrence. The accident was not survivable because of the total destruction of the cabin area. The emergency locator transmitter (ELT) was damaged on impact, eliminating the possibility to transmit a distress signal and the wreckage location. There was no indication that incapacitation or physiological factors could have affected the Bell 206 pilot’s performance.

The graphic area forecast (GFA) weather charts showed a low-pressure system moving eastwardly across Quebec, which would have affected the weather in the Réservoir Gouin area by early morning on March 19, 2008. Other aviation weather reports surrounding the region indicated marginal to below-VFR conditions with poor visibility and snow showers. (For more on those weather reports, please read the full occurrence report on the TSB Web site.)

The Canadian Aviation Regulations (CARs) applicable to minimum visual meteorological conditions (VMC) for VFR flight within uncontrolled airspace state that no person shall operate an aircraft in VFR flight within uncontrolled airspace unless the aircraft is operated clear of clouds and with visual reference to the surface. Where the aircraft is a helicopter and is operated at less than 1 000 ft AGL during the day, flight visibility should not be less than 1 SM, except if otherwise authorized in an air operator certificate (AOC) or a flight training unit (FTU) operator certificate—helicopter.

Aerial view of accident trajectory

Aerial view of accident trajectory

Click on image above to enlarge.

Réservoir Gouin is a large, irregularly shaped body of water extending 55 NM east-west and 40 NM north-south. It is situated in Class G uncontrolled domestic airspace. The irregularly shaped shoreline made up of multiple inlets, fingers, and islands makes it particularly difficult to navigate, especially in poor weather. The weather at the time of the occurrence was fluctuating between VMC and IMC. The environment was conducive to whiteout conditions where the degree of contrast was low due to the overcast, obscure sky, flat light, reduced visibility in snow showers and the snow-covered reservoir. Upon taking off in an easterly direction, the pilot had a finger of trees as a reference below the helicopter and the expanse of the white snow-covered reservoir surface in front of him.

Flight in whiteout conditions may result in a poorly defined visual horizon that will affect the pilot’s ability to judge and stabilize aircraft attitude, or reduce the pilot’s ability to detect changes in altitude, airspeed, and position. If visual cues are sufficiently degraded, the pilot may lose control of the aircraft or fly into the ground or surface of the water.

A search of the TSB database for the period from January 1998 to the end of December 2007 revealed 18 helicopter occurrences involving collision with terrain in whiteout conditions. These 18 occurrences involved 45 persons, 13 of whom were fatally injured and 23 of whom were injured. Studies have indicated that a majority of whiteout condition occurrences happen during VFR weather conditions where the pilot is justified in initiating the flight or chosen route, but where visual cues are limited due to flat light, restrictions in visibility, overcast sky conditions and snow-covered terrain. In most cases, the pilot is unaware of a loss of visual references and a loss of control of the aircraft happens insidiously. The study did not indicate that low-time pilots were more at risk of being involved in this type of occurrence in comparison with high-time pilots.

The pilot in this occurrence obtained a Canadian private helicopter pilot licence in May 2005. His helicopter training was conducted on Robinson R22 helicopters and he was endorsed on the Bell 206 helicopter in November 2005. He did not hold an instrument rating. The pilot’s Category 3 aviation medical certificate was valid at the time of the occurrence; he was restricted to day flying only, with operational two-way radio communications. It was not possible to confirm the pilot’s experience on rotary-wing aircraft, but it is estimated that he had approximately 130 hr total time, 85 of which were completed on the accident helicopter. The pilot also held a private fixed-wing licence, obtained in May 2001. The total number of hours on fixed-wing aircraft is unknown, but at the time of obtaining his helicopter licence, he had approximately 65 hr on fixed-wing aircraft.

Both the fixed-wing and helicopter training included five hours of instrument flight training, including unusual attitudes flight training. Flying in whiteout conditions is discussed during ground school training, and if weather conditions permit, is demonstrated during dual instruction on the helicopter. Because the Bell 206 pilot’s training took place from March to May, it is likely that whiteout conditions could not have been demonstrated; however, this could not be verified during the course of the investigation.

On March 13, 2008, a similar helicopter occurrence (TSB occurrence A08Q0053) took place at dusk in whiteout conditions over a large, frozen, snow-covered expanse of water. The pilot survived the accident with minor injuries; the helicopter was destroyed.

The weather at the time of the occurrence was reported to fluctuate between VMC and IMC. The minimum visibility for operating VFR in uncontrolled airspace below 1 000 ft is 1 SM. The pilot had little experience flying in marginal weather. It is possible that the pilot’s decision to take off in low visibility and low ceilings was affected by fluctuating weather conditions and that the Cessna pilot had taken off in similar conditions just minutes before.

Whiteout conditions existed at the time of the occurrence, reducing the visual cues available to the pilot to maintain control of the aircraft. The pilot had little exposure to helicopter flight in whiteout conditions and may not have known to fly close to shore in order to use the trees and shoreline as contrasting cues against the white snow of the frozen lake. Inadequate ground references prevented the pilot’s accurate perception of the helicopter height and attitude in reference to the surface. It is likely that the pilot lost control of the helicopter while flying in whiteout conditions over the vast snow-covered frozen surface of Réservoir Gouin.

The SkyTrax tracking system installed on the occurrence helicopter was programmed to record the helicopter’s last known position every 2 min, which helped reduce the search area and locate the helicopter in a timely manner.

In its findings as to causes and contributing factors, the TSB concluded that the pilot likely encountered whiteout conditions, making it difficult to maintain visual reference and causing disorientation, which resulted in impact with the frozen snow-covered lake.

In closing, the TSB mentioned in the safety action taken section that Transport Canada published the article “Coming Soon to a Theatre Near You: Whiteout” in issue 4/2008 of the Aviation Safety Letter (ASL). We sincerely hope that the article in ASL 4/2008 and this new one will serve their intended purpose: to promote awareness and prevention.

Massive headache prevention for helicopter pilots: Wear a helmet!

Takeoff in Conditions of Freezing Drizzle and/or Light Freezing Rain (Fixed-Wing Airplanes)—Part II

by Paul Carson, Flight Technical Inspector, Certification and Operational Standards, Standards, Civil Aviation, Transport Canada. This is the second of a two-part article on this critical subject. Part I appeared in Aviation Safety Letter (ASL) 4/2009.


During the winter of 2005–2006, a Transport Canada Civil Aviation (TCCA) inspector observed a number of airplanes operated by various air operators taking off in conditions of freezing drizzle (forecast and actually reported). The inspector considered that the operations were in contradiction of Canadian Aviation Regulations (CARs), specifically CAR 605.30:

De-icing or Anti-icing Equipment

605.30 No person shall conduct a take-off or continue a flight in an aircraft where icing conditions are reported to exist or are forecast to be encountered along the route of flight unless

(a) the pilot-in-command determines that the aircraft is adequately equipped to operate in icing conditions in accordance with the standards of airworthiness under which the type certificate for that aircraft was issued; or

(b) current weather reports or pilot reports indicate that icing conditions no longer exist.

Subsequent discussion identified that air operators and flight crews have insufficient information when faced with conducting a takeoff in these conditions. These discussions also identified that nothing in the current regulations and standards authorizes, nor strictly prohibits, takeoff during conditions of freezing drizzle and/or light freezing rain.

Hazards associated with in-flight operation in supercooled large drop (SLD) icing conditions

Start of contamination
Anti-ice fluids are designed to flow away from the aerofoil critical leading edge region and off the trailing edge as airspeed increases. Although this behaviour will differ for different fluids, different airfoils, different temperatures, etc., a reasonable assumption is that the critical leading edges will be free from all fluid at rotation. Once again, approval of flight in icing conditions includes demonstration of satisfactory performance of the ice protection systems (IPS) as well as demonstration of satisfactory handling qualities and a measurement of the performance degradation with the ice expected on both the unprotected surfaces and any residual ice on the protected surfaces resulting from proper operation of the IPS. Although not just limited to taking off in freezing drizzle and/or light freezing rain, approval also includes other conditions in U.S. Federal Aviation Regulation (FAR) 25, Appendix C, one being the assumption that ice accretion on surfaces begins at liftoff.

