The new head of the Boulder-based Developmental Testbed Center knows quite a bit about WRF. In fact, he helped create it. Ying-Hwa “Bill” Kuo, who’s been at NCAR since 1982, took the reins of the DTC on 1 April after the retirement of founding DTC director Robert Gall. NOAA’s Steven Koch will continue as deputy director of the DTC.
Steven Koch and Bill Kuo. (Photo by Carlye Calvin.)
Kuo and Gall were inspired to help develop what became WRF in 1996 after a program reviewer asked them about the impact of NCAR’s community modeling on operational numerical weather prediction. “Bob [Gall] looked at me, and we said, ‘It’s close to zero.’ But we agreed that NCAR’s work should benefit society.” Shortly afterward, Kuo and Gall met with Geoff DiMego (NOAA National Centers for Environmental Prediction, or NCEP) to lay the groundwork for the multiagency collaboration that led to WRF.
The twin missions of prediction and research are tightly linked at the DTC, which was championed by Gall, Koch, and WRF program coordinator Nelson Seaman (Pennsylvania State
The emphasis is on rigorous, quantifiable results, says Kuo: “We’ve focused on being a neutral party in testing and evaluation.”
Since 2006, the DTC’s Modeling Evaluation Tools (MET) package has provided a wide variety of techniques for verifying the skill of forecast models. The latest version—MET 2.0, released in April—includes new tools for verifying probabilistic forecasts, producing statistics on wind direction, and defining subdomains within models.
The DTC’s dedicated computing time allows for new code to be tested for as long as several months, helping ensure that it’s up to the rigors of day-to-day forecasting use. “Research scientists don’t always have the resources to carry out that many runs over an extended period. A new routine might work well for seven days, but on the eighth day, it might blow up,” says Kuo. Conversely, operational centers may not have the luxury of time needed for in-depth analysis of their modeling techniques.
The DTC’s two dozen staff are drawn mainly from NCAR’s Research Applications Laboratory and NOAA’s Earth System Research Laboratory (ESRL), with most holding joint appointments between the DTC and their home institutions. The center also maintains a visitor program and holds several workshops each year. Funding comes from NOAA, NSF/NCAR, and the Air Force Weather Agency.
Increasingly, the DTC is working on ways to entrain more data into models. NCEP is using the center to help build an open-access version of one of its key data assimilation tools, the Gridpoint Statistical Interpolation scheme. The DTC will also offer a tutorial for NOAA’s Hurricane WRF model, which will be released to the community in the coming year as a coupled atmosphere/ocean system, eventually to include a wave model.
“I would say the DTC is one of the shining examples of NCAR and NOAA working together,” says Kuo. “Absolutely,” adds Koch. “All of the examples cited herein—the core WRF systematic testing procedures, the verification development and applications, support of the Hurricane WRF to the community —as well as others, have fully involved NCAR and ESRL’s Global Systems Division working closely together over the years.”