Forecast "Plumes" From NWS GEFS

UPDATE - the new precipitation plumes from 06 UTC December 6th.

An interesting, experimental, forecast product is now available from NWS EMC for the GEFS (Global Ensemble Forecast System). "Plumes" refers to time series of a specific weather element that are generated directly from the model forecasts. In this case the plumes are plotted, all superposed, on a single time series background. The example above shows the current ensemble members' forecasts of accumulated precipitation for TUS (the airport) for the next 192-hours. Black curve is the ensemble mean and blue curve is the forecast from the operational version of the GFS.

The ensembles are are in perfect agreement through early next Friday with no precipitation forecast by any of the ensemble members. However, after Friday there is a wide spread among the various forecasts that ranges from about 0.00" to one outlier that predicts over 0.40". The operational forecast appears very similar to the average  amount forecast by the suite of ensemble members.

This certainly appears to be a useful product and I'll be watching as the winter progresses.
------------------------------------------------------
Corey Guastini, of NOAA, posted the following message and link to the experimental site to the MAP list:

MAP,

A GEFS upgrade was implemented this morning, and EMC's Model Evaluation Group has created a plumes page to help analyze the ensemble output:

http://www.emc.ncep.noaa.gov/mmb/cguastini/gefs/EMCGEFSplumes.html

The functionality of this GEFS plumes page is similar to that of the SREF plumes run out of SPC. There's a selection of seven fields to choose from for major stations across the country, data from each of the 20 perturbed members and the control member along with the ensemble mean and corresponding run of the GFS, and the option to view a subset of the forecast hours for a given cycle (only the first 192 hours with 3-hourly output are included). If you notice any problems or think there is an additional field/station that should be included, feel free to contact me. This page may not display properly in some older browsers.

Thanks,

Corey