Impingement limits
With SLD icing conditions, the droplets are larger and have greater momentum due to the higher mass. The droplets will impact the leading edge of an airfoil section over a greater chord-wise extent than the smaller droplets associated with FAR 25, Appendix C conditions. In addition, SLD droplets may splash and break up into smaller fragments, which may run back prior to freezing. IPS that have been designed to prevent ice build up (anti-icing systems) or remove accreted ice (de-icing systems), have not been demonstrated to be effective in SLD icing conditions.

Pneumatic boot operation in SLD icing
A problem has been identified in the design of pneumatic boot de-icing systems on some airplanes where the chord-wise extent of the boot-protected area did not consider SLD icing conditions, thus resulting in ice accretion aft of the protected area. This accretion has been particularly hazardous when a residual ridge of ice is left just aft of the boot on the upper wing after boot operation to break off ice. Flight tests on several different airplanes, using a tanker airplane to simulate in-flight SLD icing conditions, have shown that a ragged, span-wise ridge forms just aft of the protected area.

Residual icing ridge formed aft of boot-protected surface due to boot inflation

One effect of this ridge can be non-linear hinge-moment characteristics on trailing edge controls. For unpowered controls, hinge-moment anomalies at the surface can result in pulsing of the pilot’s control, and in the extreme, a reversal in the direction of the pilot’s force can occur. That is, the control can self-deflect to an extreme position, and excessive pilot effort can be required to return the control to a neutral position.

One accident and two incidents in SLD icing conditions

The section below describes one accident and two incidents where encounters with SLD have been documented. There are other encounters that have been documented in various databases where SLD was suspected, but much of the information was collected for other reasons, not specifically for SLD icing conditions.

ATR 72 accident at Roselawn (31 October 1994)
The National Transportation Safety Board (NTSB) in the United States concluded that this accident occurred due to ice accretion on the wing upper surface just aft of the leading edge pneumatic boot and in front of the trailing edge ailerons. The airplane was in autopilot control during a hold at approximately 8 000 ft with the flaps partially extended. The flaps were then retracted. The increase in the wing angle of attack (AOA) due to the flap retraction caused a flow separation at the wing tip due to the ice accretion. The flow separation caused a hinge-moment discontinuity at the aileron, which in turn caused the ailerons to self deflect to full deflection. The autopilot was unable to correct the overbalance and the airplane had a lateral departure from which recovery was not accomplished.

The icing conditions identified in this accident included SLD icing conditions. Much of the aircraft accidents in SLD conditions deal with the arrival phase, long holds at slow airspeeds similar to this accident.

Transportation Safety Board of Canada (TSB), Aviation Investigation Report, Roll Oscillations on Landing, Airbus A321-211, Toronto/Lester B. Pearson International Airport, Ontario, December 7, 2002, Report Number A02O0406

At approximately 16:07 Eastern Standard Time (EST), an Airbus A321-211 airplane was on approach to Toronto/Lester B. Pearson International Airport (LBPIA), Ont., with 123 passengers and 6 crew on board. At approximately 140 ft above ground level (AGL), on final approach to Runway 24R with full flaps selected, the airplane experienced roll oscillations. The flight crew leveled the wings, and the airplane touched down firmly. During the approach, the airplane had accumulated mixed ice on areas of the wing and the leading edge of the horizontal stabilizer that are not protected by anti-ice systems.

