Model Diagnostic Discussion
NWS Weather Prediction Center College Park MD
350 AM EDT Thu Sep 13 2018

Valid Sep 13/0000 UTC thru Sep 16/1200 UTC

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

00Z model evaluation including final preferences
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

...Overall synoptic pattern across the CONUS...
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Preference: 00Z GFS / 12Z ECMWF blend
Confidence: Slightly above average

07Z update: The 00Z ECMWF sped up its timing but the 00Z UKMET
slowed down with a shortwave into northern CA on Saturday.
Relatively minor adjustments were made with the Pacific Northwest
closed low for Saturday. Wanting to maintain the blend provided as
the preliminary preference to capture the middle ground of
solutions, the 00Z GFS/12Z ECMWF blend is preserved for the final
preference.

Longwave troughing will stay in place across the west coast of the
U.S. through Sunday morning while ridging remains roughly from the
southern High Plains into the Great Lakes. Hurricane Florence will
track into the southeastern U.S. coast late Thursday night while a
surface low/trough is expected to track toward the southern Texas
coast through Friday.

The tropical systems are discussed separately below, while the
western CONUS longwave and embedded shortwaves will be discussed
here. Generally, the 00Z GFS was faster with the individual
shortwaves, while the 12Z ECMWF was a bit slower, with these two
models roughly representing the two extremes in model timing. The
12Z ensemble spaghetti charts showed similarly with the GEFS
faster and ECMWF members slower. There is reason to believe the
middle of the two is best at this time given divergent trends
observed in the GEFS/ECMWF timing over the past 4 12/00Z cycles.
The longwave pattern from the Pacific to the Atlantic shows
meridional flow which may favor the slower solutions, however, at
this point will prefer the middle ground.


...western Gulf coast trough/low nearing south Texas Friday
morning...
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Preference: 00Z GFS/00Z ECMWF/00Z UKMET blend
Confidence: Average

07Z update: No significant changes were noted between the 00Z and
12Z ECMWF/UKMET/CMC, so the GFS/ECMWF/UKMET blend is maintained
for the final preference, just replacing the 12Z with 00Z.

The 00Z Tropical Weather Outlook from NHC shows this system has a
70 percent chance of becoming a tropical depression before
reaching the coast. The latest deterministic models are in fair
agreement, but the 00Z NAM is stronger and a bit slower with the
track, while the 12Z CMC is faster than the model consensus. The
remaining 00Z GFS, 12Z UKMET and 12Z ECMWF are in fair agreement
with a blend of these 3 models preferred given stronger support.

Please refer to the latest NHC discussions and products for more
detailed information on this system.

...Hurricane Florence...
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Preference:  00Z NAM/12Z UKMET closest to 03Z NHC advisory

07Z update: Relative to their 12Z runs, the 00Z ECMWF is similar,
the 00Z UKMET sped up and the 00Z CMC adjusted south. Given the
changes in the UKMET, the 00Z NAM was added as a second choice for
the track while maintaining the 12Z UKMET for the preference.

The 03Z NHC advisory for Florence matches well with the model
consensus until Friday (after landfall) when some model spread
begins to develop. The 12Z CMC appeared to be the farthest from
the NHC track, while the 12Z UKMET was closest. The remaining 00Z
NAM, 00Z GFS and 12Z ECMWF diverge with timing and are displaced
north or south of the NHC forecast.

Please refer to the latest NHC discussions and products for more
detailed information on Hurricane Florence.

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

Otto