Model Diagnostic Discussion NWS Weather Prediction Center College Park MD 236 AM EDT Fri Sep 07 2018 Valid Sep 07/0000 UTC thru Sep 10/1200 UTC ...See NOUS42 KWNO (ADMNFD) for the status of the upper air ingest... 00Z model evaluation...with final preferences and confidence ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ...Tropical Depression Gordon... ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Preference: General model blend with slightly less weight on the UKMET Confidence: Slightly Above Average 07Z Update: The CMC has sped up while the ECMWF has slowed such that clustering of the full suite (including 00Z GEFS mean) is very tight through 09/12Z/Sunday morning. Thereafter, the latest UKMET deepens the surface low too quickly and slows it down to become a western outlier with a significant displacement in the max QPF axis. For this reason it gets smaller weighting in the preferred blend after 60 hours. --- Prior Discussion --- 00Z NAM has slowed with its progression of the surface low associated with Gordon, while the 00Z GFS sped up from its previous solution, clustering very well with each other. Overall, this shrinks the total spread and these now lie in the middle of the guidance envelope which is well in line with the 18Z GEFS mean. The CMC remains a slow outlier with Gordon which is most evident as a deeper and further NW 700mb low/trough axis. Despite these differences, through 00Z/9th models are in very good agreement with the placement of the surface low and it is really only into day 3/Sunday when significant spread develops. A non-CMC blend is then preferred due to overall clustering of the guidance and little discrepancy in max QPF placement. ...Remainder of CONUS... ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Preference: Non-NAM blend Confidence: Slightly above average 07Z Update: 00Z suite is in reasonable agreement and some of the earlier differences have gone away. NAM is now the fast outlier with the shortwave ejecting into western Canada Saturday night into Sunday so the UKMET can be incorporated back into the blend, while the CMC has lost its tropical feature near FL by Sunday to come into line with the other guidance. For this reason a non-NAM blend is acceptable for the rest of the CONUS. --- Prior Discussion --- Elsewhere, the NAM/UKMET appear to hedge towards their known bias of fast/strong evolution with a shortwave ejecting through the Pacific NW causing subtly faster and drier progression of a cold front across the northern tier states late Saturday into Sunday. A low pressure noted near FL by the CMC on Sunday is a strong outlier from the rest of the global guidance. For these reasons a GFS/ECMWF is preferred across the rest of the CONUS. Model trends at www.wpc.ncep.noaa.gov/html/model2.shtml 500 mb forecasts at www.wpc.ncep.noaa.gov/h5pref/h5pref.shtml Weiss