Model Diagnostic Discussion NWS Weather Prediction Center College Park MD 241 PM EDT Sun Sep 15 2019 Valid Sep 15/1200 UTC thru Sep 19/0000 UTC ...See NOUS42 KWNO (ADMNFD) for the status of the upper air ingest... 12Z Model Evaluation with Preferences and Forecast Confidence ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ...Deep Latitude Western Trof Entering the West by Monday; Reinforced by Closed Low Late Tues... ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Preference: 12z GFS/ECMWF/CMC blend first shortwave; General model blend for the second. Confidence: Slightly above average 19z update: The 12z ECMWF/CMC both trended ever so slightly faster (more so for the CMC), to match the GFS fairly well. This provides additional confidence in a 12z GFS/ECMWF/CMC blend at slightly above average confidence for this initial shortwave. The UKMET remained a bit too fast and deep to incorporate at this time. The second, closed low, continues to show much better overall agreement which includes the UKMET. However, the UKMET is generally a bit weaker relative to the remaining deterministic guidance, but still is a higher QPF producer though in similar places/timing with other solutions to support its inclusion for this wave. So general model blend is preferred for this system. ---Prior Discussion... Models continue to agree fairly well in the approach and entrance of the trof into the West Coast/N CA. However, by 48hrs the spread in the models manifests. The 00z UKMET, as it has been for days, remains fastest with little other support/reasons. As the shortwave crosses the eastern Snake River Plain, the 12z NAM becomes more compact again and rapidly deepens further as it stretches across the Rockies into eastern MT, and it clearly in an unrealistic feedback deepening that the model has a bias toward. The 00z CMC is not too poor, but starts to even lag the ECMWF becoming significant by 72hrs to include. There is a typical spread between slightly faster 12z GFS and 00z ECMWF that has sped up a bit over prior runs. As such, preference is 12z GFS/00z ECMWF blend at average confidence. Upstream, the closed low dropping into the Pacific NW by late Tuesday is fairly agreeable with the only exception being the UKMET which again is fast, toward 19.00z, the CMC and ECMWF are a bit more sharp with increasing positive tilt, but this does not appear to be out of the realm of best probability given average day 3 spread. As such non-UKMET is fine for this portion of the larger scale trof. ...Southwest Shortwave, Moving through Central High Plains into Northern Plains/MN by Tuesday... ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Preference: General model blend Confidence: Slightly above average 19z update: The 12z ECMWF, CMC and UKMET all trended a bit stronger and with sizable increase in forward speed progressing into the Northern Plains/MN. This puts it on par with the GFS/NAM and provides significantly greater confidence in a general model blend for this wave. ---Prior Discussion--- The 12z GFS continues to be the strongest bringing the wave north-northeast across the Southern Rockies into the Central High Plains late Monday and up through the Northern Red River Valley by Tuesday. While, it is the strongest it has trended weaker though is still a bit faster than the ECMWF/CMC which are typical of bias a bit too slow. The faster solution and earlier return of moisture helps break out convection earlier as well, but this is supported by the UKMET/NAM to have some confidence especially now that the magnitude/coverage is reduced slightly. The 00z CMC, like the ECMWF is slow, but is also a bit too dry, favoring the approaching western trof to provide forcing, suggesting its removal or blend reduction would be best. The 00z UKMET having issues with a too fast western trof (see above), so it is also not preferred in this blend. So a GFS/NAM and 00z ECMWF blend is preferred but splitting the timing differences. Confidence is average given the moderate spread and influence of the approaching western trof. ...Western Gulf of Mexico TUTT/surface wave... ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Preference: 12z NAM/ECMWF blend Confidence: Slightly above average 19z update: The 12z UKMET and CMC continue to be quite amplified favoring the northern portion of the surface inverted trof/wave with slightly stronger deeper mid-level core to the system. The 12z ECMWF trended this direction as well and lies between the NAM and CMC/UKMET solutions, further supporting less influence of the weaker GFS. Still will favor a NAM/ECMWF solution but some very low percentage of UKMET/CMC may be incorporated to handle this reduced spread in solutions. The TUTT is expected to continue west across the Gulf with eventual support to develop a weak surface wave nearing/southeast of the mouth of the Rio Grande Monday to Tuesday. The 00z UKMET continues to be strongest, followed by the 00z CMC and so greater depth/steering draws it north. The 00z ECWMF is a bit slower and weaker but in line with the recent few NAM runs, including 12z. The 12z GFS is a bit stronger/slower but remains weaker and given downstream weakening across NW Mexico, slides more west than other guidance. All considered a blend of the 12z NAM and 00z ECMWF is middle ground and keeps with continuity, and while not aligned perfectly, other guidance provides confidence toward this blend and so the preferred blend has slightly above average confidence. ...Tropical Storm Humberto... ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Preference: Official NHC Forecast Best Proxy: 06z GFS/GEFS mean and 00z ECENS mean blend Confidence: Slightly below average 19z update: After 48hrs, the 12z GEFS solution broadens given greater number of solutions slowing (even slower than the 12z GFS itself). Likewise, the UKMET (which is on pace with the 15z NHC track but north) and ECMWF slow relative to their 00z solutions with the ECMWF lagging the GFS, still. While the 12z CMC did speed up relative to the 00z solution, it remains the slowest solution with continued suggestion that the downstream ridge is stronger and potential for the system to stall. So there remains slightly below average confidence especially after 60hrs and the 06z GFS/GEFS and 00z ECENS mean match the 15z forecast track the best. ---Prior Discussion--- Humberto continues to strengthen and surrounding steering environment suggests a stalling and slow expected turn toward the east is within the next day. Guidance continues to suggest a slowing overall. Overall, the 06z GFS, GEFS mean and the 00z ECENS mean are very close to the official NHC forecast. The main take away though, is the newest 12z guidance continues to support a slowing in the post-2day forecast period with greater ridging building toward the northeast. While clearly not to the magnitude of the 00z CMC that actually shows significant shearing and surface slowing to the east with medium range suggesting a turn back west. There are about 10-15% of the ECENS members that also suggest this eventual stall due to the developing block. So there is still modest uncertainty at 60-84hr period and beyond. The 00z ECMWF and 12z GFS are tightly packed now with the 12z NAM very close all about 12hrs slow and slightly south of the official track, So best proxy to the 15z NHC forecast is the 06z GFS/GEFS mean and 00z ECENS mean. Model trends at www.wpc.noaa.gov/html/model2.html 500 mb forecasts at www.wpc.noaa.gov/h5pref/h5pref.shtml Gallina