At 1:00 a.m. on a cold October morning in 2002, John-Paul Clarke ‘91, SM ‘92, ScD ‘97, stood in a damp field in Floyds Knobs, IN, listening for two United Parcel Service jets to roar into Louisville International Airport. The first came in flying a traditional stepped landing pattern in which it incrementally decreased altitude several times before reaching the runway. From where Clarke stood, about 650 meters below, the noise was loud enough to wake slumbering Floyds Knobs residents. The other plane followed 20 minutes later flying an experimental pattern the aeronautics and astronautics associate professor had designed: it stayed at a higher altitude longer before it descended into the airport, a procedure that pushes the limits of both onboard flight management systems and air traffic controllers. Even without the data being recorded by the sound equipment beside him, Clarke knew that his procedure was significantly quieter. “You could really hear a difference,” he remembers.
Clarke was just one of seven researchers monitoring aircraft noise levels in Floyds Knobs on that night and several others. When all of the data was analyzed, it confirmed what he had suspected: the experimental landing procedure cut noise on the ground between three and six decibels, more than enough to be noticeable. What’s more, planes flying Clarke’s procedure saved fuel. Since those first tests, Clarke’s team-several MIT aero-astro graduate students and about a dozen researchers from Boeing, NASA Ames Research Center, and NASA Langley Research Center-has refined its design so that air traffic controllers can manage multiple planes flying the procedure in moderately heavy traffic. Meanwhile, Clarke is creating a similar procedure for Gatwick Airport in London as part of the Cambridge-MIT Institute’s Silent Aircraft Initiative. If the Federal Aviation Administration and the U.K.’s National Air Traffic Services approve Clarke’s designs for Louisville and Gatwick, they will provide much-needed noise relief for the residents of Floyds Knobs and London, and will blaze the trail for other airports interested in designing similar quiet approaches.
The Quiet Challenge
As a major parcel-sorting hub for UPS, Louisville has a particularly difficult noise problem. In the wee hours of every morning, more than 90 UPS jets land at the airport. Contending with the noise those planes generate has been expensive; since 1991 the airport authority has moved more than 1,600 families at a cost of more than $180 million. Still, the airport can’t afford to move everyone who’s ever been awakened by incoming 767s. In 2000, James DeLong, then the airport’s general manager, read one of Clarke’s papers on the feasibility of quieter landing approaches. “I was a pilot, and as I read through [Clarke’s paper], I understood what he was trying to accomplish,” remembers DeLong, who has since retired. “Not only did it have great potential for noise [reduction], but it had potential to save fuel, it had potential to increase the capacity of airportsso it was a win-win as I understood his concept.” Two years later, he invited Clarke to design and test such a landing procedure that could be used by UPS’s existing fleet.
“I [was] in the airport business over 30 years and concluded that the most serious single problem facing the airport system-in what represents a threat to the ability to accommodate future demand for air travel-is noise,” says DeLong.
The idea behind the new design, called a continuous-descent approach, is deceptively simple. The standard landing procedure is a steplike approach in which planes alternately descend and level off several times over about 50 kilometers before they reach the runway. Not only does this require planes to fly at a lower altitude longer, where more of their noise filters down to communities, but it also requires additional force from the engines to keep the plane level, which creates even more noise. But Clarke’s plan called for the plane to stay at a higher altitude longer, and then descend into the airport without leveling off.
The team had to tackle a number of challenges while developing the procedure. Clarke had to make sure that it would significantly reduce noise on the ground, that it was safe under different weather conditions, and that air traffic controllers could handle multiple incoming planes flying the new pattern. “You’ve got to make sure you keep all the different objectives and constraints in view at the same time,” says Clarke. “You can’t just try to optimize on one subject, because you’ll wind up violating another.”
During the initial design stage, Clarke used a noise simulation program he developed to help him decide which physical flight path would be the quietest. The program allows researchers to plug in the exact course they want an airplane to fly, along with the topographical characteristics and the population density of the area it’s flying over. Then the program calculates the noise the airplane creates on the ground and the number of people affected by it. After Clarke evaluated several possible paths, he and researchers at Boeing Commercial Airplanes used a Boeing simulator that includes all the cockpit hardware and controls of a real 767 to create a detailed step-by-step pilot procedure. These simulator sessions also verified that the plane could handle the procedure under different wind conditions.
No matter how good a procedure looks in a simulator, however, a real-world test is the only way to see if it truly works. For eight nights in late October and early November 2002, Clarke, his MIT colleagues, the Boeing and NASA researchers, two pilots, a meteorologist from UPS, and an air traffic controller from the Louisville control tower monitored nightly flights for noise. Each night they selected two UPS planes: one to fly the traditional landing procedure and the other to fly the continuous-descent approach. Four or five crews headed for Louisville were given briefing packets describing the procedure before they took off. Once the planes were airborne, two crews were notified that they’d been chosen to participate. “We picked the pilots at random and basically told them, hey, you’re going to do this procedure tonight,” says Clarke. This method verified that UPS could implement the procedure without having to give its pilots special training.
Then the Boeing, NASA, and MIT researchers headed out to seven different locations in Floyds Knobs that were under the flight path, bringing with them noise measurement equipment. “We drove into people’s fields, or pulled up off the road, and set up microphones,” says Clarke. Was he nervous that the procedure might not reduce noise as planned? “Nah,” Clarke says with a smile. “I was nervous that the pilots would fly them right.”
Clarke needn’t have worried; after the team analyzed the data from the tests, it found that the continuous-descent approach reduced the noise from between 69 and 70 decibels to between 62 and 63 decibels. That’s a significant amount, considering that a three-decibel reduction in noise is noticeable to the average person, and a 10-decibel reduction is perceived as being half as loud. The researchers also discovered that airplanes flying the procedure saved about 225 kilograms of fuel.
Although the 2002 tests were a success, more work needs to be done to prove that the procedure is practical in moderately heavy air traffic. When several planes land in quick succession, air traffic controllers typically use the speed of the planes when they level off to approximate how far they are from the runway and each other. However, a plane using the continuous-descent approach never flies level, and its speed continually decelerates. If many planes fly this approach, it’s difficult for air traffic controllers to predict how close they will be to each other as they approach the runway. The procedure Clarke designed for the 2002 test attempted to account for this by keeping the planes’ speeds as constant as possible during the initial descent, but it was only tested in light traffic.
This September, however, Clarke will test the procedure in moderately heavy traffic. He is designing an approach that 20 UPS planes will fly as they come into the airport within two minutes of each other. Clarke hopes that this test will prove that air traffic controllers can manage multiple-aircraft landing using the new approach. If all goes well, the team will then apply to the FAA for approval of the Louisville procedures, a process that Clarke estimates could take less than a year.
Meanwhile, Clarke is bringing his expertise to the Cambridge-MIT Institute’s multidisciplinary Silent Aircraft Initiative, which was launched last fall. The three-year initiative is bringing together researchers from the University of Cambridge and MIT, as well as experts from industrial and government partners, to design an airplane so quiet that its noise would be imperceptible outside the airport. Aeronautics and astronautics assistant professor Karen Willcox, SM ‘96, PhD ‘00, is using Clarke’s noise simulator in her evaluations of the team’s experimental designs.
While most of the engineers in the initiative will be working on the new airplane, Clarke will focus on creating a continuous-descent approach for existing commercial aircraft at the Gatwick Airport in London. The team hopes to conduct a flight test in 2005 and have the procedure implemented at Gatwick in 2006.
Clarke’s work on the continuous-descent approach will help the Silent Aircraft Initiative make the case that its work will help the U.K.’s economy-a mission common to all CMI joint research projects (see “From Cambridge to Cambridge,” MIT News, May 2003). Quieter landing procedures could have broad economic benefits for airports, airlines, and communities. For example, airports that have noise curfews could operate more flights, thus cutting down on congestion and delays-problems that ultimately lead to higher airline operating costs and ticket prices. In the long term, quieter landings could help alleviate community concerns over the construction of new runways. A dramatically quieter airplane would significantly increase the number of flights that can fly in and out of airports worldwide, but it will take decades to bring to fruition. “It takes 10 years to develop an airplane, and then another 15 years before significant numbers of them get into the fleet,” says aeronautics and astronautics professor Ian Waitz, who’s leading the economics component of the initiative. “But the operation things that J-P Clarke is involved in, those, a year or two down the road, can change things.”
Indeed, if Clarke’s approach is approved for Louisville and Gatwick, the residents of Floyds Knobs and London can look forward to sleeping a lot more soundly in just a couple years.
Geoffrey Hinton tells us why he’s now scared of the tech he helped build
“I have suddenly switched my views on whether these things are going to be more intelligent than us.”
ChatGPT is going to change education, not destroy it
The narrative around cheating students doesn’t tell the whole story. Meet the teachers who think generative AI could actually make learning better.
Meet the people who use Notion to plan their whole lives
The workplace tool’s appeal extends far beyond organizing work projects. Many users find it’s just as useful for managing their free time.
Learning to code isn’t enough
Historically, learn-to-code efforts have provided opportunities for the few, but new efforts are aiming to be inclusive.
Get the latest updates from
MIT Technology Review
Discover special offers, top stories, upcoming events, and more.