There are many reasons why the actual runway environmental conditions can vary significantly from the reported values.
SNOWY RUNWAYS, LOW CEILINGS AND A CROSSWIND LANDING ARE threats that a prudent flight crew will take seriously. Now add a runway distance that is barely marginal if the subjective braking action reports are “good,” and the situation turns into one in which every decision and action by the flight crew needs to be spot on.
To further heighten the risk, let’s put a body of water at the boundaries of the marginal-length runway. That was the situation faced by the flight crew of Delta Flight 1086, a McDonnell Douglas MD-88, as it approached New York’s LaGuardia Airport (LGA) on the snowy day of March 5, 2015.
The flight left Atlanta that morning heading for LGA. While enroute, the flight crew continued to monitor the weather conditions at LGA and assessed the factors that could affect stopping performance. They closely examined company policies for landing on contaminated runways and understood that a change in runway conditions from accumulating snowfall could increase the landing distance and that a change in wind could cause the flight to exceed crosswind limits.
The flight crew asked the dispatcher and the Washington Air Route Traffic Control Center controller for braking action reports, but neither had any reports at the time because LGA operations personnel were conducting snow removal operations, and no aircraft were landing. The four previous automatic terminal information service (ATIS) reports (issued between 07:51 a.m. and 10:24 a.m.) contained outdated and contradictory field-condition information about the status of LGA’s runways. Besides company and ATIS reports, air traffic control (ATC) communications as late as 10:40 a.m. gave the impression to the flight crew that at least some patches of runway surface would be visible upon breaking out of the instrument meteorological conditions (IMC) on the approach.
But upon first seeing the runway in sight at 233 ft. above ground level (AGL), the runway appeared white. This was contrary to their expectations given the recent snow-cleaning operations and the reports of good braking action by two of the four preceding aircraft. Only 13 sec. elapsed between the time the captain called the runway in sight and the 50 ft. automated call-out, during which the flight crew intensely focused on precise control of the aircraft. It would have been difficult for the crew to visually assess the nature and depth of the snow on the runway.
A combination of factors resulted in the MD-88 veering off the runway, coming to rest with the nose of the aircraft over the berm above Flushing Bay. There were no fatalities, but 24 people were injured.
While the bulk of the official NTSB accident report focused on a phenomenon called rudder blanking, the accident amply illustrates the consequences when the actual runway environment differs from the reported conditions. Accurately predicting the effects of wind, temperature and runway surface conditions are vital to every takeoff and landing. Standard practice in the aviation industry expects a pilot to dutifully enter the performance charts for these parameters to calculate the aircraft’s performance. Yet there are many reasons why the actual runway environmental conditions can vary significantly from the reported values.
Wintry conditions prevailed in the day previous to this Citation CJ4 incident. During landing, the aircraft hit some ice and veered off the runway. Credit: Wasatch County Fire District
The situation is amplified for business aviation aircraft that operate into a wide spectrum of airports, most of which are non-towered and have limited resources for runway snow removal. The runways often do not include features such as crowning, grooves and porous filled concrete to minimize the pooling of water that exists on runways serving scheduled air carriers.
Fixed-base operator (FBO) personnel likely have little training on the accurate assessment of braking action from the perspective of an aircraft’s needs. Furthermore, transitory phenomena such as the melting action from daytime sun on snowbanks adjacent to a runway can result in a liquid that turns into black ice after sunset and will not be readily apparent.
WINDS AT THE THRESHOLDSome of business aviation’s most glamorous locations are surrounded by significant landscapes that can create their own microclimates. Notable U.S. examples are Aspen, Eagle, Telluride, Gunnison, Sun Valley, Truckee and Jackson. European examples include Gstaad, Samedan (St. Moritz) and Courchevel. These are considered some of the most challenging airports in the world because of their difficult topography and winds, as well as high altitude.
The microclimate effects produce rapidly changing localized winds that will not be detected by an airport’s automated weather observing system (AWOS). Adverse winds caused by mountain wave, diurnal “canyon” winds or convective activity can create downdrafts of significant strength.
Localized winds just short of the landing threshold can cause negative effects on an aircraft’s stability, control and performance. Even minor variations in vertical currents as the aircraft is precariously transitioning into the landing flare can cause the aircraft to balloon or dive markedly from the desired glide path. A sudden loss of headwind from windshear can cause the aircraft to nose down and temporarily lose important airspeed.
These effects can be even more pronounced when a runway’s threshold is close to vertical terrain. A classic example would be the cliff adjacent to the threshold of Telluride’s Runway 9. As the sun’s angle moves across the sky and begins to heat that slanted terrain, the air immediately adjacent to the cliff begins to heat and rises rapidly in a thermal.
