Model Diagnostic Discussion NWS Weather Prediction Center College Park MD 137 PM EST Thu Mar 08 2018 Valid Mar 08/1200 UTC thru Mar 12/0000 UTC ...See NOUS42 KWNO (ADMNFD) for the status of the upper air ingest... 12Z model evaluation with final preferences ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ...Coastal low currently near coastal Maine, weakening westward into southern Quebec through Friday morning... ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Preference: General model blend Confidence: Above average The models show similarly with this system. ...Secondary surface low near Nova Scotia Friday night... ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Preference: General model blend Confidence: Above average The 12Z CMC adjusted toward the model consensus so that by 00Z/10, the 12Z CMC no longer appears to be farther east compared to the remaining deterministic consensus with the surface low, like the 00Z CMC. The latest ensemble scatter low plots continue to show a modest amount of spread between 00Z-12Z/10, with the 12Z CMC and remaining 12Z guidance agreeing toward the middle of those ensemble differences. ...Shortwave dropping south across British Columbia Thursday night with the southern portion reaching into the Plains Saturday... ...Surface low developing over the southern Plains late Saturday and moving toward the Southeast on Sunday... ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Preference: 12Z GFS/12Z GEFS/00Z ECMWF/12Z CMC blend Confidence: Below average The shortwave coming through southwestern Canada shows some differences in the models early on with the 12Z NAM less amplified across the southern portion of the trough compared to the 12Z GFS/ECMWF/UKMET/CMC. These differences grow downstream into the Plains with little agreement seen among the models. Most model run to run continuity is poor, with perhaps the GFS showing the most consistency over the weekend. However, the 12Z GFS appears to be a stronger outlier with its surface low as it develops and tracks across the South. At 500 mb, ensembles and the majority of deterministic guidance shows a trend toward stronger with the mid-level trough east of the Mississippi River on Sunday into Monday. The is contrary to the 12Z ECMWF which trended much weaker for Sunday across the South. While ensemble scatter low plots show a moderate to large degree of spread on Sunday, a center point of the lows, when considering trends, is nearest the position of the 12Z GFS and 12Z CMC valid 00Z/11 to 00Z/12. The 12Z ECMWF appears too weak with the system at 500 mb while being too far south at the surface, south of most of the latest ensemble low plots and appearing worse than the 00Z ECMWF. The 12Z NAM is significant faster at the surface while the 12Z UKMET also appears too far to the south like the ECMWF, appearing worse-off than its 00Z cycle. No single model or two-way blend is preferred here given the large differences and poor run to run consistency. Therefore, a multi-model blend of the 12Z GFS/12Z GEFS/00Z ECMWF/12Z CMC is preferred with hopefully the strengths of each model being maximized while minimizing apparent weaknesses. ...Weak shortwave approaching California from the west Saturday evening... ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Preference: 12Z GFS/ECMWF/UKMET blend Confidence: Average The 00Z ECMWF is much faster with the shortwave coming into California than many of the latest ensemble guidance but the 12Z ECMWF slowed down some. The 12Z GFS/UKMET are slower and weaker than the ECMWF but The 12Z UKMET splits the difference between the stronger ECMWF and weaker GFS. Given the southern stream jet is rather zonal across the Pacific with a meridional northern stream jet, the preference is not be toward the slower side of the models. With a weaker amplitude system like the 12Z GFS, a slower/weaker solution does not fit the conceptual model, so a middle ground between the 12Z GFS/ECMWF/UKMET is recommended at this time which is reasonably close to the latest ensemble means. Model trends at www.wpc.ncep.noaa.gov/html/model2.shtml 500 mb forecasts at www.wpc.ncep.noaa.gov/h5pref/h5pref.shtml Otto