ATL: GERT - Post-Tropical - Discussion

[otowntiger]

I'm sorry, but the GFS is garbage. What a fold to all the other models in just one run. I'll be looking at it with a grain of salt this season until it proves me otherwise.

I think 99L is what may be garbage. Its a tease. It really goes to show that we don't know everything that's going to happen no matter how improved our technology is these days, especially in regards to meteorological models.

[wxmann_91]

I'm thinking 5-10% chance of development next 48 hrs, 20% within 120 hrs. Maybe as high as 50-60% chance beyond 5 days if it survives.

Only 5-10? Right now have it 50/80% as per NHC, so explain pleaseeee

Conditions never really supported an 80%, IMHO. This is why the GFS needs to be ignored for tropical cyclone forecasting.

[Hammy]

I'm sorry, but the GFS is garbage. What a fold to all the other models in just one run. I'll be looking at it with a grain of salt this season until it proves me otherwise.

I think 99L is what may be garbage. Its a tease. It really goes to show that we don't know everything that's going to happen no matter how improved our technology is these days, especially in regards to meteorological models.

This is nothing new from the GFS--it's not conjuring storms from nowhere anymore but it's still excessively bullish on development when something's there, and likely convective feedback issues given it was spinning up and then merging two lows. I would still put it at 50/50 to form but the GFS is basically in CMC/GEM territory prior to anything forming.

[Florida1118]

I'm thinking 5-10% chance of development next 48 hrs, 20% within 120 hrs. Maybe as high as 50-60% chance beyond 5 days if it survives.

Only 5-10? Right now have it 50/80% as per NHC, so explain pleaseeee

Conditions never really supported an 80%, IMHO. This is why the GFS needs to be ignored for tropical cyclone forecasting.

Or, instead of taking a model as gospel and "ignoring" a major model, you could take it as guidance - what it was designed for. To say the GFS should be flat out ignored in all tropical cyclone forecasting is ignorant and foolish, at best.

[Hypercane_Kyle]

I'm sorry, but the GFS is garbage. What a fold to all the other models in just one run. I'll be looking at it with a grain of salt this season until it proves me otherwise.

I think 99L is what may be garbage. Its a tease. It really goes to show that we don't know everything that's going to happen no matter how improved our technology is these days, especially in regards to meteorological models.

This is nothing new from the GFS--it's not conjuring storms from nowhere anymore but it's still excessively bullish on development when something's there, and likely convective feedback issues given it was spinning up and then merging two lows. I would still put it at 50/50 to form but the GFS is basically in CMC/GEM territory prior to anything forming.

What's funny is the CMC has been pretty conservative this year, at least compared to years prior.

[SFLcane]

 https://twitter.com/ryanmaue/status/893524970738987008

[jason1912]

 https://twitter.com/RyanMaue/status/893524970738987008

[SFLcane]

2pm conditions " Appear" favorable that wording usually means percentage downgrade forth coming. We shall see

[Nimbus]

Garbage in Garbage out with these model runs, still a little dry East of -45W apparently.
I'm kind of bothered that the tracks are all further south since the Hispaniola destruction scenario is only a narrow possibility.
Anywhere else and a weak TS can become a strong Hurricane in 36 hours if there is a window of favorable conditions.

[gatorcane]

Speaking of the NAVGEM, looks to have basically dropped development (not that it had anything significant in the first place).

[wxmann_91]

Only 5-10? Right now have it 50/80% as per NHC, so explain pleaseeee

Conditions never really supported an 80%, IMHO. This is why the GFS needs to be ignored for tropical cyclone forecasting.

Or, instead of taking a model as gospel and "ignoring" a major model, you could take it as guidance - what it was designed for. To say the GFS should be flat out ignored in all tropical cyclone forecasting is ignorant and foolish, at best.

Say I get some guidance from friends about big decisions I make. If I consistently get bad guidance from a particular friend, even if he meant well, but had other friends who gave me good guidance on average, wouldn't logic dictate that I stop listening to the bad guidance from the particular friend? Models shouldn't be taken as gospel, but they need to be weighed according to their performance. Nobody cares what the NAVGEM or CMC or the JMA or the NAM says (and to an even stronger extent, the LBAR, BAMs, etc.). If the GFS performs as badly, why should we weigh it any higher than those models?

[GCANE]

Looking at how the models runs are evolving and the persistance in COAMPS having this move west; IMHO, likleyhood is that this will run thru the Carib,

[LarryWx]

12Z Euro: even weaker than the 0Z Euro at hour 48 and the 0Z Euro didn't even do much with it. So, considering that as well as the 12Z model consensus to this point, I don't expect the 12Z Euro to do much with this.

[tolakram]

Say I get some guidance from friends about big decisions I make. If I consistently get bad guidance from a particular friend, even if he meant well, but had other friends who gave me good guidance on average, wouldn't logic dictate that I stop listening to the bad guidance from the particular friend? Models shouldn't be taken as gospel, but they need to be weighed according to their performance. Nobody cares what the NAVGEM or CMC or the JMA or the NAM says (and to an even stronger extent, the LBAR, BAMs, etc.). If the GFS performs as badly, why should we weigh it any higher than those models?

Mainly because you are wrong and there are stats to prove it. Euro has a lot of strengths but the GFS has not failed as badly as many make it out, again with statistics to prove it. If you are basing your trust of the GFS out past 120 hours then why not fault the euro for developing the same system a few days ago with a similar solution?

Models are guidance, and the blend is still the best solution. Without the GFS hurricane forecasts would be worse, not better.

[Alyono]

This may end up being more of a now-cast situation, since none of the models develop this any time soon, yet the NHC is giving this a 50/80, they must know something that we don't

forecast continuity

[Alyono]

I have been giving the NAVGEM more weight in m forecasts than the "improved" GFS. The "improvements" have degraded the model to the point it is no longer usable, and not just for the tropics

[AJC3]

This thread, like many others this season, is straying into a general discussion and comparison of the value of various global model guidance, versus discussing the model guidance for 99L.

Let's keep it on topic folks. If you want to continue a general discussion about models, you can always start a thread in TT.

Thanks.

edit: Actually, this is the general discussion thread - my bad. Still, let's keep it on topic.

[wxmann_91]

Say I get some guidance from friends about big decisions I make. If I consistently get bad guidance from a particular friend, even if he meant well, but had other friends who gave me good guidance on average, wouldn't logic dictate that I stop listening to the bad guidance from the particular friend? Models shouldn't be taken as gospel, but they need to be weighed according to their performance. Nobody cares what the NAVGEM or CMC or the JMA or the NAM says (and to an even stronger extent, the LBAR, BAMs, etc.). If the GFS performs as badly, why should we weigh it any higher than those models?

Mainly because you are wrong and there are stats to prove it. Euro has a lot of strengths but the GFS has not failed as badly as many make it out, again with statistics to prove it. If you are basing your trust of the GFS out past 120 hours then why not fault the euro for developing the same system a few days ago with a similar solution?

Models are guidance, and the blend is still the best solution. Without the GFS hurricane forecasts would be worse, not better.

I 100% agree with you -- on the *old* GFS, which was serviceable and even outperformed the Euro in a couple of instances. This version of the GFS has performed much worse with regards to tropical cyclones. I'd love to see those stats on the new GFS but it hasn't passed the eye test thus far, and this has affected forecasts... e.g. forecasts of Hilary strengthening to Cat 4 status, JTWC forecasts for Noru to restrengthen to Cat 4 status close to Kyushu, 80% 5-day probs for 99L, forecasts for Greg to strengthen into a hurricane... the list goes on. Sure, the Euro isn't perfect but it's performed much better with regards to those storms.

I'm fine with a blend but I'd put something like 0.01% stock on the latest version of the GFS.

[Frank2]

The latest TWO is hedging away from anything short-term "early or middle of next week" - sometimes the TWO"s are too "optimistic" then have to backpedal when things don't quite go as planned...

[tolakram]

What is the anomaly above PR? Seems to drift down out of nowhere.