Model Diagnostic Discussion
NWS Weather Prediction Center College Park MD
205 PM EST Sat Jan 26 2019

Valid Jan 26/1200 UTC thru Jan 30/0000 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: Multi-model blend
            (1) Great Lakes: Lean 12Z GFS, 12Z NAM, 00Z CMC
            (2) Gulf Coast: Lean 00Z ECMWF, UKMET, CMC
Confidence: Average

---18Z UPDATE---
No change to the preliminary preference as model clusters seem
relatively similar on the 12Z model cycle.

---PREVIOUS DISCUSSION---
Similar to the past several model cycles, model agreement on the
synoptic pattern remains high. There is limited model spread with
the large-scale height pattern, and there is a high degree of
confidence in the position and scope of the broad, amplified
trough over the central and eastern US.

Several important mesoscale differences remain, however, and these
are focused in two primary areas: (1) with the surface low
tracking into the Great Lakes, and (2) timing of features near the
base of the trough.

For the clipper low tracking into the Great Lakes, model spread
has continued to narrow with respect to the position of the
cyclone. Ensemble sensitivity analysis continues to show that the
majority of the spread is related to the latitudinal position of
the low as it tracks across the region. The 00Z ECMWF and its
ensemble members represent the southern end of the distribution,
and the other deterministic models are situated further north
(joined by most NAEFS members). The preference is to lean toward a
compromise, but with slightly greater weight to the non-ECMWF
cluster. The difference between the 00Z ECMWF and 12Z NAM
(extremes of the deterministic distribution) with surface low
position at 28.12Z is only 150 km. The NAM, GFS, CMC and GEFS are
strongly clustered, and the preference is to lean to an expected
surface low position only slightly south of that consensus given
the multi-model nature of the agreement. The ECMWF has shown a
slight southward bias with some surface low tracks this winter,
but still cannot be ruled out in this scenario.

Further south, there are noteworthy timing differences with the
wave kicking through the Gulf on Sunday, as well as with the cold
front trailing off the clipper low from the Great Lakes. The 12Z
GFS and 06Z GEFS represent the fast end of model spread, and the
preference is to trend slower than that toward the remaining
multi-model consensus.


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

Lamers