Approximately three hours later on the same day, another Airbus A321-211 airplane, with 165 passengers and 7 crew on board was on approach to Runway 24R at LBPIA. At 18:59 EST and approximately 50 ft AGL, the airplane experienced roll oscillations. The flight crew conducted a go-around, changed flap settings, and returned for an uneventful approach and landing. At the gate, it was noted that the airplane had accumulated ice on areas of the wing and the leading edge of the horizontal stabilizer that are not protected by anti-ice systems. There was no damage to the airplane or injury to the crew or passengers.

Given the similarities between these two occurrences, the TSB concluded (1) “It is likely that the icing conditions encountered by both aircraft were outside the Federal Aviation Regulation 25, Appendix C envelopes used for certification of the A321,” and (2) “Drizzle droplet size ranged from 100 to 500 microns. Federal Aviation Regulation (FAR) 25.1419, Appendix C envelope for certification of flight in icing conditions has maximum mean effective drop diameter between 40 and 50 microns.”

The full report can be found at the following Web site: www.tsb.gc.ca/eng/rapports-reports/aviation/2002/A02O0406/A02O0406.asp.

Meteorology measurement criteria forecasting/reporting freezing drizzle and/or light freezing rain vs. FAR 25, Appendix C

Weather forecasts are not made in terms of FAR 25, Appendix C parameters such that they would match the certification icing environment. Also, pilot reports (PIREP) of icing conditions are unique to the airplane from which they are reported—light icing to a Boeing 727 could be heavy icing to a Beech Baron.

Appendix C is not adequate for freezing drizzle and/or light freezing rain given that the maximum droplet size in the appendix is 40 microns for stratiform droplets and 50 microns for cumuliform droplets, whereas the smallest probable drizzle droplet size is 100 microns, and raindrops begin at 500 microns. Furthermore, any cumulus cloud that has a vertical extent that is greater than its horizontal base may include “appreciable numbers” of droplets that are larger than 50 microns.

A minor point, but it should be noted that maximum drop size in FAR 25, Appendix C is 40 microns (or 50 microns) median volume diameter (MVD) or mean effective diameter (MED), not absolute diameter droplet size. The “smallest” freezing drizzle and/or light freezing rain drops are actually measured in absolute diameter terms, not MVD or MED.

Transport Canada Civil Aviation Requirements

Icing certification
In general, TCCA follows the same certification requirements as the Federal Aviation Administration (FAA). These requirements include use of FAR 25, Appendix C as a definition of the in-flight icing atmosphere. TCCA does have additional guidance material on how compliance must be demonstrated for performance and handling qualities. This guidance has led to different limitations and/or configurations of IPS for many foreign airplanes, mainly turbopropeller powered airplanes. In some cases, other authorities have subsequently adopted these additional measures after accidents.

Operational requirements in CARs Part VI and Part VII
The relevant operational regulations relating to flight in icing conditions are contained in CARs Part VI—General Operating and Flight Rules and in Part VII—Commercial Air Services. The following extracts are pertinent:

(a) CARs Part VI, Subpart 2—Operating and Flight Rules
602.07 No person shall operate an aircraft unless it is operated in accordance with the operating limitations

(a) set out in the aircraft flight manual, where an aircraft flight manual is required by the applicable standards of airworthiness.

(b) CARs Part VI, Subpart 5—Aircraft Requirements
605.30 No person shall conduct a take-off or continue a flight in an aircraft where icing conditions are reported to exist or are forecast to be encountered along the route of flight unless

(a) the pilot-in-command determines that the aircraft is adequately equipped to operate in icing conditions in accordance with the standards of airworthiness under which the type certificate for that aircraft was issued; or

(b) current weather reports or pilot reports indicate that icing conditions no longer exist.

It should be noted that there is a proposal to change the content of CAR 605.30 contained in Notice of Proposed Amendment (NPA) 1998-252 to read as follows:

605.30 No person shall conduct a take-off or continue a flight in an aircraft under IFR where icing conditions are reported to exist or are forecast to be encountered along the route of flight or under VFR into known icing conditions unless

(a) the pilot-in-command determines that the aircraft is adequately equipped to operate in icing conditions in accordance with the standards of airworthiness under which the type certificate for that aircraft was issued; or

(b) current weather reports, pilot reports, or briefing information relied upon by the pilot-in-command indicate that the forecast icing conditions that would otherwise prohibit the flight will not be encountered during the flight because of changed weather conditions since the forecast.

