NEWSCLIP-2002-08-26: Overrun Highlights Hazard of Unstabilized Approaches

…an article from
Air Safety Week

One can imagine the surprised looks on those Southwest Airlines [LUV] passengers waiting to board their airplanes at California’s Burbank-Glendale- Pasadena Airport as they saw Southwest Flight 1455 obviously moving too fast to stop before running out of concrete. Their interest probably turned to alarm as they saw the B737-300 blow through a blast barrier in a shower of metal parts and finally stopping just short of a gas station immediately outside the airport perimeter.

The runway overrun that occurred in the early evening hours of March 5, 2000, illustrates a number of timeless verities about airline operations: captains should not allow themselves to get committed to fast, steep, unstabilized approaches; air traffic controllers can be more helpful by positioning aircraft so that flightcrews always have the option of a safe go- around; and airports with inadequate free space around their runways are accident factories.

The National Transportation Safety Board (NTSB) just completed its investigation of the accident. Bob Benzon, NTSB investigator in charge, said at a recent hearing on the case, “If the runway had been dry, the airplane could have been stopped.” But it wasn’t, and since the airplane touched down farther along the runway than would normally be the case, the captain, who was the pilot flying, did not have enough remaining runway to bring the airplane to a safe stop. He needed 800 more feet. He would have had 1,000 feet if the recommended safety area for just such contingencies had surrounded Burbank’s Runway 08. Because of the close proximity of a highway just outside the airport, there was only about 32 feet of space between the end of the runway and a metal wall designed to deflect jet blast. As Benzon said, “This airport has grown too big for its footprint.”

Airport authorities were acutely aware of the cramped surroundings. The passenger terminal housing Southwest’s gates was placed close to the runway, eating into the desired runway safety area (and close enough to give waiting passengers a ringside view of the accident). The airport’s executive director told the NTSB that “the greatest safety concern and risk to life and property” was the closeness of the terminal to the runway, and the closeness of parked and taxiing aircraft to both of the airport’s runways. However, he lamented, a plan to demolish the terminal and build a new one with more setback from taxiway and runway operations had been hamstrung by Burbank voters. They had stripped the city council of authority to approve any new airport construction projects unless a lengthy list of conditions was met. The inability to satisfy those conditions effectively halted all development activity at the airport.

As a side note, the case also illustrates the old saying about closing the barn door after the horses have bolted out. Some 18 months after the accident, the airport received $1.9 million from the Federal Aviation Administration (FAA) to install what is known as an engineered materials arresting system (EMAS) at the departure end of runway 08. This material, usually a special mix of concrete, will help bring an overrunning aircraft to a safe stop. “It will stop airplanes going 50 knots,” Benzon said, pointing out that “the accident airplane departed
the runway at 32 knots.”

The seeds of this accident were sown when the accident airplane departed Las Vegas two hours late because of rain and gusting winds in the Los Angeles basin. The late departure may have contributed to a subtle time pressure on the crew. About 19 miles from the airport, the crew was advised to prepare for landing on Runway 08, a change from the original plan to land on Runway 33. The controller advised the crew to maintain speed of 230 knots or greater. From that point forward, the airplane would be coming in too fast.

Other factors combined to place the pilots in a bad position for landing. Some of them were of the pilots’ own making:

* The airplane was too fast all the way to touchdown.

* The approach was considerably steeper than the desired three-degree descent angle.

* The crew had not keyed up the landing module of their computer, which would have given them timely information about landing distance performance in conditions involving a tailwind and high gross weight, both of which applied in this case. The landing module also would have
given them the airplane’s go- around performance.

* Required altitude callouts had not been made during the approach to landing.

* “Sink rate” warnings from the airplane’s ground proximity warning system (GPWS) were not acknowledged.

* The captain called for 40 flaps when the aircraft was between 20-30 knots faster than the limit speed of 158 knots for 40 flaps. When the first officer tried to point this out, the captain replied, “Put it to [flaps] 40. [I]t won’t go, I know that. [I]t’s all right.”

