Crews need specific training for flying in the mountains

 - October 18, 2006, 6:00 AM

Pulling into position for takeoff or passing the final approach fix for landing, especially at a mountain airport, presents many challenges. The accident history of corporate jets indicates an apparent lack of preparation for contingencies, such as engine failure on takeoff, mountain wave turbulence, extreme cold temperature errors in altimeters, technology glitches, changes in IFR clearance and so on. Another indication is the increase in business jet fatalities last year despite a decrease in general aviation overall.

Think this doesn’t apply to you? How about a little pop quiz? You are on approach to Aspen, Colo., with airport temperature -30 degrees C [ISA -30]. At Red Table VOR [DBL], you perform a procedure turn for the VOR approach at 14,000 feet msl [about 6,000 feet agl], then fly the step-downs to the minimum descent altitude (MDA) of 10,200 feet msl [about 2,400 feet agl]. What is your altimeter error due to the extreme cold temperature at DBL? A quick check of the chart in the AIM, paragraph 7-2-3 (shown at right), indicates that the error would be approximately 1,140 feet [the chart only goes to 5,000 feet, but extrapolation indicates a 190-foot increase in error per 1,000 feet].

At the MDA, the error would be about 456 feet. This should cause a pilot some concern, since TERPS rules on nonprecision approaches typically provide only 300 feet of vertical clearance from terrain and obstacles. What should you do? How about adding the temperature error to all of the prescribed altitudes on the approach, such as is required in Canada. So you would level off at your adjusted MDA of 10,700 feet msl, giving you a true altitude of slightly more than 10,200 feet. This way you will have the full clearance from terrain designed into the approach.

AIM paragraph 7-2-3 indicates “The possible result of the above example should be obvious, particularly if operating at the minimum altitude or when conducting an instrument approach. When operating in extreme cold temperatures, pilots may wish to compensate for the reduction in terrain clearance by adding a cold temperature correction.” [emphasis added]

Think your new RVSM, FMS or TAWS system protects against this problem? Better make sure the system you have installed compensates for it. Advisory Circular 23-18 indicates “For those [TAWS] systems that do not provide extreme cold temperature compensation of barometric altitude (corrected altitude), the AFM, AFMS or supplemental AFM should include the following note: ‘Operations at extreme cold temperatures, for example, -30 degrees C, will result in a significant reduction in terrain clearance provided by TAWS alerts.’

“To reduce or eliminate errors potentially induced in barometric altitude temperature extremes, the design should include features such as corrected altitude or special operational procedures. For operations in cold weather, either the system should be able to account for variations in extreme cold weather temperatures, for example, -30 degrees C, or additional flight-crew procedures should be considered to address pressure altitude limits for vertical position determination.”

Controlled Flight into Terrain

How about unexpected changes in IFR clearances, coupled with technology glitches? Perhaps the best known controlled flight into terrain (CFIT) accident is the December 1995 crash of American Airlines Flight 965 near Cali, Colombia. The Boeing 757 was en route from Miami to Cali. The wind was calm, and the crew expected to land on the usual Runway 01, which had an ILS.

As they approached from the north, however, the controller requested that they accept Runway 19 instead, and they agreed. In hastily setting up and executing an arrival [one of the crew remarked “…we will have to scramble to get down”], descent and VOR approach to Runway 19, they turned due east rather than maintaining the southerly headings dictated by the arrival and approach and descended into a mountain.

The crew received a ground proximity warning system (GPWS) warning but was apparently unable to avoid contacting the mountain because it had already extended the air brakes to expedite the descent. Apparently the crew misunderstood the ATC clearance and input a “direct to” Cali into the FMS, which deleted all the intermediate fixes that were part of the arrival and approach and further interfered with the situational awareness of the crew.

Risk Assessment and Pilot Judgment

The aircraft accident report from the Colombian Aeronautica Civil referred to research performed on dynamic situations where experienced people made decisions based upon cues that they matched with previous experiences. The investigators noted that both flight crew were experienced ATP-rated pilots who had taken straight-in approaches to shorten flight time on many occasions without incident.

But the report also noted that the research showed that these types of decisions could be dangerous when, despite changing circumstances or the initial assessment of the situation being incorrect, the initial decision is not reconsidered.

In fact, the crew did not have time to pull out and fully review the approach chart, locate the VOR and NDB fixes that were essential to the approach and consider the proximity of terrain. They allowed themselves to be rushed, to the demise of themselves and their crew and passengers.

A similar accident happened in 1991 near Rome, Ga., when a Beechjet crew took off in mountainous terrain in marginal VFR weather, without an IFR clearance and lost track of, or never knew, where the terrain was. According to the NTSB report, the company airplane had seven executives aboard who were making a tour of 12 facilities in different cities in the South. The time was limited at each stop. The two aircraft on approach to Rome would have resulted in a delay of IFR release for the Beechjet, and the captain elected to take off VFR and to try to obtain an IFR clearance in the air.

The Board found no evidence that the crew used a sectional chart, and cockpit voice recorder (CVR) remarks indicated that neither pilot was familiar with the terrain in the vicinity. They were basically “scud running,” which the Board found to be a particularly dangerous type of operation in any environment but especially in the mountains. The aircraft was not equipped with a GPWS.

