Model Diagnostic Discussion NWS Weather Prediction Center College Park MD 232 PM EDT Mon Apr 23 2018 Valid Apr 23/1200 UTC thru Apr 27/0000 UTC ...See NOUS42 KWNO (ADMNFD) for the status of the upper air ingest... 12Z model evaluation including final preferences ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ...Closed low over the lower OH valley today, gradually opening into a trough and lifting into the Mid Atlantic and Northeast, phasing with northern-stream shortwave digging through the Great Lakes Tuesday Night into the Northeast... ...Weak surface low over the interior Southeast, redeveloping along the Atlantic coast by mid-week... ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Preference: 50/50% 12Z GFS / 12Z ECMWF blend Confidence: Below average Ensemble spaghetti height trends show that the primary trough axis reaching the Mid-Atlantic region Wednesday morning was reasonably consistent over the past 4 cycles (valid 12Z/25) but the latest cycle adjusted slower/south. This may be a temporary shift, but currently, the 12Z NAM/GFS are faster with the progression (NAM fastest), while the 12Z ECMWF/UKMET/CMC are slower. The GEFS/ECMWF ensemble means are between the faster and slower deterministic runs, but with a weight toward the faster 12Z GFS/NAM. The trends in the 12Z ECMWF/CMC have been faster and the 12Z UKMET has adjusted away from the better ensemble/deterministic clustering with an upstream wave valid Wednesday, shifting south/deeper with the core of the 500 mb low. This timing difference is reflected in the ensemble scatter low plots as well and is greatly influenced due to timing differences with an upstream shortwave crossing the Great Lakes Wednesday into Thursday (referenced at the end of the last paragraph above) with similar timing differences as the initial wave along the East Coast (NAM/GFS faster, ECMWF/UKMET/CMC slower). Given the shift in the ensembles, previous consistency and the latest ensemble means pulling toward the middle of the spread, the preference is to be between the 12Z GFS and 12Z ECMWF, with equal weighting given recent 12Z deterministic trends. Confidence is below average given greater than average spread and complex interaction of multiple shortwaves. ...Shortwave entering northern Great Basin, amplifying into the northern Rockies through Monday Night, and then digging through the Central Plains Wed/Lower MS Valley by Thurs... ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Preference: 12Z GFS, 12Z ECMWF, 12Z CMC blend Confidence: Average Timing differences here are such that the faster and slower ends of the spectrum are not preferred with weighting in the middle best represented by a 12Z GFS...12Z ECMWF...12Z CMC blend. The 12Z UKMET remains significantly faster/flatter with the wave, with the 12Z NAM slowest. A blend of the 12Z GFS, 12Z ECMWF and 12Z CMC is toward the middle of the spread and considered best at this time, with no significant changes noted with the 12Z ECMWF/UKMET/CMC compared to their previous 00Z runs. ...Northern stream portion of Gulf of Alaska trough breaking off and digging through the Canadian Rockies Wed, continuing into the northern Plains Thursday, with associated surface cold front pushing southeast... ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Preference: non 12Z NAM blend Confidence: Average Ensemble means support the somewhat slower timing of the 12Z GFS/ECMWF/UKMET/CMC with this shortwave while the 12Z NAM is faster as the trough axis reaches the Plains. Given decent agreement in the ensemble means and with the associated cold front, a middle of the road of the ensemble spaghetti plots approach is favored at this time, best represented by a non 12Z NAM blend. ...Southern stream portion of Gulf of Alaska trough breaking off and developing into a closed low around 35N/135W by Tuesday... ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Preference: General model blend Confidence: Average The 12Z ECMWF remains slower/west with the closed low off of the West Coast of the U.S., with the 12Z GFS faster/east. This is an often seen bias between these two models, except that the ECMWF is not supported by the 12Z UKMET but the 12Z CMC adjusted to be closer to the ECMWF compared to the 00Z CMC. The ensemble spaghetti plots support a middle of the road approach, and given overall differences are not too large through 00Z/27, a general model blend is recommended at this time. Model trends at www.wpc.ncep.noaa.gov/html/model2.shtml 500 mb forecasts at www.wpc.ncep.noaa.gov/h5pref/h5pref.shtml Otto