Aspen crash prompts approach controversy

 - April 25, 2008, 5:32 AM

On March 29, 2001 a series of operational and instrument approach procedural errors led to the crash of N303GA, a Gulfstream III, just 2,400 ft short of the approach end of Aspen-Pitkin County Airport (ASE)’s Runway 15 while attempting to complete the VOR/DME C circling approach. Eighteen people, including three crewmembers lost their lives in the accident. The NTSB report of the crash (see 'Pressure on pilots a key factor in Aspen GIII crash') was riddled with lessons about what pilots flying under instrument flight rules should not do, ranging from examples of poor crew coordination, to serious misunderstandings of basic instrument procedures, such as the fact that the GIII– certified as a Category D aircraft– was not even authorized to shoot the approach to Aspen under any conditions. Descending below the MDA without adequate reference to the airport environment was also noted.

This was hardly the kind of performance expected of professional pilots operating a sophisticated jet  under Part 135 into a mountainous area at night. Adding to the irony of the accident was the pressure put on the flight crew by the customer to get the passengers to Aspen at all costs. In addition to listing a number of errors made by the crew of N303GA, the NTSB strongly recommended the FAA revise some ambiguous procedures that even the board found confusing and potentially unsafe.

But the confusion about issues related to the Aspen accident and the accompanying IFR procedures is not limited to only the pilots of the Gulfstream. What became abundantly clear on the NBAA’s aviation manager Internet discussion board in the weeks following release of the accident report and in response to a survey question posed by AIN more recently (see page 108) is just how many pilots joining the discussion might have also suffered a similar fate. Not because they were bad pilots, but simply because of the high bewilderment factor generated by some instrument approach procedures today. Especially confusing are those involving non-precision approaches, such as the one at Aspen. Some pointed an accusatory finger at the FAA’s communication methods.

Not surprisingly, few of the pilots and experts AIN interviewed wished to have their names attached to opinions about the agency that regulates their livelihoods. One pilot said bluntly, “The FAA is often a bunch of specialists who don’t proofread the procedures they write very well. Few can understand what they have written and often do not understand how people will interpret this material. Realism often doesn’t play a part in much of what the FAA does.”

What’s So Confusing?
Some pilots AIN interviewed since the Aspen accident expressed confusion about approach issues, such as category selection and minimums, as well as the nighttime restrictions at this mountain airport. The crew of N303GA departed late from LAX for the trip to Aspen with a minimal fuel load, which added pressure to the flight because the crew was aware the circling approach was not authorized at night. There were also snow showers in the Aspen area at the time of the accident, leading to numerous missed approaches by other aircraft prior to the crash.

The Aspen approach plate simply noted that circling was not authorized at night. Most private pilots can recite the regulation that says the beginning of official nighttime begins half an hour after sunset, but at Aspen there are practical circumstances that made this approach unusable long before sunset, much less the official beginning of night. Flying into mountainous terrain as the sun sets or rises, can often mask the black hole a crew would be descending into if the airport was also obscured by weather. Because of the low light levels at the time of the accident, the pilots of the GIII most likely would have been unable to see nearby unlighted terrain, even if they could have identified the runway early in the approach.

So why would an experienced crew not be aware of the confines of the approach as darkness fell? Why would they descend below the MDA for the approach more than once during the letdown? Why would they continue down past the missed approach point on what is already a demanding approach–both for them and the aircraft–in a mountainous area without adequate visual reference? And why would the tower even clear the aircraft for an approach nearly half an hour after official sunset, considering that the GIII is not authorized to make the Aspen approach under any conditions? Was the company dispatcher confused, or simply uninformed when he sent the aircraft to Aspen? The NTSB never mentioned the category prohibition issue in its report. Was that because the board, too, believed the aircraft was operating under Category C?

How large a part will air traffic controller training assume in the responsibility for this accident? A Kansas City aviation attorney who did not want his named used said, “Controllers should also not be issuing clearances for unauthorized procedures. The FAA is way behind the power curve on this.” Pilot Robert Tod quickly recalled similarly perplexing issues during a recent approach to Santa Barbara where, “The ATIS called for the ILS 7 Approach, but the glideslope was NOTAMed out of service with no further explanation.”

Is the Aspen approach unsafe at night? Perhaps. The charter company in question–Airbourne Charter–has operations specifications that prohibit landing at airports such as Aspen if the flight cannot be on the ground before sunset. An FAA flight check crew deemed the approach unusable at night, a week before the accident, which generated a NOTAM.

But the NTSB cited the FAA for sending an ambiguous NOTAM stating that circling on the Aspen VOR DME-C approach was not authorized at night. It did not state that the approach was not authorized at night, but rather that circling was not authorized after sunset. Since the final approach is actually lined up fairly close to a straight-in with Runway 15 at Aspen, the NTSB believes the Gulfstream crew may have interpreted the NOTAM to mean they could still shoot the approach if they did not need to circle. The FAA, however, intended that no one should shoot the approach after dark. The NOTAM was updated the day after the crash with the less confusing, “Procedure not authorized at night.”

