Model Diagnostic Discussion
NWS Weather Prediction Center College Park MD
232 AM EDT Sat Mar 31 2018

Valid Mar 31/0000 UTC thru Apr 03/1200 UTC

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

00Z Model Evaluation...Including Model Preferences and Confidence
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~


...Surface low pushing through the Great Lakes today with trailing
cold front sweeping across the Northeast by Sunday...
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Preference: General model blend
Confidence: Above average

---06Z UPDATE---
The UKMET has moved into better alignment with the consensus of
the remaining models. Therefore a general blend is now preferred.

---PREVIOUS DISCUSSION---
Models are in excellent agreement with their mass fields for this
system, with the most notable difference coming from the 12Z UKMET
(slower than other models). The preference is to lean toward the
larger consensus and go with a non-UKMET blend.


...Low-amplitude shortwave reaching the Oregon coast today...
...Shearing downwind toward the High Plains by Sunday and Mid
Atlantic by Sunday Night...
...Weak surface wave or low developing in the Mid Atlantic...
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Preference: General model blend; greater weight to 00Z ECMWF
Confidence: Slightly above average

---06Z UPDATE---
No change to the preliminary preference.

---PREVIOUS DISCUSSION---
Any differences with this system are very subtle, given the rapid
forward motion of the shortwave and the low amplitude nature of
the wave. Overall, the models show quite a bit of similarity, and
thus a general blend is preferred. The 00Z GFS seems to be one of
the faster models, pushing a weaker surface low out into the
Atlantic slightly faster. The 12Z ECMWF is closer to the 00Z NAM
and 12Z UKMET (the 12Z CMC is faster like the GFS), and may
represent a more likely timing given slightly better model
agreement. Therefore, although the preference is for a general
blend, greater weight to the ECMWF is suggested.


...Upper level trough digging into the Northwest by late weekend...
...Surface cyclogenesis over the Plains Monday into Tuesday...
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Preference: 00Z GFS, 00Z UKMET
Confidence: Average

---06Z UPDATE---
Models continue to show a spread in timing similar to the 12Z
cycle, and thus the preference remains largely the same as before:
a blend of the GFS and UKMET which are slightly faster than the
ECMWF, but not as fast (or strong with the trough) as the NAM.

---PREVIOUS DISCUSSION---
The most substantial model spread continues to be with this
system, although there has been some convergence over the past
several model cycles. The 12Z CMC shows an initially weaker upper
level low dropping down the coast of British Columbia, with higher
heights building into the West from the Pacific. This is not
supported by other models, and has been a recent bias in the
model. Therefore, it is excluded from the preference. The 00Z NAM,
meanwhile, shows a stronger trough and digs it southeast the
fastest. The 12Z ECMWF is slightly slower than the median model
with the 00Z GFS and 12Z UKMET in the middle. It is a usual bias
for the ECMWF to be on the slow end, and thus the preference will
be just a bit faster than that with the GFS/UKMET.


...Lingering trough and stalled surface front in the vicinity of
Florida this weekend...
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Preference: 00Z GFS, 00Z ECMWF
Confidence: Slightly above average

---06Z UPDATE---
No change to the preliminary preference.

---PREVIOUS DISCUSSION---
In the upper levels of the atmosphere, the differences in timing
of the trough as it slowly shifts east are more apparent, with the
12Z CMC and 12Z UKMET showing a less amplified and faster trough
at 250mb. The preference would be for a slower scenario, closer to
the ECMWF ensemble mean which is best represented by the 00Z GFS
and 12Z ECMWF. The 00Z NAM shows a similar timing, but has a more
amplified trough, which is a usual bias in the model.


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

Lamers