ice and rain affect normal operations-Emphasis on Icing
There is a clear inter-relation between safe and satisfactory travel by air and weather. Most of the accidents in airplanes occur due to adverse weather, and it is one among the different causes improving towards the occurrence of the accident. It can be blamed as the reason for most of the flight delays also. All flight operations are affected by unfavorable weather. This may prevent the handling of flight totally or sometimes partially. The expenses incurred due to delays and change in route due to such weather conditions is very high. Both the passenger and the aviation industry has to bear the brunt of these situations, due to the loss of time for the passenger, the hefty hotel charges, increase in fuel consumption, and additional spending on servicing, equipment and change in crew, as also they make flying more expensive beyond plans and budgets. Fogs, thunderstorms, freezing rain, snowstorms, crosswinds, poor visibility, icing, and en-route turbulence are the types of weather that make delays and changes in route unavoidable.
According to reports available the day the crash of 747-jetliner of Singapore Airlines occurred there was heavy rain in the city of Taipei. But the pilots are not prevented by such conditions. About the acceptable weather conditions for take-off and landing each airliner company has its own set guidelines, but to proceed or not is at the discretion of the pilot. Normally the aircrafts, especially the 747, fly even in heavy rain and winds. But as the jetliner sped up on the runway visibility of the pilot was, perhaps, affected by the incessant rain. The pilot had mentioned about the jetliner hitting an object on the runway during the process of the take off. It is the vision of the pilot that enable him keep the nose of the aircraft aimed straight while on ground. And when there is heavy rain outside, striking the window of the airplane it limits the visibility of the pilot. It is also possible that the engine fire of the jetliner was blown-off by the heavy rain entering its turbo engine. The fire can go off if the speed of the rain striking the engine exceeds two inches per hour. There is an intake of about 200 pounds of air in the case of a turbo jet engine, and when more than 10% of it is water the fire in the combustion chamber may go off.
All aerofoils, including the propeller, get affected and weight of the aircrafts is increased by ice, snow and frost. Even a light layer of frost can affect the airlift of the craft very much, and can be hazardous when it sticks on to the airfoils like wings, control surfaces, propeller and rotor blades. When the surface temperature of the airplane goes below dewpoint (below 32Â° F) or less than freezing point, formation of frost takes place. Though it forms early in mornings, it gets melted with sunrise and with the increase of heat and sunlight during the day. But a pilot has to take extra precautions to clear the frost before take off if the surface temperature is below freezing point and sky is not clear and heat from sun cannot be expected. When some pilots say there is no much frost on the wings, or that there is no change in shape by a simple look, they ignore the fact that the air flow across the airfoil may be blocked to a great extent by the surface roughness formed by the crystal formation of frost over them. This affects the airlift and makes take off and landing at low speeds risky. (Bernstein; Omeron; McDonough and Politovich, 1997).
While frost affects lift by reducing or destroying it, ice add weight besides affecting the lift. These factors together makes the pilot unable to maintain the required height, or prevents him from keeping the height below which he is not expected to fly for long. When thunderstorm also is present along with icing it becomes a nightmare for the pilot. Every year a number of accidents take place due to icing. 40% of the accidents are due to structural icing, which is also known as airframe icing. (Lankford, 2000). Carburetor and induction system icing, or engine instrument icing is the reason for the remaining 60%. The aircrafts may have to very often fly in such atmospheric conditions with high presence of aircraft structural icing. Most aircrafts, including even the Aerosonde, do not have equipments for proper de-icing fitted on them, to fly in such extreme conditions. And even if these are fitted, all aircrafts should avoid all icing conditions where its presence is high or very high-where ice formation can reach greater than 2 cm in a minute. (Lankford, 2000).
Ice additions even if bit by bit, can cause problems like reduced lift and speed and increased drag. Many a time icing is very risky. The sudden increase in drag due to ice accumulation on the main rotor and tail rotor blades affecting the lift necessitates additional power to drive the blades. Ice on the fuselage also demands more power as it increases the overall weight. The glassy formation of ice on the windscreens reduces or blocks visibility, making vision possible only through side or slide open windows. Vibration imbalances occur due to untimely fall of ice from blades. The engine air intake system gets blocked with ice formation over there. Ice falling from the fuselage comes like slabs or chunks and can damage rotors or enter into the air intake system blocking normal airflow. Freezing water comes in where they are normally not expected. Normal operation and control movements become difficult when ice is there on the flight controls like bell cranks, rod ends and pitch horn. Different temperature ranges are attributed to icing, in different texts. It may take place at temperatures of 0Â°C or below, and may vary based on a number of other factors.