Tod said, “Just saying the Aspen circling approach was not authorized gives the impression there may have been another option. But at Aspen, a pilot must be smart enough to recognize more than just what the book says about defining night. Although it may not be official nighttime, it might be just as dark. Safety begins with being able to correctly interpret known procedures, such as at Aspen, where the terrain affects maneuvering ability during approach and departure.” The NTSB put it this way: “Nighttime restrictions [at Aspen] may not sufficiently mitigate potential hazards associated with flight operations into ASE and other airports with mountainous terrain during periods of darkness.”

But one NBAA pilot wondered, “What exactly does it mean when a procedure is N/A at night? Is the procedure available as long as ATC clears you for the approach before official nighttime? Does it mean that the procedure must begin–cross the IAF–before then? Or does it mean that all elements of the approach must be completed before official night?” Another said, “There is plenty to be learned from this discussion, as well as a chance to reflect on our own abilities, procedures and decision making.” The cockpit voice recorder of N303GA recorded no conversation from the crew on the specifics of the Aspen approach.

As in most accidents, a combination of factors is responsible. But no one can ignore the fact that so many pilots seem to be unclear about circling approaches and the Aspen approach specifically. But how many other pilots might have suffered the same fate as the Gulfstream crew, not because they were distracted, but simply because of a lack of knowledge?

Under the best of conditions, the Aspen approach is also highly unstabilized, something that is counter to every aspect of flight training today. Measure the distance from the missed approach point to the threshold at Aspen, some 1.4 nm in which an aircraft must lose approximately 2,400 ft. At 123 kt, this equates to a descent of more than a 3,000 fpm, hardly stabilized in anyone’s book.

A number of pilots said that in recurrent sessions, the major training organizations focus on aircraft systems and flying the approaches, with only a small amount of time spent on aircraft performance and no time spent exploring TERPS, the U.S. standard for Terminal Instrument Procedures. Pilots reaching this level of their professional lives are expected to know what makes an approach safe and how to determine the particulars for flying them. On type-rating check rides, however, it would be highly unusual for an examiner to test an applicant about protected airspace or anything more detailed than asking the particular category and airspeed at which a pilot might choose to fly a circling approach. Dave Stohr, director of training services for Air Routing International said, “Pilots are often not well versed in these areas, because in training, we don’t spend adequate time on subjects other than flying.”

Robert Wright, FAA’s manager of general aviation and commercial division in Washington, said basic instrument training and a crash like Aspen leads one to ask some pretty fundamental questions. “We are not preparing people to operate in the [ATC] system today. We may only be getting them ready to take and pass the knowledge test.” Mac McWhinney, a San Diego flight instructor and aviation curriculum development specialist, said some flight instruction programs address TERPS, a bit, but none to his knowledge offer an in-depth explanation of these building-block procedures. He added, “The FAA is not looking for the right things on the knowledge tests anyway. They are not thinking about what knowledge a pilot needs to be safer. All they are trying to do is make the tests hard.”

The Cessna Pilot Center’s instrument training program uses the book, Cleared for Approach, that offers some insights into how approaches are developed and should be flown. But in general, the publications necessary to find the specifics of instrument approach procedural answers are not easy to locate. An Internet search by AIN revealed a few Web sites that offered links to TERPS-related information, such as www.terps.com, but only the U.S. Government Printing Office offered the TERPS publication for sale. FAR Part 97, Instrument Approach Procedures, is not normally included in the Jeppesen regulations available on a subscription basis, although that company offers an FAR/AIM book that does include this information. By contrast, international instrument procedures documents normally include ICAO’s Procedure for Air Navigation Services Operations (PANSOPS), that is divided into two publications. One volume outlines how approaches are developed, while the other includes the nuts and bolts about how procedures should be safely flown. TERPS does not offer a similar “On The Job Training” publication.

Pilot Tod said, “The category issue also confused me. If you read about straight-in approaches, it is very easy to understand. But circling approaches are referenced to speed and don’t talk about the categories at all.” Tod recalled that his instrument training discussed the fact that categories existed, but not how they all fit into the instrument approach package.

Bob Johnson of Air Castle Corporation said, however, “I think the approach category system we use right now is fairly straightforward, but misunderstood.”

And John Williams, a Houston-based Falcon 900 pilot, said, “I think the associated regulations are fine as they presently exist. I don’t believe there is any ambiguity in our present state of [approach] design and certification. I’m not an aeronautical
engineer, but I have no problem recognizing the difference between a certified value and the necessity of using values greater than the minimum certified when circumstances dictate.”

“I was just at FlightSafety for recurrent,” said one NBAA member. “We did an ‘enrichment’ course on this [Aspen] accident. The Category C versus Category D question came up and the instructor’s response was that at the accident weight and the configuration they were in, they were actually legal to use Category C minimums.”

Another experienced GIII pilot, responded. “There is not any circumstance or weight that a Gulfstream GII or GIII can be Category C. Never. Circling with full flaps is not authorized in a G1159 aircraft either. The GIV and GV can be Category C because they are authorized to circle with full flaps, and due to their great wing, they fly slower. No matter what category the aircraft is in, if you fly faster than 141 knots, you are Category D.”

