Model Diagnostic Discussion
NWS Weather Prediction Center College Park MD
144 PM EST Mon Dec 10 2018

Valid Dec 10/1200 UTC thru Dec 14/0000 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: Blend of 12Z GFS, 00Z ECMWF, 00Z EC ENS, 12Z UKMET
            Least weight placed on the GFS
Confidence: Average

---18Z UPDATE---
No change to the preliminary preference. The 12Z ECMWF continued
to slow down the low over Texas late in the week, and remains on
the slow end of model spread. The preference is to maintain the
00Z ECMWF, and blend it with the 12Z UKMET and 12Z GFS, which will
provide better clustering. However, the model composition of the
preference remains the same as the preliminary version.

---PREVIOUS DISCUSSION---
The greatest uncertainty in the forecast over the next few days
lies with the sharply digging wave that should result in a closed
low and occluding cyclone over the Arklatex region by Thursday.
Ensemble spread is highest with this wave, particularly as it
relates to timing. An ensemble sensitivity analysis shows around
half the variance can be explained by timing differences.

Earlier runs of the GFS and the NAM, as well as the 00Z CMC showed
a slower occlusion, and thus a faster progression of the trough
axis to the east. However, the 12Z runs of the GFS and NAM have
trended slower. This makes sense conceptually, as the sharply
digging wave through the Rockies is expected to have a powerful
100+ knot upper level jet on the back side. These situations tend
to favor strong closed low development aloft, and rapid
development and occlusion of a surface cyclone. The stronger
closed low aloft would also tend to lead to slower eastward
progression in the broader westerlies.

Therefore, the 00Z UKMET and ECMWF were favored from the 00Z
cycle, but the preference is also to blend in some 12Z GFS given
the recent trend and to counteract some of the typical slow bias
in the ECMWF. This also has the benefit of addressing some other
timing differences, mainly with waves moving through the northern
states in the northern stream flow.


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

Lamers