Recently, NTSB Chairman Deborah A. P. Hersman issued a warning that those in the government and aviation industry who are enamored of the planned Next Generation Air Transportation System (NextGen) should get their heads out of the clouds and realize that the people who will have to use the system—i.e., the air traffic controllers—are as important to safety as the automation itself.
Speaking at the Air Traffic Control Association’s Annual Conference in October 2009, Hersman said that when investigators conduct their accident investigations, they often learn that while controllers know how to push buttons for routine tasks, they don’t have a clue as to how things actually work. “Without a clear understanding of what the system is doing, the controllers will no longer be in charge [with NextGen],” asserted Hersman. “They’ll just be along for the ride, and that’s no better in ATC than it is in a cockpit.”
The Air Traffic Control Association (ATCA) was formed in 1956, and is a nonprofit international organization representing the interests of professionals in the ATC industry. ATCA has been working with the FAA and others in the development of NextGen.
Hersman indicated that in the event of a system failure, dependence on automation can lead to unintended consequences, and that backup systems often aren’t as capable as the primary system. This was evident in mid-November 2009, when the FAA experienced a software problem with its computer system’s data router in Salt Lake City. The problem prevented automatic filing of flight plans, forcing controllers to manually manage flight plan data, which resulted in nationwide delays and cancellations affecting thousands of flights.
Hersman expressed concern about NextGen’s advanced automation capabilities to squeeze more air traffic into finite airspace. The NTSB wants assurances that the system “will, in all cases, avoid putting aircraft and passengers at risk by placing aircraft in unsafe proximity to convective weather—even when there appear to be efficiency or capacity benefits in doing so.” She cited situations in which controllers have refused to give pilots permission to deviate around thunderstorms, telling them they’d have to declare an emergency to deviate from a clearance. Hersman was hopeful that improved weather displays would eliminate situations in which controllers believed their information was inaccurate and failed to issue weather warnings to pilots, who then entered convective activity and crashed.
One accident involving some of the factors highlighted by Hersman involved a Mitsubishi MU-2B flown by an airline transport pilot with 30,708 hours. The airplane was destroyed when it impacted terrain near Argyle, Fla. The pilot, who was the only person on board, was killed. The flight was on an IFR flight plan from Tulsa, Okla., to Panama City–Bay County International Airport (PFN) in Panama City, Fla.
The pilot phoned flight service at 7:26 a.m., and got a weather briefing. The briefer noted no adverse weather conditions along the route, stating that the pilot could expect to encounter only some showers between Tulsa and Fort Smith, Ark. The airplane departed Tulsa at 8:53 a.m., and then climbed to FL190. Two hours later, it was handed off to Jacksonville Air Route Traffic Control Center. At 11 a.m., the pilot was instructed to descend to 15,000 feet MSL. Two minutes later, the controller broadcast an alert that a SIGMET had been issued pertaining to thunderstorms in portions of Florida, southwest of the pilot’s route. At 11:03 a.m., the controller cleared the pilot to descend to 11,000 feet. At 11:10:21 a.m., the pilot was instructed to contact Tyndall Approach on 125.2 MHz.
Review of the interphone communications between the Tyndall Radar Approach Control (RAPCON) North Approach radar assistant and a controller at Eglin Air Force Base showed that at 11:10:18 a.m., the Tyndall radar assistant asked Eglin for information on the intensity of the weather on Eglin’s radar. The Eglin controller responded that his display was showing intensities one through six. The Tyndall radar assistant replied, “One through six? Nothing specific? Okay, thanks,” and concluded the call.
The pilot checked in with the Tyndall RAPCON North Approach controller at 11:10:39 a.m., and reported having ATIS information Tango. The Tyndall controller advised that Uniform ATIS was current, and provided the pilot with the updated information: PFN was reporting estimated wind from 250 degrees at five knots, visibility 10 miles and few clouds at 3,000 feet. The pilot was told to expect a visual approach. He then transmitted, “We’re at 11,000, like to get down lower so we can get underneath this stuff.” The controller told the pilot to stand by and expect lower altitude in three miles. About 15 seconds later, the controller cleared the pilot to descend to 6,000 feet.
At 11:12:27 a.m., the pilot was instructed to contact a different Tyndall Approach controller. The new controller cleared the pilot to descend to 3,000 feet at his discretion, and the pilot acknowledged. There was no further contact with the airplane. At 11:15:40 a.m., the Panama controller attempted to advise the pilot that radar contact was lost, but repeated attempts to establish communications and locate the airplane were unsuccessful.
About one mile south of the accident site, a witness reported hearing a “loud bang” and then seeing the airplane in a nose-down spiral as parts separated from it. At the time of the accident, the witness stated that it was raining with lightning and thunder. Local authorities responding to the accident site reported hard rain and that “the thunderstorm materialized very quickly.”
Tyndall RAPCON is a U.S. Air Force approach facility; it lacks displays for controllers that show weather intensity. Instead, when a controller begins a shift, he or she receives initial weather information from the watch supervisor and also from looking at a wall display that loops a weather radar picture from AccuWeather.com. The wall display doesn’t provide close-up coverage of the Tyndall/Panama City area. On the day of the accident, weather at Panama City was VFR, but there were thunderstorms in the vicinity. The first RAPCON controller who handled the flight noted the planned route would take the airplane through an area of weather. After the airplane started descending to 6,000 feet, the controller started a handoff to the Panama sector of the RAPCON. The first controller had no discussions with the Panama controllers about weather information.
When the pilot checked in with the Panama sector, that controller didn’t know if the pilot had been advised of the weather ahead. The airplane was still proceeding direct to PFN, descending to 6,000 feet. The radar controller issued a “pilot’s discretion” descent to 3,000 feet, which the pilot acknowledged. The controller stated that when he took the handoff, the airplane was already on the eastern edge of weather that was about 10 miles deep. The pilot didn’t ask about weather or deviations. The airplane strip used by the controller showed that the pilot had received the ATIS information, but there were no other notations about weather. He wasn’t sure if the airplane actually was in the precipitation area at the time. When he transmitted that radar contact was lost and received no response when asking the pilot for a position report, he initially thought the pilot had gone off frequency on his own.
An NTSB investigation showed that the weather along and near the route of flight included moderate and extreme cells. A controller’s assistant explained that Tyndall controllers communicate with Eglin Air Force Base controllers to obtain weather information because Eglin has better weather displays, but he wasn’t sure what they looked like. Another assistant said he didn’t like NEXRAD displays, which showed only three weather intensities. He was aware that cyan is used to depict the worst weather, but was unable to describe the appearance of the other two weather intensities (royal blue for least intense and checkered cyan for moderate). Investigation showed that the Tallahassee and North Florida NEXRAD weather radars were out of service at the time of the accident. They are among eight radar sites providing controllers with part of their weather coverage in the accident area. Tyndall RAPCON’s manager told investigators that when he learned of the accident, he became concerned about what the controllers might have been seeing or not seeing at the time of the accident. Pictures showing a fast-growing area of weather may not necessarily have been on ATC displays. Operations managers weren’t always told that NEXRAD radars were out of service, but the manager believed that this would be useful information.
The NTSB determined that the probable cause of this accident was the pilot’s inadvertent flight into thunderstorm activity, which resulted in a loss of control, the exceeding of the airplane’s design limits and its subsequent in-flight breakup. A contributing factor was the failure of ATC to use available radar information to warn the pilot he was about to encounter moderate, heavy and extreme precipitation along his route of flight.