Ed Lizard Posted June 20, 2011 Share Posted June 20, 2011 I'm not getting in between two red taggers fighting, engaged in spirited academic debate, but the models just seem to always have hints of stuff 10 days or so away this Spring that haven't quite materialized. Scientific proof/documentation- not a bit. Link to comment Share on other sites More sharing options...
am19psu Posted June 20, 2011 Share Posted June 20, 2011 I'm not getting in between two red taggers fighting, engaged in spirited academic debate, but the models just seem to always have hints of stuff 10 days or so away this Spring that haven't quite materialized. Scientific proof/documentation- not a bit. Again, taking any particular model run verbatim at that time range is folly. Using other tools, such as the MJO/shallow water waves/SOI/etc., can help you decide if a particular storm threat is erroneous or not. But to say there is no value or that it is simply producing climatology is wrong as well. Link to comment Share on other sites More sharing options...
Srain Posted June 20, 2011 Share Posted June 20, 2011 Arlene. Is that you...? Link to comment Share on other sites More sharing options...
Ed Lizard Posted June 20, 2011 Share Posted June 20, 2011 Arlene. Is that you...? Just me- I trust nothing GFS past Hour 192. But we have another 10 day storm to look forward too not happening around June 30th. Link to comment Share on other sites More sharing options...
HurricaneJosh Posted June 20, 2011 Author Share Posted June 20, 2011 Just me- I trust nothing GFS past Hour 192. But we have another 10 day storm to look forward too not happening around June 30th. I don't think anyone here is suggesting that a model run showing a cyclone 10 days from now should be "trusted", and that we should count on that exact scenario. No one is that naive. What it does show is that the model is sniffing out a more favorable pattern next week. Link to comment Share on other sites More sharing options...
Ed Lizard Posted June 20, 2011 Share Posted June 20, 2011 I don't think anyone here is suggesting that a model run showing a cyclone 10 days from now should be "trusted", and that we should count on that exact scenario. No one is that naive. What it does show is that the model is sniffing out a more favorable pattern next week. Just waiting for a storm a week (or more) after mid-August and a few weeks of Rainstorms season cancel posts with NAO's and Russian heat waves. Then no more petting the sweaty stuff. Ok, people are confusing me now about Kelvin and MJO phases and the Ukie and the other models. Does Steve's GFS la-la land storm have pattern support from the smart people? It did seem to have some Euro support, so I'm at least glass 1/64th optimistic... Link to comment Share on other sites More sharing options...
wxmx Posted June 20, 2011 Share Posted June 20, 2011 I'm with Adam here. Seeing a hurricane in the 276hr GFS tells me nothing. But sometimes I will say that the GFS may be into something, or say that is out to lunch... based on many other things. Take Steve's 240hr image above. This time I think it might be something to watch. The Euro agrees with the timing and location, it has been a persistent feature around that time in the models, the Euro/GFS/Ukie MJO models show that by that time an MJO wave should be resurfacing into phase 1 or 2, which are favorable phases for NAtl cyclones. Obviously there's a limited confidence at that time range, but it's significant and above climo, IMO Link to comment Share on other sites More sharing options...
VAwxman Posted June 20, 2011 Share Posted June 20, 2011 Again, taking any particular model run verbatim at that time range is folly. Using other tools, such as the MJO/shallow water waves/SOI/etc., can help you decide if a particular storm threat is erroneous or not. But to say there is no value or that it is simply producing climatology is wrong as well. Yep. Models do NOT simply produce climatology. That's not even close to being the case. Link to comment Share on other sites More sharing options...
HurricaneJosh Posted June 20, 2011 Author Share Posted June 20, 2011 Ok, people are confusing me now about Kelvin and MJO phases and the Ukie and the other models. Does Steve's GFS la-la land storm have pattern support from the smart people? It did seem to have some Euro support, so I'm at least glass 1/64th optimistic... My point was simply that Steve understands that a 240-hr model run is not something to "trust". Link to comment Share on other sites More sharing options...
CUmet Posted June 20, 2011 Share Posted June 20, 2011 No, you clearly do not. If you did, you would recognize how much better the GFS has become at sniffing out medium range tropical cyclone threats. There's been quite a bit of online discussion about its increased prowess in the medium range since its upgrade last summer. You're also ignoring the ECMWF ensemble, which has skill at forecasting TC genesis and MJO evolution up to 15 days in advance. Didn't you watch how the ECMWF ensemble performed last summer during PREDICT? Again, taking any particular model run verbatim at that time range is folly. Using other tools, such as the MJO/shallow water waves/SOI/etc., can help you decide if a particular storm threat is erroneous or not. But to say there is no value or that it is simply producing climatology is wrong as well. This. Nobody's saying it's perfect, but using these methods has some skill over climatology at the 2-week time scale. Link to comment Share on other sites More sharing options...
wxmx Posted June 20, 2011 Share Posted June 20, 2011 Arlene. Is that you...? Same day and time but exactly one year before: Link to comment Share on other sites More sharing options...
Ivanhater Posted June 20, 2011 Share Posted June 20, 2011 12z Canadian is interesting around the Yucatan Link to comment Share on other sites More sharing options...
Srain Posted June 20, 2011 Share Posted June 20, 2011 My point was simply that Steve understands that a 240-hr model run is not something to "trust". Bingo. As many of us have been noticing, there are changes in the pattern ahead. We've seen a very persistent Ridge killing any chance for tropical development in the Western Basin. The MJO has been hinting a more favorable setup during that time and we are now seeing that via the guidance. There are some around here that do see the need for longer range forecasting and follow such trends. Link to comment Share on other sites More sharing options...
