Understanding Air France 447 (5 page)

The night of the accident, AF447 was in radar and VHF radio contact with the Recife controller as they left the coast of Brazil. The controller instructed the crew to contact Atlantico over the INTOL waypoint (the boundary between Recife and Atlantico), and Dakar over TASIL waypoint (the boundary between Atlantico and Dakar) and issued the frequencies for each.

At 01:34 the crew contacted Atlantico and successfully completed a SELCAL check, which verified that the controller could call the flight. That was the last radio voice contact with Air France 447.

At 01:35 the Atlantico controller asked the crew three times for its estimate for passing TASIL. The crew did not answer, nor was the call noted on the voice recording.

One disadvantage of both HF with SELCAL, and CPDLC is that you cannot hear what the pilots of other aircraft are doing and saying. To somewhat make up for that, pilots will often communicate on a designated air-to-air frequency to exchange such information. The night of the accident other flights along the route were diverting around the worst areas of weather, but no air-to-air calls are in the transcript. The crew of AF447 was on their own to determine if a deviation around weather was warranted.

Chapter 4: Intertropical Convergence

As the flight progressed north of the Brazilian coastline, it entered a band of weather fueled by global circulation patterns known as the intertropical convergence zone (ITCZ).

Soon after the accident occurred, it was a reasonable guess that they had flown into a large thunderstorm and suffered severe damage, which brought the airplane down. After all, the airplane flew through an area of known severe thunderstorms and was lost minutes later.

A detailed meteorological analysis of the flight was done by Tim Vasquez, a meteorologist who did weather route forecasting for the US Air Force in the mid 1990’s. His excellent analysis can be found at his Weather Graphics website:
http://www.weathergraphics.com/tim/af447/.
If you are at all interested in the more technical aspects of the weather, do not miss this site. Several graphics in this publication are provided courtesy of Mr. Vasquez.

Weather does play a part in this accident. The crew flew into an area of heavy weather, with only a slight deviation from course. The pitot tubes became clogged and shortly thereafter the crew was unable to maintain control of the airplane.

But airplanes have been dealing with thunderstorms and icing for the better part of 100 years, and in high altitude jets for over 50. So what happened here?

The evidence points to a unique combination of the characteristics of the inter-tropical convergence zone, and the specifics of particular models of pitot tube used by Airbus.

The A330-200 and -300 models were launched in 1994 and 1998 respectively. It would be 2008 before this pitot tube issue was known and well after the accident before it was understood.

The ITCZ

The ITCZ is a region that circles the globe near the equator. The zone moves slightly north of the equator in the northern-latitude summer (May-September), as it was on the night of the accident.
⁸
Unlike land-mass thunderstorms which are often driven by convection from below or by frontal action, the thunderstorms of the ITCZ are driven by global circulation patterns with warm moist air coming from the equatorial region. The prevailing easterly trade winds of the Northern and Southern hemispheres converge provide the lifting action required to create a storm. As with all thunderstorms, when the air rises it expands and cools leading to cloud formation.

The tropopause is the dividing line between the troposphere and stratosphere and acts like a ceiling on vertical storm development. It is the point at which an anvil head forms on many thunderstorms, as they cannot grow any higher. Near the equator, where the ITCZ is, the tropopause is typically in the 50-60,000 foot range, as opposed to 30-40,000 feet typical for the mid latitudes. The high tropopause at these low latitudes means that the storms can grow to great heights leaving little prospect of flying over them.

The significance of the ITCZ for aviators is that the oceanic thunderstorms within it show up poorly on weather radar. These equatorial storms also tend to produce less lightning than higher latitude storms, which may tend to mask their severity - especially at night. A moonless night and lack of lightning makes it difficult to make a visual evaluation of the storm. For AF447, the half moon was setting in the west, off the aft left of the airplane. The storm tops, by some estimates, towered another four miles above them.

Studies of storms in this region have shown a weakening of updrafts in the 20,000 foot range, which may account for the lesser amounts of lightning produced. Above approximately 20,000 feet ice crystals form. This shift to a lower energy state of matter (water to ice) gives off a small amount of heat which then adds to the updraft’s upward velocity to reach, and often penetrate, into the stratosphere.

A meteorological analysis of the flight 447 theorizes that the thunderstorm tops reached 56,000 feet, with updrafts strong enough to penetrate the stratosphere by about 6,000 feet.

The final accident report states, “The Captain appeared very unresponsive to the concerns expressed by the PF about the ITCZ. He did not respond to his worry by making a firm, clear decision, by applying a strategy, or giving instructions or a recommendation for action to continue the flight. He favored waiting and responding to any turbulence noticed. He vaguely rejected the PF’s suggestion to climb, by mentioning that if “we don’t get out of it at three six (36,000 ft), it might be bad”.

An enhanced satellite image, courtesy of Tim Vasquez’s site, shows the track of AF447 in relation to the storm. The image was captured about 5 minutes after the airplane entered the storm. The flight path is noted as a yellow line. The small deviation from their course did little to avoid the worst part of the storm.

An animation is also available from the BEA which shows the deviation of other aircraft through that area, including AF459 which was on the same assigned track as AF447, but about 37 minutes in trail.
⁹

The graphic below is an annotated snapshot from the animation. The deviation paths of other aircraft can be seen in blue, purple, and orange. AF447’s track is in yellow.

In the minutes prior to the accident, the crew discussed the appearance of St. Elmo’s fire. St. Elmo’s fire appears as a glowing static discharge, often accompanied by small lightning like discharges on the radome and windscreen. On the A330, St Elmo’s fire is often seen as a hazy glowing ball on the ice probe that protrudes forward between the windshields. The ice probe is where pilots check for airframe ice accumulation because it is very difficult to see any other part of the airframe. A YouTube.com search for “St. Elmo’s fire cockpit A330” will yield numerous examples of this for you to see. In my experience whenever I have seen St. Elmo’s fire, turning on the exterior lights has revealed snow conditions, which can lead to the static charge build-up causing the phenomenon. At 01:37 the captain remarked, “it’s snowing.”

The following illustration, from weathergraphics.com, is the result of analysis of satellite and other data on the storm’s profile and the flight’s progress through it. Light shading is precipitation near the surface, medium shading is cloud material, and dark shading is suspected updraft areas. The green line (not part of the original image) is an approximate vertical path through the storm.

Air France 447 encountered conditions that clogged its heated pitot tubes with frozen material. The question is-“How could heated pitot tubes ice over?”

One early theory, and the subject of a NOVA television production,
The Crash of Flight 447,
was one of supercooled water - water cooled to below its normal freezing temperature, yet remaining as water. When disturbed, the water freezes almost instantly.

Supercooled water is not difficult to produce in your own freezer at home. Take a bottle of purified water and put it in the freezer. Hours later you may find the water still in liquid form, but if you agitate it, the ice crystals will grow and it will freeze solid within seconds. This can happen in the air too. If an airplane encounters supercooled water, significant ice accumulation will rapidly occur.

Many experts discounted the likelihood of supercooled water in this type of storm. Supercooled water in the atmosphere would not only have iced over the probes, but the entire airplane, and that did not happen. We know this because the A330 is equipped with two very sensitive ice detectors, and the flight data recorder revealed that at no time during the flight did they detect any ice accumulation.

One commenter on the Weather Graphics website’s AF447 article provided this interesting observation: “ I'm an aircraft icing specialist and wanted to point out a factor that hasn't been discussed much...high ice crystal concentrations. I've seen flight test data from power rollbacks due to flight in high ice crystal environments … In our case, the crystals collected within heated, aspirated Ram Air Temperature sensors, forming a 0°C slush…”

Seconds before the pitot tubes clogged, ice crystals hitting the exterior or the airplane are heard on the voice recorder. Ice crystals bounce off the exterior of an airplane and cause no visible ice accretion, but they can enter the probe inlets. When highly specific climatic conditions exist in combination with certain combinations of altitude, temperature, and Mach, the concentration of ice crystals entering a probe can exceed its capacity to melt and evacuate the moisture through its drain holes. The result is that the ice crystals form a physical barrier within the probe that disrupts the measurement of total pressure.

The final report states, “As soon as the concentration of ice crystals is lower than the de-icing capacity of the probe, the physical barrier created by the accumulation of crystals disappears and measurement of the total pressure becomes correct again. Experience and follow-up of these phenomena in very severe conditions show that this loss of function is of limited duration, in general around 1 or 2 minutes.
¹⁰
”