Jump to content

bluewave

Members
 View Profile  See their activity

  • Posts

    34,383

  • Joined

  • Last visited

 Content Type 

Everything posted by bluewave

  1. bluewave
    The models did fine when you account for their northward correction short term with amped systems. I pointed this out over the weekend. That’s the reason you never want to be in the jackpot zone 5 days out. Model snowfall forecasts and exact location of the mixing line are seldom accurate beyond 24-48 hours. That’s why I am not a fan of posting all these longer range snowfall maps on social media.
  2. bluewave
    I think the last time we really had to worry about confluence here was 2-6-10. But that was a -5 AO vs -3 now. Most of the time for us the correction is more N and W with amped systems. That’s why some of our best snowstorms on record were modeled 3-5 days out to crush Philly or DC. But then corrected north in the short term. So a day 5 forecast with the mixing just to our south usually shifts further north in the short term. So now it looks like mixing will be an issue near the coast following the initial front end thump.
  3. bluewave
    It looks like one of the features on the v16 is a fix for the cold bias. Do you know if it’s still scheduled to go operational in February? https://www.emc.ncep.noaa.gov/gmb/STATS_vsdb/
  4. bluewave
    The models are more tucked in with the low than just a few days ago. The ridge east of New England is a little stronger. This has been a recent theme. So the trough goes neg tilt quicker allowing mixing issues to arise after the initial front end thump. New run Old run
  5. bluewave
    This could be the first time that the AO was positive for nearly every day in November and negative for so many days in December.
  6. bluewave
    It’s probably just the suppression bias in the v15. We can only hope that v16 has fixed this issue.
  7. bluewave
    We haven’t had coastal systems in so long, that the EE rule has been in hibernation.
  8. bluewave
    That H500 closing off that the NAM has near 84 hrs would be an insurance policy against the high over New England suppressing this to the south.
  9. bluewave
    I once spoke to someone that was asking me about online weather forums. They asked me why the people on there keep hoping that somebody would get crushed by a storm.
  10. bluewave
    I wish they could all be this easy. But I think this was the last real long range snowstorm grad slam hit by a member ECMWF family of products.
  11. bluewave
    The UKMET can also be pretty erratic from storm to storm and run to run. It’s usually a rear view mirror assessment on which model did the best with each individual storm. I am going to wait until the UL gets fully sampled tonight over Oregon. Day 3-4 snowfall forecasts have always been a challenge since so many important things usually happen in under 48 or 36 hours. The general public doesn’t mind if you forecast 1.00” of rain and .50 verifies. But you would probably hear about it if it was snow especially on this forum.
  12. bluewave
    I posted the GFS cold bias earlier in this thread. The model loves to push the baroclinic zone too far S and E. So the storms end up too suppressed.
  13. bluewave
    It’s telling us that the GFS is probably too flat again.
  14. bluewave
    The UKMET looks further NW than the GFS. But these early lo res charts aren’t the easiest to make out. https://www.meteociel.fr/modeles/ukmo2.php?nh=1&ech=72&carte=1021
  15. bluewave
    The NAM actually looks like it could be one of the EPS members more than the GEFS.
  16. bluewave
    Great day to finish any yard clean ups with a top 5 warmest 12-13 in tap. Data for December 13 - NY CITY CENTRAL PARK, NY Click column heading to sort ascending, click again to sort descending. Date Max Temperature Min Temperature Precipitation Snowfall Snow Depth 2015-12-13 67 55 0.00 0.0 0 1923-12-13 64 42 0.10 0.0 0 1984-12-13 63 46 0.00 0.0 0 1946-12-13 63 36 T 0.0 0 1919-12-13 61 42 0.07 0.0 0 1990-12-13 60 41 0.00 0.0 0 1968-12-13 60 36 0.00 0.0 0 1949-12-13 60 36 0.17 0.0 0
  17. bluewave
    The models that handle the 500 mb heights near Oregon and general PNA best will get the storm track correct. Bottom line: I'll be really curious to see tonight's MFR sounding. That 500mb height observation will perhaps help give us (and models) a clue where the storm might go later on. 5570m vs 5590m may seem small, but in a nonlinear system those tiny differences grow rapidly!
  18. bluewave
    This storm will push up against one of the strongest 50/50 lows on record.
  19. bluewave
    The next big leap may be AI. https://www.ecmwf.int/en/newsletter/163/news/ai-and-machine-learning-ecmwf ECMWF is currently making a significant effort to support applications of artificial intelligence and machine learning and to identify how such applications may improve numerical weather prediction at the Centre. Many standard methods used by ECMWF scientists on a daily basis can be regarded as examples of machine learning. However, there has recently been a surge in new methods which have the potential to revolutionise the work of operational weather prediction centres. Such methods include the use of deep neural networks, which can learn the dynamics of very complex non-linear systems from data.
  20. bluewave
    Today should be the warmest day of the month so far with widespread 60s. KEWR GFS MOS GUIDANCE 12/13/2020 0000 UTC DT /DEC 13 /DEC 14 /DEC 15 / HR 06 09 12 15 18 21 00 03 06 09 12 15 18 21 00 03 06 09 12 18 00 X/N 63 37 44 29 40 TMP 51 50 50 55 60 59 53 47 42 40 39 39 41 41 38 37 34 32 30 38 32
  21. bluewave
    I don’t know. But it seems like when you try to fix one model bias or error, it creates another one. The real model data that you want are local model model biases for the individual NWS offices. So you can say x model is having this bias for storms with y teleconnections in place around the NY Bight. But these general model statistics don’t go that local to include elements such as storm track.
  22. bluewave
    They fixed the GFS cold bias with the v16. But now it has a warm bias compared to the other guidance. The 500 MB skill scores are a little lower than the current OP. That being said, I have no idea how it will do with this storm since it’s just 1 case. https://www.emc.ncep.noaa.gov/gmb/STATS_vsdb/
  23. bluewave
    It’s probably more accurate to use the term model biases. Seasonal storm track biases have been well documented. But we would need to use machine learning for real time model bias correction to the forecast maps.
  24. bluewave
    Looks like the 12z EPS was a little more tucked today vs yesterday due to the slightly weaker +PNA. The first really important run should be 12z Sunday when the energy is fully ashore in California. Then we watch to see what type of short term trends develop.
  25. bluewave
    The AO volatility continues to make headlines. This could be one of the greatest +November to -December reversals on record. Notice how a very +AO in November usually carried over to December. The one exception was 1978. 2020 2.419 3.417 2.641 0.928 -0.027 -0.122 -0.412 -0.381 0.631 -0.072 2.086 2015 1.092 1.043 1.837 1.216 0.763 0.427 -1.108 -0.689 -0.165 -0.250 1.945 1.444 2013 -0.610 -1.007 -3.185 0.322 0.494 0.549 -0.011 0.154 -0.461 0.263 2.029 1.475 2011 -1.683 1.575 1.424 2.275 -0.035 -0.858 -0.472 -1.063 0.665 0.800 1.459 2.221 1994 -0.288 -0.862 1.881 0.225 -0.115 1.606 0.351 0.828 -0.084 0.174 1.779 0.894 1978 -0.347 -3.014 0.502 -0.967 0.059 0.635 -0.604 -0.354 -0.099 0.895 2.470 -0.980
×
×
  • Create New...