Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

WeatherNext 2: Our most advanced weather forecasting model

blog.google

Post reply on HN
Print view View original

71–80 of 140 posts

Re: WeatherNext 2: Our most advanced weather forecasting model

#71

Post by OisinMoran »

empath75 wrote: post #65
lampiaio wrote: post #54

Reminds me of a funny WWII story: Kenneth Arrow and his statisticians found that their long-range forecasts were no better than numbers pulled out of a hat. The forecasters agreed and asked their superiors to be relieved of this duty. The reply was: "The Commanding General is well aware that the forecasts are no good. However he needs them for planning purposes."

There is a fairly compelling argument that divination in the ancient world was not a useless waste of time, as is commonly assumed, but that having either a process or a person that can make essentially random choices for them allowed people to make hard, consequential decisions where they might otherwise be paralyzed, especially when the penalty for not acting was worse than making a mistake.

I've also read that a source of randomness like that could help prevent things like over-extracting some land

Re: WeatherNext 2: Our most advanced weather forecasting model

#72

Post by carabiner »

plantain wrote: post #67
bilsbie wrote: post #38

Where can I use this? I’ve been trying to find hyperlocal forecasts like darksky used to be.

Darksky was only ever good marketing.

The UX was great but predictions were terrible. I swear the only people who liked it did so out of confirmation bias, which can affect anyone. Just a week ago here on HN, there were users here claiming Farmer's Almanac was accurate.

Re: WeatherNext 2: Our most advanced weather forecasting model

#74

Post by counters »

jakewins wrote: post #7

Is anyone aware of good sources of higher resolution models? Hourly resolution like this model provides doesn’t help much now that energy markets have moved to 15-min and 5-min resolution.

Sure - you'd simply use a regional, high-resolution model. In some parts of the world, these exist for free (e.g. NOAA runs the HRRR [and soon the RRFS] over CONUS, which is re-run every hour and outputs data on a ~3km grid at up to 15 minute temporal resolution). There exist vendors that will run a custom NWP simulation over a region-of-interest for clients, typically forced by GFS or ECMWF forecasts at the boundaries; some power users of these type of data even have internal teams that will do this. And in this arena are models like StormCast and CorrDiff from NVIDIA - which a few weather companies have white-labeled to replace the NWP models they used to run as mentioned above.

Re: WeatherNext 2: Our most advanced weather forecasting model

#75

Post by culi »

trillic wrote: post #51
bilsbie wrote: post #38

Where can I use this? I’ve been trying to find hyperlocal forecasts like darksky used to be.

The HRRR is VERY good in my opinion. It updates hourly with a 15-minute resolution 18 hours out and hourly 48 hours out. https://rapidrefresh.noaa.gov/hrrr/

HRRR only works for the US though. Windy.com is great for comparing different models. (switcher is in the bottom right hand corner)

https://www.windy.com/?hrrrConus

Also checkout HRDPS model if you're in Canada/northern US

https://www.windy.com/?canHrdps

Re: WeatherNext 2: Our most advanced weather forecasting model

#76

Post by culi »

baxtr wrote: post #47
jachee wrote: post #44

Earlier quoted context omitted.

Apple integrated the hyperlocal darksky stuff into their native Weather app. It had a few growing pains, but it's as good as it ever was, imho.

Agreed. The one thing I’d like them to improve are the precipitation maps though. They just feel awkward and unreliable.

I've been burned by Apple's rain forecast many times causing me to time my bike ride home at the worst possible time

I don't think DarkSky was any better though to be fair. It's just a hard problem

Re: WeatherNext 2: Our most advanced weather forecasting model

#77

Post by rootusrootus »

toephu2 wrote: post #73

I can't wait! So where's the app? could not find 'WeatherNext 2' in the appstore. iOS default weather app is notoriously very inaccurate.

Doesn't the iOS app mostly just channel info directly from other sources like the local gov't weather service? I suppose maybe they tried to put some intelligence into it when they bought Dark Sky. That seems about the time it started trying to predict rainfall in the next few minutes. Which hasn't ever worked for me.

Re: WeatherNext 2: Our most advanced weather forecasting model

#78

Post by culi »

jakewins wrote: post #7

Is anyone aware of good sources of higher resolution models? Hourly resolution like this model provides doesn’t help much now that energy markets have moved to 15-min and 5-min resolution.

You need a premium subscription but Windy.com has a pretty neat API for devs

https://www.windy.com/

In the bottom right hand corner you can switch between different models and it points out their resolution levels

Re: WeatherNext 2: Our most advanced weather forecasting model

#79

Post by counters »

DoctorOetker wrote: post #60
meandthewallaby wrote: post #28

Earlier quoted context omitted.

This is really important: You're not the end user of this product. These types of models are not built for laypeople to access them. You're an end user of a product that may use and process this data, but the CRPS scorecard, for example, should mean nothing to you. This is specifically addressing an under-dispersion problem in traditional ensemble models, due to a limited number (~50) and limited set of perturbed ini…

Sorry to hijack you: I have some questions regarding current weather models: I am personally not interested in predicting the weather as end users expect it, rather I am interested in representative evolutions of wind patterns. I.e. specify some location (say somewhere in the North Sea, or perhaps on mainland Western Europe), and a date (say Nov 12) without specifying a year, and would like to have the wind patterns…

I mean - you don't need any ML for that. Just go grab random samples from a ~30 day window centered on your day of interest over the region of interest from a reanalysis product like ERA5. If the duration of ERA5 isn't sufficient (e.g. you wouldn't expect on average to see events with a >100 year return period given the limited temporal extent of the dataset) then you could take one step further and pull from an equilibrium climate model simulation - some of these are published as part of the CMIP inter-comparison, or you could go to special-built ensembles like the CESM LENS [1]. You could also use a generative climate downscaling model like NVIDIA's Climate-in-a-bottle, but that's almost certainly overkill for your application.

[1]: https://www.cesm.ucar.edu/community-projects/lens

Re: WeatherNext 2: Our most advanced weather forecasting model

#80

Post by cogman10 »

lampiaio wrote: post #54

Reminds me of a funny WWII story: Kenneth Arrow and his statisticians found that their long-range forecasts were no better than numbers pulled out of a hat. The forecasters agreed and asked their superiors to be relieved of this duty. The reply was: "The Commanding General is well aware that the forecasts are no good. However he needs them for planning purposes."

I think it was a stats class where I learned this, but as it turns out bad weather is less common than good weather. To be a fairly accurate weather person, you merely need to say "there will be no precipitation" and you'll be right like 90% of the time anywhere on earth.

What makes that funny is that historically, weather forecasters have been less than 90% accurate.

Now, I will say that today's weather models are pretty dang amazing. The 10 day forecast rarely wrong for me.

Post reply on HN