I could barely care less because NOAA weather models are garbage. The NOAA GFS and NAM models are basically dinosaurs at this point compared to what Meteoblue is running. I've been tracking this stuff for ten years and it's not even close anymore. GFS still limps along at 13km resolution and only updates four times a day. Like, are we still in 2010? It completely screws up thunderstorm timing and can't handle mountains for shit. NAM isn't much better at 12km resolution, constantly busting rainfall forecasts especially in complex terrain.
Meanwhile, Meteoblue's models are on another level. NEMS cranks at 3km resolution and nails thunderstorms like 30% better than NAM. The German ICON model with its weird icosahedral grid is crushing GFS by 15-25% on wind forecasts. Then there's the European ECMWF model everyone knows it consistently beats GFS by at least half a day in accuracy. AROME is insane for mountain forecasts with resolution down to 1.3km, making NAM look like a joke with 40% better precipitation predictions. The real game-changer though is Meteoblue's multi-model approach, which averages out the errors from individual models and boosts accuracy by 20-35% over the old school GFS/NAM combo. I don't know why anyone still relies on GFS or NAM except for historical consistency or if you need a really broad overview. For anything specific or local, the newer models demolish them.
Can we get a disclosure on your experience and who you work for, if you’re going to weigh in so heavily on important public vs private sector issues like this?
It may be worth mentioning that NEMS and AROME are only available for parts of Europe. I'm willing to believe they are better, but here in Northeastern America I typically find that NAM is the best I have available for 3-day, and GFS is usually slightly better than ECMWF and ICON for long term (although none of the long term are great).
* cut the CDC, specifically the overseas pandemic tracking and response units
* withdrew the CDC Representative in China
* shutdown the PREDICT Program tracking coronavirus and zoonotic diseases
* Eliminated the Global Health Security Office
Basically T-balled Covid-19.
These people slash, burn and dismantle organizations that have successfully prevented disaster, then disaster happens.
We're going to see staggering levels of death and destruction from things like hurricanes because of this wild corrupt foolishness and I'm sure they'll blame it on some black person or woman somewhere, like they did with the bridge collapse and California fires
Meanwhile, Meteoblue's models are on another level. NEMS cranks at 3km resolution and nails thunderstorms like 30% better than NAM. The German ICON model with its weird icosahedral grid is crushing GFS by 15-25% on wind forecasts. Then there's the European ECMWF model everyone knows it consistently beats GFS by at least half a day in accuracy. AROME is insane for mountain forecasts with resolution down to 1.3km, making NAM look like a joke with 40% better precipitation predictions. The real game-changer though is Meteoblue's multi-model approach, which averages out the errors from individual models and boosts accuracy by 20-35% over the old school GFS/NAM combo. I don't know why anyone still relies on GFS or NAM except for historical consistency or if you need a really broad overview. For anything specific or local, the newer models demolish them.