Resistance to the forward motion of the aircraft through the air, known as the drag, is greatly influenced by the dirty formation of ice on the aircraft wing. There are two types of drags viz. The parasite drag and the induced drag. Induced drag is the one formed on lift, and builds up with the increase in angle of attack. Hence the induced drag will be higher in the case of aircrafts with contaminated wings than that with the normal ones as the ones with contaminated wings have to fly at higher angle of attack for the required lift at a particular air speed. More over there is a higher induced drag value at any angle of attack as the air flow is separated earlier from the upper surface of the wings due to the dirty form of ice on them. The increase in induced drag due to ice contamination is higher when compared to increase in parasite drag. The normal stall progression of a swept wing gets varied on a contaminated wing, which is influenced by surface roughness.
With a contaminated wing, the normal nose-down pitching moment in the direction of stall recovery, which accompanies a stall, is reduced. The effects of the degraded pitching moment quality vary with variation in angle of attack from an out-of- trim condition which can have abnormal responses to control column inputs. The sensitivity to ice contamination is high at the leading edge portion of the wing. The more the forward most extent of contamination is further away from the leading edge the less is the effects of this contamination. At temperatures just below the freezing point the slow accumulation of ice takes place and this calls for severe loss in aircraft movements. It is said that the major cause for take off accidents, specially in the case of jet transport aircrafts is the damaging effects on lift and drag due to ice accumulation and the related surface roughness of the wings.
To save aviation from risky icing conditions, proper and timely icing predictions are necessary. (Rasmussen et al., 1992). But it is not easy to give such a prediction of the icing conditions. Hence it becomes difficult to give flight plans leaving out areas prone to be of icing hazards. And the few available predictions give an exaggerated view of icing factors and on its severity. They are also limited to mesoscale icing conditions. These predictions can be of great use aircrafts, which are large in size. They are also of use in determining microphysical parameters acting upon the icing conditions. They would also be of great use to UAV and smaller aircraft operations along with of being benefit for Aerosonde operations. On noticing any adverse icing conditions while in flight the controller of the Aerosonde is to change the flight plan accordingly to save it from any possible loss or damage.
Predictions on icing may be verified with the pilot’s report on icing including its type and severity (Bernstein; Omeron; McDonough and Politovich, 1997). The probability of icing can be found out by observations of wind direction, temperature and dew point depression as also with hand analyzed surface charts, and satellite imagery. But using each of them can be time consuming. Thompson et al. (1997a) states: “in the real atmosphere, icing cannot exist in sub-saturated environments (RH<100%), except within precipitation or a decaying cloud.” This was followed by another study, which had the purpose of locating regions of cloud using NOVA-AVHRR satellite data. (Thompson; Bruintjes, Brown and Hage, Thompson, G., R.T. Bruintjes, B.G. Brown, and F. Hage, 1997).
Then a comparative study on the accuracy of the data on different regions as to which of them matched with predictions and actual conditions was conducted. And it is found that the satellite imagery enables a better isolation of probable icing area, except in the cloudless regions. The method of elimination would not be able to determine the icing takes place in cloudless regions. But in regions where cloud does not exist, it is unlikely that icing would occur in those regions. But it is possible that if an aircraft moves from an area with water drops present to a cloudless area where the temperature is below freezing point, then some icing occurs. For aircraft icing to occur there should be the presence of at least SLW, if not cloud.
Even though it is a rare possibility, it is possible for icing to occur at areas outside the ones with predicted icing conditions, when the scheme developed by Thompson et al. (1997) is used. Many findings of these algorithms may give near accurate predictions on the probable icing regions; but the area specified by these findings could be very wide. The icing predictions are necessary to isolate or limit the area of icing probabilities and to give wider area for the aircraft to fly, and this is the main aim of such predictions. But another study by Brown observed the fact that by algorithms using temperature and relative damp conditions it is difficult to arrive at the task. As on the factors leading to correct icing predictions there is much dispute among researchers today. Whatever are those factors responsible, it is an established fact that ice is “the most difficult systematic threat to flight safety today.” (Sand; Cooper; Politovich and Veal, 1984)
We shall now analyze the aircraft accidents, which occurred as a result of icing. Crash of Ozark Airline Flight 982, which was a Douglas DC-9-15, occurred on take off from the Sioux City Airport, Sioux City, Iowa on 27th Dec.1968. The cause of the accident was related to a stall near the upper limits of ground effect, and thereby losing control, as a result of aerodynamics and change in weight due to airfoil icing. The aircraft was not de-iced before take off. Another crash occurred on take off from Newark International Airport, Newark, New Jersey, on 27th Nov. 1978, of Trans World Airways Flight 505, which was a Douglas DC-9-10. At an altitude of about 65 feet above ground level, and speed of 154 knots, it lost control shortly after take off. Here also non-de-icing before take off and airframe icing were attributed to be the cause for the accident. Yet another crash took place on 5th Feb.1985, of an Airborne Express Douglas DC-9-15 while take off from Philadelphia International Airport, Philadelphia, Pennsylvania. Here too the cause found was due to non-de-icing before take off and airfoil icing. There are some common factors in each of these accidents, viz. 1.Below normal angle of attack stalling of aircraft shortly after take off was there in each of the crashes. 2.Freezing rain and/or snow was present. 3. De-icing of the aircraft before take off was not done.4. The aircrafts didn’t have leading edge devices or de-icing devices fitted on them.
Scandinavian Airlines System (SAS) McDonnell Douglas MD-81, registering OY-KHO and departing Stockholm (ESSA) for a flight to Copenhagen (EKCH), had a forced landing in a field, immediately after take-off, as both of its engines failed. The impact of the landing was such that the aircraft broke into three, and this occurred on 27th Dec.1991. It is a known fact that if the atmospheric humidity is high or if it rains, and simultaneously if the wings also are chilled, then ice formation can take place on the upper surfaces of the wings. It is another known fact that such ice get broken off on lift off due to the swift movements of the wings. During flight from Zurich the fuel had become very chilled. There were 2550 kg of fuel in each wing tank on landing, nearly 60% of the tank volume, sufficient enough to chill the upper surfaces of the wings. Weather conditions there was such that ice formation could take place. The flight technician noticed ice formation on the wings while on inspection of the aircraft at night. The passengers noticed that indication tufts were not moving on de-icing, and on take off the ice came peeling off the wings. 2.3.2 showed that the engine damage started with “soft” objects getting sucked into them. Considering all these facts the Board of Accident Investigation reached a clear conclusion that ice peeled off the wings on take off, which then was sucked into by the engines, and caused the damage. It was known for years that this type of engines sucked ice in. In the year 1985, a DC-9-51 had similar engine damages. This risk is even greater in MD-80 series of engines because of the wing tank configuration and larger intake area for air in the engine. (Air Traffic Accident on 27 December 1991 at Gottrora)
Another aircraft occurred when Japan airlines flight 8054, a Douglas DC-8-62-F crashed on take off from Anchorage International Airport, Anchorage, Alaska. The aircraft stalled at a height of about 60 feet AGL, at or shortly after reaching V2. The airframe icing was considered to be the cause behind this accident. This aircraft also was not de-iced before take of, as in the previous crashes. As the aircraft approached land, the icing conditions were such that ice accumulated on the wings. The United States had introduced regulations banning take off, if the wings, propellers or control surfaces of the aircraft had frost, snow or ice sticking on to them. Even today these regulations are in force, as cited under Federal Aviation Regulations (FAR) 121.629,135.227, and 91.209. Known as “clean aircraft concept” these regulations, were based on the fact that in the presence of ice formation of any type, changes of aircraft flight characteristics and degraded performance of aircraft were inevitable.
Seeing the increasing number of accidents of large transport aircrafts and small general aviation aircrafts and to clear the wrong ideas prevalent on the effects of slight surface roughness by ice accumulations on the flight characteristics and aircraft performance, and the good results of ground de-icing fluids, the United States FAA published Advisory Circular (AC) 20-117, in Dec, 1982. It was meant mainly to emphasize the ‘clean aircraft concept’ following ground operations proper to aircraft icing and to inform on help and guidance available for going in accordance. AC20-117 stated the wide- ranging effects ice formation on the aircraft due to differing nature of effects on the different designs of aircraft, and of their unpredictable nature of the formation of ice. It also stated that reduction of wing lift of up to 30% and increase in drag up to 40% can be the result when ice, frost, or snow with a thickness and surface roughness as that of a medium or coarse sand paper, accumulated on to the leading edge and upper surface of its wings, as indicated by wind tunnel and flight tests. Such changes in lift and drag, reduce control, increase stall speed and also change flight characteristics of aircraft. (Lankford, 2000)
AC -20-117 has given first place to surface roughness in the increase in drag and decrease in lift, and advices not try to take-off until it is confirmed that there is no snow, frost or other ice formation accumulating on to any of the major parts of the aircraft. Further AC -20-117 warns that the ice, frost or snow developed on ground, changes its effect on the flight characteristics unlike the ones developed during flight, and that any aircraft that can go through the icing conditions in forward movement get a certification for flight in icing conditions, ignoring the ice formed on ground. It also called for a close observation before take off as this only can clearly confirm that the aircraft is free of such ice accumulations. There are many factors influencing the accumulation of ice frost or snow. If moisture is splashed over, blown into, or sublimated onto the aircraft surface, it can cause surface roughness, as in the case of rain, snow or similar conditions. AC-20-117 vested the ultimate responsibility on the pilot-in-command to see that the clean wing concept is adhered to. (Lankford, 2000)
