Technology unties the air traffic knot

Dallas-Ft. Worth airport was the scene of one of the opening battles in a war the Federal Aviation Administration (FAA) is waging against air traffic delays.

By using lasers to validate predictions of aircraft wake turbulence, NASA and Department of Transportation (DOT) engineers advanced the technology that would allow two streams of aircraft to use parallel runways as close as 750 ft, as compared to the current 4,300-ft unrestricted spacing limit. NASA Administrator Dan Goldin told Design News in January (see DN 2/5/2001, p. 110) that the program could also change aircraft spacing in the landing queue, with the potential to increase arrival capacity on average 6% (6 aircraft per hour at DFW).

Nature of the beast. A visit to DOT's Volpe National Transportation Systems Center (Cambridge, MA) provided insight into such technologies' potential. Volpe is a research facility dealing with all aspects of the DOT, from air traffic to the Coast Guard and highways.

Ed Spitzer, acting Volpe director of traffic and operations management, highlights the fact that, in general, the air traffic system flows fairly freely under normal conditions-but it is local delays (chiefly in arrivals) in the airport areas that usually cause the bulk of national system back ups. The Chief of the Airport Surface Div., John LoBue, likened this to a hose that is free flowing, but has its flow restricted when the ending nozzles are tightened, a literal bottleneck situation.

Currently, aircraft are spaced 3-6 nautical miles apart, depending on aircraft type, when approaching an airport because of wake turbulence. (Aero engineers work in the classical naval units of nautical miles (1.1 statute miles) and knots (nautical miles per hour).) This turbulence comes about from the physics of a wing's lift-the high-pressure air under a wing curls up and around the wing tip into the low-pressure region on the upper surface, generating a horizontal vortex trailing from each wing tip (see figure, p. 66). Because the wing pushes down on the air it passes through to generate lift (action equals reaction), the flow field behind the airplane moves lower relative to the airplane and carries the tip vortices with it. These in turn, mutually interact to add to the descending effect. Since lift is equal to aircraft weight in steady cruise flight, the heavier the aircraft the greater the strength of the trailing vortices.

It's less of a problem at high speed at altitude, because vortex energy is spread along the flight path and aircraft are spaced farther apart. But at lower speeds it is more concentrated, and in the airport area aircraft are closer to each other so the chance of a wake turbulence encounter increases. Such turbulence can cause a choppy flight and loss of control of small aircraft-some business jets have even crashed due to it. Encountering wake turbulence when landing is particularly perilous, because a pilot has little or no recovery margin.

Parallel universe. Many airports have parallel runways along the direction of the generally prevailing wind to boost landing and departure traffic. But a large number of these runway pairs cannot be used to host independent operations (where pilots and controllers do not have to consider what is happening on the other runway) that potentially maximize free traffic flow. The reason is that the runways are too close, increasing the chance of a midair collision or a wake turbulence encounter due to cross winds blowing the vortices onto the adjacent flight path (see chart).

Up until now, a lateral runway separation of 4,300 ft was considered the minimum spacing limit to prevent a "blunder" maneuver by an aircraft approaching one runway from causing a midair collision with one using the adjacent strip. Help is already at hand here, however. That limit was set because airport control radars are only updated (scanned) once every five seconds. The advent of the Precision Runway Monitor (PRM), a high-update radar, scanning once a second, allows this limit to drop to 3,400 ft for straight-in approaches. This value can be reduced to 3,000 ft if an angled (offset) approach path to one runway and an Improved Monitor Controller display are used, because there's little time that flights can be running close-by and parallel just before touchdown.

If simultaneous dependent approaches, coordinated between runways, are used, the separation limit drops to 2,500 ft. But these approaches must stagger the aircraft streams to alternate touchdowns on each runway so that any blunder will not cause a midair collision with an aircraft approaching the adjacent runway. Less than 2,500 ft runway separation puts the runways into the "close-spaced" category-they are treated as a single runway for instrument (lowered visibility) operations to avoid wake turbulence encounters-whose effect decays to insignificance after it has traveled 2,500 ft laterally close to the ground. Significantly, five of the top eight airports for delays fall into this class.

But the Volpe folks are enthusiastic about an FAA lead program late this year in San Francisco, with 750-ft runway separation. Part of it aims at allowing paired approaches used under visual flight conditions in good weather to be possible under less than ideal weather, down to a 1,600 ft ceiling (cloud bottom height) from the current instrument limit of 3,500 ft. With a paired approach, controllers and pilots coordinate to bring two aircraft in on adjacent runways so that the longitudinal difference in time along the flightpath is shorter than the potential wake travel time to the adjacent runway. Thus by the time any wake could be transported by cross winds to the other flightpath, the aircraft has passed. To do this safely, the two aircraft must "visually merge" at an altitude below the ceiling altitude-the lower this altitude, the greater the pilot and controller workload, however. Precision approach radar and displays aid in achieving this, but complexity arises from the fact that not just one aircraft pair is involved, but rather a string of planes on both runways needs choreographing.

One technology to aid this management is Automatic Dependent Surveillance-Broadcast (ADS-B), planned to be implemented nationwide starting in 2003. Aircraft equipped with ADS-B and special situational awareness displays will automatically broadcast parameters such as precise position and identification via a data link without any ground radar cueing. The in-formation transmitted will improve situational awareness in the cockpit or for ground users similarly equipped. Potential applications are as disparate as non-radar covered areas-including much of Alaska where it is being used already to aid mountain-flying bush pilots to match their GPS positions to hazardous terrain-map data bases-as well as airport ground vehicles o avoid colliding with aircraft.

But for such direct improvement of parallel runway use, Volpe's Spitzer highlights development of vortex behavior prediction models, based on the Dallas work which ended last year. There NASA Langley Research Center's (Hampton, VA) laser-based Aircraft Vortex Spacing System (AVOSS) determined how wind and other atmospheric conditions affect the vortex behavior of different aircraft. Such data will go into vortex prediction models validated with wind and laser sensors to detect and track vortices. Combined with short-term runway weather predictions, this information will aid controllers in spacing aircraft on parallel runway approaches.

If the separation between aircraft in each runway string could be tightened up, even more aircraft could be landed in a given time, or released for take-off. However, Spitzer says, given the state of the prediction methods, "Reduced longitudinal separation to a single runway is not around the corner. We need to better understand vortex encounter behavior. Vortex behavior depends on weather and we have to be able to predict at least a half hour in advance. You can't line up aircraft and then have to change. AVOSS gave us a behavior model, but the operational details still have to be worked out."

On the ground. Getting aircraft on the ground more rapidly is fine, provided they don't then get bottled up on the taxiways waiting for crossing traffic or gate assignments. John LoBue notes that a Surface Management System (SMS) developed by NASA Ames (Mountain View, CA) looks to "maximize utilization of concrete." SMS is perfecting a set of automation tools to minimize delays and maximize flow. "Based on a runway configuration and where a flight is going, or coming in from, aircraft movements on the field could be optimally sequenced for best routing and minimal time," he says. "It's a multivariable problem and involves seeing what controllers do now to develop models to mimic and improve results. The issue is developing confidence in the automation tool so that the controller would let the computer do the work with the controller making the final judgement."