Prognostic Meteorological Discussion
Issued by NWS

Current Version | Previous Version | Text Only | Print | Product List | Glossary Off
Versions: 1 2 3 4 5 6

FXUS10 KWNH 161653

Model Diagnostic Discussion
NWS Weather Prediction Center College Park MD
1253 PM EDT Fri Mar 16 2018

Valid Mar 16/1200 UTC thru Mar 20/0000 UTC

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

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

...Deep Cyclone over the Northeast U.S. and Eastern Canada...
...Digging Shortwaves Wrapping Through South Side of Cyclone...
...Gradual Shift to the E/NE this Weekend and Early Next Week...
Preference: 06Z GEFS Mean, 00Z EC ENS Mean
Confidence: Average

As the broad cyclone continues to gradually shift east and
northeast into the Canadian Maritimes, the Northeast U.S. should
largely be in a northwesterly flow regime with a surface ridge
building in by Monday. Models are in good agreement with this
general progression, but show quite a bit more variability to the
northeast closer to the center of the cyclone. A general,
broad-based model blend would probably work best in the Northeast.
Further to the northeast, models are struggling with how to handle
the interactions between several waves that will be digging
southeast and wrapping into the overall circulation. The overall
preference for the system, therefore, is to rely on the GEFS and
ECMWF ensemble means, which will best account for the variety of
detail differences. The 12Z GFS and 00Z ECMWF are probably the
closest operational model approximations of these means.

...Strong Plains Cyclone Weakening as it Shifts East into the Ohio
Valley and Carolinas this Weekend...
...Upper Level Shortwave Shearing Out and Deamplifying...
Preference: General Model Blend
Confidence: Above Average

Over the past several model cycles, models have moved into much
better agreement with this system and now show tight clustering in
many of the mass fields. Most of the surface low positions by
18/00Z are within 150-200km of each other in the Carolinas. Aloft,
the 12Z NAM is still the fastest model and shows the strongest
wave as it is being sheared out, but has trended substantially
toward the global models. The NAM is close enough that it can`t be
discounted as a reasonable possibility, but should represent the
stronger end of depictions of the wave aloft.

...Large Closed Low over the Northwest U.S....
...Breaking Down into Peripheral Shortwaves on Sunday that Slowly
Meander in a Lingering Shear Axis on Monday...
Preference: General Blend (0-48h); GEFS/EC ENS Means (48-84h)
Confidence: Above Average early, trending to Below Average late

Through around 18/12Z (48 hour forecast), the models are in quite
good agreement overall -- even with the beginning stages of the
breakdown of the upper level closed low. The global models and NAM
all have similar positions and amplitudes of peripheral waves
until that time, so the preference through Saturday Night is for a
general model blend with relatively high confidence. After that
point, the peripheral waves get caught in a residual shear axis as
an upper level ridge rapidly builds into the Gulf of Alaska and
British Columbia over the top of a strong closed low around
37N/145W in the Pacific, and shortwave ridging also builds into
California and Nevada behind a departing trough (discussed below).
The result will be relatively small-scale waves slowly meandering
in weak flow and confidence in the forecast specifics is lower.
Beyond 48 hours, the preference is to trend toward a blend of the
GEFS Mean and ECMWF Ensemble Mean. As with the cyclone over
Eastern Canada, the GFS and ECMWF are probably the operational
models that most closely approximate these ensemble means, but the
ensembles will be better able to capture the variety of

...Stronger Peripheral Wave on the Western Periphery of the Closed
Low moves into California on Saturday...
...Wave Ejects into the Plains on Sunday with Lee Cyclogenesis...
...Eastward Progress into the Mid-Mississippi Valley on Monday...
Preference: 12Z GFS, 00Z ECMWF, 00Z CMC
Confidence: Average

As this strong wave kicks through the Southwest, all the models
show strong lee cyclogenesis and all models also show a closed low
forming aloft over the Plains (including the ensemble means). In
general, with this sort of process, the preference would be to
lean on the slower end of guidance as systems with a closed low
aloft tend to progress more slowly. The 12Z NAM and 00Z UKMET tend
to spread height falls into the Plains the fastest, and this may
be due to them indicating considerably more (around 1 degree
latitude) amplification to the trough as it reaches the lower
Colorado River Valley early Sunday, as compared to other global
models and ensemble means (which are tightly clustered). As the
low closes off aloft by early Monday, the 12Z GFS is about 150km
north of all other models, and appears to be even north of the
GEFS mean. The GFS is, however, also on the slower end of the
model spread. Therefore, the preference is to include the GFS to
represent the potential for a slower progression to the wave, but
the greater weight will be placed on the 00Z ECMWF and 00Z CMC
which may have the low position at a more likely latitude.

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.