Amid a memorable winter, the snowstorm that crossed the Chicago area on December 20, 2010 was not especially strong. The city’s two largest airports received two to three inches of snow, and the timing and amount of precipitation were well-handled by the traditional forecasts issued by the local NWS office.
Even so, it was an expensive evening for the city and the airlines. After gorging on caffeine and doughnuts, representatives of UAL and SWA revealed the truth of their companies’ failures during the night of study. UAL had numerous diversions since they had expected an acceptance rate of 48-56/hr, which was for a time 48-56/hr too high. SWA had zero diversions but made numerous overflights, which is not a diversion in the same way that a sort conducted without kilos of late packages is not late.
SWA has a policy that they forbid travel by unaccompanied minors during weather disruptions like this one in order to avoid ugly confrontations with angry parents of missing children. It cuts down on Amber Alerts.
Al Perez of O’Hare City Operations is the manager in charge of snow removal at ORD. He once held the same position at MDW and has been doing this work for decades. His goal is sort of a “no runway left behind” ideal where closures are brief and affect as few runways as possible. It is not uncommon for different parts of the airport to experience different precipitation rates and this adds to his tactical planning load. He has trained his group to use the wind to help remove snow. He also considers moisture content since he can omit the anti-ice treatments if the snow is dry enough (20:1 to 15:1).
He shared his appreciation of the challenges along with the pieces of information he could use to make his job easier. Perez currently has four sources of weather information, including one person from UAL who goes to the tower as a consultant. He discussed his suffering from the obvious complication of enjoying multiple sources of forecasts. Oddly his next remark was that he was considering hiring his own meteorologist to add to this mix.
Principal expenses are people and equipment, naturally. Total costs from the December 20th storm was around $10 million. Bad manpower forecasts result in added costs, whether he has too many or too few. The night before this conference the forecast called for three to five inches but only .2 inches fell. If the forecast had verified he would have spent $250 thousand less than he actually did for the dusting.
Perez cares most about precipitation start/stop times, temperatures and wind direction. But the biggest shortcoming in his information is detailed snowfall rate forecasts through the course of a storm. He said this would be especially helpful for planning equipment usage.
Personnel from ARTCC, TRACON, and the ORD Tower presented their complex range of issues from the traffic management perspective. Their most difficult problem among many is planning for brap/bran reports. The city closes a runway on the first bran report and the tower people have a rule of thumb that two brap obs rapidly lead to a bran report.
Putting a stop to landings means takeoffs have to be stopped due to all the aircraft holding.
The tower needs 15-20 minutes lead-time for turning the airport around. If they get that, the operation is no big deal.
Traffic flow is affected by where the planes exit the runway.
They expressed their wish that tafs and metars evolve into more useful sources of information, in the spirit of NextGen. This led to a discussion of restrictions on what can and cannot go into a taf and a metar, due to the WMO.
Like Al Perez, the people who came to the workshop from traffic management clearly enjoy solving problems.
Lewis University’s dining hall followed this with a terrific lunch. That was fortunate since the featured luncheon speaker read every PowerPoint® slide verbatim. It is possible that Lewis is affiliated with the Reading is Fundamental program. Anyway, the food was too good to throw.
The broad picture I got throughout the day was that the NWS has a hodge-podge (I mean this in a positive sense) of local initiatives that deal with problems specific to different terminal areas. Some of these also address enroute matters.
A probabilistic snow accumulation program for Denver is in its third year. This was a result of DIA Ground operations struggling to extract more information than was possible from the traditional taf. The technique relies on the SREF and was developed by the SOO at NWS Boulder. It is available at http://www.crh.noaa.gov/bou/?n=aviation.
The majority of delays in the NAS arise from the NY/NJ/PA airspace. A similar aviation conference hosted by NWS in New York highlighted the lack of coordination between the airlines, FAA, and NWS. NWS conducted a 60-day demonstration to enhance weather services by dedicating an aviation forecaster to planning telephone conferences. Customer feedback was overwhelmingly positive. They have found some success in using Bufkit’s momentum transfer product in forecasting gusts at the surface. It takes the thickness of the boundary layer into account when determining how much of the energy mixes down. Science.
In Atlanta, tafs go out every two hours. They are using NWSChat, and emphasize forecaster confidence in each afd.
Chicago has put together a one-stop website for aviation customers. They showed how forecasting deep, moist convection only at arrival and departure points can be misleading for the terminal area, so they are revamping that product. They have an experimental winter precipitation outlook with hour-by-hour parameters and a discussion. http://www.weather.gov/zau.
AWC in Kansas City summarized their voluminous offerings along with changes that are upcoming. Bar bet: the number of convective sigmets, around 30,000 annually, is their only product that outnumbers the CCFP (25,000 polygons served each year), which survives bafflingly under a shinier acronym.
These efforts are done with an eye toward the megacolossal NextGen project and will test some concepts and products in advance.
The mets at the Romeoville NWS, adjacent to Lewis U., have learned a lot from the planning phone calls. That office added three forecasters who have been cross-trained in CWSU operations. They also have an Aviation Services Improvement meteorologist.
Their postmortem of the December storm made them realize that visibilities alone told them little about the snowfall rate. ORD reported SNINCR during two separate intervals but the visibility never fell below 3/4sm. They speculate that smaller flakes yielded a better visibility during the heaviest snow. They are going to study this possible relationship. I suggested they include sleet in their work but that precipitation type does not fall for long in their area, climatologically.
This was an exceptional workshop. It was one of the few times where a faces-to-faces gathering made sense. The material presented was not just a re-hash of Powerpoint slides, mostly.