Model Diagnostic Discussion NWS Weather Prediction Center College Park MD 146 PM EST Fri Feb 22 2019 Valid Feb 22/1200 UTC thru Feb 26/0000 UTC ...See NOUS42 KWNO (ADMNFD) for the status of the upper air ingest... 12Z Model Evaluation...with Final Preferences and Confidence ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ...Closed low ejecting out of the Southwest... ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Preference: near the 12Z ECMWF Confidence: Average ...19Z update... Adjustments made by the 12Z ECMWF/UKMET/CMC were greatest in the 12Z ECMWF with a shift in the direction of the 12Z GFS regarding the 500 mb low track across the Midwest into the Great Lakes, with lesser adjustments toward the 12Z GFS from the 12Z UKMET/CMC. Given the shift and tighter model clustering, near the 12Z ECMWF appears best at this time for the surface to mid-level low track across the Great Lakes and regarding the warm frontal placement moving north along the East Coast. ...previous discussion follows... The 12Z NAM/GFS were noted to be faster than the 00Z ECMWF/UKMET/CMC with the ejection of the closed low across the Midwest and toward the Great Lakes region which translates into some minor differences with the surface-850 mb low track. By Sunday though, the 00Z ECMWF shows slowest with the 500 mb low compared to the remaining deterministic and ensemble mean guidance, likely meaning the 00Z ECMWF could be displaced a bit too far to the west. These differences are also tied to an upstream mid-level vorticity max moving across the Dakotas Sunday morning with the ECMWF showing more interaction between this feature and the closed low to the east. However, as the upper low advances into southeastern Canada, the 12Z NAM races well out ahead of the guidance followed next by the 12Z GFS, 00Z UKMET/CMC and the 00Z ECMWF lagging behind. Regarding the upper low/surface low track into the Great Lakes, a blend between the 00Z ECMWF and 12Z NAM/GFS is preferred but with more weight toward the slower ECMWF, just not going full ECMWF given it appears to be a little too slow. Across the East Coast, with a secondary low lifting north and a warm front extending to the southeast, the 12Z GFS is a weaker outlier with the surface low and is farther south/slower with the warm front extending into the western Atlantic. A 12Z NAM/00Z ECMWF blend is preferred along the East Coast for Saturday night into Sunday night. ...Pacific Northwest trough/closed low Sun/Mon... ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Preference: 12Z GFS / 12Z ECMWF blend Confidence: Average ...19Z update... Across the West, small adjustments were made, some toward the 12Z GFS, some away. Overall, model spread remains about the same as earlier, but somewhere between the 12Z GFS and 12Z ECMWF continues to represent the middle of the latest spread and preferred middle ground. ...previous discussion follows... Latitude differences appear with a slow moving, positively tilted upper trough off of the Pacific Northwest by Sunday and a related surface low off of the West Coast. Regarding the surface low, the 12Z NAM/GFS are farther north when compared with the 00Z ECMWF/UKMET/CMC at 00Z/25. Ensemble scatter low plots show stronger support for a more ECMWF-like position, but there is some support for the faster/northern NAM/GFS by the GEFS and some Canadian ensemble members. By 12Z/25 and beyond, the 12Z NAM begins to stand out more significantly with a farther north latitude 500 mb trough axis off of the West Coast due to differences in its handling of vorticity energy moving around the trough axis. While there is high confidence in leaning against the 12Z NAM for Sunday night into Monday, it is less clear which direction to lean between the GFS/ECMWF. Given relatively good agreement aloft among the non-NAM deterministic guidance, a blend between the 12Z GFS/00Z ECMWF appears best at this time to capture the middle of the model spread. Model trends at www.wpc.ncep.noaa.gov/html/model2.shtml 500 mb forecasts at www.wpc.ncep.noaa.gov/h5pref/h5pref.shtml Otto