Wakeup call on wakes

Air traffic has increased greatly over the last 20 years. The most straightforward ways to deal with this increase are: increase the number of aircraft that can land at our existing airports; build more airports; or build bigger airplanes.

Building more airports is a non-starter. Such construction would cost many billions of dollars (and no one wants an airport next door). Landing more aircraft seems possible, but will definitely require that air traffic controllers reduce the spacing between aircraft.

George C. Greene, a researcher at NASA Langley Research Center, Hampton, VA, says the current state of our understanding of wake turbulence makes spacing reductions difficult. And the situation will become even more complex if aerospace companies produce new, super-heavy aircraft.

Today, many airports are capacity-constrained. So putting the maximum number of people into a given airport may involve a tradeoff between size and wake turbulence. "If you could replace a couple of heavy aircraft with a single one, there could be a tremendous economic incentive to buy it. But if the airplane doesn't buy you more people on the ground per hour, because of wake turbulence, then it doesn't make as much sense."

In flight, aircraft generate a vortex off each wingtip. Each coil-like cylinder of rapidly spiraling air develops when air from the high-pressure region below the wings curls around the tip of the wing into the low-pressure area above the wing. The core of the vortex is at a lower pressure than the air circulating around it.

Vortices trail out behind the aircraft and cause wake turbulence. If another aircraft flies into the vortices, it can experience a roll violent enough to cause an accident.

Goals and problems. Wake turbulence studies for NASA's Terminal Area Productivity Program (TAP), which is now underway at NASA Langley, are part of the joint NASA/FAA Wake Vortex Program. TAP seeks to achieve clear-weather airport capacities in instrument-weather conditions.

Researchers at NASA Langley want to increase airport capacity as much as 15% by finding ways to safely reduce wake-vortex-imposed separation standards. Doing so requires vortex hazard characterization, development of vortex-detection technology, and system concept development.

If successful, their work will result in the demonstration of a concept called Aircraft Vortex Spacing System (AVOSS). Ultimately, AVOSS will feed data into the Terminal Radar Area Control (TRACON) automation system. "What the AVOSS system is about," says NASA researcher Leonard Credeur, "is to try to determine what the wake vortex is doing today, right now, when this next airplane is going to land.

"Controllers now do separation by distance," says Credeur. "Actually, wake vortex is a time-based phenomenon. It takes a certain time for a wake vortex to blow out of an aircraft's approach corridor, descend out of the way, or decay." NASA seeks to define an acceptable level of wake strength that translates to a time separation for one specific type of airplane following another: for example a DC-9 following a Boeing 757.

The NASA researchers point out that pilots must be involved to help define what constitutes a wake hazard. "If you've got a young, bold pilot, he may think a big upset's okay; a more conservative pilot may favor a smaller number," Greene observes. "It's difficult to get pilots to agree, since many haven't experienced a serious vortex encounter."

AVOSS will include several subsystems. One, a weather subsystem, will use several measurement techniques to determine what's happening to the weather now, and what will happen in the next 30 minutes. "AVOSS will probably require wind information from the surface to about 1,500 ft," says NASA Langley Aerospace Technologist David A. Hinton. Algorithms now being written at Langley will convert sensor data into wake-vortex behavior predictions. This predictive subsystem will produce an estimate of the time separation required between aircraft for safe landing.

"The real killer problem is the knowledge base needed to write the predictive subsystem. Once we have that, hopefully the software won't be very intensive," says Hinton.

On the ground, a real-time sensor subsystem will detect, track, and quantify vortices. Data from this sensor subsystem go to the predictive subsystem to refine its output. If measured conditions on the field don't correspond to the predicted conditions, the air traffic control system will go back to conservative default aircraft spacings.

Below about 200 ft, Hinton explains, "there are ground interactions that cause more-rapid vortex decay, and also make predicting transport more difficult." More data on vortex behavior must be collected in the low-altitude region. Getting the atmospheric data won't necessarily involve a technical stretch.

"There are several ways of doing it. Existing sodar systems will ping the atmosphere and give you the winds every 20 meters through 400 or 500 meters above the ground. You can buy radar profilers that do the same thing up to several kilometers above the ground. And there are acoustic systems that measure temperature aloft over the first few hundred meters" says Hinton.

It may prove possible to track wake turbulence with radar. "People have confidence that they'll be able to find the vortices with lidar and get a signature," says Robert T. Neece, senior research engineer at NASA Langley. "Lidar's likely to have trouble in fog or light rain; whether or not radar could provide an all-weather capability remains a question."

Theory predicts that researchers should be able to detect vortices with X-band radar, and achieve higher resolution than is possible at longer wavelengths. Also, weather radars now in use operate at that frequency. "We might be able to incorporate an X-band wake vortex detection function into an existing radar," says Neece.

The present thrust of Neece's research is to demonstrate vortex reflectivity at X-band. "Within six months we're going to have to determine what we're going to do relative to X-band. That may not determine what we're going to do with radar. There's a possibility that we may use radar to complement lidar."

In some weather conditions, radar may provide the best means of detecting vortices. In clear air, Bragg scattering is the source of vortex reflectivity. To get Bragg scattering, eddys with a diameter of about a half-wavelength at the radar frequency must exist in the vortices. But if there are water droplets in the vortices, they can provide reflectivity. Thus, Neece explains, if clear-air vortex-detection capability can't be demonstrated, a radar could still provide that capability in foul weather.

To make a radar a good detector for wake vortex, Neece remarks, requires a large antenna to get the necessary resolution. "You must do some things to improve your signal/noise ratio. And beyond that the main changes will be in how you process doppler data," says Neece. Asked what constitutes the most interesting or challenging part of the radar problem, he responds: "It's trying to discover a vortex signature. How do we identify a vortex within this radar data? Once we crack that nut, we'll have it."

Equipment designed to observe vortices in real time is needed because the ATC system, the pilot community, and the travelling public might find it difficult to trust a purely algorithmic solution to the wake-vortex problem. So AVOSS will combine the predictive subsystem with the sensor subsystem. The NASA researchers point out that an AVOSS might be feasible without a dedicated wake sensor. But the system would require more conservative wake predictions, and the gain in airport capacity it could offer would be less than that achievable with a system that includes a sensor.

Real-world information. Data on wakes are now being gathered for NASA at a facility built by MIT's Lincoln Laboratories at the Memphis International Airport. It employs a 10.6-micron continuous-wave laser and a signal processor to track vortices in real time, and watch them decay. The Memphis facility also includes a radar profiler, a sodar, and a 150-ft tower instrumented with wind vanes, temperature and humidity sensors, barometers, and other devices.

The two primary carriers at Memphis, FEDEX and Northwest Airlines, provide aircraft data for every arrival. So researchers can track each airplane that lands, and each vortex, and correlate it with ATC data. The flight's operator then tells researchers what the airplane weighed--an important parameter in estimating initial vortex strength. This approach generates a data set that describes the aircraft that made the wake, the weather, and the wake's behavior.

"Roughly 100 wakes have been tracked. We're now starting to process them," says Hinton. "We're looking at a demonstration of the predictive algorithms in 1999. We plan to go to the field this year with a Version One predictor. We will run that predictor on roughly a daily basis here at Langley, using data transmitted back from Memphis. After we get the laser tracking data, probably a month or two later, we'll do a comparison to see how well we track."

As the modelling proceeds, Hinton and his colleagues expect to see field data and the predictive algorithms converge. "I believe we'll be converging by 1999. I have a lot of confidence we'll be able to converge on a product that will be useful to the AVOSS system," Hinton asserts.

Looking at weather represents a new engineering approach to wake-turbulence studies. "On the operational side, in the past," says Greene, "it hasn't been recognized how important weather effects are on vortices. On the science side it was always recognized as important, but it has taken a long time to get these ideas merged. You can predict the initial strength of a vortex. The hard thing is to predict its decay."

Improving our understanding of wake turbulence can enable existing capacity-limited airports to safely handle more traffic at a reasonable cost. By the end of the century, if NASA's researchers are successful, aerospace engineers and air traffic controllers will have a new battery of sensors and data with which to battle this aviation hazard.