Model Diagnostic Discussion
NWS Weather Prediction Center College Park MD
1119 AM EST Sun Jan 27 2019

Valid Jan 27/1200 UTC thru Jan 31/0000 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: General model blend
Confidence: Above Average

The models remain in excellent agreement on the synoptic pattern,
and have moved into much better alignment on the mesoscale details
specific to individual waves moving through the amplified flow
pattern. The 12Z GFS and 12Z NAM offer comparable forecasts to the
CMC, ECMWF and UKMET for most of the features across the country.
Although the models do continue to differ on precisely the
location of a specific front or low center, there is no systematic
bias that would require the exclusion of a particular model from
the preference today. Given the strong gradients associated with
the Arctic air mass pushing into the Great Lakes, slight model
disagreements on position and timing of fronts may lead to narrow
zones of more significant model spread. However, a blend seems to
be the prudent approach at this time given a lack of systematic
bias from one or two individual models.

One exception to the model similarity today is along the West
Coast near the end of the discussion period. Models show a
mid-upper level trough closing off into a low near 150W on Monday,
and then approaching the West Coast on Wednesday. The models do
show a decent amount of variability as to the speed of the
approach and (to a lesser extent) the latitude of the approach.
The 00Z ECMWF was on the fast end, which is contrary to its usual
bias. Given the synoptic situation -- the development of a closed
low moving through a ridge axis -- the preference would be toward
slower forecasts, and thus more of a GFS, UKMET and NAM blend is
preferred in these areas at this time.

A final note that the ECMWF QPF seemed too light along the East
Coast relative to the other models, so greater weight was given to
those models for the forecast east of the Appalachian Mountains.


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

Lamers