To create training content, we will collaborate closely with NCEP model developers, NWS trainers, and NWS end users. The training format used will be more adaptable to NWS work schedules: short, focused multimedia lessons and webcasts. Two 2-4 minute audiovisual presentations were produced during 2012, and were well-received. Since restarting the NWP training effort, a variant of this approach has been used to produce a NWS webcast introducing the High Resolution Rapid Refresh (HRRR), and to produce a joint COMET/NWS webcast on a major update to the Global Forecast System (GFS).
For the rest of fiscal year 2015, Blend training will combine work by the NWS, COMET, and the Cooperative Institute for Research in the Atmosphere (CIRA). NWS subject matter experts (SMEs) will lead real-time teletrainings on general aspects of the Blend, followed by question and answer periods. As a companion effort, COMET will publish a lesson to introduce operational forecasters to the Blend and its use in the forecast process. COMET and CIRA will update pre-existing NWP and EPS information tables on the MetEd website (the Matrices), to reflect current model configurations. Any components used in the Blend but not found in the matrices, will be added (ECMWF's NWP and EPS). One or more short, directed lessons of 20 minutes or less on the Global Ensemble Forecast System (GEFS), and/or other scheduled model upgrades, will be produced, as resources allow. We also hope to be able to implement a new, more user-friendly interface to replace the NWP and EPS Matrices.
In 2016, major updates to NWP, EPS, and multi-ensemble North American Ensemble Forecast System (NAEFS) will require new NWP/EPS information content. Short, directed training lessons will be published for upgrades that significantly change NWP or EPS behavior. As the Blend moves from experimental status to operations by the end of 2016. additional training lessons are already planned for statistical techniques, Blend generation, and Blend product verification. The NWS will add training content on Blend grid edits and quality control, Blend product reliability, and the day 1 mean Blend product.