Statistical study of aircraft accidents and incidents related to de/anti-icing process in Canada between 2009 and 2014

Numerous research studies exist on USA aircraft icing situation and are based mainly on NTSB Accident Database and Synopses, FAA Accidents/Incidents Data System and the NASA Aviation Safety Reporting System. The aim of this... [ view full abstract ]

Numerous research studies exist on USA aircraft icing situation and are based mainly on NTSB Accident Database and Synopses, FAA Accidents/Incidents Data System and the NASA Aviation Safety Reporting System. The aim of this study is to show the existing Canadian situation of de/anti-icing ground accidents and incidents, by taking a critical look at Transport Canada’s database.
Accidents and incidents reports published in CADORS Transport Canada database were studied. Criteria related to ground, take-off and landing were used to sort interesting data. Every runway related events were excluded to focus on ground aircraft de/anti-icing process.

The analysis was completed in five steps: a) select icing related data from CADORS database, b) sort files related to the de/anti-icing process, c) read all selected files to determine those relevant to ground de/anti-icing process, d) remove those that don’t, e) and complete a qualitative analysis. The qualitative analysis is based on three main concerns: a) identification of contextual information as geographical dispersion or phase of flight, b) identification of accident/incident causes, c) and identification of consequences.
Results show that frequency of ground de/anti-icing events is most important in Québec and British Columbia. Those two provinces undergo 26% of total events. This phenomenon is not directly linked to traffic volume. Ontario presents the highest air traffic of all Canadian provinces, but one of the lowest accidents/incidents rate. Québec arrives in 4th position in regards to air traffic and presents a high percentage of accidents/incidents. It can be partly explained by airport and aircraft size. Small airports (less than 60,000 movements per year) are more prone to incidents and so do small airplanes (less than twenty-two seats). The turboprop engine powered airplanes are more affected by icing issues. The most critical period identified is from December 10th to January 10th. Incidents peaks match the cold peak (temperatures below -20°C), combined with a significant snow cover (daily snowfall superior to 10cm). 90% of incidents happen during take-off and initial climb phases. A typical scenario would be: identification of an abnormal situation, followed by detection of ice accretion on critical surfaces by the maintenance crew. Most incidents didn’t result in damage to aircraft. Most of the time, one would notice premature landings or rejected take-offs. On occasion, emergency is declared, a flight attendant may be injured, mayday declared, loss of control or aircraft rescue & fire fighting requested.
To the best of our knowledge, this study is the first Canadian database analysis, and allows us to describe more clearly the Canadian de/anti-icing accidents/incidents situation. The subject seems to call for more thorough studies to better understand existing interactions between the different factors of concern.