Model Diagnostic Discussion
NWS Weather Prediction Center College Park MD
1245 PM EDT Wed Mar 21 2018

Valid Mar 21/1200 UTC thru Mar 25/0000 UTC

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

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

...Nor'easter Along the East Coast Today...
...Occluded Surface Low Reaching Newfoundland By Friday...
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Preference: 12Z GFS and 00Z ECMWF
Confidence: Above Average

Model mass field agreement is now quite good over the next 24-36
hours, and a general model blend could probably be used as the low
passes by the Northeast. However, timing differences do emerge as
the low continues to push northeastward, and the preference for
this system is based on that. The 00Z UKMET and CMC tend to
accelerate the surface low fastest by Friday and Friday Night,
while the 12Z NAM is the slowest. The models that are closest to
the multi-model ensemble mean are the 12Z GFS and 00Z ECMWF. Most
models have a closed height contour at 500mb into Friday morning,
suggesting a deep, occluded cyclone that would tend to resist
significant acceleration. There is also a lack of a significant
upstream wave in Canada or strong flow wrapping in that would tend
to shear out the circulation and move it along faster. The result
is that a low track that is slightly slower than the ensemble
means is preferred. This would be the ECMWF and GFS.


...Broad, Low-Amplitude trough in the Northern Plains Digging
through the Great Lakes Thursday and Mid Atlantic Friday...
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Preference: Non-UKMET Blend
Confidence: Slightly Above Average

The global models and the 12Z NAM all show similarity with their
mass fields for this system, with overall good agreement on trough
amplitude and timing. With time (by Friday) the UKMET increasingly
diverges, slowing down the trough more than other global models
and ensembles, and by 24/12Z the trough axis is about 400 km west
of the consensus of remaining models. Given that, the preference
is to lean toward the significant cluster of models and exclude
the UKMET.


...Deep Trough / Closed Low offshore the Northwest thru Friday...
...Opening into a Trough and Moving into the Northwest Saturday...
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Preference: Non-CMC Blend
Confidence: Slightly Above Average

The 00Z CMC shows some more substantial differences with the other
models regarding this system. It tends to move the closed low
further to the south initially, and then has the trough axis
reaching the Pacific Northwest coast ahead of the other models.
All the other models move the closed low more slowly, which makes
sense conceptually, and then kick a shortwave trough into the
Northwest ahead of the low on Friday evening. The CMC makes an
attempt to do this as well, but is much more suppressed with the
shortwave, and eventually has shortwave ridging where the other
models place the shortwave trough. Otherwise, models are in
relatively good agreement. They do have some detail differences,
but mostly have a similar timing of the closed low progression,
amplitude of the resulting trough, and arrival of that trough in
the Northwest. Therefore, a non-CMC blend is the preference.


...Ongoing Atmospheric River Event Focused into California...
...Related Shortwaves Ejecting into the West Thursday and Friday...
...Lee Cyclogenesis on the Plains Consolidating into a Surface Low
near Kansas by Friday Evening...
...Moving East into the Ohio Valley on Saturday...
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Preference: General Model Blend through Thursday Night
            Blend of 12Z GFS, 06Z GEFS Mean Afterward
Confidence: Average

Models show reasonable similarity in their mass fields through
Thursday Night, and given the limited differences in the synoptic
features a general model blend is preferred through that time.
Afterward, most of the variance appears to be related to timing of
the shortwave as it moves through the West and out into the
Plains, and the associated surface low that will be developing in
the lee of the Rockies. the 12Z NAM shows the fastest progression
overall, while the 00Z CMC, UKMET, and ECMWF are all slower. The
12Z GFS seems to be closest to the ensemble means of the GEFS and
ECMWF, and provides a reasonable intermediate scenario. Thus, the
preference is increasingly weighted to the 12Z GFS as the forecast
progresses.


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

Lamers