Model Diagnostic Discussion
NWS Weather Prediction Center College Park MD
1141 PM EST Fri Jan 11 2019

Valid Jan 12/0000 UTC thru Jan 15/1200 UTC

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

00Z Model Evaluation...with Preferences and Confidence
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

...Overall Pattern Across the CONUS...

Preference: 12Z ECMWF, EC ENS, UKMET; 18Z GEFS Mean
Confidence: Average

The greatest model spread and resulting QPF differences over the
next several days are along the West Coast near California. Models
have slight variations with how they handle the rex block over the
interior West, but more with the eventual position of the closed
low dropping southeast toward California. The 12Z CMC is situated
on the south end of model spread by Monday morning, further south
than all but one GEFS and one CMC Ensemble members. Given a
significant lack of ensemble support, the CMC was excluded from
the preference.

Meanwhile, the 00Z GFS and 00Z NAM show the position of the
primary surface low by Tuesday morning (15.12Z) well south of the
12Z UKMET and 12Z ECMWF by a few degrees of latitude. The models
have been somewhat erratic without a clear run-to-run trend. An
examination of the ensemble members shows a significant amount of
spread that has failed to shrink over the past several model
cycles. The feeling is that the ECMWF may be a bit far north with
its surface low position (given a significant number of EC ENS
members south of that position), however, the GFS and NAM keep a
considerable portion of the QPF offshore, which does not seem to
be the scenario most supported by ensembles. For this reason, the
preference leans in the direction of the GEFS and EC ENS means for
the West Coast system.

The 12Z ECMWF and UKMET are also included in the preference as
they seem reasonable elsewhere in the country and are fairly
consistent with the developing cyclone along the East Coast. The
00Z GFS seems too suppressed, and has a surface low offshore that
is further south than the other models initially, and eventually
lags well behind the other models by Monday Night (15.06Z). The
GEFS Mean is much closer to the broader model consensus, so the
GFS is excluded from the preference at this time too.


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

Lamers