Prognosticating the Bird Strike Hazard

I don’t get it; why is predicting the hazard of bird strikes so difficult for the Federal Aviation Administration (FAA)? The U.S. Air Force, which does a lot of low level flying, has provided pilots reports of the hazard for years. The FAA could take the Air Force methodology and graft it onto the civilian system.

Instead, the FAA is engaged in a media blitz, pushing articles and television news reports of various pilot programs to install bird detection radar at airports. Let the record reflect, radar is good at detecting water, and birds can be described as little (and not so little) bags of water wrapped in feathers. Flocks of birds that hazard aviation are going to show up on radar. But pilot programs buy time for an FAA where stately progress is the norm even in the face of an urgent need (witness the FAA’s slow roll on fuel tank inerting to prevent aircraft explosions).

To be sure, airport radars designed specifically to detect birds are a worthy investment, but the Air Force has been using routine air traffic control and weather radars for years to detect the flight paths of flocking birds. This is not a technological challenge that passeth all understanding.

The issue of bird strikes is on everybody’s mind because of at least two crashes caused by birds: 1) the downing earlier this year of US Air flight 1549 into New York’s Hudson River after the airplane flew into a flock of Canada geese and the engines, unable to digest carcasses weighing up to 26 pounds, quit, turning the airplane into a high-tech, fly-by-wire glider, and 2) the 2008 crash of a Cessna charter airplane at Oklahoma City, killing the five persons aboard. (See Aviation Safety Journal, ‘Hearings Reveal Definition of Ditching Needs Upgrading’ and ‘Bird Strike Investigation Reveals Loosy-Goosy Operation’)

In the case of the Hudson River crash, the National Transportation Board (NTSB) presented a very interesting illustration, obviously taken from two separate radars: one showed a plot of the US Airways airplane, flying smack into a bunch of Canada geese, whose path was derived from another ground radar. How do we know this? Because the periodic dots on the display equate to the times at which the rotating radar antennas illuminated the target aircraft and birds.

Radar plot of the US Airways flight path (green dots) intersecting the Canada geese (red dots).

Radar plot of the US Airways flight path (green dots) intersecting the Canada geese (red dots).

A similar plot was shown at the NTSB’s 28 July hearing into the Cessna crash, where a single pelican flying as part of a flock struck the aircraft, and a radar plot was shown of the Cessna’s flight path and that of the pelicans.

The derivation of the Hudson River plot was not discussed, but it was discussed in the case of the Cessna crash. Investigators said the airport radar recorded some 6,000 primary radar returns at the time of the accident (a window of about 10 minutes). Picking out the birds would have been impossible if done manually – which is why airplanes are tracked via their secondary, transponder returns. But the technology could be automated (i.e., that’s what ‘computer programming’ is for). To be sure, birds don’t broadcast their altitude, like an airplane’s transponder, but any primary returns above a certain size, indicating many birds in close formation, could be singled out. Basically, if such pictures can be produced after the fact for accident investigation purposes, it ought to be possible to tweak them into near real-time bird predictions. Then, if the hazard is present, it can be included in weather reports to the pilots and the tower controller. This reporting could be really simple: birds at such-and-such an azimuth, distance and elevation (if known) from the airport. Pilots could delay takeoff or landing based on the reported threats, or controllers could curtail airport operations until the avian threat had passed.

This sort of reporting and action is done all the time for thunderstorms. Birds pose an equal if not greater hazard.

If the FAA wants to deploy new, high-resolution radar for bird tracking, fine, but these radar pilot programs are years from universal deployment at, say, the 50 busiest passenger airports. Many airports feature grassland and/or bodies of water nearby, providing natural habitat that attracts birds. Start somewhere, and if an airplane strikes birds not detected by radar, that provides an opportunity to refine the radars.

It is obvious from the NTSB graphics that the flight paths of airplanes and the tracks of flocking birds reside in separate databases that need to be combined to produce a real net benefit to safety.