Model Diagnostic Discussion
NWS Weather Prediction Center College Park MD
1217 AM EDT Wed May 16 2018

Valid May 16/0000 UTC thru May 19/1200 UTC

...See NOUS42 KWNO (ADMNFD) for the status of the upper air
ingest...

00Z NAM/GFS evaluation with preliminary model preferences
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
GFS/NAM initialization errors do not seem to degrade their short
range forecasts.


Northwest/West/Central Plains
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
A compromise of the 12z UKMET/12z Canadian/12z UKMET/00z GFS/00z
NAM appears good here -- with confidence above average -- as the
guidance shows good agreement.


Northern Plains
~~~~~~~~~~~~~~~
Preference: Non-NAM blend; confidence above average

From Friday morning onward, the 00z NAM becomes slow with a trough
crossing southern Canada and a shortwave moving out into the
central Plains when compared to the other guidance -- its usual
bias.  A compromise of the 00z GFS, 12z Canadian, 12z UKMET, and
12z ECMWF is preferred here with above average confidence.


OH Valley/Southeast/East
~~~~~~~~~~~~~~~~~~~~~~~~
Preference: Model Blend; confidence average

The guidance has trended stronger with a 500 hPa closed cyclone in
the lower OH valley, which has led to a more northerly warm front
in VA Friday into early Saturday and a somewhat quicker movement
to the deep layer Gulf low, when compared to last night's
preferences.  The 00z GFS appears too far north/too defined with a
surface low in southern OH, but this is a minor detail.  A
compromise of the 00z GFS, 00z NAM, 12z UKMET, 12z ECMWF, and 12z
Canadian should work out fine.  There is concern that the boundary
in VA may be too far north considering the strength of the surface
high moving through southeast Canadian and New England, which
bears watch in later model runs.  With that in mind, kept
confidence no better than average.


Model trends at www.wpc.ncep.noaa.gov/html/model2.shtml
500 mb forecasts at www.wpc.ncep.noaa.gov/h5pref/h5pref.shtml

Roth