Wednesday, January 23, 2008
Air Traffic Safety vs. Capacity
Air Traffic Controller -- "Avianca 052 heavy roger what is your alternate ?"
Avianca 052 -- “ It is Boston but we can’t do it now we, we will run out of fuel now. “
That was a radio exchange between Avianca Airlines flight 052 and an air traffic controller, 47 minutes before the aircraft ran out of fuel and crashed. The flight was trying to land at New York’s John F. Kennedy (JFK) airport. Avianca 052 had been placed into different holding patterns on three separate occasions during its flight. It was repeatedly placed in holding for one simple reason -- the number of airplanes scheduled to land at JFK exceeded the capacity of the airport. That was a fact in 1990 when this accident occurred and it is still a fact today.
The press runs a story about air traffic control (ATC) virtually every day now. The stories cover a variety of issues. The Bush Administration wants funding for a new ATC system called “NextGen”, various groups are suing the FAA for redesigning the departure and arrival routes into New York (thereby increasing the noise in their neighborhoods) and it seems as if everyone is upset by the FAA’s proposal to impose limitations on the number of flights into JFK.
It appears as if all have forgotten the lessons of Avianca 052.
Various entities in aviation are using the complexities found in air traffic control to obscure the basic facts in order to further their agendas. The basic fact I would like to clarify is the finite capacity of a runway. I too have an agenda. My agenda is safety.
Any runway has a finite capacity. The key to understanding this is in understanding time. Only one aircraft is allowed to use the runway at any time. It takes a certain amount of time for a departing aircraft to taxi onto a runway, accelerate to flying speed and lift off. Likewise, it takes a certain amount of time for an arriving aircraft to touch down, slow down and taxi off the runway. The time it takes the typical airliner to do either one -- land or takeoff -- is roughly one minute.
The math is as simple as it is inescapable. Roughly 60 airliners can use a runway in one hour if conditions are absolutely perfect. It is physically impossible to improve that number. However, it can get a lot worse.
It helps in understanding all this if you’ll focus on the simple, physical acts required in aviation instead of dwelling on the confusing complexities. Imagine it is foggy at the airport. An aircraft landing on the runway has to find a taxiway to get off the runway just as you must find the entrance to the parking lot when you’re driving down a foggy street. In good visibility, you can see the entrance from a distance and slow down at the last second. In poor visibility (rain, fog, snow) you must slow down to ensure you have enough time to spot the entrance and then make the turn. Every extra second an aircraft spends on the runway -- searching for the taxiway -- decreases the capacity of the runway. There is another aircraft flying towards the same runway and air traffic controllers cannot allow it to land until the first aircraft is clear of the runway.
For air traffic controllers, time equates to distance. If an aircraft is over the threshold of the runway and about to touch down, we know that it will take approximately one minute for that aircraft to touch down, slow down and taxi off the runway. That means that the next aircraft to land must be at least one minute behind the one about to land. To keep matters simple, I’ll use an average aircraft approach speed of 180 mph -- or 3 miles per minute. If you could visualize an air traffic controllers’s radar scope in these optimal conditions you would see a string of aircraft aligned with the runway and spaced out at 3 mile intervals. Not so coincidentally, the legal separation controllers must use in this situation -- the distance required between each aircraft -- is 3 miles.
Let’s take this optimal situation one step further. If you could line up an hour’s worth of arrivals, one minute apart -- 3 miles -- then you would have 60 aircraft in a straight line. That line would stretch out 180 miles from the airport. Furthermore, the last aircraft in that line would be limited to flying no faster than 180 mph lest it catch the aircraft in front of it. Again, the important concept to grasp is that the distance and the speed puts the airplane one hour away from the airport. For aircraft landing at JFK, the line would stretch past Philadelphia, almost to Baltimore. Aircraft flying from either city to JFK would be waiting in line before they even left the ground.
Pilot --“...you happy with that distance?”
Co-pilot -- ”aah, he's.... we'll be all right once we get rollin'. he's supposed to be five miles by the time we're airborne, that's the idea.”
That was the conversation between the pilot and co-pilot of an airliner about to depart JFK airport. They’re discussing the distance between their aircraft and the Boeing 747 that had just departed in front of them. The pilots are concerned about wake turbulence -- the violently disturbed air found behind large aircraft. Just two minutes later, their aircraft did indeed encounter wake turbulence. When the pilot tried to compensate for the turbulence, the tail of the airplane snapped off. All 260 people aboard American Airlines flight 587 -- and 5 people on the ground in Belle Harbor, NY -- perished that day. It all started with wake turbulence.
What the accident investigations, the cockpit voice recorders and the air traffic control tapes don’t reveal to you is the enormous pressure the people in aviation work under. Pilots, air traffic controllers and even airline CEOs are under constant pressure to make the airplanes fly and to make sure they fly on time. The pressure to fly in poor weather, to tighten up the spacing between aircraft and to wring every last drop of efficiency out of the system is incredible.
The aviation industry can’t afford to let these economic pressures erase the memory of the expensive safety lessons we’ve learned in the past. The cost in lives and money is simply too high.
