Model Diagnostic Discussion
NWS Weather Prediction Center College Park MD
107 PM EDT Mon Sep 17 2018

Valid Sep 17/1200 UTC thru Sep 21/0000 UTC

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

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

...Overall preference for the entire CONUS...
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Preference: Blend of 00Z ECMWF, 12Z UKMET
Confidence: Average

Models begin to show some larger differences in their mass fields
by Tuesday Night and Wednesday, which should have the greatest
impact on WPC QPF for the deepening low from the central Plains
into the Upper Midwest Wednesday into Thursday. The 12Z GFS and
00Z CMC were furthest northwest with the position of the surface
low, and the GFS deepens it much faster than the other models.
This is not preferred, as the large-scale flow pattern is still
relatively flat. The models that show more modest deepening seem
more reasonable. The 12Z NAM, however, is much faster and further
east than the deterministic global models, which is also on the
edge of the model spread envelope.

Therefore, the preference is to lean closest to the 00Z ECMWF and
12Z UKMET, which have displayed a greater degree of consistency,
and are also generally closer to the 00Z ensemble means. The
ECMWF/UKMET blend also has the added benefit of being much closer
to ensemble means with the departing trough in the Northeast, and
the base of the broad longwave trough in the West. Therefore, the
preference seems reasonable to apply across the CONUS, although
the other models may be incorporated more outside of the central
Plains and Midwest into the WPC QPF.

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

Lamers