Ed Lizard Posted June 20, 2011 Share Posted June 20, 2011 I so want a TC befire July 4th so the fireworks will not burn down half of SETX... Link to comment Share on other sites More sharing options...
The Dude Posted June 20, 2011 Share Posted June 20, 2011 Yep. Models do NOT simply produce climatology. That's not even close to being the case. I dunno. That CLP seems to always hug climatology... Link to comment Share on other sites More sharing options...
VAwxman Posted June 20, 2011 Share Posted June 20, 2011 I dunno. That CLP seems to always hug climatology... Thanks for giving me a dose of FAIL. Link to comment Share on other sites More sharing options...
Ed Lizard Posted June 20, 2011 Share Posted June 20, 2011 Thanks for giving me a dose of FAIL. Subtle humor there... Link to comment Share on other sites More sharing options...
HM Posted June 20, 2011 Share Posted June 20, 2011 No, just as a researcher who deals with TC modeling for a few years now. How are those turbulent HBL rolls coming along? Link to comment Share on other sites More sharing options...
turtlehurricane Posted June 20, 2011 Share Posted June 20, 2011 How are those turbulent HBL rolls coming along? They have become helical convective vortices (HCVs), the paper is in internal review and heading for AMS review within a month. It ends up they're not really related to standard PBL rolls at all, they seem to be generated by bouyant forcing combined with the shear profile of the HBL, which leads to a DNA-esque flow structure. They may be important in understanding VHTs, since where HCVs meet the eyewall updraft and turn upwards is where you find the VHTs. Essentially, HCVs may be the mechanism by which high vorticity and low angular momentum air is injected into the eyewall, which is what VHTs are proposed to do. Link to comment Share on other sites More sharing options...
Eyewall2005 Posted June 20, 2011 Share Posted June 20, 2011 Any prog over 84 hrs is highly likely not to verify. Link to comment Share on other sites More sharing options...
wxmx Posted June 20, 2011 Share Posted June 20, 2011 Any prog over 84 hrs is highly likely not to verify. Totally not the point being discussed Link to comment Share on other sites More sharing options...
Eyewall2005 Posted June 20, 2011 Share Posted June 20, 2011 Totally not the point being discussed Just dont understand why some put any belief in 200hr+ model runs.. The skill level that far out is quite poor. Link to comment Share on other sites More sharing options...
wxmx Posted June 20, 2011 Share Posted June 20, 2011 Just dont understand why some put any belief in 200hr+ model runs.. The skill level at that timeframe is quite poor. Just wrong usage on your part. Models in the long range are not for a precise pinpoint of a discrete entity. They give hints on patterns, teleconnections, interactions between discrete entities, etc. You don't use it to forecast a "hurricane in the Bay of Campeche", but rather to upgrade your awareness for "possible enhanced tropical activity in the western Atlantic basin for late June, early July". Link to comment Share on other sites More sharing options...
wxmx Posted June 20, 2011 Share Posted June 20, 2011 OK, weird! Which gives a little support to the undercounting of non US landfalling cyclones for those born in the far E Atlantic Link to comment Share on other sites More sharing options...
Eyewall2005 Posted June 20, 2011 Share Posted June 20, 2011 Just wrong usage on your part. Models in the long range are not for a precise pinpoint of a discrete entity. They give hints on patterns, teleconnections, interactions between discrete entities, etc. You don't use it to forecast a "hurricane in the Bay of Campeche", but rather to upgrade your awareness for "possible enhanced tropical activity in the western Atlantic basin for late June, early July". Understood. I Just dont like to look past 80-120hrs in any model. Link to comment Share on other sites More sharing options...
wxmx Posted June 20, 2011 Share Posted June 20, 2011 Using data from 1966-2010, I expected to see the probability lower for <40W cyclones...and it did lower to 13.2% ... but for cyclones <50W it did lower even more to 11.4% Link to comment Share on other sites More sharing options...
Ed Lizard Posted June 20, 2011 Share Posted June 20, 2011 Quite a few GFS ensemble members 1008 mb or less in the BoC, 3 with a 1004 or below, not bad considering reduced grid resolution of the ensembles. I'm not gonna automatically bah humbug this one, and nobody has actually answered the question about the Ukie MJO forecast asked earlier today referring to the discussion yesterday, and its 5 to 20 minutes to 12Z Euro time. Link to comment Share on other sites More sharing options...
Srain Posted June 20, 2011 Share Posted June 20, 2011 The 12Z Euro continues the trend for the Yucatan area. That models also suggests a robust wave in the Caribbean in the longer range. Link to comment Share on other sites More sharing options...
wxmx Posted June 20, 2011 Share Posted June 20, 2011 Quite a few GFS ensemble members 1008 mb or less in the BoC, 3 with a 1004 or below, not bad considering reduced grid resolution of the ensembles. I'm not gonna automatically bah humbug this one, and nobody has actually answered the question about the Ukie MJO forecast asked earlier today referring to the discussion yesterday, and its 5 to 20 minutes to 12Z Euro time. I actually unintendently responded to it, it's the post just after your question... I think. Link to comment Share on other sites More sharing options...
am19psu Posted June 20, 2011 Share Posted June 20, 2011 Quite a few GFS ensemble members 1008 mb or less in the BoC, 3 with a 1004 or below, not bad considering reduced grid resolution of the ensembles. It's funny you bring up the GEFS. I explicity avoid that model for medium range genesis forecasting. It's using the old GFS model physics on a coarse grid. Until the PREDICT pages start up again (I hope they do), I don't think you can get ECMWF ensemble data for free in the Atlantic. Link to comment Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.