Friday, August 26, 2016

Downscaled Forecasts from the Short-Range Ensemble Forecast (SREF) System

Alex Weech, an extremely capable undergraduate who has been working with me this summer, recently added downscaled precipitation forecasts from the National Centers for Environmental Prediction (NCEP) Short-Range Ensemble Forecast (SREF) system to the suite of products we have available at

The SREF is a 26 member ensemble that includes forecast from two different modeling systems, the NCEP Eulerian non-hydrostatic multi-scale model (NMMB) and the Advanced Research version of the Weather Research and Forecasting Model (WRF-ARW).  Yes, I know these are terrible acronyms.  They don't call NOAA the National Organization for the Advancement of Acronyms for nothing (technically the WRF-ARW is not a NOAA product, but that doesn't make it any easier).

The SREF is initialized at 03, 09, 15, and 21 UTC and produces 87 hour forecasts at 16-km grid spacing.  We then downscale those to 800-m grid spacing using climatological precipitation analyses.  For now, we are generating plots showing precipitation forecasts for the entire 87-hour period.  An example from this morning's 15 UTC run is below.

We are also generating plume diagrams.  Since the weather in northern Utah is pretty uninteresting, here's the plume for Wolf Creek Pass, Colorado.

You will notice in that plume diagram that there is a strong clustering of forecasts by model, with the ARW being wetter and the NMMB being drier.  That's a fairly common characteristic of the SREF and one that we will have to examine to see if it skews the probability statistics for the forecasts.  And, with that being said, I really don't know how good these forecasts will be.  We'll start taking a look and perhaps by ski season we'll know whether or not to continue looking or just come up with a large sum of money to buy the ECMWF ensemble forecasts...

1 comment: