This year, the wild variance in hourly weather reports on my phone has really been something. I attributed it to likely budget cuts as a result of DOGE, but if those forecasts came from Google itself the whole time, all is clear now.
I find it interesting that they quantify the improvement on speed and number of forecast-ed scenarios but lack details on how it results in improved accuracy of the forecast per:
```
WeatherNext 2 can generate forecasts 8x faster and with resolution up to 1-hour. This breakthrough is enabled by a new model that can provide hundreds of possible scenarios.
```
As an end user, all I care is that there's one accurate forecasted scenario.
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.
It feels like real weather AI|Forecast|whatever_you_want_to_call_it is still far, far away. Maybe it's just the consumer aspect of weather apps but I don't feel as if I get any more accurate data now than I did back when my parents turned to the daily weather channel for the forecast. Still a lot of clear days when rain was predicted or the even more dreaded torrential downpour when it was supposed to be sunny and clear.
Obviously all I have is anecdata for what I'm mentioning here but from a consumer perspective I don't feel like these model enhancements are really making average folks feel as if weather is any more understood than it was decades ago.
The forecasts are being actively improved, it's just not an overnight step change.
For example, I have just added rainbow.ai short term precipitation forecast into https://weathergraph.app, and it's the best short term forecast I have ever used - based on radar data + AI prediction based on wind etc.
It sounds simple, but there is surprising complexity even just getting (in fast predicting) the 'ground truth' from the radar data, as each radar is noisy, is updated at a different time, might not work at a time ... so even the "current precipitation according to radars" is not a reading, but a result of ML model.
Anyone know whether we can use this to simulate hurricanes/floods in particular areas, instead of looking at real existing data and helping model an existing hurricane as it's happening? (which is definitely more important and impactful, but the simulation angle is the one I happen to be curious about at the moment).
Like if I wanted to simulate whether something like Hurricane Melissa would've gone through a handful of southern US states, what would the effect have been, from an insurance or resiliency standpoint.
15 years later and still no word from Google if they will use the barometers in Android devices to assimilate surface pressure data. It has been shown that this can improve forecast accuracy. I think IBM may be doing it with their weather apps, but Google/Apple would have dramatically more data available.
Apple even bought Dark Sky, which purported to do this but never released any information - so I doubt they really did do it. And if they did, I doubt Apple continued the practice.
Been waiting a long time to hear Google announce they'll use your barometer to give you a better forecast. Still waiting I guess.
Im pretty deep into this topic and what might be interesting to an outsider is that the leading models like neuralgcm/weathernext 1 before as well as this model now are all trained with a "crps" objective which I haven't seen at all outside of ml weather prediction.
Essentially you add random noise to the inputs and train by minimizing the regular loss (like l1) and at the same time maximizing the difference between 2 members with different random noise initialisations.
I wonder if this will be applied to more traditional genai at some point.
Windy is my favorite weather app. The forecast is usually very good and I can switch underlying models to see how much they disagree (an indication that the weather might be erratic). Plus the wind vector animation is mesmerizing and it's fun to look at all the satellite overlays and webcam feeds.
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."
Googles weather prediction engine is already very good, and the new hurricane model was breathtakingly good this season when tested against actual hurricane paths. Meanwhile, the US Government Global Forecasting System continues to get worse.
For folks who are interested, I suggest checking out "The weather machine: a journey inside the forecast" by Andrew Blum[0]. It's a great read into the history of weather forecasting pre-Covid.
On what geometric surfaces do weather models run? Spheriods? Spheres? Projections on planes? Geoids??
Weather is three-dimensional and I would guess that the difference between sphere and (appropriate) spheroid could impact predictions. It seems possible that, at least for local and hyperlocal forecasts, geoids would be worthwhile. But as you go from plane -> sphere -> spheroid -> geoid, computing resources must increase pretty quickly.
And even if a geoid is used, that doesn't mean the weather user sees a geoid or section of geoid. Every consumer weather application displays a plane, afaict. Maybe nautical or aeronatautical weather maps display spheres?
The German “ICON” model uses a spheroid shape that escapes me right now, but each grid cell is effectively a triangle. Most others are gridded squares.
I've found Google's default Weather app to be quite poor starting around six months ago. Consistently off by two to five degrees.
On the advice of someone here on hackernews I tried out weawow, and though it is a terrible name it is _very_ accurate. So much better and consistent. Love it so far.
I noticed my local weather forecasts from Google search have gotten significantly less accurate these days.
Like, they consistenly called for freezing seasonal overnight lows many weeks before it was remotely probable. You'd get better predictions asking anyone who's lived here a couple years. In fairness, I'm in a region that's notoriously difficult to forecast, but the popular non-Google sources seem to be generating better predictions.
I wonder if the rollout of this new model is related (either occurred and made it worse, or will come and make it better).
I'd love to get some hard data. Are there any sites out there where you can compare past performance of different prediction models at a very localized scale?
I’ve read the article and skimmed the paper, but they don’t seem to say if the model has to be retrained every few days or every now and then. Because if it’s trained on marginals it seems to me that the model essentially learns to forecast weather at each location (a bit like locals do), but weather patterns might change over time so I guess they’ll have to retrain it. The retraining frequency might be a deal breaker if for example you have to train it for 3 wall lock days every, say, 7 days.
29 comments
[ 2.3 ms ] story [ 58.6 ms ] threadhttps://developers.google.com/maps/billing-and-pricing/prici...
``` WeatherNext 2 can generate forecasts 8x faster and with resolution up to 1-hour. This breakthrough is enabled by a new model that can provide hundreds of possible scenarios. ```
As an end user, all I care is that there's one accurate forecasted scenario.
Obviously all I have is anecdata for what I'm mentioning here but from a consumer perspective I don't feel like these model enhancements are really making average folks feel as if weather is any more understood than it was decades ago.
For example, I have just added rainbow.ai short term precipitation forecast into https://weathergraph.app, and it's the best short term forecast I have ever used - based on radar data + AI prediction based on wind etc.
It sounds simple, but there is surprising complexity even just getting (in fast predicting) the 'ground truth' from the radar data, as each radar is noisy, is updated at a different time, might not work at a time ... so even the "current precipitation according to radars" is not a reading, but a result of ML model.
Like if I wanted to simulate whether something like Hurricane Melissa would've gone through a handful of southern US states, what would the effect have been, from an insurance or resiliency standpoint.
Apple even bought Dark Sky, which purported to do this but never released any information - so I doubt they really did do it. And if they did, I doubt Apple continued the practice.
Been waiting a long time to hear Google announce they'll use your barometer to give you a better forecast. Still waiting I guess.
https://arstechnica.com/science/2025/11/googles-new-weather-...
Essentially you add random noise to the inputs and train by minimizing the regular loss (like l1) and at the same time maximizing the difference between 2 members with different random noise initialisations. I wonder if this will be applied to more traditional genai at some point.
This reminds me of variational noise (https://www.cs.toronto.edu/~graves/nips_2011.pdf).
If it is random noise on the input, it would be like many of the SSL methods, e.g. DINO (https://arxiv.org/abs/2104.14294), right?
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."
https://arstechnica.com/science/2025/11/googles-new-weather-...
[0]: https://search.worldcat.org/title/1153659005
Weather is three-dimensional and I would guess that the difference between sphere and (appropriate) spheroid could impact predictions. It seems possible that, at least for local and hyperlocal forecasts, geoids would be worthwhile. But as you go from plane -> sphere -> spheroid -> geoid, computing resources must increase pretty quickly.
And even if a geoid is used, that doesn't mean the weather user sees a geoid or section of geoid. Every consumer weather application displays a plane, afaict. Maybe nautical or aeronatautical weather maps display spheres?
On the advice of someone here on hackernews I tried out weawow, and though it is a terrible name it is _very_ accurate. So much better and consistent. Love it so far.
Like, they consistenly called for freezing seasonal overnight lows many weeks before it was remotely probable. You'd get better predictions asking anyone who's lived here a couple years. In fairness, I'm in a region that's notoriously difficult to forecast, but the popular non-Google sources seem to be generating better predictions.
I wonder if the rollout of this new model is related (either occurred and made it worse, or will come and make it better).
I'd love to get some hard data. Are there any sites out there where you can compare past performance of different prediction models at a very localized scale?