Session 16A.3 High-resolution winter-season NWP: Preliminary evaluation of the WRF ARW and NMM cores in the DWFE forecast experiment

Friday, 5 August 2005: 11:00 AM
Empire Ballroom (Omni Shoreham Hotel Washington D.C.)
William C. Skamarock, NCAR, Boulder, CO; and D. Dempsey

Presentation PDF (2.1 MB)

The Developmental Test Center (DTC) conducted a winter NWP forecast experiment from 10 January to 31 March 2005 - the DTC Winter Forecast Experiment (DWFE). The experiment consisted of running daily 48h forecasts using the WRF ARW and NMM dynamical cores at high resolution (dx ~ 5 km) for the continental US. The full model fields at 3 hour intervals during the forecast are available to the community on the NCAR mass storage system.

We have been evaluating the forecasts from the two cores by computing kinetic energy spectra and by examining the fine-scale structure in the solutions. The spectra in the two models differ substantially in the mesoscale. The WRF-ARW forecast spectra clearly show the observed transition from a wavenumber (k) dependence of k**(-3) at large scales to k**(-5/3) in the mesoscale at all levels in the free troposphere and lower stratosphere. Conversely, the WRF-NMM spectra exhibit only a subtle transition in the lower troposphere and a k**(-3) spectra (i.e., no transition) in the upper troposphere. These and other differences in the spectra reflect notable differences in the forecast fields and arise primarily from differences in the dynamical core dissipation and filtering formulations and configurations. We will present these spectra and forecast field examples, and we will outline the formulation and configuration choices that produce these differences in addition to discussing the implications of these forecasts experiment results for high-resolution winter season NWP.

- Indicates paper has been withdrawn from meeting
- Indicates an Award Winner