Model Diagnostic Discussion
NWS Weather Prediction Center College Park MD
255 AM EDT Thu Apr 02 2020
Valid Apr 02/0000 UTC thru Apr 05/1200 UTC
...See NOUS42 KWNO (ADMNFD) for the status of the upper air
ingest...
00Z Final Model Evaluation with Preferences and Confidence
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
...Western Atlantic closed low and offshore surface low...
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Preference: General Model Blend
Confidence: Above average
...07Z Update...
The 00Z UKMET/CMC/EC are all in fairly decent agreement with the
rest of the guidance and thus a general model blend should
suffice.
...Previous Discussion...
An elongated trough will move off the Mid-Atlantic/Northeast coast
through the period, eventually consolidating into a strong deeply
stacked ocean cyclone NW of Bermuda. There is a highly amplified
synoptic blocking pattern that will set up over the higher
latitudes resulting in a deep and slow moving 500mb trough over
the Western Atlantic Ocean. A surface low will develop offshore
and retrograde as strong mid level vorticity rounds the stacked
low. The surface low will make its closest approach to land on
Friday. Model differences are fairly minor overall with the 12Z
GFS a bit farther west with its surface low feature as compared to
the rest of the deterministic models. However, the latest 00Z GFS
has come more in line and could be used within the blend. There is
some variability with respect to QPF, but this is largely a result
of a warm frontal placement, low level jet influences and
mid-level impulse propagation.
...Upper Midwest/Great Plains/Mississippi Valley...
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Final Preference: General Model Blend with lesser weight toward
the 00Z NAM/GFS
Previous Preference: 12Z EC/ECENSmean/GEFSmean
Confidence: Average
...07Z Update...
The 00Z EC slowed a bit with the first impulse moving through the
Upper Midwest which is evident by the precipitation trending west.
The 00Z CMC has stayed consistent with the 00Z UKMET advancing a
bit faster. All this to say that a general model blend with
lesser weight put on the fast 00Z NAM/GFS is ideal. The frontal
passage farther south through the Plains and MS Valley is in
fairly decent agreement among the models overall. It is the
convection along the boundary that will promote differences among
the QPF guidance.
...Previous Discussion...
A large scale trough across the western US associated with a gyre
anchored in Alberta will help to eject multiple impulses across
the northern Rockies/Plains and into the Upper Midwest/Great Lakes
region. The first impulse will shift from the northern
Rockies/Plains into the Upper Midwest through early Sunday and the
second impacting the northern Rockies Saturday into Sunday.
Within this section we will discuss the first impulse that will
help to develop a weak surface low in the Upper Midwest region
with a trailing cold front through the Plains edging east into the
MS Valley region Friday through Sunday. The 00Z NAM/GFS is a bit
too fast with both the impulse and thus surface features.
Additionally, the axis of vorticity from the 00Z GFS is more
longitudinal in nature which results in the energy propagating too
quickly by the end of the period. Meanwhile the 12Z UKMET/CMC are
too slow. Therefore, the EC/ECENSmean/GEFSmean seems like a
reasonable middle ground and is the suggested model blend of
choice. There are differences among the models in terms of
mesoscale features that will result in variability with respect to
QPF.
As an aside...another aspect of this broad scale trough is the
increasing Gulf moisture interacting with a front and seasonably
strong westerly flow/southern stream jet which will yield
thunderstorms in the southern Plains. The 00Z NAM QPF intensity
continues to drop off with southward extent. The other models do
display some spatial spread which is a product of the surface
boundary placement and mid-level energy interaction, but all are
more indicative of robust precipitation, so the NAM appears to be
a dry outlier, especially across south Texas. So while the 00Z NAM
should be removed from any blend within this region, the remaining
QPF differences will be improved by higher resolution guidance
that can handle these mesoscale features. For now, the 00Z EC
seems a bit too far south the the higher QPF amounts, thus, in
terms of a model blend for QPF, the 00Z GFS/00Z UKMET/00Z CMC
seems like a logical choice Friday into Saturday. Leaned on the
ensemble means to assist with the frontal progression and
convection Saturday into early Sunday.
...Northern Rockies...
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Preference: Non-00Z GFS Blend
Confidence: Average
...07Z Update...
The 00Z EC/UKMET/CMC changed little with regard to this feature
and thus the same blend of choice remains.
...Previous Discussion...
With respect to the large scale synoptic trough pattern situated
across the western US as discussed above, the second impulse
rounding the trough will swing through the northern Rockies/Plains
Friday through Saturday, lifting north by Sunday. Models are in
fairly good agreement with this feature. However, the 00Z GFS is
slightly farther south and is thus producing more precipitation
south into the ranges of NW WY. Therefore, with regard to this
features moving through, would opt for a non-00Z GFS blend.
...West Coast...
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Final Preference: 00Z EC/NAM/ECENSmean/GEFSmean
Previous Preference: Blend of the 12Z ECMWF/ECENSmean and 00Z
GFS/GEFSmean/NAM
Confidence: Average
...07 Update...
00Z EC made little adjustment to its previous run with both the
00Z UKMET and 00Z CMC illustrating a more broadly defined trough
axis as opposed to a closed low shifting off the west coast.
Though their first impulse ejecting into CA/OR is aligned with the
rest of the guidance. When comparing the 00Z GFS with the first
impulse, it is now too fast and outside the model consensus.
Therefore, removed it from the blend.
...Previous Discussion...
A series of impulses will drop south and east toward the west
coast as energy rounds the aforementioned trough anchored over the
western half of the country. All models depict these feature
well, but there are differences with respect to intensity and
overall placement of these vorticity maxims. The first smaller
impulse will move inland near southern OR/northern CA late
Saturday into early Sunday. A strengthening upper level jet stream
will support the secondary piece of energy dipping south to deepen
into a mid-level closed low by the time is approaches the West
Coast on Sunday. The 12Z UKMET is too slow and weak with this
energy. Meanwhile, the 12Z CMC is too far south, especially with
respect to energy ejecting west into CA. The 00Z NAM seems
reasonable. Thus, this leaves the 12Z EC/00Z GFS/NAM and their
respective means for the model blend of choice. There are subtle
hints the GFS is starting to become too fast with this feature,
which is a known bias among the model. But it is within
tolerance. Therefore, these models are a good middle ground for
now.
Model trends at www.wpc.ncep.noaa.gov/html/model2.shtml
500 mb forecasts at www.wpc.ncep.noaa.gov/h5pref/h5pref.shtml
Pagano