Model Diagnostic Discussion NWS Weather Prediction Center College Park MD 302 AM EDT Sat Mar 28 2020 Valid Mar 28/0000 UTC thru Mar 31/1200 UTC ...See NOUS42 KWNO (ADMNFD) for the status of the upper air ingest... 00Z Model Evaluation with Final Preferences and Confidence ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Southwest to Great Lakes strong cyclone through Monday...Secondary Mid-Atlantic coastal low late Sun/Mon... ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Preference: General Model Blend Confidence: Above average ...07Z Update... The latest 00Z EC looks very similar to its previous run and is within the ensemble cluster. Also, the 00Z CMC and UKMET have speed up and are in better agreement with the overall model suite. No change to the preferences. ...Previous Discussion... Models continue to trend closer together especially from the Plains into the Great Lakes by late Sunday. While the 00z CMC/UKMET showed similar evolution with timing remaining slow relative to the increased clustering in the ECENS mean/GEFS and the operational NAM/GFS and ECMWF, the latest 12Z CMC/UKMET have come into better agreement. The track of the surface low/mid-level closed low is very well aligned among the guidance, though the 00Z NAM and now the 12Z UKMET seem to be a tad too strong which may have minor implications on the QPF and thermal profiles. Speaking of thermal profiles, while the 00Z GFS is very much within the model/ensemble cluster, the thermal structure seems to be at odds with the strength of the surface low. Regardless, these differences are quite minimal and are within the ensemble spread. Thus, our model preference of choice for this system is a general model blend. Shortwave tracking across the Southwest through Mon...TX/OK Panhandle surface low Mon ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Preference: Non-CMC Blend Confidence: Average ...07Z Update... The 00Z EC is tracking a bit too quickly compared to its previous run and it now just outside of the model clustering. Therefore, would rather there be more weight given to the 12Z EC. With respect to the 00Z CMC/UKMET, they are shown improvements overall in timing/progression and track. However, the intensity of the mid/surface low and the synoptic scale and mesoscale features have resulted in higher QPF values that are outside the bounds of the model consensus, especially the CMC. Therefore, feel the 00Z CMC should not be used at this time. ...Previous Discussion... A compact shortwave west of WA state will dig southwest toward the Central CA coast early Sunday. Upstream a fast wave through the Gulf of AK will interact and carve out a sharper/full-latitude wave along the West Coast. Models are in fairly good agreement with respect to the mass fields and expected surface low development in the southern Plains. The 12z CMC continues to struggle the most with this evolution, especially within the mid-levels. As discussed below in the next section, because the CMC is too fast with the shortwave and surface low, it is interacting with the southern stream system. Therefore, will remove from the model blend. In addition, other model differences exist - the 00z GFS is a tad weaker, 12Z EC is a bit too strong and the 12Z UKMET a bit too slow as compared to the ensemble means/clusters. However, regardless of these differences, these models are still in fairly good agreement with no model severely out of tolerance. Therefore, with respect to the mass field evaluation, a general model blend, apart from the 12Z CMC, is supported. Deep Gulf of AK gyre with first peripheral shortwave/associated surface triple-point/front entering PacNW Monday... ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Preference: General Model Blend Confidence: Average ...07Z Update.. The 00Z EC is still in line with its previous run and the ensemble means. The 00Z CMC is still a tad too fast, but in much better alignment with the model consensus. The 00Z UKMET has also sped up a bit, making both the 00Z UKMET/CMC usable, supporting a general model blend approach. ...Previous Discussion... A favorable large scale set-up for broadening/upscale development across the Gulf of AK will be occurring through the weekend as a strong shortwave in broad cyclonic flow allows for a broader/larger cyclonic gyre to form across the Gulf of AK late Sunday. Along the periphery of the developing large scale wave, strong speed vorticity with embedded shortwave will press strong convergence along the occluded front near an enhancing triple-point near Vancouver Island toward the Juan de Fuca Strait. There is large model variance in the timing/strength of the wave and the gyre itself. The 12Z CMC is less favorable/less cyclonic across AK from the polar vortex to remain more flat and progressive to be more favorable especially given longer term trends in the ensemble suite toward a broader cyclone evolution. The 00z NAM continues to suggest this as well, but is quite fast and shows some typical day 3 over-amplification of the wave compared to the GFS/ECMWF solutions. The 12Z UKMET is far too slow with the overall progressive nature of this system. An improvement from the previous model run was the 12z ECMWF which did become better aligned with bulk of ECENS members which are closer to the GFS/GEFS overall. The timing/placement of the ECMWF along with the GFS/GEFS and ECENS mean suggest some confidence for a blend to be preferred with less confidence on the faster/flatter 12Z CMC output and the slower 12Z UKMET. Confidence in the evolution is solid, but internal changes, run to run variations are likely to suggest confidence overall in the blend is average to slightly below average. Model trends at www.wpc.ncep.noaa.gov/html/model2.shtml 500 mb forecasts at www.wpc.ncep.noaa.gov/h5pref/h5pref.shtml Pagano