Prognostic Meteorological Discussion
Issued by NWS

Current Version | Previous Version | Graphics & Text | Print | Product List | Glossary On
Versions: 1 2 3 4 5 6
FXUS10 KWNH 231650

Model Diagnostic Discussion
NWS Weather Prediction Center College Park MD
1249 PM EDT Fri Mar 23 2018

Valid Mar 23/1200 UTC thru Mar 27/0000 UTC

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

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

...Deep trough; closed low offshore the Northwest today...
...Opening into a series of shortwaves moving into the
Intermountain West this weekend...
...Shortwave currently near the Aleutians digging down the West
Coast Sunday and leading to trough amplification and renewed
Plains lee cyclogenesis by Monday...
Preference: 00Z ECMWF, 00Z UKMET
Confidence: Slightly Below Average

Given the complexity of how the closed low near the Northwest
opens up into a trough, with several embedded shortwaves in the
larger circulation separating from the original circulation,
confidence is slightly below average in the detail and specifics
with the Western trough. Additionally, the shortwave that will
play a big role in the further amplification from Sunday into
Monday is currently over the Aleutians.

For this reason, have used ensemble cluster analysis to evaluate
the more likely trough configuration by Monday evening (27/00Z).
The 12Z GFS shows a much broader trough than other models, with
height falls spreading further out into the Plains than the other
global models, which show the base of the trough lagging over the
Desert Southwest. With a sharply digging shortwave into the base
of the trough, the latter option seems to be more likely from a
conceptual standpoint and this is also the configuration that is
most popular among the GEFS and ECMWF ensemble clusters. The 12Z
NAM has a narrower and sharper trough like the non-GFS global
models, however its trough axis is displaced further to the east
so it too spreads lower heights into the Plains.

Therefore, the 00Z ECMWF, UKMET and CMC seem to have more robust
support from available ensemble clusters. The 00Z CMC does show a
high height bias in the first 24-36 hours over the east Pacific
relative to all other models, and it is systematic from low
latitudes to around 50-55N. The preference is thus for a blend of
the 00Z ECMWF and UKMET.

...Shortwave traversing the Four Corners today and reaching the
Plains by tonight with surface cyclogenesis...
...Becoming sheared out as it pushes through the Ohio Valley and
Carolinas with surface low transferring to the coastal Carolinas...
Preference: Non-CMC Blend
Confidence: Slightly Above Average

Models have moved into much better agreement with this system, and
show fairly tight clustering of surface low and frontal positions
as well as height fields aloft. There are some small-scale
differences, but these are not systematic in a way that would rule
out any one model. The exception may be the 00Z CMC which
deamplifies the shortwave much quicker (6-12 hours) and as a
result begins to weaken and lag the surface low more behind the
other models. The preference is therefore for a non-CMC blend.

...Arctic shortwave digging sharply south through Quebec on
Saturday with closed low developing off the Northeast by Sunday...
...Associated strong surface high settling into New England...
Preference: 12Z GFS, 12Z NAM, 00Z ECMWF
Confidence: Average

The 00Z UKMET shows a much stronger shortwave digging south out of
Quebec relative to the other models, and as a result ends up with
a stronger upper level low over the Atlantic as it closes off on
Sunday. This was the most notable difference among the models for
this system, and most of the precipitation impacts should be
offshore. The 00Z CMC also showed a high height bias over the
Atlantic in the base of the trough, and that may affect its
forecast somewhat.

Model trends at
500 mb forecasts at


$$ is the U.S. government's official web portal to all federal, state and local government web resources and services.