Typhoon Tip Posted February 26, 2016 Share Posted February 26, 2016 I'll take P007 for 156 please ... Link to comment Share on other sites More sharing options...
OceanStWx Posted February 26, 2016 Share Posted February 26, 2016 gotcha - wonder if just the hardware alone might appease some. huh, 4D though? ... I read a long time ago that NCEP did a cross-model experiment between the GFS and Ecm; i think they took the Euro's initialization and fed it to the GFS and the GFS scored a lot better. either that or I have a remarkable talent for science fiction. be that as it may, it'll be fascinating to see what comes out of the May upgrade. wonder why they chose that month? was that deliberate - heh, i could surmise a scenario where they want put it together during the off calendar time of the year like just in case - lol You're correct. The 4DVAR is superior to the current data assimilation. And this upgrade won't be all the way there either, but at least the GFSX scores show significant improvement over the current 3DVAR. I'm guessing late spring was chosen in order to get a full cold season of parallel ops. Link to comment Share on other sites More sharing options...
JC-CT Posted February 26, 2016 Share Posted February 26, 2016 Where do people find the gfsx output? Link to comment Share on other sites More sharing options...
OceanStWx Posted February 26, 2016 Share Posted February 26, 2016 yup...I read the same thing Which is maddening. I mean why do you think Cliff Mass gets so upset about the US handling of NWP? Basically we have a solid model core that we don't invest data assimilation in. Garbage in, garbage out. It's part of the reason why you see wild swings run to run, versus subtle ECMWF movement. Link to comment Share on other sites More sharing options...
OceanStWx Posted February 26, 2016 Share Posted February 26, 2016 Where do people find the gfsx output? Some of the pay sites have it, like Weatherbell, similar to the Euro it's usually listed as parallel. Link to comment Share on other sites More sharing options...
JC-CT Posted February 26, 2016 Share Posted February 26, 2016 Which is maddening. I mean why do you think Cliff Mass gets so upset about the US handling of NWP? Basically we have a solid model core that we don't invest data assimilation in. Garbage in, garbage out. It's part of the reason why you see wild swings run to run, versus subtle ECMWF movement. The navgem is 4dvar, isn't it? But people bash the heck out of that model Link to comment Share on other sites More sharing options...
CoastalWx Posted February 26, 2016 Share Posted February 26, 2016 EC ensembles like the day 7.5 deal that is offshore on the op. We take. Link to comment Share on other sites More sharing options...
ORH_wxman Posted February 26, 2016 Share Posted February 26, 2016 EC ensembles like the day 7.5 deal that is offshore on the op. We take. That's a hefty mean for Mar 4th. Big mean precip too for that far out. Link to comment Share on other sites More sharing options...
OceanStWx Posted February 26, 2016 Share Posted February 26, 2016 The navgem is 4dvar, isn't it? But people bash the heck out of that model Yes it is, but it's also the poorest resolution of the "major" models and thus suffers in the more robust verification statistics (i.e. anomaly correlation since it won't capture the anomalies to the extent higher res models will). There has also not been as much optimization of that model compared to the GFS either. Regardless, a "3.5" DVAR already makes the GFS better. 4DVAR would only further that. That's not to say it improves East Coast cyclogenesis any, but overall the model would get better. Link to comment Share on other sites More sharing options...
JC-CT Posted February 26, 2016 Share Posted February 26, 2016 A nice multi-model compromise should leave us with a moderate cutter followed by a whiff to Bermuda. Link to comment Share on other sites More sharing options...
ORH_wxman Posted February 26, 2016 Share Posted February 26, 2016 EC ensembles are interested in the Mar 7th idea too. Link to comment Share on other sites More sharing options...
Typhoon Tip Posted February 26, 2016 Share Posted February 26, 2016 You're correct. The 4DVAR is superior to the current data assimilation. And this upgrade won't be all the way there either, but at least the GFSX scores show significant improvement over the current 3DVAR. I'm guessing late spring was chosen in order to get a full cold season of parallel ops. yeah that makes sense ... plus, (hahaha) if it suddenly f's up all the time ... no one will notice 'till next October! just kidding Link to comment Share on other sites More sharing options...
JC-CT Posted February 26, 2016 Share Posted February 26, 2016 A nice multi-model compromise should leave us with a moderate cutter followed by a whiff to Bermuda. Yes it is, but it's also the poorest resolution of the "major" models and thus suffers in the more robust verification statistics (i.e. anomaly correlation since it won't capture the anomalies to the extent higher res models will). There has also not been as much optimization of that model compared to the GFS either. Regardless, a "3.5" DVAR already makes the GFS better. 4DVAR would only further that. That's not to say it improves East Coast cyclogenesis any, but overall the model would get better. Thanks. And hey, I just checked our good friend Navgem at 12z and it shows pretty much exactly the scenario I described above. Good thing it sucks Link to comment Share on other sites More sharing options...
Typhoon Tip Posted February 26, 2016 Share Posted February 26, 2016 A nice multi-model compromise should leave us with a moderate cutter followed by a whiff to Bermuda. yep ... may as well just line 'em up for their regularly scheduled cosmic dildo flogging - Link to comment Share on other sites More sharing options...
MJO812 Posted February 26, 2016 Share Posted February 26, 2016 Euro mean is east of the op for the cuttet It is also more northwest with the coastal . Euro control crushes the coast. Link to comment Share on other sites More sharing options...
