Model Diagnostic Discussion
NWS Weather Prediction Center College Park MD
112 AM EDT Wed Apr 24 2019

Valid Apr 24/0000 UTC thru Apr 27/1200 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: 00Z GFS, 12Z ECMWF (greater weight ECMWF)
Confidence: Average

Models demonstrated some significant differences after 26.00Z
(Thursday evening), with the main points of contention the
strength of the southern stream wave ejecting toward the Mid
Atlantic coast (and associated East Coast cyclogenesis), and
details with a Northern Plains shortwave and low. The preference
is generally to lean toward a compromise of the ECMWF and GFS
given some important differences in the other models, but to place
a greater weight on the 12Z ECMWF.

Variation with the East Coast cyclogenesis seems to stem from the
strength of the southern stream wave and the phasing with a
northern stream wave digging through the Great Lakes. The 00Z GFS
and NAM show more substantial phasing and thus a stronger cyclone
off the Northeast by Friday Night. The 12Z CMC, as it showed
elsewhere across the CONUS, seems to have much higher heights
through the base of the trough than the other models and this may
affect the forecast through the Southeast. The 00Z UKMET and 12Z
ECMWF are much closer to the 18Z GEFS Mean and 12Z ECMWF ENS Mean.
Therefore, for this system, the GEFS may be an appropriate
substitute for the deterministic GFS, with greater weight still
applied to the ECMWF.

For the Plains system, the GFS and ECMWF are reasonably similar
and close to ensemble means. The 00Z NAM appears too cold and is
displaced slightly south with its surface low track. The CMC and
UKMET are further south with the surface low in the Plains -- the
UKMET due to a more southward trajectory of the shortwave, and the
CMC due to a weaker low (upper level heights higher than the other
models).

Overall, a compromise between the GFS and ECMWF should lead to a
forecast that has fairly good support from the ensembles, with the
GEFS a potential substitute for the 00Z GFS along the East Coast
by Day 3.

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

Lamers