Part 1
Safety Alarms Sound On FAA Speed Test
After thousands of flights, agency has no data on safety or benefits of Houston experiment
For nearly four years, the FAA has been quietly conducting a "test" in the skies above one of the busiest airports in the U.S. in which airliners are routinely allowed to exceed the normal 250-kt. speed limit in congested terminal areas.
Houston controllers can legally offer pilots the chance to break the agency's speed limits in a unique, but little-known FAA four-year-old ongoing test.
Known as the Houston Speed Exemption Test, or Houston 250, the experiment is designed to test whether accelerating arrivals and departures at the George Bush Intercontinental Airport (IAH) can cut delays. Air traffic controllers there say the pilots of hundreds of aircraft taking off accept the voluntary clearance every day, and that "99.9%" of them like it.
But an independent panel of advisors to the FAA has been warning the agency for three years that its experiment is fraught with risks--not the least of which might be increased odds of mid-air collisions or substantially greater damage from bird strikes.
The Air Traffic Procedures Advisory Committee, a 23-member group representing military, airline, general aviation and air traffic control organizations, is frustrated at what members believe to be the FAA's glossing over of its past concerns that safety may be taking a back seat to the capacity enhancements. Atpac told the agency in its most recent statement that it is taking "serious exception" to the continuation of the experiment.
So far, the FAA's response to the complaints of its own watchdog group has been that the airline industry has asked the agency to keep the test going--and that safety concerns will be considered in the next phase of the test program.
THE FAA STILL CONSIDERS the program to be in the feasibility stage, which is why officials say they have yet to gather any cost-benefit data that could be used to justify the benefits of the test in light of the real or perceived dangers.
While the agency did recently analyze one of Atpac's primary worries--that higher speeds make it easier for airliners to mistakenly depart the test area and clash with slower-moving traffic--the FAA admits it has not implemented safety recommendations generated by an internal study.
Regardless, FAA officials said the agency remains committed to extending the practice to other airports if a planned risk assessment at Houston proves favorable.
Atpac said more airports would already be using the procedure had the group not protested. The FAA had planned to introduce the procedure in Atlanta, Los Angeles, Dallas/Fort Worth, San Francisco, Denver, San Diego, Memphis and eight other airports, but delayed the plan after Atpac raised objections in April 1998.
Meanwhile, at Intercontinental, the nation's 13th busiest airport, controllers say the test is "used frequently" and is "doing what it was designed to do."
But that's one of Atpac's basic problems with Houston 250--how can a test be doing what it's supposed to if FAA has not defined test parameters, success criteria or a cutoff date?
From all appearances, speeding up arrivals and departures at an airport to make room for more arrivals--and therefore adding capacity to the facility--seems to make a lot of sense.
At least it did in 1995 when RTCA Inc., a federal advisory panel, suggested that the FAA investigate lifting or eliminating the agency's 250-kt. speed restriction within Class B airspace around major airports. As a general rule, the FAA restricts aircraft flying below 10,000 ft. MSL to 250 kt. or less--but an RTCA recommendation is considered technically astute.
Class B zones, formerly called Terminal Control Areas, originated in 1970 as a way to keep airliners and non-airliners from mixing near airports. The structuring was partly spurred by an NTSB investigation into a fatal 1969 mid-air collision between an Allegheny Airlines DC-9 and a single-engine Piper PA-28. The Board's probable cause of the Fairland, Ind., crash listed "deficiencies of the air traffic control system in a terminal area."
Today's Class B airspace, typically shaped like an upsidedown wedding cake, is a tightly controlled zone centered over major U.S. airports. It extends 10,000 ft. vertically and out as far as 30 naut. mi. from the airport, but is individually tailored to each airport. Inside Class B, aircraft are required to have altitude-reporting transponders and be in voice and radar contact with air traffic controllers.
Just past Class B boundaries, however, aircraft are not held to those requirements and radar returns on the planes could be sketchy at best.
The FAA implemented RTCA's speed exemption at the IAH Terminal Radar Approach Control on June 26, 1997, offering pilots the option to increase speed "at their discretion" above the 250-kt. limit.
