Model Diagnostic Discussion NWS Weather Prediction Center College Park MD 237 AM EST Sat Feb 16 2019 Valid Feb 16/0000 UTC thru Feb 19/1200 UTC ...See NOUS42 KWNO (ADMNFD) for the status of the upper air ingest... 00Z Model Evaluation...with Preferences and Confidences ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ...Overall Pattern Across the CONUS... ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ A difficult forecast will set up over the lower 48 as the mean upper trough axis currently just off of the West Coast shifts to the desert Southwest for Monday evening. As a shortwave emerges into the Central Plains on Saturday/Sunday, it will become sheared with a powerful upper level jet axis oriented from the Southwest to the Northeast by Sunday evening, located between anomalous ridging over the Caribbean and upper troughing over southeastern Canada. ...Low amplitude shortwave reaching the Mid-Atlantic on Saturday with corresponding surface low across the Southeast... ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Preference: General Model Blend Confidence: Slightly Above Average The weak shortwave trough, currently moving through the mid-MS River Valley will trek eastward toward the Mid-Atlantic by Saturday afternoon. A weak surface reflection, currently in MS, will move eastward off the Southeast US coast. A broad area of precipitation from the lower Ohio Valley into the Deep South will shift toward the Carolinas and Virginia tonight before moving offshore by Saturday afternoon. The near-term model guidance is in good agreement with this feature, with just a subtle shift to the south in the last model cycle. This should put the heaviest QPF from southwest KY into portions of middle TN, northern GA, western/central NC, and southern VA. A general model blend, leaning on the hi-res model data and NAM will be preferred. ...Lead mid-level shortwave crossing through the Central Plains and Midwest... ...Weak surface low tracking across the Appalachians Sunday night ... ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Preference: Blend of 00z GFS, 00z NAM, 12z ECMWF Confidence: Slightly below Average A closed low currently over the interior Rockies will open up as it moves into the central Plains tomorrow, eventually shearing out by Sunday as it moves into the lower Great Lakes region. Model differences are seen in the speed and placement of the main energy, especially later Saturday into Sunday. The GFS is on the faster side of the model spread, while the NAM and UKMET to some degree are slower. The CMC seems to be the weakest and also the furthest north/west with its placement, and generally looks like an outlier by Sunday (and therefore will not be included). At the surface, a weak low will track from CO to north TX to KY, generally weakening as it does so. The NAM is furthest to the north with the low position by 18z Sunday, while the ECMWF is east and the GFS south. But all 3 models are relatively similar for 36-48 hours out. As it moves out of the Appalachians, there are differences noted whether a secondary low develops across the Carolinas or offshore. Forecast confidence is below average here. All the operational models have some similar large scale characteristics, so a blend of the 00z GFS, 00z NAM, 12z ECMWF looks good at this point. ...Mid-level trough re-amplification across the West Coast on Sunday... ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Preference: General Model Blend Confidence: Average Large troughing is expected to develop across the western US late in the weekend through early next week, with shortwave energy from Washington State through southern California. In the large scale sense, the operational models exhibit fairly good agreement with the placement and timing of the main features. There are some subtle differences noted with smaller scale vorticity maximums moving through the flow, as expected by 48-60 hours. An impressive push of upper level ridging is expected to shunt the troughing toward the southwest US by Monday while the whole system shears out and energy gets ejected out into the Rockies and Plains. With no significant differences seen, a general model blend will suffice here for this time period. ...Low amplitude mid-level shortwave trough nearing southern British Columbia late Monday... ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Preference: 00z ECMWF/ECENS, 00z UKMET blend Confidence: Below Average 0730z update: The 00z guidance offered little forecast confidence improvement with the compact shortwave trough expected early next week across the Pacific Northwest. The CMC maintained itself from its 12z run and shows similarity to the 00z GFS (stronger solution) while the 00z UKMET and ECMWF now show a much weaker shortwave (like the 00z NAM). This now puts the GFS and CMC as more outliers given the trend. As such, will lean on the ECMWF/UKMET solutions for this time period and region. ---previous discussion--- As alluded to in the previous section, a shot of upper level ridging will move across the B.C. into Washington State early next week. Quick on its heels will be a very strong shortwave trough, as advertised by all of the global operational models. There are however significant model differences with the placement, strength, and evolution of the feature Monday into Tuesday. The 00z NAM completely weakens the feature into a weak open wave as it crosses into Washington State, with barely any mid-level reflection by Monday night. Meanwhile, the CMC takes the feature well inland, moving it into eastern WA, Idaho by Tuesday before weakening. The GFS/ECMWF/UKMET show a compact closed low forming, staying close to the coast as it rounds the trough. The GFS/UKMET are faster than the ECMWF by 84 hours (19.12z) but not by a large amount. The ensemble means from the GFS and ECMWF are a good compromise, as they don't show nearly the strong signal seen in the operational models, but do show a mid-level shortwave near the coast during this time. The inland/east CMC solution seems like an outlier. So the WPC preference is for a GFS/ECMWF (and their means) and UKMET blend. Model trends at www.wpc.ncep.noaa.gov/html/model2.shtml 500 mb forecasts at www.wpc.ncep.noaa.gov/h5pref/h5pref.shtml Taylor