Localized winds just short of the landing threshold can cause negative effects on an aircraft’s stability, control and performance. Aircraft flight path maintenance can instantly be compromised by these sudden and surprisingly strong vertical air currents. A potent example of this is the warning for strong vertical turbulence at the edge of the mesa collocated with the threshold of Telluride’s Runway 09. Credit: Kim Henneman
For those lacking the benefit of a soaring background, thermals are rising parcels of air that continue to rise as long as the surrounding air is cooler. It is not uncommon for the strength of the cores of these thermals to exceed 2,000 fpm in the western U.S. states. Conversely, the outer portion of these rising bubbles—indeed, imagine the shape of a doughnut, with the middle rising and the outside descending—can be nearly as strong.
Aircraft control and flight path maintenance can instantly be compromised by these sudden and surprisingly strong vertical air currents. Incidentally, there is a warning for pilots that the Telluride airport sits on a 1,000-ft. mesa, with the precaution of strong vertical turbulence along the mesa’s edge.
What is the FAA criteria for the siting of a wind sensor? According to Order JO 6560.20C, “Siting Criteria for Automated Weather Observing Systems,” the preferred siting of the wind sensor at an airport with only a visual or non-precision runway is adjacent to the primary runway 1,000 ft. to 3,000 ft. down runway from the threshold.
Clearly, these indicators are not able to accurately sense the shifting wind currents in the threshold of a runway such as Telluride’s Runway 9.
This type of rapidly changing adverse wind close to the approach end of the runway was a contributing factor in the crash of a Socata TBM 700 on Feb. 15, 2003, at Aspen-Pitkin County Airport, Colorado. The approach was stabilized at 100 kt. with landing gear and flaps in the landing position. The approach was normal until approximately 100 ft. above the runway, at which point the airplane encountered a turbulence condition, causing rapid-roll tendencies right and left.
As the pilot began his landing flare at about 15 ft. above the runway, the left wing dropped rapidly—combined with a sudden high sink rate—and struck the runway. Fortunately, none of the four individuals in the aircraft were injured. Winds at the time were reported 310 deg. at 6 kt. Records suggest that the winds were variable throughout the day. The NTSB determined the pilots had failed to maintain aircraft control. Contributing factors include the tailwind and the turbulence.
TEMPERATURE AT THE RUNWAYThe heat on the ramp was unbearable while walking out to the aircraft on a hot August afternoon in Lincoln, Nebraska. ATIS was reporting 108F, but it felt much worse than that on the ramp.
Mechanics from Duncan Aviation walked out to the aircraft with their recently acquired infrared temperature detector. Their “temperature shot” from the cement showed a reading of 127F. The blacktop was even worse. It showed 143F.
As per company operating procedures, takeoff performance was calculated using the reported ATIS temperature. Fortunately, we had no passengers and only a modest amount of fuel for the post-maintenance test flight. Normally the takeoff distance would be relatively short at that light weight and low altitude, but the end of the runway seemed unusually close when we rotated for takeoff.
Months later, I was flying with a colleague whose primary passion in life is competitive racing of high-performance automobiles. He informed me that the auto racing industry is cognizant of the difference between the racetrack’s temperature and the reported air temperature. In fact, teams will purposely tune-up their engine performance in conditions as close as possible to the track conditions, replicating the time of their race.
Certainly, this same principle applies to aircraft. When the temperature of the air at the height of our engines and wings is significantly hot, we should expect longer takeoff runs, anemic climb rates, higher speeds for takeoff, reduced engine longevity and reduced climb gradients. Excessive temperatures will undoubtedly bake the tires and brakes during ground operations, increasing the risk of high-speed tire failure and overheating wheel and brake assemblies.
Small vortices shed from the lip of the helipad at Air Zermatt create a unique, localized airflow. It is necessary to keep a windsock and any other obstructions clear of the Final Approach and Takeoff (FATO) and Touchdown and Liftoff (TLOF) areas of a helipad—thus the windsock is not located to sense these localized air flow structures. Credit: Kim Henneman
According to FAA Order JO 6560.20C, “Siting Criteria for Automated Weather Observing Systems,” the temperature sensor must be mounted so that the aspirator intake is 5 plus or minus 1 ft. above ground level or 2 ft. above the average maximum snow depth, whichever is higher. It can be placed at any convenient location on the airport that is protected from radiation from the sun, sky, earth and any other surrounding objects, but at the same time, must be properly aspirated.
The sensors must be installed in such a manner as to ensure that measurements are representative of the free air circulating in the locality and not influenced by artificial conditions such as large buildings, cooling towers and expanses of concrete and tarmac to minimize the effect that the underlying ground itself might have on temperature.
I emphasize those final words with italics in the hopes that you might reach the same question I have. For the record, heat transfer is not within my engineering specialty. Many of you with soaring backgrounds will recognize the drawings in training manuals of the warmer air over heat-soaked ground to include large expanses of concrete or asphalt becoming more buoyant than air over adjacent grass-covered landscape and eventually rising as a thermal. This further reinforces my curiosity in the micro-scale temperature differences around an airport.
When will this adverse heat problem over the runway be most problematic? The amount of solar radiation absorbed by the ramp depends on various factors, such as the angle of the sun with respect to the ramp—given that the noontime sun directly overhead bombards the ramp with the highest ratio of sunshine. Clear skies and cloudy days can also contribute, as can numerous other factors. Dark surfaces, such as asphalt, absorb more radiation than lighter-colored surfaces, which tend to reflect some of the radiant energy.
