Model Diagnostic Discussion NWS Weather Prediction Center College Park MD 257 AM EDT Wed Mar 20 2019 Valid Mar 20/0000 UTC thru Mar 23/1200 UTC ...See NOUS42 KWNO (ADMNFD) for the status of the upper air ingest... 00Z Model Evaluation Including Preferences and Confidence ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ...Trough and Developing Low in the East... ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Preference: Blend of 00Z GFS and 12Z ECMWF Confidence: Average Considerable model spread continues with the evolution of a strong low developing in the Northeast late this week. Complexity is added by two phasing events: (1) on Thursday with a southern stream shortwave and another wave digging sharply through the Great Lakes, and (2) on Friday when more substantial cyclogenesis will begin over the Northeast. At that point the initial shortwave will be ejecting in a negatively tilted fashion up the coast and will phase with yet another sharply digging shortwave from the Great Lakes. Subtle detail differences in this sequence of events do lead to some important variation in the strength and timing of the low. In general, the NCEP models (00Z NAM and GFS, 18Z GEFS) are on the faster end of the spread; by 12Z Friday the GFS has the surface low further north than over 90 percent of ECMWF ensemble members. The 12Z ECMWF is actually faster than the associated ECMWF Ensemble Mean, 12Z CMC and 12Z UKMET, but all of these models are on the slower end of the spread. For now, the preference was to opt for a blend in timing between the GFS and ECMWF, which are in the middle of the deterministic model spread and should provide something close to a median forecast in terms of timing. However, given the considerable number of ECMWF ensemble members slower than even the deterministic ECMWF, it should be noted that the timing could slow down further than the official forecast. ...Remainder of the CONUS... ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Preference: General Model Blend Confidence: Above Average Elsewhere across the CONUS, models are generally in good agreement, with relatively low model spread. The preference in these regions is for a general model blend. The biggest differences appear to be with the amplitude of the trough on Friday as it begins to kick out of the West. This leads to larger height spreads aloft over Arizona, New Mexico and Texas. The NCEP models generally show more amplification, which leads to higher QPF. Model trends at www.wpc.ncep.noaa.gov/html/model2.shtml 500 mb forecasts at www.wpc.ncep.noaa.gov/h5pref/h5pref.shtml Lamers