Model Diagnostic Discussion
NWS Weather Prediction Center College Park MD
226 PM EDT Wed Aug 15 2018

Valid Aug 15/1200 UTC thru Aug 19/0000 UTC

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

12Z model evaluation including final preferences
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

...Overall Synoptic Pattern across the CONUS...
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Preference: General model blend
 Exception: Non-CMC in Great Lakes/Northeast after 17/12z
Confidence: Slightly above average

19z update: The 12z ECMWF shifted away from the compact shortwave
allowing for a slightly faster elongation to the positive tilt
trof over New England/Mid-Atlantic by Sat, while a shade further
south it is in much better agreement with the GFS/NAM/UKMET.  The
12z CMC though, trended slower and is now further west with the
trof through Day 3.  Elsewhere, the models remain agreeable to
support a general model blend with the exception of the 12z CMC
over the Great Lakes and Northeast after 17/12z.  Confidence is
above average in blend.


---Prior Discussion---
Models are congruent with nearly all systems crossing the CONUS
for the short range forecast period.  The greatest differences
remain with the merging/binary interaction between the weakening
upper low/opening trof through the Lower Ohio River currently and
the compact wave in the Black Hills.  The differences in the mass
fields comes down to small timing differences, where the GFS is a
bit quicker overall typical of its bias, with the upstream wave
reaching IL by early Friday which leads to the lag of the surface
wave lifting across the Great Lakes compared to the other
guidance, but this is a small variation in the mass fields and
well within the ensemble cluster.  On the other side of the
spectrum, the 00z ECMWF is slower with the initial wave, slowing
the upstream wave leading to the upstream wave remaining a bit
more compact in the mid-levels leading to a delaying of the
surface wave across Ontario into the Northeast, fairly typical
with its bias.  The ECMWF though does lag significantly late Fri
into Sat, even compared to the 00z ECENS mean...suggesting slowly
phasing the 00z ECMWF out any blend by Day 3.   The 12z NAM and
00z UKMET/CMC are a bit more progressive than the ECMWF matching
closer to the GFS by the end of the forecast period as well, but
are nice middle ground solutions.   All in all, a general model
blend should suffice for this system, but hedging away from the
00z ECMWF toward the 00z ECENS mean after 12z Fri is WPC
preference.

Elsewhere, models are in strong agreement with the timing and
general evolution of the wave crossing the Northwest into the
Northern US Rockies and High Plains by the end of the forecast
period.  Further north in central Canada, there are some
timing/mid-level trof orientation differences, especially with the
slow/flat CMC, but overall do not seem to affect/leak into the
CONUS by 19/00z, enough to support a general model blend at above
average confidence for this system.


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

Gallina