Less than a month later, Atpac was briefed on the new program--and concerns began to emerge. "The committee expressed many questions regarding the test criteria, the objectives and the conditions governing the test participants," the FAA wrote in the meeting minutes.
At an April 1998 meeting, the level of concern grew when the Mitre Corp. briefed Atpac members on an FAA-funded study of the test. The study looked at air traffic operations at the airport during two month-long windows, one before and one after the test started.
The main effect Mitre noted was that some aircraft were now exiting the side of Class B airspace prematurely at speeds greater than 250 kt., spilling out into the speed-restricted region under 10,000 ft.
It took the FAA three years to analyze the Mitre data to find out why aircraft were having difficulty staying in the test area.
The Flight Procedure Standards Branch in June 2000 issued a report with five recommendations aimed at keeping airliners within the bounds of IAH-controlled airspace, based on tests with a Boeing 727-200 simulator flying Houston-like departures at speeds of 250-310 kt.
Using 14 FAA flight crews and one Continental Airlines crew flying Houston-like departures in the simulator, investigators found that crews exited Class B airspace too fast and too low in 22 out of 55 attempts. Successful crews, according to the study, used "significantly higher power settings, reconfigured their aircraft and accelerated to their climb speed in a more expeditious manner than pilots of the unsuccessful flights."
TO PREVENT THE POP-OUTS, investigators recommended five options, including: Telling pilots "in some appropriate manner" of the need to perform the departure climb in the "most expeditious and efficient" way before clearing them for high-speed climb profiles; expanding Class B airspace with climb corridors and instructing pilots to maintain maximum take-off thrust to 3,000 ft., then to reconfigure for climb with maximum power "appropriate for atmospheric conditions" until 10,000 ft. is reached.
Despite the findings of the FAA's own internal study, regulators say the Houston test has not been modified because it remains in the preliminary, or feasibility stage. Modifications, officials say, will be made only after the test enters phase two when a risk or safety assessment is completed.
Eric Harrell, the FAA's liaison with Atpac, says the agency iscompleting a test plan and identifying a "funding stream" for the safety assessment.
It's a Catch-22 that's frustrating to Atpac as millions of passengers unwittingly take part in a "test" that some say is now considered standard procedure, but has been implemented without the required waivers and safety analyses.
Atpac's primary concern is that see-and-avoid collision tactics might not work when an airliner flying too fast mistakenly enters the realm of local traffic outside of the Class B zone, regardless of whether the pilots see the conflict arising.
But high closure speeds could also be a problem when bird strikes occur. Atpac wants to know whether airliners' windshields and airframes were designed to handle bird or waterfowl strikes at speeds higher than 250 kt.
IN JANUARY 1998, an in-service Delta Air Lines 727 departing IAH in the test struck six snow geese at 7,000 ft., severely damaging the aircraft and engines, according to FAA records.
Though no connection was made between the aircraft's speed and the collision damage, some Atpac members suspected a link. FAA officials say the proper place to study the issue is in the upcoming safety assessment.
Atpac's primary concern continues to be the specter of a collision not with a bird, but with another plane. The committee has asked the FAA to either assure that airliners using the procedure stay within protected airspace or put speed limits back in place; in other words, end the test.
The agency, however, said airlines using the procedure are happy with the results--and that safety issues would be considered in the near future.
After meeting with representatives from airlines, general aviation organizations and pilot unions last October--and getting positive feedback on Houston 250--Harrell said the FAA agreed to keep the test going and to proceed with risk assessment, though no formal announcement was made.
And based on a "favorable result of the risk assessment," he added, "it would be our plan to use this [procedure] at other airports."
Asked whether the airlines attending the meeting presented any quantitative data proving the cost-effectiveness of the added speed, Harrell replied, "No."
A Continental representative said the test is safe and helps the airline "sell a schedule" by reducing delays, but that the airline has no quantitative data proving the link. Continental has a major hub at Intercontinental with 353 daily departures.