The Board noted that the captain appeared, from his behavior on this flight, and from statements they received, to have a tendency not to “employ good operating practices,” and that the first officer was aware of this tendency and brought it to the attention of management, but to no avail.

This sort of accident should serve as a warning to any company that completely relies on its flight crew for assurance of flight safety. Clear and open lines of communication should be maintained so that junior flight crewmembers and others can make management aware of shortcomings in decision-making and procedures without repercussions.

The days of the seasoned old captain being the unchallenged authority on all matters in the cockpit should be put behind us, the report noted, and the Board recommended that companies should be urged to establish policies and procedures that “encourage first officers to play an active role in cockpit decision-making.”

Engine-out Procedures

So let’s consider a sample plan for departure contingencies. It’s a cold and windy winter day at Aspen, Colo. [ASE], elevation 7,815 feet. Ceiling is 4,000 feet, visibility seven miles. A look at the Lindz Four Departure and takeoff minimums indicate that a non-standard climb gradient is not legally required for departure under these weather conditions.

But the mountains are still looming on either side and ahead of departure Runway 33. The performance data may indicate that your jet can make the climb gradient of 460 feet per nm [7.6 percent] up to 14,000 feet the departure procedure calls for with all engines running, so you assume you are safe to climb above and around the obstacles and terrain using the departure procedure. But what if you lose an engine just after V1? It is doubtful your performance data will reflect adequate performance to make the 7.6-percent climb gradient up to 14,000 feet on one engine. Do you have a backup plan? The answer too often is no.

A little background: all transport-category aircraft must meet the minimum “second segment” climb requirement under Part 25 of the FARs of 152 feet per nm [2.4 percent] up to 400 feet. If applicable, Part 135 prohibits takeoff at a weight and temperature that does not permit a climb that will clear obstacles by 35 feet vertically, plus a safety margin of .8 percent of distance traveled [48 feet at first nm] within 300 feet horizontally.

Part 91 has no such specific requirement beyond the second segment and departure procedure requirements, but pilots might be wise to observe the same or similar guidelines for safety. And it may be considered careless or reckless operation under Part 91 not to do so.

One method of planning for the loss of an engine on takeoff would be to become familiar enough with the departure area to permit visual separation from the mountains or other obstacles such as by following roads and rivers. This would require sufficient ceiling and visibility to permit visually seeing and avoiding the obstacles and terrain, of course. One guideline for Aspen some pilots use is a minimum 5,000-foot ceiling and 10 miles visibility, for either arrival or departure.

One could also develop or obtain a specific instrument procedure to be used in the event of loss of an engine, which provides greater horizontal separation from obstacles and terrain than the departure procedure provides.

Airlines and other commercial operators use these so-called “emergency escape procedures,” which are available from several vendors for most transport-category airplanes and mountainous airports–to satisfy Part 121 and 135 requirements. They specify maximum weights for takeoff at given temperatures. And they often specify turns, climb and navigational procedures different from those of the departure procedure to provide obstacle and terrain clearance based upon the specific climb capability of your aircraft on one engine.

Mountain Wave

And what about the gusty wind conditions pilots encounter when departing from Aspen? How can you avoid finding yourself unable to outclimb extreme downdrafts coupled with turbulence on your departure climb, approach or missed approach? The AIM, paragraph 7-5-5, warns that any time winds aloft at mountain peaks exceed 15 knots, mountain wave activity might be found on the leeward side of the mountains. It states: “Many pilots go all their lives without understanding what a mountain wave is. Quite a few have lost their lives because of this lack of understanding. One need not be a licensed meteorologist to understand the mountain wave phenomenon.”

While it might seem unrealistic to cancel flights into or out of a mountainous airport whenever the winds aloft at the peaks exceed 15 knots, there certainly should be some limit. Some professional pilots set this limit at as low as 30 knots, while others make it as high as 50 knots.

Whatever limits you decide upon, you should observe them on each flight. And in addition to checking winds-aloft forecasts and AWOS reports in the vicinity of your destination mountainous airport, you should look carefully for pilot reports, convective sigmets and other weather reports for indications of either reported or forecast wind shear, or moderate or severe turbulence, especially if associated with lenticular or rotor/roll clouds within a few hundred miles of the front range of the mountains. If it exists or is forecast in that area, chances are excellent it will be even worse within the mountainous area.

An illustration of this pitfall can be found in the 2003 crash of a Westwind near Taos, N.M. The NTSB found that there was a convective sigmet covering the area for severe turbulence and mountain wave activity. Later review of satellite imagery revealed that bands of roll clouds were in the vicinity of the airport.

The crew was either unaware of or ignored the sigmet, and began the flight from Las Vegas to Taos. On approach to Taos, shooting the VOR approach on autopilot, the aircraft apparently pitched up dramatically when the autopilot disconnected, causing an upset from which the crew was unable to recover.

Mountain wave activity is not always something the aircraft and crew can handle. In the extreme, and especially close to the ground or terrain, an aircraft under the effects of this phenomenon is often unrecoverable. The only safe course is avoidance.