OceanStWx Posted February 26, 2016 Share Posted February 26, 2016 EC ensembles are interested in the Mar 7th idea too. Improved shuffle towards the benchmark on the 5th too. Link to comment Share on other sites More sharing options...
Typhoon Tip Posted February 26, 2016 Share Posted February 26, 2016 One thing I'm going to be fascinated to see is how the bias spectrum of the GFS may or may not change once all that happens. 'Magine how weird that would be if it suddenly started doing the same thing as the ECM ? Link to comment Share on other sites More sharing options...
Typhoon Tip Posted February 26, 2016 Share Posted February 26, 2016 Euro mean is east of the op for the cuttet It is also more northwest with the coastal . Euro control crushes the coast. know what - I'm embarrassed to say I want to see that solution for my self. why? because - I've heard that stated so many times this winter and it's annoying. "control run crushes" "control run destroys" ... yet, nothing has yet crushed and/or destroyed anything yet - leaving me to ponder what use beyond futility the 'legendary control run' really has. Link to comment Share on other sites More sharing options...
JC-CT Posted February 26, 2016 Share Posted February 26, 2016 One thing I'm going to be fascinated to see is how the bias spectrum of the GFS may or may not change once all that happens. 'Magine how weird that would be if it suddenly started doing the same thing as the ECM ? I would not at all be surprised if that happened. Look at the JMA, another 4dvar model. It follows the Euro, like, a lot. know what - I'm embarrassed to say I want to see that solution for my self. why? because - I've heard that stated so many times this winter and it's annoying. "control run crushes" "control run destroys" ... yet, nothing has yet crushed and/or destroyed anything yet - leaving me to ponder what use beyond futility the 'legendary control run' really has. You know what its purpose is, stop pretending. Link to comment Share on other sites More sharing options...
Typhoon Tip Posted February 26, 2016 Share Posted February 26, 2016 basically, the higher you lift any object the more potential energy it stores... same thing here: the higher these runs manage to lift your hopes and spirits, the greater amount they get to claim in sadness and rage when all that good vibe gets instantly converted to kinetic madness as you fall from great great heights and expectations. don't buy into anything 'positive' looking - how many times must we collectively be beaten over the heads with model lies before we begin to just stop listening to them. hahaaa Link to comment Share on other sites More sharing options...
CoastalWx Posted February 26, 2016 Share Posted February 26, 2016 Improved shuffle towards the benchmark on the 5th too. Would you say having the EC ensemble improves grids beyond day 4 or 5? Wonder if there is justification for seeing those for you guys. Link to comment Share on other sites More sharing options...
Typhoon Tip Posted February 26, 2016 Share Posted February 26, 2016 I would not at all be surprised if that happened. Look at the JMA, another 4dvar model. It follows the Euro, like, a lot. You know what its purpose is, stop pretending. oh i'm not pretend jack - haha, exactly my point. Link to comment Share on other sites More sharing options...
dendrite Posted February 26, 2016 Share Posted February 26, 2016 Euro mean is east of the op for the cuttet It is also more northwest with the coastal . Euro control crushes the coast. It takes it over Lake Ontario like the op. Link to comment Share on other sites More sharing options...
JC-CT Posted February 26, 2016 Share Posted February 26, 2016 oh i'm not pretend jack - haha, exactly my point.The ensemble members run off of it? Link to comment Share on other sites More sharing options...
OceanStWx Posted February 26, 2016 Share Posted February 26, 2016 Would you say having the EC ensemble improves grids beyond day 4 or 5? Wonder if there is justification for seeing those for you guys. Totally useful for me and X to look at. But the unfortunate part is that we have to take what we see on the web and move the grids towards that, instead of having an EPS grid to blend into the forecast. That would be ideal. Link to comment Share on other sites More sharing options...
CoastalWx Posted February 26, 2016 Share Posted February 26, 2016 Totally useful for me and X to look at. But the unfortunate part is that we have to take what we see on the web and move the grids towards that, instead of having an EPS grid to blend into the forecast. That would be ideal. I hear ya. Just trying to see if you can justify that or at least not paying for a vendor to see it lol. Link to comment Share on other sites More sharing options...
CoastalWx Posted February 26, 2016 Share Posted February 26, 2016 I'm interested late next week. Feels like it's not a clown threat for now. Link to comment Share on other sites More sharing options...
JC-CT Posted February 26, 2016 Share Posted February 26, 2016 I'm interested late next week. Feels like it's not a clown threat for now. Thursday or Saturday? Saturday I'm guessing Link to comment Share on other sites More sharing options...
OceanStWx Posted February 26, 2016 Share Posted February 26, 2016 I hear ya. Just trying to see if you can justify that or at least not paying for a vendor to see it lol. I can almost guarantee we're never getting the EPS locally. Nationally WPC may be able to keep it in their system, but even that's not rock solid. Sadly the WeatherBell subscription might be one of the best uses of the office budget in a long time give the amount we use it. Link to comment Share on other sites More sharing options...
OceanStWx Posted February 26, 2016 Share Posted February 26, 2016 I'm interested late next week. Feels like it's not a clown threat for now. Yeah, it's looking like we have to suffer the cutter early in the week to carve out a seasonably deep trough for the next couple of threats. Link to comment Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.