Planned Operational Implementation of the High Resolution Rapid Refresh at NCEP

- Indicates paper has been withdrawn from meeting
- Indicates an Award Winner
Tuesday, 4 February 2014: 4:00 PM
Room C203 (The Georgia World Congress Center )
Stephen S. Weygandt, NOAA/Earth System Research Laboratory, Boulder, CO; and C. Alexander, S. G. Benjamin, M. Hu, D. C. Dowell, T. G. Smirnova, J. B. Olson, E. P. James, P. Hofmann, G. S. Manikin, J. M. Brown, B. D. Jamison, and H. Lin

The High Resolution Rapid Refresh (HRRR), an hourly updated 3-km atmospheric prediction system developed by the NOAA ESRL Global Systems Division (GSD), is planned for operational implementation at the National Centers for Environmental Prediction (NCEP) in 2014. This implementation will follow an upgrade to the NCEP operational RAP slated for late 2013 or early 2014. The HRRR system currently is run by GSD in a real-time experimental configuration, nested within a real-time experimental version of the upgraded RAP (RAPv2) and is used by many aviation, severe weather, and general forecast meteorologists. The HRRR is especially skillful at predicting the timing, location, evolution, and structure of convective systems. The HRRR also does well at capturing a variety of other small-scale weather phenomena, including terrain related features, frontal passages and wind shifts, and precipitation details within large storm systems. The HRRR benefits greatly from the use of RAP background fields (which incorporate the latest observations on an hourly basis via a hybrid ensemble analysis technique), but also includes a one hour 3-km pre-forecast data assimilation cycle.

In this presentation, the current HRRR configuration that will be implemented at NCEP will be described, including the WRF-ARW model formulation and the 3-km data assimilation procedure. This 3-km data assimilation procedure includes a one-hour pre-forecast period, in which four 15-min. cycles of radar reflectivity assimilation are completed, followed by application of the Global Statistical Interpolation (GSI) variational analysis package at 3-km to ensure close fit to the latest conventional observations. The WRF-ARW model formulation used for both the upgraded RAP and the HRRR utilizes a specially adapted formulation of the MYNN boundary-layer scheme and a 9-level soil model. These changes have contributed to improved near-surface forecasts. Additional work to speed up the 3-km analysis and post-processing will also be described. This work has reduced the model latency, a key factor for a guidance product such as this.

In addition to summarizing the RAPv2 and HRRR operational implementation process and challenges associated with it, we will describe ongoing developmental work toward ensemble-based, hourly updated weather guidance from the North American Rapid Refresh Ensemble (NARRE). The NARRE will be based on a six member ensemble with both WRF-ARW and NMMB dynamic core members and co-development by NOAA/ESRL and NOAA/NCEP/EMC.