It takes a lot of incoming radiation to heat up concrete, but once it does reach a warm temperature, it tends to retain that heat for quite some time.
RUNWAY SURFACE CONDITIONSFor obvious reasons, it is important for a pilot to have an accurate report on the runway surface conditions to properly perform a Landing Performance Assessment. Unfortunately, the flexibility of business and emergency medical services (EMS) aircraft to operate into a wide spectrum of airports creates the distinct disadvantage of uncertainty in the runway surface conditions.
The Flight Safety Foundation’s study of fixed-wing EMS accidents found that critical information regarding runway conditions was not transmitted to pilots in 14 of 36 accidents during landing.
One of those accidents occurred on Jan. 31, 1995, as the pilot of a Cessna 421 attempted to land at the remote airstrip in Chinle, Arizona. The airplane was dispatched in day visual meteorological conditions (VMC), and local police reported that the runway was dry, despite a recent snowstorm.
On touchdown, the pilot discovered that the runway felt softer than usual, and shortly afterward encountered a dip in the runway that sent the aircraft slightly airborne—and then off the runway through a barbed-wire fence. The three occupants were uninjured, but the aircraft was substantially damaged. The NTSB report noted that although the runway surface appeared dry, there was dry dirt about 1-2 in. deep, with a soft layer underneath.
A Flight Safety Foundation study of business jet safety reviewed 287 NASA Aviation Safety Reporting System (ASRS) reports in which pilots noted problems with runway conditions. Poor runway conditions were cited in 33% of the 287 reports, and lack of adequate runway condition reports was cited in 18%. It should be no surprise that contaminated runway conditions were present in 71% of the runway overrun accidents and incidents reviewed in the sample.
Unreported or inaccurate weather conditions and braking reports were factors in a landing overrun at Ohio State University Airport (OSU) by the flight crew of a Learjet 23. Light drizzle was reported by ATIS. No braking action advisories or reports were given. The Learjet touched down in the touchdown zone, and the crew immediately applied thrust reversers and spoilers along with maximum braking. Much to their unwelcome surprise, the braking action was nil. As the jet neared the end of the runway, the crew secured the engines, and the aircraft came to a rest 75 ft. off the end. As the pilots waited for emergency vehicles to respond, they noted that the ground became covered with clear ice due to freezing rain.
Acres of pavement with scant shade turn an airport ramp and runway into its own “heat island.” Single-engine air tankers on standby at this Rock Springs, Wyoming, air tanker base utilize reflective shades in the windows to lessen the radiative heat into the cockpit. Credit: National Interagency Fire Center
What can a pilot do to better prepare for a landing or takeoff given the possibility of uncertainty in the reported runway conditions? In an ASRS report, the Learjet pilot wrote: “If we had more information, we would have acted differently. My recommendation is this: If there is any precipitation at all in the winter months, regardless of the temperature, plan on poor braking action at best, replan your landing distance and divert if necessary” (NASA ASRS Report No. 293469, January 1995).
Experience can be an unforgiving teacher. The previous examples illustrate the pitfalls of relying on reports about the runway environment. This conundrum also applies during dynamic changes in precipitation and winds during thunderstorms or heavy snowfall events. Runway conditions and wind direction can rapidly change from the conditions used to conduct a thorough Landing Performance Assessment just 20 min. prior.
CONSIDER THE UNCERTAINTIESAviation training has failed to introduce pilots to the possibility of uncertainty in these reported values. In contrast, it is standard practice in engineering to include possible errors such as instrument error, position error and reading error into a formal analysis of the uncertainty. A draft report would be sternly tossed back if an engineering apprentice failed to perform a formal analysis of the uncertainty.
It is also standard practice in engineering to include a safety factor for the unknowns. Our safety factors in aviation can quickly dwindle given the uncertainties and inaccuracies with reported runway environmental conditions. Yes, there are safety margins “sort of” built into the landing performance data for transport aircraft. I purposefully use the caveat “sort of” due to the inherent differences in the techniques used by flight-test crews to establish the landing distances versus the method used by proficient transport crews in normal flight operations.
Thus, as you can see, accurate prediction of the effects of wind, temperature and runway surface conditions on takeoffs and landings can be prone to varying degrees of uncertainty. Furthermore, at uncontrolled airports there can be a lack of credibly measured conditions. This further complicates the task of a flight crew attempting to get the most accurate information possible.
Astute flight crews should scrutinize the possible sources of uncertainty when planning a takeoff or landing, contemplate the possibility that the runway environment could be worse than reported and consider applying prudent safety factors into their decision-making.
—Upon his retirement as a non-routine flight operations captain from a fractional operator in 2015, Dr. Veillette had accumulated more than 20,000 hours of flight experience in 240 types of aircraft, from balloons, rotorcraft, sea planes, gliders, war birds, supersonic jets and large commercial transports. He is an adjunct professor at Utah Valley University.
