Jump to content
  • Member Statistics

    17,515
    Total Members
    7,904
    Most Online
    amirah5
    Newest Member
    amirah5
    Joined

Election week coastal storm threat


Midlo Snow Maker

Recommended Posts

  • Replies 1.3k
  • Created
  • Last Reply

snow mentioned in my forecast

  • Tuesday Night A chance of rain and snow showers. Mostly cloudy, with a low around 33. Chance of precipitation is 30%.
  • Wednesday A chance of rain and snow showers. Cloudy, with a high near 44. Chance of precipitation is 40%.

Link to comment
Share on other sites

It's pretty sad that the EURO kicks the GFS around on a daily bases. We may as well just stop wasting government money on the GFS and just use the EURO long range and still use the NAM short range.

It doesn't always kick the GFS's butt, no one model is good enough to use in the longer range and the gfs and the other models along with the ensembles from all the models offer the best approach to forecast. Lest you forget how sometimes the euro can be massively wrong, here's an example when the euro was forecasting a significant storm and the gfs wasn't, the latter was right.

http://www.washingtonpost.com/blogs/capital-weather-gang/post/another-iffy-snow-threat-in-the-offing/2012/02/14/gIQA6yHqDR_blog.html#pagebreak

Link to comment
Share on other sites

I can't see the maps that well on my phone. Any change in strength with the hp to the n on the latest gfs? Still slides east the same from what I can see but I can't tell of if the strength is any different.

It is more of a phased solution a little too far east for us but much closer to the euro than the previous run. I think something wrapped up is correct but whether it gets us or not is still up in the air. This is a tougher call than Sandy as the blocking is not as defined so you are more susceptible to timing issues between the various waves.

Link to comment
Share on other sites

It doesn't always kick the GFS's butt, no one model is good enough to use in the longer range and the gfs and the other models along with the ensembles from all the models offer the best approach to forecast. Lest you forget how sometimes the euro can be massively wrong, here's an example when the euro was forecasting a significant storm and the gfs wasn't, the latter was right.

http://www.washingto....html#pagebreak

I'd say the EURO on most occasions is more right than wrong. There are a few times when the GFS out did the EURO but, it is not as often as the other way around. Ok don't get rid of the GFS but, geez improve it. We are only 96 hrs out from this next storm as the GFS still can't figure it out.

acz5.gif

Link to comment
Share on other sites

Where were these charts when the GFS beat the euro with Debbie?

It's called selective memory. Without doubt the euro wins in the verification game over the long haul but the verification scores are not that different. In this case I suspect the euro is more right than the earlier GFS runs but the new GFS has moved in that direction and may not have a bad solution. Neither is good for up towards NYC...too much wind.

Link to comment
Share on other sites

It's called selective memory. Without doubt the euro wins in the verification game over the long haul but the verification scores are not that different. In this case I suspect the euro is more right than the earlier GFS runs but the new GFS has moved in that direction and may not have a bad solution. Neither is good for up towards NYC...too much wind.

I always view deterministic models as something to take with a grain of salt beyond day 6 and even day 5. So IMHO...I don't really care what deterministic runs are showing at day 7 or 8. I'll keep those runs in mind, but look more towards ensemble guidance and see how those solutions look. When we get to within 96 hrs and both models start to converge on a solution, that's fine by me. I don't need to know that the euro had it at day 8 or whenever it was. Lord knows the day 8 euro op has had some clown solutions too. Where the euro seems to be lethal, are the details within 72-96 hrs. I do think it's pretty good in that range as far as deterministic guidance goes.

Link to comment
Share on other sites

I'd say the EURO on most occasions is more right than wrong. There are a few times when the GFS out did the EURO but, it is not as often as the other way around. Ok don't get rid of the GFS but, geez improve it. We are only 96 hrs out from this next storm as the GFS still can't figure it out.

You don't think they're CONSTANTLY trying to improve it? There's only so much you can do, especially with the time, money and computing power restrictions placed on the folks who work on the GFS. Study up on what's behind the GFS and the Euro and then tell me what you think can realistically be done to improve the GFS with the restrictions on time, budget, etc.

Link to comment
Share on other sites

I always view deterministic models as something to take with a grain of salt beyond day 6 and even day 5. So IMHO...I don't really care what deterministic runs are showing at day 7 or 8. I'll keep those runs in mind, but look more towards ensemble guidance and see how those solutions look. When we get to within 96 hrs and both models start to converge on a solution, that's fine by me. I don't need to know that the euro had it at day 8 or whenever it was. Lord knows the day 8 euro op has had some clown solutions too. Where the euro seems to be lethal, are the details within 72-96 hrs. I do think it's pretty good in that range as far as deterministic guidance goes.

Even within 96 hrs in complex situations, the euro can occasionally be pretty wrong so ensemble forecasting certainty still has plenty of merit. People, especially weenies, seem to have unrealistic expectations of deterministic model forecasts. Heck, if a model gets it right by 48 hours I'm a happy camper, that still gives most people time to take action.

Link to comment
Share on other sites

You don't think they're CONSTANTLY trying to improve it? There's only so much you can do, especially with the time, money and computing power restrictions placed on the folks to work on the GFS. Study up on what's behind the GFS and the Euro and then tell me what you think can realistically be done to improve the GFS with the restrictions on time, budget, etc.

I understand the time and money limitations, I just can't understand with todays technology how we don't have something better. Someday I guess.

I am not thinking the EURO is 100% correct on this storm, I am just saying it had the right idea of a coastal before any model.

Link to comment
Share on other sites

Even within 96 hrs in complex situations, the euro can occasionally be pretty wrong so ensemble forecasting certainty still has plenty of merit. People, especially weenies, seem to have unrealistic expectations of deterministic model forecasts. Heck, if a model gets it right by 48 hours I'm a happy camper, that still gives most people time to take action.

Yeah even at 96 hrs we certainly have discrepancies and busts. But for a deterministic model, it seems like that day 3-4 is pretty good all things considered.

But the euro had some busts this summer, even within 48hrs.

Link to comment
Share on other sites

I understand the time and money limitations, I just can't understand with todays technology how we don't have something better. Someday I guess.

I am not thinking the EURO is 100% correct on this storm, I am just saying it had the right idea of a coastal before any model.

Take some computer modeling classes. That oughta open your eyes :P

Link to comment
Share on other sites

You don't think they're CONSTANTLY trying to improve it? There's only so much you can do, especially with the time, money and computing power restrictions placed on the folks to work on the GFS. Study up on what's behind the GFS and the Euro and then tell me what you think can realistically be done to improve the GFS with the restrictions on time, budget, etc.

DT explained the problem perfectly. "You can only shine up s*&@ to a certain point". I fear most of the GFS problems reside in the initialization scheme.

hotdog.gif

Link to comment
Share on other sites

DT explained the problem perfectly. "You can only shine up s*&@ to a certain point". I fear most of the GFS problems reside in the initialization scheme.

hotdog.gif

He's not the best source of info about the model and he is much too hard on the gfs, sometimes to his own detriment. DTK is the best source here on modeling and what the problems are.

Link to comment
Share on other sites

DT explained the problem perfectly. "You can only shine up s*&@ to a certain point". I fear most of the GFS problems reside in the initialization scheme.

hotdog.gif

When mets talk like that it mostly shows off they are kinda clueless.

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

Guest
This topic is now closed to further replies.
  • Recently Browsing   0 members

    • No registered users viewing this page.

×
×
  • Create New...