We touched down beneath a dismal overcast at the Detroit Metro Airport, the runway a mixture of compacted snow and patches of bare, black concrete. The irregular pattern of snow removal indicated that the slippery area was mostly at the departure end of the runway, beyond the spot where the main wheels contacted the surface upon our landing.
As the 727 slowed below 60 knots to a safe taxi speed, I applied some pressure to the nosewheel tiller and turned the airplane to the left. The copilot and flight engineer were sharing a lighthearted moment while I navigated onto the taxiway as per ground control’s instructions. Unfortunately, the airplane wasn’t cooperating. It was defiantly attempting to continue straight ahead off the taxiway into a snowbank.
Further movement of the tiller was making the problem worse, and braking only dramatized the skid. A taxiway excursion flashed before my eyes — along with my short career as a 33-year-old captain. My heart rate accelerated. In a moment of pure instinct, I raised the power levers of the two pod engines and threw them backward into reverse thrust.
It wasn’t until that moment did my crew react with raised eyebrows. The airplane reluctantly ceased its forward motion, an embarrassing disaster averted. When I explained the rude interruption of jet-engine noise, the response was a nod and, “Oh.”
In contrast to my crew’s nonchalance, I contemplated the condition of my underwear and resumed the taxi to our gate without further incident.
The event made an impression on my psyche. Years later, I am still overly cautious about maneuvering on slippery surfaces. The fact that Detroit’s arrival runway had variable conditions throughout its length, despite a report of a “good” braking action, gave me pause for concern. What if our landing rollout had continued into the snow-covered area? What if an aborted takeoff was attempted on the same runway? Which brings me to this column’s discussion.
Fast forward to present day. Realistic calculations based on actual runway conditions have become a focus of the FAA and the airlines. The catalyst for this focus was Southwest Airlines’ runway overrun accident at Chicago Midway Airport on December 8, 2005. The 737 landed in a snowstorm, skidded off the end of Runway 31C and came to a stop on a busy roadway after hitting multiple vehicles. Tragically, a 6-year-old boy was killed, making the event the airline’s only fatal accident.
Although numerous circumstances contributed to the event, one of the significant culprits was the FAA legal methodology used to calculate runway length required for landing. It was determined that performance charts used by U.S. airlines for dispatch purposes did not take into detailed consideration contaminated conditions, such as snow, wet runways and so on.
The engineered performance charts were appropriate for dispatching airplanes to a particular destination, but the actual determination to land safely was left to the discretion of the captain. Without guidelines, that determination can be subjective, based on experience.
As a result of the Southwest accident, an official FAA aviation rule-making committee was created, involving industry experts who re-engineered the performance charts to account for various runway conditions. The calculations are called advisory landing distance charts. In addition to environmental parameters (wind, temperature, etc.) other variables include but are not limited to: reverse-thrust availability, mechanical abnormalities and flap position. Crews now have a better foundation on which to base their landing decision.
Beyond advisory landing distance charts, a standardized system for determining and communicating runway conditions called the Runway Condition Assessment Matrix (RCAM) was established. I know … yikes. The matrix considers the type of contamination, the contamination depth, the temperature and the resulting potential braking action on the runway. The RCAM associates an assessment number from 6 to zero, with zero being a braking action “nil” report. The assessment number can be different for each third of the runway; for example, 5-5-2. Having armed you with all of this information, I bring you to my arrival story at JFK.
On one of my typical return flights from London, approximately two hours before our touchdown, my copilot and I read the ATIS report via the ACARS printout. In a nutshell, the weather was good VFR with a temperature of 43 degrees Fahrenheit and light winds out of the east. JFK was using runways 22L and 22R for arrivals. The airport was departing Runway 31L. A nasty snowstorm had passed through the area approximately 48 hours prior.
The ATIS reported the RCAM for all operational runways as 3-3-3. Knowing that we are always assigned Runway 22L, calculations via our authorized iPad landing-assessment app indicated that our 777-300 would have about 100 feet of the 8,400-foot runway remaining when we came to a stop. I got a similar answer with the old-fashioned performance charts. Not good.
It seemed the solution was simple. With two hours’ advance notice, using our dispatcher, we requested Runway 31L — the longest available piece of concrete. The answer came back through Boston Center just as quickly. “Sure. Expect a one-hour hold.” My copilot and I shook our heads. Really? A one-hour hold would most likely involve a diversion. We didn’t have enough gas, nor do most other international arrivals. Great. But why?
I got the inside scoop through a high-level ATC friend. First, the Port Authority has to complete a runway “sweep” inspection before the landing, even though 31L was being used for departures. Next, the Tracon controllers have to coordinate a gap in the arrivals for the parallel runways 22L and 22R to protect the airspace for a possible missed approach while we’re landing on the intersecting runway, a complicated separation process.
Flights are entering the arrival gates from about five different waypoints. The localizer has to be activated, a procedure that involves both ATC and the Port Authority. It’s a busy time of night, so most likely other airplanes would have to hold until we completed our arrival. And apparently the temporary stop of departures on 31L has to be explained to upper-level management for its temporary lack of utilization. Good grief.
Fortunately, our story had a happy ending. As we suspected, the RCAM report did not accurately reflect the actual runway conditions. We received a verbal report from the final JFK controller: “Runway is dry. Braking action reported as good.”
So how could this potentially expensive and inconvenient situation have been avoided? The Port Authority insisted on reporting the assessment as 3-3-3 on subsequent ATIS reports, with no regard to actual conditions. I think it’s time for the Port Authority to consider its own assessments. If indeed the runway assessment was accurate, we may very well have landed our 777 with 300-plus passengers elsewhere.
And that still would have been a satisfactory ending in my book.
