Model Diagnostic Discussion
NWS Weather Prediction Center College Park MD
1150 AM EST Wed Dec 12 2018

Valid Dec 12/1200 UTC thru Dec 16/0000 UTC

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

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

...Overall Pattern Across the CONUS...

Preference: 00Z ECMWF ENS, 00Z ECMWF, 12Z NAM
            Greatest weight on ECMWF Ensemble Mean
            Least weight on 12Z NAM (especially after 14/00Z)
Confidence: Slightly below average

Model consensus is strong on an amplification of the pattern over
the CONUS in the next 36-48 hours, with ridges in the West and
East, and a pronounced trough in the Plains. In general, models
are showing fairly good similarity in this same time frame, but
model spread becomes quite large in the subsequent one to two
days. The majority of model spread is related to continued timing
issues with the larger closed low that develops over the Southern
Plains, as well as a combination of timing and amplification
differences related to a trough that will be arriving in the
Pacific Northwest Friday.

Ensemble sensitivity analysis for the southern low show that the
majority of variance is related to timing issues, and that this is
sensitive to the strength of the digging wave, as well as the
amplification of a secondary wave in the northern stream near the
US-Canada border. A more amplified wave in the northern Plains
tends to lead to more interaction with the closed low, and a
faster ejection. A stronger digging shortwave into the southern
Plains also tends to lead to a stronger closed low that develops
further west and is slower to move out.

The preference is generally to continue to lean toward the slower
end of the model spectrum, toward the ECMWF and ECMWF Ensemble. It
has been consistent, and other models have generally trended
closer to it in the late Thursday to early Friday time frame.
Spread increases after that, but given such a strong closed low,
the preference would be toward a slower ejection of the low given
its separation from the northern stream flow.

The ECMWF ensemble mean is slightly faster than the deterministic
ECMWF, and thus more weight was placed on the ECMWF ensemble mean
as a compromise with some of the faster models. The 12Z NAM also
seems reasonable with timing and placement through Thursday
afternoon, but then begins to eject the low considerably further
north than the other deterministic models and most ensemble
members. Therefore, progressively less weight was placed on the
NAM, beginning Thursday afternoon.

A blend of the ECMWF, EC ENS, and NAM also provides a reasonable
forecast over the rest of the CONUS and allows for a compromise
with the timing spread for multiple waves, with a lean toward the
slower end of progression overall.


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

Lamers