Model Diagnostic Discussion NWS Weather Prediction Center College Park MD 1112 AM EST Sat Feb 17 2018 Valid Feb 17/1200 UTC through Feb 21/0000 UTC ...See NOUS42 KWNO (ADMNFD) for the status of the upper air ingest... 12Z GFS/NAM evaluation with preliminary model preferences ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ NAM/GFS initialization errors do not seem to degrade their short range forecasts. Shortwave moving offshore the Northwest in three days ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Preference: Model blend; confidence average There is a bit of dispersion in possible low pressure placement and depth associated with a southward moving shortwave moving down the western side of the upper level trough. Over the last several days, the guidance has been slowly dampening this feature. Currently the 12z GFS is a deep outlier and the multi-day trend suggests something weaker both aloft and at the surface. The 00z global ensemble guidance has no defined surface low to track in this region, implying that a weaker solution would be best. Blending the 12z GFS with the 12z NAM, 00z ECMWF, 00z UKMET, and 00z Canadian will accomplish this task. Confidence in this choice is no better than average. Upper trough digging into the West Pressure pattern in the Plains/Midwest/Great Lakes ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Preference: Non UKMET-blend; confidence average Because the 00z UKMET has a slightly sharper trough across the northern High Plains and West, its surface low position in MI late Tuesday is more southwesterly than the remainder of the deterministic guidance. While there is broad spread in the possible low position, the best clustering lies near the 00z Canadian, 00z ECMWF, 12z GFS, and 12z NAM. A blend of these pieces of guidance is preferred with average confidence. Low moving from the Southeast offshore New England and its associated shortwave aloft ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Preference: Model blend; confidence slightly above average While the 12z NAM is stronger aloft and a little more northwesterly with the surface low tonight, this type of solution has randomly seen over the last several days on various pieces of guidance, lies within the ensemble clustering, and cannot be fully ruled out. Because of this, a compromise of the 12z GFS, 12z NAM, 00z UKMET, 00z Canadian, and 00z ECMWF is preferred with this system with slightly above average confidence. Model trends at www.wpc.ncep.noaa.gov/html/model2.shtml 500 mb forecasts at www.wpc.ncep.noaa.gov/h5pref/h5pref.shtml Roth