The media have devoted many stories to the FAA’s future air traffic control system called NextGen -- Next Generation Air Transportation System. It is the “satellite-based” air traffic control system that is touted as the cure-all for the current airline delays. One of the basic premises of the system is that -- because of its greater accuracy -- it will allow airplanes to fly closer to each other. Admittedly, there are some (very limited) instances in air traffic control where this would be desirable. But it won’t solve the core problem -- runway capacity. Remembering the lessons above, you still need a minimum of three miles (or one minute) between landing airliners. Controllers are capable of running airplanes closer together now -- with the current radar-based system -- but safety won’t allow them to do it. Until the safety-mandated rule that only allows one airplane on the runway at a time changes, a system that allows controllers to run aircraft closer together won’t increase the runway’s capacity.
While we are on this subject, I need to call your attention to another point about wake turbulence. Wake turbulence exists behind departing and landing airliners. Behind the largest aircraft -- classified as “heavy” aircraft -- the spacing requirements increase to 5 miles. Remember that theoretical line of aircraft stretching out 180 miles from JFK airport ? Throw in a couple of “heavy” airliners and the line will stretch past Washington, D.C. In that these “heavy” type aircraft are typically the ones used on overseas routes, a large number of them use JFK. All three aircraft involved in the accident scenarios noted above were indeed “heavies.”
“Air Traffic Controller -- Avianca 052 heavy roger what is your alternate ?”
These basic facts should give you some insight as to the complexities involved in air traffic control and airport capacity. If you see a comparison in the press between traffic at Kennedy (JFK) and LaGuardia (LGA), for instance, armed with the knowledge that JFK has many, many more “heavy” aircraft than LGA, you will understand that LGA can handle more airplanes per runway than JFK can. Most of LGA’s arrivals only need three miles of spacing while many of JFK’s will have to be spaced five miles apart.
The most complex issue in determining runway capacity is weather. Under the best weather conditions, JFK airport can handle 68 arrivals per hour -- using 3 of its 4 runways simultaneously. If low clouds or visibility -- or even unfavorable winds -- force the controllers to use different runways, the arrival rate drops dramatically. Even in relatively good weather the rate can drop to 51 arrivals per hour. During typical bad weather the rate can drop to 38. And as we all know, during a thunderstorm or an ice storm, the rate can drop to virtually zero.
This brings us to the heart of the problem. How do we determine a realistic capacity number for an airport when something as unpredictable as weather is the major variable ? I propose the we use safety as the determining criteria.
First and foremost, as I hope I have shown you, there is an absolute limit to the number of airplanes any runway can handle, per hour, even in perfect weather. At an absolute minimum that limit should be enforced -- by rule and regulation -- for every commercial airport in the country. Currently it is not and -- unbelievably -- airlines are allowed to schedule more flights than the runways can handle in even perfect weather. It is madness.
Let me use our theoretical runway capacity of 60 planes per hour once again. Imagine if we could schedule it perfectly. Imagine if we had 60 airplanes per hour scheduled for every hour of the day. Then imagine if there was just one tiny glitch -- weather, a flat tire or an airport vehicle mistakenly wanders onto the runway. An arrival has to go around and make another attempt to land. Where do you find the extra slot for that aircraft ? Every slot is taken for every minute of the day. But the airplane must land or it will run out of fuel just like Avianca 052. That makes the decision easy for air traffic controllers. An arrival takes priority over a departure. They will use a departure slot for the arrival. Now the departures will be delayed for the rest of the day -- or in this theoretical case -- for eternity.
The runway -- the system -- must have some “give” to it. It must have some unused capacity in order to ensure safety. In reality it does. That unused capacity is after midnight. That is the reason you see flights arriving at 1 or 2 AM after a day of bad weather. There is no “give” -- during normal hours -- in the system when there is adverse weather. This leads to the massive flight delays Americans suffered through this summer.
The reason is as old as it is simple -- greed. Airlines can make more money selling 70 airplanes worth of tickets per hour than they could if they limited themselves to the 60 airplanes per hour that the runway can handle. In fairness to the airlines, it’s not in their interest to limit themselves. It is easier to sell the tickets and blame the delays on the weather or the “antiquated” air traffic control system. Especially if the flying public doesn’t understand runway capacity limits and therefore fails to notice that the “antiquated” air traffic control system is delivering more airplanes to the runways than the runways can handle.
The government has abdicated its responsibility in this area. The economic portion of the aviation industry was deregulated in 1978. The safety portion -- supposedly -- was not. The Federal Aviation Administration has the legal authority to limit the number of flights into Kennedy, Newark, LaGuardia, Chicago O’Hare and Washington (D.C.) National airports. The FAA should be given the same authority for all commercial airports. And Congress should compel them to use that authority. Currently they do not. The FAA was forced into lifting the slots restrictions at JFK and the result was predictable -- massive delays. The FAA reimposed slot restrictions at Chicago O’Hare (after the last public outcry about delays) and delays went down. They are currently being pressured to relax or lift those restrictions at O’Hare. If they do, you can be assured that increased delays at O’Hare will return.
Congress should pass legislation mandating that each commercial airport’s maximum hourly capacity be established and published. Furthermore, the FAA should impose limitations on the number of flights that can be scheduled at each commercial airport. That number should be less than the maximum capacity, taking into account such factors as typical weather patterns for the airport, routine maintenance and any other factor that typically limits capacity. It is time to recognize the inherent limits imposed upon the National Airspace System by runway capacity. It is time to put safety ahead of economic considerations before we relearn the lessons of Avianca 052. It is time to put safety back in it’s rightful place -- First.
January 23, 2008