The intent of the proposed change is to permit more flexibility in operating in reported icing conditions. However, it does not clarify the situation with respect to taking off in freezing drizzle and/or light freezing rain. In addition, the present status of the NPA is with the Regulatory Unit (RU) of TCCA pending publication in Canada Gazette, Part 1.

(c) CARs Part VII, Subpart 4—Commuter Operations
704.63(1) No person shall conduct a take-off or continue a flight in an aircraft when icing conditions are reported to exist or are forecast to be encountered along the route to be flown unless the aircraft is equipped to be operated in those conditions and the aircraft type certificate authorizes flight in those conditions.

(d) CARs Part VII, Subpart 5—Airline Operations
705.69(1) is identical to 704.63(1).

Interpretation of operational requirements
As noted above, the aircraft flight manuals (AFM) of currently certified airplanes do not contain any specific limitations prohibiting takeoff in SLD icing conditions. The Type Certificate may or may not reflect the wording in the AFM, but will specify whether the certification basis includes the applicable FAR paragraphs relating to ice protection. Also, the Type Certificate is not a document that is generally familiar to air operators and flight crews. It is possible through a Supplemental Type Certificate (STC) to have an IPS (more frequently seen for small airplanes) added to airplanes that would include additional limitations regarding flight in icing conditions.

The AFMs of some airplanes do contain a limitation indicating that if severe icing conditions occur (as identified by various visual cues), the airplane must immediately exit these icing conditions. Severe icing is noted as including freezing drizzle and/or light freezing rain. The differences in measurement criteria between FAR 25, Appendix C and aviation meteorological reports remain.

TCCA continues to collect and analyze data in consultation with other authorities worldwide in an effort to enhance present day knowledge regarding the safety of flight in conditions of freezing drizzle and/or freezing rain.


  1. J. C. T. Martin, Transport Canada Aircraft Certification Flight Test, Discussion Paper No. 41, The Adverse Effects of Ice on Aeroplane Operation, Issue 2, 4 July 2006.
  2. J. C. T. Martin, Transport Canada Aircraft Certification Flight Test, Discussion Paper No. 50, Takeoff in Conditions of Freezing Drizzle or Freezing Rain (Fixed-Wing Aircraft), Issue 2, 29 September 2006.

Author’s note: Part 1 of the above article was published in the ASL 4/2009. It contained the following conclusion. 

Takeoff into known freezing drizzle and/or light freezing rain is outside of the flight envelope for which any airplane currently operating today is certificated. Not only is it unwise to operate in such conditions, it is also unsafe, and based on the best information available at this time, also illegal.

Transport Canada (TC) has undertaken a review of the current practice of taking off in freezing precipitation to assess potential hazards and determine whether any regulatory or safety action is required. TC has not reached a final conclusion on this issue, but after reviewing current practices, it has identified important safety information to share in this ASL article and the previous one.

The article is intended to inform operators and flight crews of the potential hazards of taking off in conditions of freezing drizzle and light freezing rain. This article stresses the importance of understanding the hazards associated with operating in icing conditions and the limitations associated with the certification of airplanes for flight into known icing conditions.

At this time, TC has not drawn any firm conclusions on the safety of taking off in freezing drizzle or light freezing rain. However, TC is of the opinion that taking off in freezing drizzle and light or greater intensity freezing rain may be hazardous and, in the case of moderate or heavy freezing rain conditions, these fall outside the protection afforded by de-icing and anti-icing fluids. TC will consult the aviation industry to consider the effectiveness of current regulations and standards.

TC therefore retracts the last sentence of the conclusion and replaces it with a reiteration of the current guidance on this subject. Specifically, operation of an aircraft in conditions of freezing drizzle or freezing rain should be avoided whenever possible.

Date modified: