Model Diagnostic Discussion
NWS Weather Prediction Center College Park MD
1229 PM EDT Fri May 31 2019

Valid May 31/1200 UTC thru Jun 04/0000 UTC

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

12Z Model Evaluation Including Preferences and Forecast Confidence
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

...Overall Pattern Across the CONUS...
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Preference: Non-CMC blend
 Exception: 00z ECWMF/12z NAM/GFS blend in East/Northeast
            (Low weighting to GFS)
Confidence: Average

Over the next few days, deterministic model mass fields are in
relatively good agreement across the country and their QPF
patterns are fairly similar. The greatest amount of spread
continues to be with a broad trough pushing from the Midwest to
the Mid Atlantic today and its subsequent developing off-shore
coastal surface low, as well as with the trough amplifying into
the Great Lakes later in the weekend. 

The West and Southern High Plains are most agreeable with the mass
fields, particularly with the broad N-S elongated but weak closed
low in the West.  Only toward the end of day 3, do the 00z UKMET
and CMC both generally weaken and start sliding northeastward into
the Central High Plains, pressing the ridge and persistent upslope
flow regime eastward as well.  This is less favored solution but
is only a very mild difference at the end of the forecast period,
so a general model blend could be implemented or one favoring
weighting toward the GFS/NAM and the ECMWF (though it does remain
a bit drier in QPF even though instability/moisture is ample).

The shearing shortwave in the Ohio Valley will continue to slide
east toward the mouth of the Chesapeake Bay by Sat.  This spurs a
surface wave in the Carolinas tonight that will slide northeast. 
Here the spread rapidly increases with the approach of stronger
northern stream digging trof in the Great Lakes.  The 00z CMC is a
clear outlier from this point forward being slow in the northern
stream and not shearing/absorbing the off-shore wave northward. 
The 12z GFS shows some similarities, but has trended toward
lifting the wave out.  On the other side of the guidance, the
00UKMET is good with the off-shore wave, but the influence of a
faster, but stronger closed low in the Great Lakes allows for
northward shift with a secondary surface development well south
and east of the Benchmark.  The 12z NAM is also a bit stronger
with this surface wave, but has much better timing within the
ensemble suite to support its inclusion; though due to a stronger
response may curve west a bit too much with increased QPF in ME
late Sun into Mon which seems less likely.  The 00z ECMWF is most
central to the ensemble suite and has been fairly consistent run
to run for quite a few cycles providing an anchor to the preferred
blend of the 12z NAM/GFS and 00z ECMWF.

As alluded to, the closed low developing over the Northern Great
Lakes/Ontario/Quebec has quite a few binary interaction with the
exiting/weakening closed low currently in Quebec, which makes
predictability a bit less with increased spread.   The issues with
the CMC being slow and very compact as it digs into the base of
the trof, are also supported by the GFS, though it is slowly
trending weaker compared to the 00z/06z runs.  This remains not
favored in the mass fields being a bit too slow, too far south. 
However, this does not appear to negatively affect the QPF fields,
placement relative to the ECMWF.  The UKMET is much too strong and
fast, with the surface wave fairly east of the otherwise good
cluster across the Great Lakes early Sunday, though comes back
into line with the NAM/ECMWF/ECENS mean/GEFS mean and slightly
slower GFS into Sun/Mon.  Still, think some inclusion is ok in a
blend but will favor a ECMWF-heavy weighting followed by the 12z
NAM with lesser inclusion of the GFS and UKMET to round out the
blend preference.  Confidence is moderate to high in QPF, but the
modest mass field differences only suggest average confidence
overall.

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

Gallina