It is seen that some pilots expect a number of warning signals before anything serious due to icing occurs. But it may vary based on the speed of ice accumulation on the aircraft and very often it can be swift too. Often there are enough forewarnings like loss of climb rate, decreasing airspeed, difficulty in keeping the required height and similar problems related to overall performance seen. Swift actions of corrective measures are to be taken here to prevent total loss of control and to make a safe landing. Immediate action is necessary as icing starts. Climbing down to safer areas, going up if situation favors, or taking 180Â° turning to go back to safer areas are all possible course of actions. While considering ice, pilots often ignore windshield of aircraft. Many a piston aircraft defroster fails in removing ice from the windshield. On entering into air with temperatures at or below freezing level the pilot will have to fly blindfold unless the windscreen is made “anti-ice” (through heating or with chemical spray). It can be hazardous as at an altitude, when other aircrafts are not visible, and on landing the runway is invisible, and if the freezing temperatures extend to ground level. (Symons, 1996)
Thus in conclusion it can be said that icing has been a major cause for aircraft accidents, causing personal tragedy and agony, as also severe economic loss to aviation industry. The aviation industry goes on collecting data and improving on its findings on the hazards of icing, and it should go on for, ever to get better knowledge on icing impacts and to reduce such ice-related hazards. The need for additional training for pilots in the areas of standard operating procedures, cockpit resource management and decision-making was stressed by a recent U.S. General Accounting Office report (GAO). It also recommends, standards of minimum training program requirements to be incorporated, after considering a number of past accidents. There are different options to escape icing conditions, and success is in opting the best. Coming down to heights of warmer temperature is a good option in many cases. Climbing up, to safer heights, is another option, if there is enough power to do so, and also if clearness of skies above is certain. But here there is some risk involved as ice could form on the undersides of the wings, and at aft of any boot-or-bleed-air-protected leading edge wing panels, if more time is spent at climb angles of attack. This reduces lift to a great extent and hence some manufacturers publish minimum airspeeds to be used on climbing in icing conditions.
Hence A 180Â° turn is the best option. For a flight that came from ice- free area, a return to that area could be a safer move. And if there is an icing condition all around, then a landing at the nearest airport, or if none of them is close by, then an off-airport landing could be the best option. The pilot has to have a clear idea in mind as to what he would do if he enters an icing area all upon a sudden. And if a satisfactory plan with good chance of success cannot be reached at, then it is better not to fly. With the manifold growth of aviation industry, it is highly essential to get improved weather information, pass it quickly to aviation decision makers with accuracy, who in turn should be well equipped to understand its meanings in deep, make fast decisions to ensure safety of the aircrafts. The nation needs a well crafted, and well-coordinated Aviation Weather Program to meet the current needs, as also to prepare for the future necessities.
Bernstein, B.C., Omeron, T.A. McDonough, F. And Politovich, M.K. 1997: The relationship between aircraft icing and synoptic-scale weather conditions. Weather & Forecasting, 12, 742-762.
Lankford, T.T., 2000: Aircraft Icing: A pilot’s guide to supercooled drizzle droplets, icing accident case studies and cold weather techniques. Practical Flying Series, McGraw-Hill Companies, Inc., pp.336 pp.
Rasmussen, R., M. Politovich, J. Marwitz, W.R. Sand, J. McGinley, J. Smart, R. Pielke, S. Rutledge, D. Wesley, G. Stossmeister, B. Bernstein, K. Elmore, N. Powell, E. Westwater, B. Boba Stankov, and Burrows, D. 1992: Winter Icing and Storms Project (WISP). Bulletin of the American Meteorological Society, 73, 951-974.
Sand, W.R., Cooper, W.A. Politovich, M. K and Veal, D.L. 1984: Icing conditions encountered by a research aircraft. Journal of Climate and Applied Meteorology, 23, 1427-1440.
Thompson, G., Bruintjes, R.T. Brown, B.G. And Hage, F. 1997: Intercomparison of in-flight icing algorithms – Part I – WISP94 real-time icing prediction and evaluation program. Weather & Forecasting, 12, 878-889.
Symons, L. December 1996, Weather, Volume 51, pp.419-425
Author Unknown) (n.d) Air Traffic Accident on 27 December 1991 at Gottrra, AB country: report C. 1993:57, Case L-124/91