Others speculated that because the flight was operated under FAR 135, it may have been over maximum landing weight for that runway to comply with the 60 percent landing runway rule, not to mention the fact that the runway was contaminated. Was the GIII crew confused, or simply under that much pressure to land?

Most people believe a Hawker 800, for example, is certified under Category D for circling, but are often not sure why. FAR Part 97, Instrument Approach Procedures, says that the aircraft category is determined by multiplying the aircraft’s stalling speed in the landing configuration by 1.3, at maximum certified gross landing weight. So indeed, the manufacturer determines the base category of the aircraft, since that weight cannot change. But circling approaches are based upon speed, which can change in the landing configuration on the basis of weight. Explaining that there are both certification and operational categories, the FAA’s Tom Penland, air carrier branch manager in Washington said, “Categories may be perceived as coming with the airplane, but if you go outside of those criteria, you have to make appropriate changes to ensure adequate clearances [from terrain].” He added, “But big airplanes don’t circle. There’s a reason they don’t.” The rule of thumb according to Penland is simply, “Fly faster and fly higher.”

There is also a difference of opinion on which direction categories change. Most pilots believe categories can vary upwardly, such as a Category C aircraft falling into Category D for circling, but not the other way. One FAA spokesman said, “You can go up a category, but never down from the way the aircraft is certified.”

Wright disagreed, saying, “If you can go up a category, you can also go down one,” from the one in which an aircraft is certified. Some pilots believe, too, that if they can fly a category D aircraft slow enough to remain in Category C’s protected airspace, they may fly an approach under Category C.

Falcon 900 pilot John Williams said, “An aircraft is certified in only one category, but may be operated using minimums associated with a higher category, but never a lower one.” FAR Part 97 also seems to support this opinion, defining the approach category as downwardly unalterable: “Aircraft approach category means a grouping of aircraft based on a speed of 1.3 Vso (at maximum certificated landing weight). Vso and the maximum certificated landing weight are those values as established for the aircraft by the certificating authority of the country of registry.”

The Kansas City attorney explained the overall dilemma. “It’s up to the crew to reconcile the differences between the certification category and the operational category. There is no way for the FAA to give a final word without being unduly restrictive. You have to bring training, practice, standardization, briefing and discipline all together to get the right answer.”

If you agree that the operational category chosen for a circling maneuver is flexible, depending upon the speed necessary to perform the maneuver as long as the aircraft remains within the protected airspace, there is also a question on whether the speed used to circle is indicated, ground or true airspeed. Consider again the approach into Aspen. The flight data recorder showed the aircraft holding approximately 130 knots IAS on the way inbound from the final approach fix, Red Table VOR. However, at nearly 14,000 ft msl, the aircraft’s TAS would have been considerably higher than 130 kt, as much as 13 kt in one expert’s opinion, further reducing the safety margins on an already tricky approach in IFR weather. Did the crew not care about this issue or were they simply not aware of the affects of altitude on their approach speed? FAA’s Dave Cook, a helpful guy who designs instrument approaches in Oklahoma City said, “I only develop protected [approach] airspace. We don’t make our calculations based upon speed.”

AIN asked some pilots what speed they reference in a circling maneuver. Two replied groundspeed, three said indicated and one answered indicated but accounting for true airspeed factors. When defining approach categories, the FAA references only the word “speed,” although on straight-in non-precision approaches, groundspeed is the guideline. The Airman’s Information Manual says that if a pilot is circling at a higher speed than the normal approach category, the higher approach speed should be used. One pilot said, “The regulation should state that determining your approach category should be based upon the actual fuel load versus the maximum fuel load and weight. That would be a much better way of determining what approach category an aircraft should be in. However, that is not what the regulation says.”

If pilots don’t learn the details, nuances and blind alleys of various kinds of instrument procedures and their related categories at initial or recurrent or when working on their instrument ratings, where do they learn? Some pilots claimed to have learned on the job or through their own reading. At issue however, is that a pilot doesn’t know if the person teaching understands instrument approaches any better than their teacher taught them. Stohr said, “Pilots should be learning this aircraft category and minimum information during their instrument rating.”

Wright, the FAA manager of GA and commercial aviation, said he thinks it is time to start filling in the cracks in the system. He is at work on a white paper that addresses training issues in relation to real world performance in the evolving ATC environment. “I’m looking at technology and airspace and how this all may affect the current training system,” he said. “It is time we stimulate some new thinking on the subject.”

Apparently, many pilots share many different perspectives on instrument approach procedures. Williams said, “Finding good guidance on any aviation-related question can be a challenge. There’s not usually a single source you can go to in order to find answers. In most cases, simply asking the FAA’s FSDO inspector won’t be satisfactory either. These are pilots, who likely came from the same pool of confusion as other pilots in the industry.”

Williams offered a few useful tips to break through the regulation and procedural clutter. “Never take what the FAA has interpreted and told you over the phone as error free. Get second opinions and verify the sources. Professional pilots must be responsible enough to pursue clarification of the things they find confusing. Sometimes applying a little logic can go a long, long way.”