Model Diagnostic Discussion
NWS Weather Prediction Center College Park MD
237 PM EDT Fri Mar 27 2020
Valid Mar 27/1200 UTC thru Mar 31/0000 UTC
...See NOUS42 KWNO (ADMNFD) for the status of the upper air
ingest...
12Z Model Evaluation with Final Preferences and Confidence
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Southwest to Great Lakes strong cyclone today to
Monday...Secondary Mid-Atlantic coastal low late Sun/Mon...
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Preference: Non-CMC blend
Confidence: Slightly above average
19z update: Small adjustments of the ECMWF, a tad faster and
southward overall, combined with the faster solution of the 12z
UKMET suggest maybe the NAM is a tad too strong/slow to break down
and the GFS is more within overall tolerance after the breakdown
through the Great Lakes. The only significant outlier at this
point, still remains the slow 00z CMC. As such will favor a
Non-CMC forecast blend at slightly above average confidence. This
may smooth out the solution (less than ideal) but also
encapsulates the run to run variation and uncertainty remaining in
the transfer to the coastal surface low for a solid compromise.
---Prior Discussion---
Models continue to trend together especially through the Plains
into the Great Lakes by late Sunday. The 00z CMC/UKMET show
similar evolution but the timing remains slow relative to the
increased clustering in the ECENS mean/GEFS and the operational
NAM/GFS and ECMWF, with the 12z NAM, 00z ECMWF/ECENS mean showing
the greatest congruence. The NAM does depict a slightly
faster/stronger solution as well as the typical cold bias through
the lowest layers, so this may be a p-type issue,(please refer to
QPFHSD for additional winter weather); but will add the GFS/GEFS
into the blend to account for some slower/southward members of the
suite, including some older continuity. So a 12z GFS/NAM and 00z
ECWMF blend preferred at slightly above average confidence.
By late Sunday into Monday, the deep cyclone over the Great Lakes
has fully matured and is starting to fill/weaken and shift toward
coastal surface cyclogenesis. The typically faster GFS leads the
charge again and may be a bit too fast initially allowing for the
surface cyclone it wrap up closer to the coast/Gulf of Maine than
other guidance members, with the NAM/ECENS mean more centralized
in the overall guidance suite. As such will shift toward a 12z
NAM/00z ECMWF/ECENS preference for the coastal low portion on Day
3. Confidence is average for this portion given model spread and
sensitivity of timing difference in the weakening/filling phase
transferring eastward.
Combining pair of shortwaves along West Coast late Sat/early Sun,
tracking through Southwest Mon...TX/OK Panhandle surface low Mon
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Preference: General model blend
Confidence: Slightly above average
19z update: The 12z UKMET/CMC still slightly lag the mid to upper
level mass pattern compared to the GFS/NAM and ECMWF, but the low
levels/surface flow to have confidence in a general model blend
for mass fields. There is some solid uncertainty to location of
mesoscale boundaries and nose of moisture/instability
intersections across the Southern Plains and therefore QPF is a
bit off, at least less certain in latitude. Thinking the
CMC/UKMET/GEFS position of the QPF maxima looks solid, within the
general model blend.
---Prior Discussion---
Currently, GOES-W WV depicts a compact shortwave west of WA state
that is expected to dig southwest toward the Central CA coast
early Sunday. Upstream a fast wave through the Gulf of AK will
take the prior's place and shear into a sharper/full-latitude wave
along the West Coast. The 00z CMC is the only deterministic
member that is struggling with this evolution. The 12z GFS is a
tad weaker than the remaining guidance but this does not seem to
be a major issue, as the wave combines and progresses into an
average length shortwave trof through the Southwest. The 12z
NAM/GFS eventually show some faster progression, relative to the
ECMWF. The timing of the jet streak undercutting and providing a
broad diffluent pattern over the High Plains late Monday supports
lee cyclogenesis across the TX/OK panhandle by 00z Tuesday. The
00z UKMET however, is slower with the jet exit region but also
stronger with the low level return flow, supporting increased
convective activity and upscale growth further south over the
Permian Basin to Central TX, which may occur but given better
agreement on day 3, further north...will support a blend of the
12z NAM/GFS and 00z ECMWF at slightly above average confidence.
Deep Gulf of AK gyre with first peripheral shortwave/associated
surface triple-point/front entering PacNW Monday...
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Preference: 12z GFS/ECMWF blend
Confidence: Average
19z update: The 00z CMC and UKMET both trended toward a
stronger/broader low in the Gulf of AK, though the CMC, like the
NAM is fast and a tad flatter than the GFS/ECMWF and less favored.
The UKMET seems to be a compromise between the NAM/CMC and the
ECMWF/GFS...but will little large scale difference in the 12z
ECMWF... will continue to favor a 12z GFS/ECMWF blend at average
confidence.
---Prior Discussion---
A favorable large scale set-up for broadening/upscale development
across the Gulf of AK will be occurring through the weekend as a
strong shortwave in broad cyclonic flow allows for a
broader/larger cyclonic gyre to form across the Gulf of AK late
Sunday. Along the periphery of the developing large scale wave,
strong speed vorticity with embedded shortwave will press strong
convergence along the occluded front near an enhancing
triple-point near Vancouver Island toward the Juan de Fuca Strait.
There is large model variance in the timing/strength of the wave
and the gyre itself. The 00z UKMET/CMC are less favorable/less
cyclonic across AK from the polar vortex to remain more flat and
progressive to be more favorable especially given longer term
trends in the ensemble suite toward a broader cyclone evolution.
The 12z NAM suggest this as well, but is quite fast and shows some
typical day 3 over-amplification of the wave compared to the
GFS/ECMWF solutions. The 00z ECMWF is strangely much stronger and
forward compared to the bulk of ECENS members which are closer to
the GFS/GEFS overall. Still the timing/placement of the ECMWF
along with the GFS/GEFS and ECENS mean suggest some confidence for
a blend to be preferred. Confidence in the evolution is solid, but
internal changes, run to run variations are likely to suggest
confidence overall in the blend is average to slightly below
average
Model trends at www.wpc.ncep.noaa.gov/html/model2.shtml
500 mb forecasts at www.wpc.ncep.noaa.gov/h5pref/h5pref.shtml
Gallina