National Blend of Models Update and Performance during High-Impact Events

The National Weather Service (NWS) is developing the National Blend of Models (NBM) to provide a nationally consistent and skillful suite of calibrated forecast guidance based on a blend of NWS and non-NWS deterministic, ensemble, and statistically post-processed model output. NBM version 3.2 (NBM V3.2) will be placed into operations at the National Centers for Environmental Prediction (NCEP) around November 2019.

The NBM performs well against other forms of guidance and individual Numerical Weather Prediction (NWP) when considering bulk verification statistics over long periods of time. The consensus approach attempts to minimize error, so it is reasonable to expect that the NBM may not be the best guidance for each individual weather event. However, by design, it should also rarely (if ever) be the lowest performing guidance.

This presentation will show that the performance for extreme or high impact events is comparable to other forms of guidance, including official NWS forecasts. This is in part to dismiss a commonly held belief that the NBM is designed to help primarily when the weather is benign. Evidence suggests that the NBM is every bit as useful during extreme events when it really counts.

Several case studies involving extreme temperatures, heavy rain events, significant winter weather, and strong gradient winds will be presented. A combination of subjective and objective verification of these high impact events will be shown. Although the NBM may not always be the best guidance available in each of these events, it will certainly show that it is a viable starting point for gridded forecast collaboration in the NWS.