Model Diagnostic Discussion
NWS Weather Prediction Center College Park MD
209 AM EST Mon Jan 21 2019

Valid Jan 21/0000 UTC thru Jan 24/1200 UTC

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

Final 00z Model Evaluation
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

...East Coast...
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Preference: Any of the operational models
Confidence: Above Average

The deep cyclone over New England as of Sunday evening will
gradually pull away to the north and east. Guidance is in strong
agreement in handling this large scale, well-defined system.


...Pacific System Traversing the Nation Through Wednesday...
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Preference: Blend of the 00Z UKMET / Canadian
Confidence: Slightly Below Average

A full latitude, positively tilted trough will move steadily
across the country, beginning in the western U.S. on Monday. Over
recent runs there has been some systematic bias showing up, with
ECMWF and its ensemble being slower and NCEP models being faster.
At present time the slowness of the ECMWF is more pronounced on
Day 1 into early Day 2, holding cold air back behind a stronger
low level wave in the central Plains states.  The ECMWF then syncs
a little better with the GFS at later times, although the
fast/slow biases still show up at the surface through Day 3 - and
the ECMWF remains considerably slower than the ECMWF ensemble mean
with the low tracking through the Great Lakes. The GFS, on the
other hand, becomes so fast, especially in moving the surface
front toward the East Coast on Day 3, that it starts to move away
from ensemble consensus. The NAM also produced an odd solution
that looks pretty good through Day 2, and then inexplicably
accelerates the low level frontal zone at break-neck speed on Day
3. Sorting out all of these oddities, we are left with the unusual
preferred combination of UKMET and Canadian. These two models
smooth out some of the low confidence shortwave details aloft,
while their surface solutions are better timed with the multi-run
ensemble consensus - and especially with the ECMWF Ensemble mean
which resembled the 24/12Z forecast from the WPC Medium Range desk.


...Pacific Northwest Tue / Wed...
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Preference: Blend of the 00Z operational models, excluding the
Canadian
Confidence: Average

A concentration of low amplitude troughing is expected to move
through a mean ridge position and onto the West Coast Wednesday.
Model spread has been noteworthy for several consecutive model
cycles, with some solutions like the 12Z UKMET being remarkably
fast, and others like the recent ECMWF runs holding onto a more
pronounced vort max at a more southerly latitude around 45 deg
north. The better looking consensus over the long haul of multiple
model cycles has been solutions more like the 00Z NAM/GFS in terms
of forward speed and in keeping most of the energy concentrated
farther north toward Washington and British Columbia. At 00Z the
UKMET and ECMWF became a little more similar to the NAM/GFS, while
the Canadian fell slightly out of favor with respect to the
shortwave details, but not by much. Overall the models do seem to
be converging.

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

Burke