Jump to content
  • Member Statistics

    15,514
    Total Members
    7,904
    Most Online
    Raifu
    Newest Member
    Raifu
    Joined
NJwx85

Major Hurricane Irma

Recommended Posts

1 minute ago, DCAlexandria said:


emoji849.pngemoji849.pngemoji849.pngplease. Euro has gone from OTS to gulf to Cuba to east coast of FL.


Sent from my iPhone using Tapatalk

Yeah, at hours T+192, T+216, and T+240....I suspect that if the U of Albany graphic went out to those leads, Euro would continue it's dominance.  The stats don't lie.

  • Like 2

Share this post


Link to post
Share on other sites
2 minutes ago, DCAlexandria said:


emoji849.pngemoji849.pngemoji849.pngplease. Euro has gone from OTS to gulf to Cuba to east coast of FL.


Sent from my iPhone using Tapatalk

Verification scores through 120HR tell the story though. It's really not close.

Share this post


Link to post
Share on other sites
2 minutes ago, DCAlexandria said:


emoji849.pngemoji849.pngemoji849.pngplease. Euro has gone from OTS to gulf to Cuba to east coast of FL.


Sent from my iPhone using Tapatalk

 

The GFS was one of the only models to oppose the Cuba landfall scenario, and has been generally more consistent with the medium term track into the east coast. Yes, the euro has superior verification possibly with short term track, but as far as practical application of model data, I think the GFS has been superior with long term track with Irma.

  • Like 6

Share this post


Link to post
Share on other sites

Remember, there is only verification to this point.  So, the Euro has been by far the best model at the track up until now.  That doesn't speak to any flip-flopping in the future (such as landfall points).  But it does speak to the inadequacy of the other models in comparison.

Share this post


Link to post
Share on other sites
2 minutes ago, ag3 said:

 

LOL. Euro has been terrible past hour 72. Yes, it's scoring well inside of that but it's been the worst model by far outside of that. Very erratic with thousand mile shifts each run.

So you are saying the calculations at http://www.atmos.albany.edu/facstaff/tang/tcguidance/al112017/ are incorrect?  It has Euro with by far the lowest error at hour 120. 

You might want to send them an e-mail explaining in detail how your own Mean Absolute Error calculations are correct. You have done calculations on your own, right? 

  • Like 2

Share this post


Link to post
Share on other sites
3 minutes ago, Z-Cast said:

Verification scores through 120HR tell the story though. It's really not close.

The argument is post the D4-5 period. Models are now congregating toward a track which the GFS has been mostly indicating over the balance of this storm's life (an east coast landfall).

  • Like 1

Share this post


Link to post
Share on other sites

Wasn't the GFS first to take this route, and then also first to move east yesterday? Not saying that second part will even end up being right, but it seems like the GFS has been ahead of the Euro in trends for this storm.

Share this post


Link to post
Share on other sites
4 minutes ago, ag3 said:

 

LOL. Euro has been terrible past hour 72. Yes, it's scoring well inside of that but it's been the worst model by far outside of that. Very erratic with thousand mile shifts each run.

Are we looking at the same chart? Euro has the best scores at 96 and 120.

Share this post


Link to post
Share on other sites
Just now, Isotherm said:

The argument is post the D4-5 period. Models are now congregating toward a track whicj the GFS has been mostly indicating over the balance of this storm's life (an east coast landfall).

It's people anecdotally handwaving what they think verification is past 120 based on their personal biases. I'd be extraordinarily shocked if verification leaders suddenly flipped post 120. That's not how this works. 

  • Like 4

Share this post


Link to post
Share on other sites
2 minutes ago, DCAlexandria said:


emoji849.pngemoji849.pngemoji849.pngplease. Euro has gone from OTS to gulf to Cuba to east coast of FL.


Sent from my iPhone using Tapatalk

Yeah the amount of credit the euro gets, including during the winter, is a bit over the top at times. Sure it may do better than other models but the idea it's perfect, without its flaws, or it doesnt shift around a lot too is often greatly exaggerated 

  • Like 3

Share this post


Link to post
Share on other sites
3 minutes ago, Isotherm said:

 

The GFS was one of the only models to oppose the Cuba landfall scenario, and has been generally more consistent with the medium term track into the east coast. Yes, the euro has superior verification possibly with short term track, but as far as practical application of model data, I think the GFS has been superior with long term track with Irma.

I agree with this interpretation -- so far the euro gets the short term details best, but has been lagging to the GFS in terms of the big picture outcome.

  • Like 1

Share this post


Link to post
Share on other sites
4 minutes ago, Isotherm said:

The argument is post the D4-5 period. Models are now congregating toward a track which the GFS has been mostly indicating over the balance of this storm's life (an east coast landfall).

For now...until verification ends up in S. Fl... lol!!

  • Like 1

Share this post


Link to post
Share on other sites

I see everyone talking about the possibilities of a FL landfall, a possible Gulf landfall which is almost out of the question now or a Georgia/Carolinas landfall. What is the real possibility this thing curves OTS?

Share this post


Link to post
Share on other sites
4 minutes ago, Derecho! said:

It's people anecdotally handwaving what they think verification is past 120 based on their personal biases. I'd be extraordinarily shocked if verification leaders suddenly flipped post 120. That's not how this works. 

Bingo!  If at T+120 Euro is crushing the GFS and others...longer leads, there is no way a model's verification ranking flips...it would have to basically acquire Tourette's syndrome.

  • Like 1

Share this post


Link to post
Share on other sites
1 minute ago, AcePuppy said:

I see everyone talking about the possibilities of a FL landfall, a possible Gulf landfall which is almost out of the question now or a Georgia/Carolinas landfall. What is the real possibility this thing curves OTS?

15%, and I guarantee this percentage will verify.

It's still possible if that is what you are looking for.

Share this post


Link to post
Share on other sites

I would like to empathize that while most of the guidance shows some weakening over the next day or two, all guidance that keeps Irma away from Cuba shows significant strengthening once in the Southern Bahamas. I wouldn't be lowering awareness if recon finds that Irma has weakened slightly. It should still be a cat 4 on approach to Florida, and that could be conservative. 

  • Like 1

Share this post


Link to post
Share on other sites
18 minutes ago, LakeEffectKing said:

Yeah, at hours T+192, T+216, and T+240....I suspect that if the U of Albany graphic went out to those leads, Euro would continue it's dominance.  The stats don't lie.

Not just hour 192.

It's been erratic with anything past hour 72.

Share this post


Link to post
Share on other sites
25 minutes ago, Derecho! said:

It's getting to the Point where some of the consensus models don't matter because the Euro is so dominant (I really wish the Euro ensemble mean was on that chart, though). 

It's so good it's getting boring - witness the immense resistance in this thread to Euro dominance. 

When a given run has a divergence between the OP and EPS, the very next OP run moves to what the EPS had in the run 12 hrs earlier. Like we saw how the 0z OP shifted just north of Cuba like the 12Z EPS mean had.

Share this post


Link to post
Share on other sites
3 minutes ago, ag3 said:

Not just hour 192.

It's been erratic with anything past hour 72.

Except for the fact that it's wrecking everything else in verification, even at hour 120. Does the verification link really need posted for the thousandth time?  The math says ECMWF is best, so I trust it the most. 

Share this post


Link to post
Share on other sites
Guest
This topic is now closed to further replies.

  • Recently Browsing   0 members

    No registered users viewing this page.

×