* The crew did not execute a go-around, as was required by Southwest policy if they were not “in the slot.” By this term, a missed approach must be made if the airplane is not at the target airspeed, the airplane is on glideslope with a rate of descent less than 1,000 feet per minute.
None of these conditions were met. As an example, although the target airspeed was 138 knots, the airplane touched the tarmac at 182 knots – 44 knots above the reference airspeed.

* Although the Southwest Flight Operations Manual (FOM) indicated that touchdown should occur between 1,000 and 1,500 feet from the landing threshold, the accident airplane ate up more than 2,000 feet of the runway, a third of the available length, before the wheels contacted

* In a post-accident interview, the captain acknowledged that the approach was not stabilized, saying he became “fixated on the runway.” Indeed, the GPWS warnings – “sink rate” and “Whoop, whoop, pull up” – gave the flight crew obvious indications of an unstabilized approach.

The controller’s actions were not helpful:

* He told the crew of Flight 1455 to keep the speed up to 230 knots to place them in between Southwest Flight 1713 (behind them) and Executive Jet Flight 278 (ahead of them).

* He did not cancel the 230-knot speed mandate when the approach clearance was issued.

* He directed the pilots closer in and at a more acute angle than other aircraft. Indeed, this direction was unnecessary, NTSB investigators concluded. The controller could have sequenced Flight 1455 to follow Flight 1713, thereby providing more room to slow down and

* He told the crew to “cross Van Nuys at or above three thousand” feet, although on the 160 heading he directed, the airplane would pass abeam of Van Nuys, not over it.

As the NTSB report concluded, the controller “positioned the aircraft too fast, too high, and too close to the runway threshold to leave any safe options other than a go-around maneuver.”

Throw in a relatively short wet runway, and the number of latent hazards is a veritable minefield waiting to catch the unwary or the harried. Such was the case in this accident.

We asked John Sampson, a transport-category pilot who prepares the Accident & Incident table for this publication, to comment on the case. From his flying experience, Sampson confesses to “scary visions of that runway end looming up out of the mist at a great rate of knots.” Consider the pressures on the pilot, he suggests. Any go-around due to mismanagement/disorganization (and not weather) is likely to gain an interview with the chief pilot, and “no self- respecting ‘stick’ really wants to eat that sort of crow.” Consider also that a go-around may affect one’s gate allocation and many passengers’ onwards connections.

Besides, Sampson added, the stopping power afforded by modern braking systems is often sufficient to salvage landing long. “Eventually, the day comes when you’re so hot [fast] and high, thanks to a complacent and complaisant ATC operator, that you cannot even get your flap down. I’ll bet the captain never realized that he would – or even could – go off the end of the runway until about 10 seconds before he hit the blast fence,” Sampson surmised.

“So what’s the answer?” he asked, going on to answer his own question:

“An air traffic controller should have the real authority to use his judgment and gauge when someone is just trying too hard – and send him around. Time and again they will have seen it in its early happenstance and done nothing about it. Sometimes the outcome is just cooked brakes and a lot of scared passengers, but oftentimes it’s more than that.

“I’m all for ATCOs [air traffic controller officers] having the authority to reject that which they ‘just don’t like the look of.’ Most of them have the clear ability to make that judgment. Very rarely nowadays can you find one that will. But it solves a whole raft of problems because the onus is then off the pilot and all his temptations to salvage a bad situation. He will have no questions to answer from the chief pilot because ‘ATC sent him around.’

“If the ATCO’s prerogative judgement is enshrined in regulations, then he cannot be questioned. I would like to see it done that way, as part of the ATC credo. Pilots have proven that due to fatigue, bravado or consequences that they are capable of occasionally going that ‘bridge too far.’ ”

(ASW note: only highlights of the case have been presented here. The full NTSB report, published Aug. 15, may be viewed at

copied 7/31/13 from: