GraphCast: AI model for weather forecasting

https://deepmind.google/discover/blog/graphcast-ai-model-for-faster-and-more-accurate-global-weather-forecasting/

by bretthoerner • 3 years ago

630 290 3 years ago

xnx2• 3 years ago

I continue to be a little confused by the distinction between Google, Google Research and DeepMind. Google Research, had made this announcement about 24-hour forecasting just 2 weeks ago: https://blog.research.google/2023/11/metnet-3-state-of-art-n... (which is also mentioned in the GraphCast announcement from today)

mukara2• 3 years ago

DeepMind recently merged with the Brain team from Google Research to form `Google DeepMind`. It seems this was done to have Google DeepMind focused primarily (only?) on AI research, leaving Google Research to work on other things in more than 20 research areas. Still, some AI research involves both orgs, including MetNet in weather forecasting.

In any case, GraphCast is a 10-day global model, whereas MetNet is a 24-hour regional model, among other differences.

xnx2• 3 years ago

Good explanation. Now that both the 24-hour regional and 10-day global models have been announced in technical/research detail, I supposed there might still be a general blog post about how improved forecasting is when you search for "weather" or check the forecast on Android.

danielmarkbruce4• 3 years ago

Is there a colab example (and/or have they released the models) for MetNet like they have here for GraphCast?

robertlagrant• 3 years ago

This is fascinating:

For inputs, GraphCast requires just two sets of data: the state of the weather 6 hours ago, and the current state of the weather. The model then predicts the weather 6 hours in the future. This process can then be rolled forward in 6-hour increments to provide state-of-the-art forecasts up to 10 days in advance.

broast• 3 years ago

Weather is markovian

hakuseki• 3 years ago

That is not strictly true. The weather at time t0 may affect non-weather phenomena at time t1 (e.g. traffic), which in turn may affect weather at time t2.

Furthermore, a predictive model is not working with a complete picture of the weather, but rather some limited-resolution measurements. So, even ignoring non-weather, there may be local weather phenomena detected at time t0, escaping detection at time t1, but still affecting weather at time t2.

Imanari• 3 years ago

Interesting indeed, only one lagged feature for time series forecasting? I’d imagine that including more lagged inputs would increase performance. Rolling the forecasts forward to get n-step-ahead forecasts is a common approach. I’d be interested in how they mitigated the problem of the errors accumulating/compounding.

Al-Khwarizmi• 3 years ago

I don't know much about weather prediction, but if a model can improve the state of the art only with that data as input, my conclusion is that previous models were crap... or am I missing something?

postalrat• 3 years ago

Read the other comments.

counters12• 3 years ago

It's worth pointing out that "state of the weather" is a little bit hand-wavy. The GraphCast model requires a fully-assimilated 3D atmospheric state - which means you still need to run a full-complexity numerical weather prediction system with a massive amount of inputs to actually get to the starting line for using this forecast tool.

Initializing directly from, say, geostationary and LEO satellite data with complementary surface station observations - skipping the assimilation step entirely - is clearly where this revolution is headed, but it's very important to explicitly note that we're not there yet (even in a research capacity).

baq• 3 years ago

Yeah current models aren’t quite ready to ingest real time noisy data like the actual weather… I hear they go off the rails if preprocessing is skipped (outliers, etc)

lispisok• 3 years ago

I've been following these global ML weather models. The fact they make good forecasts at all was very impressive. What is blowing my mind is how fast they run. It takes hours on giant super computers for numerical weather prediction models to forecast the entire globe. These ML models are taking minutes or seconds. This is potentially huge for operational forecasting.

Weather forecasting has been moving focus towards ensembles to account for uncertainty in forecasts. I see a future of large ensembles of ML models being ran hourly incorporating the latest measurements

wenc2• 3 years ago

Not to take away from the excitement but ML weather prediction builds upon the years of data produced by numerical models on supercomputers. It cannot do anything without that computation and its forecasts are dependent on the quality of that computation. Ensemble models are already used to quantify uncertainty (it’s referenced in their paper).

But it is exciting that they are able to recognize patterns in multi year and produce medium term forecasts.

Some comments here suggest this replaces supercomputers models. This would a wrong conclusion.It does not (the paper explicitly states this). It uses their output as input data.

boxed4• 3 years ago

I don't get this. Surely past and real weather should be the input training data, not the output of numerical modeling?

mnky9800n2• 3 years ago

It uses era5 data which is reanalysis. These models will always need the numerical training data. What's impressive is how well the emulate the physics in those models so cheaply. But since the climate changes there will eventually be different weather in different places.

https://www.ecmwf.int/en/forecasts/documentation-and-support

counters12• 3 years ago

Absolutely - but large ensembles are just the tip of the iceberg. Why bother producing an ensemble when you could just output the posterior distribution of many forecast predictands on a dense grid? One could generate the entire ensemble-derived probabilities from a single forward model run.

Another very cool application could incorporate generative modeling. Inject a bit of uncertainty in a some observations and study how the manifold of forecast outputs changes... ultimately, you could tackle things like studying the sensitivity of forecast uncertainty for, say, a tropical cyclone or nor'easter relative to targeted observations. Imagine a tool where you could optimize where a Global Hawk should drop rawindsondes over the Pacific Ocean to maximally decrease forecast uncertainty for a big winter storm impacting New England...

We may not be able to engineer the weather anytime soon, but in the next few years we may have a new type of crystal ball for anticipating its nuances with far more fidelity than ever before.

kridsdale37• 3 years ago

This is basically equivalent to NVIDIA's DLSS machine learning running on Tensor Cores to "up-res" or "frame-interpolate" the extremely computationally intensive job the traditional GPU rasterizer does to simulate a world.

You could numerically render a 4k scene at 120FPS at extreme cost, or you could render a 2k scene at 60FPS, then feed that to DLSS to get a close-enough approximation of the former at enormous energy and hardware savings.

Gys2• 3 years ago

I live in an area which regularly has a climate differently then forecasted: often less rain and more sunny. Would be great if I can connect my local weather station (and/or its history) to some model and have more accurate forecasts.

speps• 3 years ago

Because weather data is interpolated between multiple stations, you wouldn't even need the local station position, your own position would be more accurate as it'd take a lot more parameters into account.

tash93• 3 years ago

One piece of context to note here is that models like ECMWF are used by forecasters as a tool to make predictions - they aren't taken as gospel, just another input.

The global models tend to consistently miss in places that have local weather "quirks" - which is why local forecasters tend to do better than, say, accuweather, where it just posts what the models say.

Local forecasters might have learned over time that, in early Autumn, the models tend to overpredict rain, and so when they give their forecasts, they'll tweak the predictions based on the model tendencies.

dist-epoch3• 3 years ago

There are models which take as input both global forecasts and local ones, and which then can transpose a global forecast into a local one.

National weather institutions sometimes do this, since they don't have the resources to run a massive supercomputer model.

Gys2• 3 years ago

Interesting. So what I am looking for is probably an even more scaled down version? Or something that runs in the cloud with an api to upload my local measurements.

freedomben• 3 years ago

weather prediction seems to me like a terrific use of machine learning aka statistics. The challenge I suppose is in the data. To get perfect predictions you'd need to have a mapping of what conditions were like 6 hours, 12 hours, etc before, and what the various outcomes were, which butterflies flapped their wings and where (this last one is a joke about how hard this data would be). Hard but not impossible. Maybe impossible. I know very little about weather data though. Is there already such a format?

tash93• 3 years ago

It's been a while since I was a grad student but I think the raw station/radiosonde data is interpolated into a grid format before it's put into the standard models.

kridsdale37• 3 years ago

This was also in the article. It splits the sphere surface in to 1M grids (not actually grids in the cartesian sense of a plane, these are radial units). Then there's 37 altitude layers.

So there's radial-coordinate voxels that represent a low resolution of the physical state of the entire atmosphere.

serjester2• 3 years ago

To call this impressive is an understatement. Using a single GPU, outperforms models that run on the world's largest super computers. Completely open sourced - not just model weights. And fairly simple training / input data.

... with the current version being the largest we can practically fit under current engineering constraints, but which have potential to scale much further in the future with greater compute resources and higher resolution data.

I can't wait to see how far other people take this.

thatguysaguy• 3 years ago

They said single TPU machine to be fair, which means like 8 TPUs (still impressive)

wenc2• 3 years ago

It builds on top of supercomputer model output and does better at the specific task of medium term forecasts.

It is a kind of iterative refinement on the data that supercomputers produce — it doesn’t supplant supercomputers. In fact the paper calls out that it has a hard dependency on the output produced by supercomputers.

carbocation• 3 years ago

I don't understand why this is downvoted. This is a classic thing to do with deep learning: take something that has a solution that is expensive to compute, and then train a deep learning model from that. And along the way, your model might yield improvements, too, and you can layer in additional features, interpolate at finer-grained resolution, etc. If nothing else, the forward pass in a deep learning model is almost certainly way faster than simulating the next step in a numerical simulation, but there is room for improvement as they show here. Doesn't invalidate the input data!

silveraxe932• 3 years ago

Could you point me to the part where it says it depends on supercomputer output?

I didn't read the paper but the linked post seems to say otherwise? It mentions it used the supercomputer output to impute data during training. But for prediction it just needs:

For inputs, GraphCast requires just two sets of data: the state of the weather 6 hours ago, and the current state of the weather. The model then predicts the weather 6 hours in the future. This process can then be rolled forward in 6-hour increments to provide state-of-the-art forecasts up to 10 days in advance.

whatever1• 3 years ago

So best case scenario we can avoid some computation for inference, assuming that historical system dynamics are still valid. This model needs to be constantly monitored by full scale simulations and rectified over time.

westurner• 3 years ago

"BLD,ENH: Dask-scheduler (SLURM,)," https://github.com/NOAA-EMC/global-workflow/issues/796

Dask-jobqueue https://jobqueue.dask.org/ :

provides cluster managers for PBS, SLURM, LSF, SGE and other [HPC supercomputer] resource managers

Helpful tools for this work: Dask-labextension, DaskML, CuPY, SymPy's lambdify(), Parquet, Arrow

GFS: Global Forecast System: https://en.wikipedia.org/wiki/Global_Forecast_System

TIL about Raspberry-NOAA and pywws in researching and summarizing for a comment on "Nrsc5: Receive NRSC-5 digital radio stations using an RTL-SDR dongle" (2023) https://news.ycombinator.com/item?id=38158091

pkulak• 3 years ago

Why can't they just train on historical data?

meteo-jeff7• 3 years ago

In case someone is looking for historical weather data for ML training and prediction, I created an open-source weather API which continuously archives weather data.

Using past and forecast data from multiple numerical weather models can be combined using ML to achieve better forecast skill than any individual model. Because each model is physically bound, the resulting ML model should be stable.

See: https://open-meteo.com

boxed4• 3 years ago

Open-Meteo has a great API too. I used it to build my iOS weather app Frej (open source and free: https://github.com/boxed/frej)

It was super easy and the responses are very fast.

mdbmdb2• 3 years ago

Is it able to provide data on extreme events. Say, the current and potential path of a hurricane? similar to .kml that NOAA provides

meteo-jeff7• 3 years ago

Extreme weather is predicted by numerical weather models. Correctly representing hurricanes has driven development on the NOAA GFS model for centuries.

Open-Meteo focuses on providing access to weather data for single locations or small areas. If you look at data for coastal areas, forecast and past weather data will show severe winds. Storm tracks or maps are not available, but might be implemented in the future.

comment_ran2• 3 years ago

How about https://pirateweather.net/en/latest/ ?

Does anyone have a compare this API with the latest API we have here?

meteo-jeff7• 3 years ago

Both APIs use weather models from NOAA GFS and HRRR, providing accurate forecasts in North America. HRRR updates every hour, capturing recent showers and storms in the upcoming hours. PirateWeather gained popularity last year as a replacement for the Dark Sky API when Dark Sky servers were shut down.

With Open-Meteo, I'm working to integrate more weather models, offering access not only to current forecasts but also past data. For Europe and South-East Asia, high-resolution models from 7 different weather services improve forecast accuracy compared to global models. The data covers not only common weather variables like temperature, wind, and precipitation but also includes information on wind at higher altitudes, solar radiation forecasts, and soil properties.

Using custom compression methods, large historical weather datasets like ERA5 are compressed from 20 TB to 4 TB, making them accessible through a time-series API. All data is stored in local files; no database set-up required. If you're interested in creating your own weather API, Docker images are provided, and you can download open data from NOAA GFS or other weather models.

Fatnino• 3 years ago

Is there somewhere to see historical forecasts?

So not "the weather on 25 December 2022 was such and such" but rather "on 20 December 2022 the forecast for 25 December 2022 was such and such"

meteo-jeff7• 3 years ago

Not yet, but I am working towards it: https://github.com/open-meteo/open-meteo/issues/206

berniedurfee• 3 years ago

I’ve always wanted to see something like that. I always wonder if forecasts are a coin flip beyond a window of a few hours.

jjp• 3 years ago

Are you thinking something like https://www.forecastadvisor.com/?

Vagantem4• 3 years ago

That’s awesome! I’ve hooked something similar up to my service - https://dropory.com which predicts which day it will rain the least for any location

Based on historical data!

polygamous_bat2• 3 years ago

Yikes, after completed three steps I was asked for my email. No to your bait and switch, thanks!

willsmith722• 3 years ago

this is really cool, I've been looking for good snow-related weather APIs for my business. I tried looking on the site, but how does it work, being coordinates-based?

I'm used to working with different weather stations, e.g. seeing different snowfall prediction at the bottom of a mountain, halfway up, and at the top, where the coordinates are quite similar.

ryanlitalien• 3 years ago

You'll need a local weather expert to assist, as terrain, geography and other hyper-local factors create forecasting unpredictability. For example, Jay Peak in VT has its own weather, the road in has no snow, but it's a raging snowstorm on the mountain.

just_testing• 3 years ago

I was going to ask about air quality, but just opened the site and you have air quality as well! Thanks!

_visgean7• 3 years ago

There is also https://github.com/google-research/weatherbench2 which has baselines of numerical weather models.

Omnipresent• 3 years ago

This is great. I am very curious about the architectural decisions you've taken here. Is there a blog post / article about them? 80 yrs of historical data -- are you storing that somewhere in PG and the APIs are just fetching it? If so, what indices have you set up to make APIs fetch faster etc. I just fetched 1960 to 2022 in about 12 secs.

meteo-jeff7• 3 years ago

Traditional database systems struggle to handle gridded data efficiently. Using PG with time-based indices is memory and storage extensive. It works well for a limited number of locations, but global weather models at 9-12 km resolution have 4 to 6 million grid-cells.

I am exploiting on the homogeneity of gridded data. In a 2D field, calculating the data position for a graphical coordinate is straightforward. Once you add time as a third dimension, you can pick any timestamp at any point on earth. To optimize read speed, all time steps are stored sequentially on disk in a rotated/transposed OLAP cube.

Although the data now consists of millions of floating-point values without accompanying attributes like timestamps or geographical coordinates, the storage requirements are still high. Open-Meteo chunks data into small portions, each covering 10 locations and 2 weeks of data. Each block is individually compressed using an optimized compression scheme.

While this process isn't groundbreaking and is supported by file systems like NetCDF, Zarr, or HDF5, the challenge lies in efficiently working with multiple weather models and updating data with each new weather model run every few hours.

You can find more information here: https://openmeteo.substack.com/i/64601201/how-data-are-store...

caseyf7• 3 years ago

How did you handle missing data? I’ve used NOAA data a few times and I’m always surprised at how many days of historical data are missing. They have also stopped recording in certain locations and then start in new locations over time making it hard to get solid historical weather information.

Guestmodinfo• 3 years ago

I always suspect that they don't tell me the actual temperature. Maybe I am totally wrong but I suspect. I need to get my own physical thermometer not the digital one in my room and outside my house and have a camera focussed on it. So that later I can speed up the video and see how much the weather varied the previous night.

kubiton• 3 years ago

What? Why?

3abiton• 3 years ago

Are multiple data sources supported?

brna2• 3 years ago

Hi Jeff, Great work, Respect!

I just hit the daily limit on the second request at https://climate-api.open-meteo.com/v1/climate

I see the limit for non-commercial use should be "less than 10.000 daily API calls". Technically 2 is less than 10.000, I know, but still I decided to drop you a comment. :)

wodenokoto• 3 years ago

10.000 requests / (24 hours * 60 minutes * 60 seconds) = 0.11 requests / second

or 1 request every ~9 seconds.

Maybe you just didn't space them enough.

tomaskafka2• 3 years ago

I confirm, open-meteo is awesome and has a great API (and API playground!). And is the only source I know to offer 2 weeks of hourly forecasts (I understand at that point they are more likely to just show a general trend, but it still looks spectacular).

It's a pleasure being able to use it in https://weathergraph.app

brahbrah• 3 years ago

And is the only source I know to offer 2 weeks of hourly forecasts

Enjoy the data directly from the source producing them.

American weather agency: https://www.nco.ncep.noaa.gov/pmb/products/gfs/

European weather agency: https://www.ecmwf.int/en/forecasts/datasets/open-data

The data’s not necessarily east to work with, but it’s all there, and you get all the forecast ensembles (potential forecasted weather paths) too

aaarrm• 3 years ago

This is awesome. I was trying to do a weather project a while ago, but couldn't find an API to suit my needs for the life of me. It looks like yours still doesn't have exactly everything I'd want but it still has plenty. Mainly UV index is something I've been trying to find wide historical data for, but it seems like it just might not be out there. I do see you have solar radiation, so I wonder if I could calculate it using that data. But I believe UV index also takes into account things like local air pollution and ozone forecast as well.

amluto3• 3 years ago

I've never studied weather forecasting, but I can't say I'm surprised. All of these models, AFAICT, are based on the "state" of the weather, but "state" deserves massive scare quotes: it's a bunch of 2D fields (wind speed, pressure, etc) -- note the 2D. Actual weather dynamics happen in three dimensions, and three dimensional land features, buildings, etc as well as gnarly 2D surface phenomena (ocean surface temperature, ground surface temperature, etc) surely have strong effects.

On top of this, surely the actual observations that feed into the model are terrible -- they come from weather stations, sounding rockets, balloons, radar, etc, none of which seem likely to be especially accurate in all locations. Except that, where a weather station exists, the output of that station is the observation that people care about -- unless you're in an airplane, you don't personally care about the geopotential, but you do care about how windy it is, what the temperature and humidity are, and how much precipitation there is.

ISTM these dynamics ought to be better captured by learning them from actual observations than from trying to map physics both ways onto the rather limited datasets that are available. And a trained model could also learn about the idiosyncrasies of the observation and the extra bits of forcing (buildings, etc) that simply are not captured by the inputs.

(Heck, my personal in-my-head neural network can learn a mapping from NWS forecasts to NWS observations later in the same day that seems better than what the NWS itself produces. Surely someone could train a very simple model that takes NWS forecasts as inputs and produces its estimates of NWS observations during the forecast period as outputs, thus handling things like "the NWS consistently underestimates the daily high temperature at such-and-such location during a summer heat wave.")

Difwif• 3 years ago

I'm not sure why you're emphasizing that weather forecasting is just 2D fields. Even in the article they mention GraphCast predicts multiple data points at each global location across a variety of altitudes. All existing global computational forecast models work the same way. They're all 3d spherical coordinate systems.

amluto3• 3 years ago

See page three, table 1 of the paper. The model has 48 2D fields, on a grid, where the grid is a spherical thing wrapped around the surface of the Earth.

There is not what I would call a 3D spherical coordinate system. There’s no field f defined as f(theta, phi, r) — ther are 48 fields that are functions of theta and phi.

WhitneyLand• 3 years ago

How does it make sense to say this is something you’ve “never studied”, followed by how they “ought to be” doing it better?

It also seems like some of your facts differ from theirs, may I ask how far you read into the paper?

kridsdale37• 3 years ago

No need, they're a software engineer (presumably). That just means they're better than everyone.

amluto3• 3 years ago

I read a decent amount of the paper, although not the specific details of the model they used. And when I say I "never studied" it, I mean that I never took a class or read a textbook. I do, in fact, know something about physics and fluids, and I have even personally done some fluid simulation work.

There are perfectly good models for weather in an abstract sense: Navier-Stokes plus various chemical models plus heat transfer plus radiation plus however you feel like modeling the effect of the ground and the ocean surface. (Or use Navier-Stokes for the ocean too!)

But this is wildly impractical. The Earth is too big. The relevant distance and time scales are pretty short, and the resulting grid would be too large. Not to mention that we have no way of actually measuring the whole atmosphere or even large sections of it in its full 3D glory in anything remotely close to the necessary amount of detail.

Go read the Wikipedia article, and contemplate the "Computation" and "Parameterization" sections. This works, but it's horrible. It's doing something akin to making an effective theory (the model actually solved) out of a larger theory (Navier-Stokes+), but we can't even measure the fields in the effective theory. We might want to model a handful of fields at 0.25 degrees (of lat/long) resolution, but we're getting the data from a detailed vertical slice every time someone launches a weather balloon. Which happens quite frequently, but not continuously and not at 0.25 degree spatial increments.

Hence my point: Google's model is sort of learning an effective theory instead of developing one from first principles based on the laws of physics and chemistry.

edit: I once worked in a fluid dynamics lab on something that was a bit analogous. My part of the lab was characterizing actual experiments (burning liquids and mixing of gas jets). Another group was trying to simulate related systems on supercomputers. (This was a while ago. The supercomputers were not very capable by modern standards.)

The simulation side used a 3D grid fine enough (hopefully) to capture the relevant dynamics but not so fine that the simulation would never finish. Meanwhile, we measured everything in 1D 2D! We took pictures and videos with cameras at various wavelengths. We injected things into the fluids for better visualization. We measured the actual velocity at one location (with decent temporal resolution) and hoped our instrumentation for that didn’t mess up the experiment too much. We tried to arrange to know the pressure field in the experiment by setting it up right.

With the goal of understanding the phenomena, I think this was the right approach. But if we just wanted to predict future frames of video from past frames, I would expect a nice ML model to work better. (Well, I would expect it to work better now. The state of the art was not so great at the time.)

cryptoz• 3 years ago

Again haha! Still no mention of using barometers in phones. Maybe some day.

EricLeer3• 3 years ago

The weather company claims to do this (they are also the main provider of weather data for apple).

user_7832• 3 years ago

(If someone with knowledge or experience can chime in, please feel free.)

To the best of my knowledge, poor weather (especially wind shear/microbursts) are one of the most dangerous things possible in aviation. Is there any chance, or plans, to implement this in the current weather radars in planes?

tash93• 3 years ago

If you're talking about small scale phenomena (less than 1km), then this wouldn't help other than to be able to signal when the conditions are such that these phenomena are more likely to happen.

jauntywundrkind2• 3 years ago

From what I can tell from reading & based off https://colab.research.google.com/github/deepmind/graphcast/... , one needs access to ECMWF Era5 or HRES data-sets or something similar to be able to run and use this model.

Unknown what licensing options ECMWF offers for Era5, but to use this model in any live fashion, I think one is probably going to need a small fortune. Maybe some other dataset can be adapted (likely at great pain)...

sunshinesnacks• 3 years ago

ERA5 is free. The API is a bit slow.

I think that only some variables from the HRES are free, but not 100% sure.

hokkos• 3 years ago

The API is unusably slow, the only way is to use the AWS, GCP or Azure mirrors, but they miss a lot of variables and are updated sparingly or with a delay.

jauntywundrkind2• 3 years ago

I created an account on ECMWF but I still dont have access to the ERA5 page, just a big permissions denied message. :/

Any pointers?

_visgean7• 3 years ago

You can get some of the historical data also from here: https://cloud.google.com/storage/docs/public-datasets/era5 (if the official API is too slow. )

To use the data in live fashion I think you would need to get license from ECMWF...

sagarpatil• 3 years ago

How does one go about hosting this and using this as an API?

syntaxing• 3 years ago

Maybe I missed it but does anyone know what it will take to run this model? Seems something fun to try out but not sure if 24GB of VRAM is suffice.

kridsdale37• 3 years ago

It says in the article that it runs on Google's tensor units. So, go down to your nearest Google data center, dodge security, and grab one. Then escape the cops.

azeirah• 3 years ago

You could also just buy a very large amount of their coral consumer TPUs :D

comment_ran2• 3 years ago

So for a daily user, to make it a practical usage, let's say if I have a local measurement of X, I can predict, let's say, 10 days later, or even just tomorrow, or the day after tomorrow, let's say the wind direction, is it possible to do that?

If it is possible, then I will try using the sensor to measure my velocity at some place where I live, and I can run the model and see how the results look like. I don't know if it's going to accurately predict the future or within a 10% error bar range.

dist-epoch3• 3 years ago

No, this model uses as input the current state of the weather across the whole planet.

carabiner2• 3 years ago

GraphCast makes forecasts at the high resolution of 0.25 degrees longitude/latitude (28km x 28km at the equator).

Any way to run this at even higher resolution, like 1 km? Could this resolve terrain forced effects like lenticular clouds on mountain tops?

dist-epoch3• 3 years ago

One big problem is input weather data. It's resolution is poor.

carabiner2• 3 years ago

Yeah, not to mention trying to validate results. Unless we grid install weather stations every 200 m on a mountain top...

max_3• 3 years ago

I have far more respect for the AI team at DeepMind even thou they may be less popular than say OpenAI or "Grok".

Why? Other AI studios seem to work on gimmicks while DeepMind seems to work on genuinely useful AI applications [0].

Thanks for the good work!

[0] Not to say that Chat GPT & Midjourney are not useful, I just find DeepMinds quality of research more interesting.

max_3• 3 years ago

Has anyone here heard of "Numerical Forecasting" models for weather? I heard they "work so well".

Does GraphCast come close to them?

max_3• 3 years ago

What's the difference between a "Graph Neural Network" and a deep neural network?

dil8• 3 years ago

Graph neural networks are deep learning models that trained on graph data.

RandomWorker• 3 years ago

Do you have any resources where I could learn more about these networks?

haolez• 3 years ago

Are there any experts around that can chime in on the possible impacts of this technology if widely adopted?

supdudesupdude4• 3 years ago

It doesnt predict rainfall so i doubt most of us will actually care about it until then. Still it depends on input data (the current state of weather etc). How are we supposed to accurately model the weather at every point in the world? Especially when tech bro Joe living in San Fran expects things to be accurate to a meter within his doorstep

counters12• 3 years ago

GraphCast does predict rainfall - see https://charts.ecmwf.int/products/graphcast_medium-rain-acc?... for example.

_visgean7• 3 years ago

It will get adopted, eventually we will have more accurate weather forecasts. Thats good for anything that depends on weather - e.g. energy consumption and production, transportation costs...

miserableuse2• 3 years ago

Does anybody know if its possible to initialize the model using GFS initial conditions used for the GFS HRES model? If so, where can I find this file and how can I use it? Any help would be greatly appreciated!

counters12• 3 years ago

You can try, but other models in this class have struggled when initialized using model states pulled from other analysis systems.

ECMWF publishes a tool that can help bootstrap simple inference runs with different AI models [1] (they have plugins for several). You could write a tool that re-maps a GDAS analysis to "look like" ERA-5 or IFS analysis, and then try feeding it into GraphCast. But YMMV if the integration is stable or not - models like PanguWx do not work off-the-shelf with this approach.

[1] https://github.com/ecmwf-lab/ai-models

miserableuse2• 3 years ago

Thank you for your response. Are these ML models initialized by gridded initial conditions measurements (such as the GDAS pointed out) or by NWP model forecast results (such as hour-zero forecast from the GFS)? Or are those one and the same?

pyb4• 3 years ago

Curious. How can AI/ML perform on a problem that is, as far as I understand, inherently chaotic / unpredictable ? It sounds like a fundamental contradiction to me.

vosper• 3 years ago

Weather isn’t fundamentally unpredictable. We predict weather with a fairly high degree of accuracy (for most practical uses), and the accuracy getting better all the time.

https://scijinks.gov/forecast-reliability

sosodev• 3 years ago

I'm kinda surprised that this government science website doesn't seem to link sources. I'd like to read the research to understand how they're measuring the accuracy.

keule2• 3 years ago

IMO a chaotic system will not allow for long-term forecast, but if there is any type of pattern to recognize (and I would assume there are plenty), an AI/ML model should be able to create short-term prediction with high accuracy.

pyb4• 3 years ago

Not an expert, but "Up to 10 days in advance" sounds like long-term to me ?

kouru2253• 3 years ago

But AI/ML models require good data and the issue with chaotic systems like weather is that we don’t have good enough data.

kouru2253• 3 years ago

Yes. Very accurate as long as you don’t need to predict the unpredictable. So it’s useless.

Edit: I do see a benefit to the idea if you compare it to the Chaos Theorists “gaining intuition” about systems.

pyb4• 3 years ago

IDK if it's useless, but it's counter-intuitive to me.

crazygringo4• 3 years ago

Because there are tons of parts of weather where chaos isn't the limiting factor currently.

There are a limited number of weather stations producing measurements, and a limited "cell size" for being able to calculate forecasts quickly enough, and geographical factors that aren't perfectly accounted for in models.

AI is able to help substantially with all of these -- from interpolation to computational complexity to geography effects.

simonebrunozzi• 3 years ago

Amazing. Is there an easy way to run this on a local laptop?

dnlkwk2• 3 years ago

Curious how this factors in long-range shifts or patterns eg el nino. Most accurate is a bold claim

stabbles• 3 years ago

If you live in a country where local, short-term rain / shower forecast is essential (like [1] [2]), it's funny to see how incredibly bad radar forecast is.

There are really convenient apps that show an animated map with radar data of rain, historical data + prediction (typically).

The prediction is always completely bonkers.

You can eyeball it better.

No wonder "AI" can improve that. Even linear extrapolation is better.

Yes, local rain prediction is a different thing from global forecasting.

[1] https://www.buienradar.nl [2] https://www.meteoschweiz.admin.ch/service-und-publikationen/...

bberenberg• 3 years ago

Interesting that you say this. I spent in month in AMS 7-8 years ago and buienradar was accurate down to the minute when I used it. Has something changed?

bobviolier• 3 years ago

I don't know how or why, but yes, it has become less accurate over at least the last year or so.

supdudesupdude4• 3 years ago

Funny to mention. None of the AI forecasts can actually predict precip. None of them mention this and i assume everyone thinks this means the rain forecasts are better. Nope just temperature and humidity and wind. Important but come on, it's a bunch of shite

je42• 3 years ago

However, tools like buienrader seem to have trouble in the recent months/years to accurately predict local weather.

brap2• 3 years ago

Beyond the difficulty of running calculations (or even accurately measuring the current state), is there a reason to believe weather is unpredictable?

I would imagine we probably have a solid mathematical model of how weather behaves, so given enough resources to measure and calculate, could you, in theory, predict the daily weather going 10 years into the future? Or is there something inherently “random” there?

danbrooks• 3 years ago

Small changes in initial state can lead to huge changes down the line. See: the butterfly effect or chaos theory.

https://en.wikipedia.org/wiki/Chaos_theory

ethanbond• 3 years ago

AFAIK there's nothing random anywhere except near atomic/subatomic scale. Everything else is just highly chaotic/hard-to-forecast deterministic causal chains.

mesoman2• 3 years ago

Cloud formation is affected by cosmic ray flux. It's effectively random.

But the real problem is chaos - which says that even with perfect data, unless you also have computations with infinite precision and time/spatial/temperature/pressure/etc resolution, eventually you wind up far from reality.

The use of ensembles reduces the effect of chaos a bit, although they tend to smooth it out - so your broad pattern 12 days out might be more accurately forecast than without them, but the weather at your house may not be.

Iterative DL models tend to smooth it faster, according to a recent paper.

counters12• 3 years ago

What you're describing is effectively how climate models work; we run a physical model which solves the equations that govern how the atmosphere works out forward in time for very long time integrations. You get "daily weather" out as far as you choose to run the model.

But this isn't a "weather forecast." Weather forecasting is an initial value problem - you care a great deal about how the weather will evolve from the current atmospheric conditions. Precisely because weather is a result of what happens in this complex, 3D fluid atmosphere surrounding the Earth, it happens that small changes in those initial conditions can have a very big impact on the forecast on relatively short time-periods - as little as 6-12 hours. Small perturbations grow into larger ones and feedback across spatial scales. Ultimately, by day ~3-7, you wind up with a very different atmospheric state than what you'd have if you undid those small changes in the initial conditions.

This is the essence of what "chaos" means in the context of weather prediction; we can't perfectly know the initial conditions we feed into the model, so over some relatively short time, the "model world" will start to look very different than the "real world." Even if we had perfect models - capable of representing all the physics in the atmosphere - we'd still have this issue as long as we had to imperfectly sample the atmosphere for our initial conditions.

So weather isn't inherently "unpredictable." And in fact, by running lots of weather models simultaneously with slightly perturbed initial conditions, we can suss out this uncertainty and improve our estimate of the forecast weather. In fact, this is what's so exciting to meteorologists about the new AI models - they're so much cheaper to run that we can much more effectively explore this uncertainty in initial conditions, which will indirectly lead to improved forecasts.

willsmith722• 3 years ago

is it possible to self-correct, looking at initial value errors in the past? Is it too hard to prescribe the error in the initial value?

brap2• 3 years ago

So isn’t it just a problem of measurement then?

Say you had a massive array of billions of perfect sensors in different locations, and had all the computing power to process this data, would an N year daily forecast then be a solved problem?

For the sake of the argument I’m ignoring ”external” factors that could affect the weather (e.g meteors hitting earth, changes in man-made pollution, etc)

_visgean7• 3 years ago

See https://en.wikipedia.org/wiki/Numerical_weather_prediction

Present understanding is that this chaotic behavior limits accurate forecasts to about 14 days even with accurate input data and a flawless model. In addition, the partial differential equations used in the model need to be supplemented with parameterizations for solar radiation, moist processes (clouds and precipitation), heat exchange, soil, vegetation, surface water, and the effects of terrain.

I think there is a hope that DL models wont have this problem.

supdudesupdude4• 3 years ago

I'll be impressed when it can predict rainfall better than GFS / HRRR / EURO etc

Vagantem4• 3 years ago

Related to this, I built a service that shows what day it has rained the least on in the last 10 years - for any location and month! Perfect to find your perfect wedding date. Feel free to check out :)

https://dropory.com

helloplanets• 3 years ago

Was interested to check this out for Helsinki, but site loads blank on Safari :(

Vagantem4• 3 years ago

Oh, yea spotted now - I’ll have a look as soon as I’m at my computer, will fix. Until then, I think you’ll have to use it on a desktop - thanks for spotting!

knicholes• 3 years ago

What are the similarities between weather forecasting and financial market forecasting?

KRAKRISMOTT• 3 years ago

Both are complex systems traditionally modeled with differential equations and statistics.

sonya-ai• 3 years ago

Well it's a start, but weather forecasting is far more predictable imo

csours• 3 years ago

Makes me wonder how much it would take to do this for a city at something like 100 meter resolution.

layoric• 3 years ago

I can't see any citation to accuracy comparisons, or maybe I just missed them? Given the amount of data, and complexity of the domain, it would be good to see a much more detailed breakdown of their performance vs other models.

My experience in this space is that I was first employee at Solcast building a live 'nowcast' system for 4+ years (left ~2021) targeting solar radiation and cloud opacity initially, but expanding into all aspects of weather, focusing on the use of the newer generation of satellites, but also heavily using NWP models like ECMWF. Last I knew,nowcasts were made in minutes on a decent size cluster of systems, and has been shown in various studies and comparisons to produce extremely accurate data (This article claims 'the best' without links which is weird..), be interesting on how many TPUsv4 were used to produce these forecasts and how quickly? Solcast used ML as a part of their systems, but when it comes down to it, there is a lot more operationally to producing accurate and reliable forecasts, eg it would be arrogant to say the least to switch from something like ECMWF to this black box anytime soon.

Something I said as just before I left Solcast was that their biggest competition would come from Amazon/Google/Microsoft and not other incumbent weather companies. They have some really smart modelers, but its hard to compete with big tech resources. I believe Amazon has been acquiring power usage IoT related companies over the past few years, I can see AI heavily moving into that space as well.. for better or worse.

shmageggy• 3 years ago

I think the paper has what you are looking for. Several figures comparing performance to HRES, and "GraphCast... took roughly four weeks on 32 Cloud TPU v4 devices using batch parallelism. See supplementary materials section 4 for further training details."

alxmrs• 3 years ago

I’m so happy you asked about this! Check out https://sites.research.google/weatherbench/

crazygringo4• 3 years ago

Making progress on weather forecasting is amazing, and it's been interesting to see the big tech companies get into this space.

Apple moved from using The Weather Channel to their own forecasting a year ago [1].

Using AI to produce better weather forecasts is exactly the kind of thing that is right up Google's alley -- I'm very happy to see this, and can't wait for this to get built into our weather apps.

[1] https://en.wikipedia.org/wiki/Weather_(Apple)

blacksmith_tb• 3 years ago

Well, Apple acquired Dark Sky and then shut it down for Android users[1], and then eventually for iOS users as well (but rolled it into the built in weather app, I think).

1: https://www.theverge.com/2020/3/31/21201666/apple-acquires-w...

_visgean7• 3 years ago

Apple moved from using The Weather Channel to their own forecasting a year ago [1].

AFAIK they don't have their own forecasting models, they use same data sources as everyone else: https://support.apple.com/en-us/HT211777

crazygringo4• 3 years ago

Your linked article says they use their own, if you're on a version later than iOS 15.2.

joegibbs• 3 years ago

When will we have enough data that we will be able to apply this to everything? Imagine a model that can predict all kinds of trends - what new consumer good will be the most likely to succeed, where the next war is most likely to break out, who will win the next election, which stocks are going to break out. One gigantic black box with a massive state, with input from everything - planning approvals, social media posts, solar activity, air travel numbers, seismic readings, TV feeds.

drakenot• 3 years ago

Sounds a bit like the premise for the Asimov series, "The Foundation"

hammad93• 3 years ago

I think it's irresponsible to call first on this because it will hinder scientific collaboration. I appreciate this contribution but the journalism was sloppy.

whoislewys_1• 3 years ago

Predicting weather and stock prices don't seem too far apart.

Is it inevitable that all market alpha gets mined by AI?

HereBePandas• 3 years ago

I'd be shocked - given the incentives - if it hasn't already happened to a great extent. Many of the types of people Google DeepMind hires are also the types of people hedge funds hire.

rottc0dd• 3 years ago

How long does this forecasting hold, given butterfly effect et al?

EricLeer3• 3 years ago

I am in the power forecasting domain, where weather forecasts are one of the most important inputs. What I find surprising is that with all the papers and publications from google in the past years, there seems to be no way to get access to these forecasts! We've now evaluated numerous of the ai weather forecasting startups that are popping up everywhere and so far for all of them their claims fall flat on their face when you actually start comparing their quality in a production setting next to the HRES model from ECMWF.

scellus• 3 years ago

GraphCast, Pangu-Weather from Huawei, FourCastNet and EC's own AIFS are available on the ECMWF chart website https://charts.ecmwf.int, click "Machine learning models" on the left tab. (Clicking anything makes the URL very long.)

Some of these forecasts are also downloadable as data, but I don't know whether GraphCast is. Alternatively, if forecasts have a big economic value to you, loading latest ERA5 and the model code, and running it yourself should be relatively trivial? (I'm no expert on this, but I think that is ECMWF's aim, to distribute some of the models and initial states as easily runnable.)

isaacfrond• 3 years ago

I find this quite surprising actually.

You'd think predicting the weather is mostly a matter of fast computation. The physical rules are well understood, so to get a better estimate use a finer mesh in your finite element computation and use a smaller time scale in estimating your differential equations.

Neural networks are notoriously bad at exact approximation. I mean you can never beat a calculator when the issue is doing calculations.

So apparently the AI found some shortcut for doing the actual computational work. That is also surprising as weather is a chaotic system. Shortcuts should not exist.

Long story short, I don't get what's going on here.

uoaei3• 3 years ago

The physical rules are well understood

Nope. They're constantly updating these models with really finnicky things like cloud nucleation rates that differ depending on which tree species's pollen is in the air. They've gotten a lot better (~2 day to ~7 day hi-res forecasts) but they're still wrong a lot of the time. The reason is the chaos as you say, however, chaos is deterministic, so, that a deterministic method can approximate a deterministic system is really not the surprising part.

You don't get what's going on here because your baseline understanding is a lot worse than you think it is.

What they're doing is skipping literal numerical simulation in favor of graph- (attention-) based approaches. Typical weather models simulate pretty fine resolution and return hourly forecasts. Google's new approach is learning an approximate Markov model at 6 hours resolution directly so they don't need to run on massive supercomputers.

flir• 3 years ago

It's a model of a model?

And it turns out to be better?

That's so counter-intuitive I'm kinda amazed anyone even bothered to research it, let alone that it worked.

Uh..... now do horse racing.

karaterobot• 3 years ago

I used to think so too, but evidently weather forecasting is a much harder problem than it seems from the outside. I was talking to a physicist who told me who had first wanted to get into weather modeling, but that it was too hard. I think his quote was something like: "those guys are hard. core."

boxed4• 3 years ago

My grandfather specifically chose meteorology as a field because it had the most numbers heh.

SoftTalker• 3 years ago

Too many variables. The combinatorial complexity exceeds what any computational model can deal with.

giovannibonetti• 3 years ago

AI/ML's bitter lesson [1] applies again. In this case, the AI model may have learned a more practical model than the one human researchers painstakingly came up with by applying piles and piles of physics research.

[1] http://www.incompleteideas.net/IncIdeas/BitterLesson.html

savanaly• 3 years ago

That’s only superficially similar to ai’s bitter lesson. The bitter lesson is about methods to achieve results in AI, not about comparing AI methods to non-AI methods.

uoaei3• 3 years ago

Hardly. They trained on the output of numerical simulations, so it's basically a method for summarizing approximate dynamics of numerical simulations themselves.

amelius2• 3 years ago

by applying piles and piles of physics research.

You mean by remembering piles and piles of example data and interpolating between it.

3cats-in-a-coat• 3 years ago

This is essentially the same exact problem as a classic chess playing program, recursively computing all possibilities N moves ahead, and an AI which "groks" the game's patterns and knows where to focus fewer resources with greater success.

This translates especially well to games like Go, where computing all moves is not even pragmatically possible the classic way. But AI beats the best Go players.

Raw models are excellent for establishing the theory, and for training the AI. But... the AI is better at figuring out more effective, precise, and efficient model within itself, based on both synthetic (based on models) and real data (actual weather patterns).

EDIT: And just to point out, this is not just an AI phenomenon. You are a neural network. And "intuition" is the sense of predicting outcomes you develop, without knowing how and why precisely. This is why I frown upon people with academic knowledge who dismiss people with say engineering or other practical experience in a field. A farmer may not tell you why doing things a weird way results in amazing crop yields, but he gets the gains, and when theory doesn't correlate with reality, it's not reality that's wrong, but the theory.

To recap, nothing beats "learning by example". And AI learns by example. Of course, the formal theoretic models that we can collectively share, explain, and evolve over time have their own strong benefits and have allowed us to grow as a civilization. Computers are in effect "formal computation machines". I don't think we'll run AI for long on digital circuits and it's a clumsy workaround. Computers will have analog processing units for AI and digital processing units for cold, hard logic and data. And the combination is the most powerful approach of all.

sheepshear• 3 years ago

The accuracy improvement boils down to representing more salient features in the model. The humans got a head start figuring out what to model, but the machine figures it out faster, so it caught up and surpassed them. Now it models more important stuff.

The speed difference is a side effect of completely different implementations. One is a step-by-step simulator, the other is an input/output pattern matcher.

mnw21cam2• 3 years ago

I did my doctorate in the Met Office.

Weather forecasting is two separate problems. The first of these is physics - given the state of the atmosphere right now, what will it do. And this is hard, because there are so many different effects, combined with the fact that our computational models have a limited resolution. There's a huge amount of work that goes into making the simulation behave like a real atmosphere does, and a lot of that is faking what is going on at a smaller scale than the model grid.

The second part is to work out what the current state of the atmosphere is. This is what takes vast amounts of computing power. We don't have an observation station at every grid point and at every altitude in the atmospheric model, so we need to find some other way to infer what the atmospheric state is from the observations that we can from it. Many of these observations are limited in locality, like weather stations, or are a complex function of the atmospheric state, like satellite imagery. The light reaching a satellite has been affected by all the layers of the atmosphere it passes through, and sometimes in a highly nonlinear way. In order to calculate the atmospheric state, we need to take the previous forecast of the current atmospheric state, compare it to the observations, then find the first derivative (as in calculus) of the observation function so that we can adjust the atmospheric state estimate to the new best estimate. This is then complicated by the fact that the observations were not all taken at a single time snapshot - for instance polar orbiting satellites will be taking observations spread out in time. So, we need to use the physics model to wind the atmospheric state back in time to when the observation was taken, find the first derivative of that too, and use it to reconcile the observations with the atmospheric state.

It's a massive minimisation/optimisation problem with millions of free variables, and in some cases we need the second derivative of all these functions too in order to make the whole thing converge correctly and within a reasonable amount of time. It takes a reasonable number of iterations of the minimisation algorithm to get it settle on a solution. The problem is that these minimisation methods often assume that the function being minimised is reasonably linear, which certain atmospheric phenomena are not (such as clouds), so certain observations have to be left out of the analysis to avoid the whole thing blowing up.

My doctorate was looking to see if the nonlinearity involved in a cloud forming as air was moving upwards could be used to translate a time-series of satellite infra-red observations into a measurement of vertical air velocity. The answer was that this single form of nonlinearity made the whole minimisation process fairly dire. I implemented a fairly simple not-quite-machine-learning approach, and it was able to find a solution that was almost as accurate but much more reliable than the traditional minimisation method.

Also, to answer the dead sibling comment asking whether weather is really a chaotic system - yes it is. The definition of a chaotic system is that a small change in current state results in a very large change in outcome, and that's definitely the case. The improvements in weather forecasting over the last few decades have been due to improvements in solving both of the above problems - the physics has been pinned down better, but we're also better as working out the current atmospheric state fairly accurately, and that has added something like a day of forecasting accuracy each decade we have been working on it.

sgt101• 3 years ago

Seems like you know what you are talking about!

What's your take on GraphCast - do you see it as a step forward?

londons_explore3• 3 years ago

Shortcuts 100% exist.

Imagine another physical problem. Simulating a sand grain and how it bounces off other sand grains or lodges against them. If you wanted to simulate a sand mountain, you could use a massive amount of compute and predict the location and behaviour of every single grain.

Or, you could take a bunch of well-known shortcuts and just know that sand sits in a heap at the angle-of-repose. That angle decides how steep the mountain will be. any steeper and it will tumble till it's at that angle.

Suddenly, the computation is dramatically reduced, and you get pretty much the same result.

tovej2• 3 years ago

The point was that weather, unlike a sandheap, is a chaotic hydrodynamic system with turbulent flows, that means it's computationally intractable to do exactly, which is why weather forecasts are only good for a few days anyway.

The example you gave does not really explain anything.

twayt• 3 years ago

You get the same result in a short span of time, heck you may even get a reliable error bound.

Where this falls apart is that error accumulates over time and not just for one heap of sand but for many such heaps of sand that also interact with other heaps of sand.

Predicting weather for the next hour is trivial. Aviation runs on the fact that you can forecast fairly accurately into the next hour most of the time.

The difficulty scales superlinearly over time due to the error accumulation over predictions

empath-nirvana• 3 years ago

Neural networks are notoriously bad at exact approximation.

Neural networks can compute pretty much anything. There's no reason, given the same inputs and with enough trainining data that it shouldn't be able to discover the same physical laws that were hard-coded previously.

T-A• 3 years ago

https://www.science.org/content/article/models-galaxies-atom...

https://towardsdatascience.com/physics-informed-neural-netwo...

https://maziarraissi.github.io/PINNs/

https://arxiv.org/abs/2001.08055

https://arxiv.org/abs/2009.11990

amelius2• 3 years ago

It's more like table lookup or interpolation than actual computation.

cowboysauce• 3 years ago

So apparently the AI found some shortcut for doing the actual computational work. That is also surprising as weather is a chaotic system. Shortcuts should not exist.

Why do you say that shortcuts should not exist? Even very basic statements like "falling pressure and increasing humidity indicate a storm is coming" are generally valid. I've done a little bit of storm-chasing and I'm able to point out areas that are likely to experience severe thunderstorms based on a few values (CAPE, dew point, wind shear, etc). I'm sure forecast meteorologists have even better skills. Are those not shortcuts?

greatpostman• 3 years ago

OpenAI is releasing legitimate AGI, google puts out a weather prediction model lol.

dnlkwk2• 3 years ago

I do love Google Maps more than any other product they have lol

lambda_garden• 3 years ago

OpenAI has not released AGI.

postexitus• 3 years ago

You assuming OpenAI's models are AGI tells more about you than anything else.

rational_indian2• 3 years ago

If Alan Turing says ChatGPT is an AGI, it's good enough for me.

tantalor• 3 years ago

One of these things is useful.

mg• 3 years ago

It's interesting, that Google keeps publishing AI research papers. Is there a business rationale behind it?

OpenAI has become one of the fastest growing companies of all time. And much of it is based on Google's "Attention is all you need" and other papers.

Since Microsoft added the Dall-E 3 image creator to Bing, Bing saw a huge inflow of new users. Dall-E is also a technology rooted in Google papers.

I wonder how Google thinks about this internally.

ArnoVW• 3 years ago

It’s difficult to retain top talent if you do not allow them to publish.

DaiPlusPlus• 3 years ago

How does Apple do it, if anyone knows? Apple is so loathe to keep their potential product plans hidden that AAPL employees aren’t even allowed to have GitHub accounts without mgr approval… but they have to be employing serious researchers, but they’ll never get to publish on volition.

jedberg• 3 years ago

For every paper they publish, they have three others that they are keeping to themselves. Publishing papers is a recruiting technique.

matsemann• 3 years ago

How's the distribution of the errors? For instance I don't care if it's better on average by 1 Celsius each day for normal weather, if it once every month is off by 10 Celsius when there is a drastic weather event, for instance.

I'm all for better weather data, it's quite critical up in the mountains, so that's why my question about how reliable it is in life&death situations.

CorrectHorseBat• 3 years ago

https://www.science.org/doi/10.1126/science.adi2336

Seems like it's better at predicting extreme weather events

tony_cannistra3• 3 years ago

Similar methodologies are being applied to climate modeling, too.

The Allen Institute has worked on it for a while, and has hired quite a few PhDs (https://allenai.org/climate-modeling).

lainga• 3 years ago

How long? The cloud microparameterisation looks really exciting, but 10-year stability for a GCM (and "nearly conserving" water) is not great

tony_cannistra3• 3 years ago

I'm not sure. NVIDIA is also working on it (with, interestingly, some of the original AI2 folks).

Similar to the DeepMind effort, the ACE ML model that AI2+others developed is really just looking for parity with physical models at this stage. It looks like they've almost achieved this, with similar massive improvements in compute time + resource needs.

hackernewds2• 3 years ago

why is hiring phds a measure?

Lacerda69• 3 years ago

if you don't hire phds you're not serious about it

tony_cannistra3• 3 years ago

in this particular case, most of the important/needle-moving work being done in climate modeling is done with a hell of a lot of context about prior work. PhDs have that, by necessity.

They're also good at prioritizing outcomes, rather than other stuff.

devindotcom• 3 years ago

yep I just talked to one of the climsim guys and included the project in my writeup of this news:

https://techcrunch.com/2023/11/14/courtesy-of-ai-weather-for...

tokai2• 3 years ago

Just like their flu modelling outperformed conventional models right?

Anon843• 3 years ago

Humm... are you referring to Google Flu? [1]

That was a very different beast. It relied on using Google searches to infer the prevalence of various Influenza Like Illnesses in real time, while the CDC reports data with a 2-week lag. Notably, some of the queries they found to be correlated were... strange... like NBA results.

Not unsurprisingly (in hindsight, at least) [2], this eventually broke down when epidemics and flu symptoms got in the news and completely changed what people were searching for.

[1] https://www.nature.com/articles/nature07634

[2] https://www.science.org/doi/10.1126/science.1248506

tokai2• 3 years ago

Yeah I know its way different methods. Sorry for being disingenuous. The point of my snarking was that google made a lot of noise about Google Flu but then quietly got rid of it when it didn't work. To me Googles research has a tendency to be more about headlines than actually solving problems.

lesuorac• 3 years ago

Notably, some of the queries they found to be correlated were... strange... like NBA results.

Doesn't seem that strange to me.

The presence of a professional sports team in your area is correlated with an increase in flu rates. Getting an ice hockey (NHL) is pretty much the worst.

https://www.upi.com/Health_News/2023/08/08/flu-deaths-sports...

acolderentity• 3 years ago

How could an ai, programmed with the bias of people that already suck at predicting the weather, even get close to being accurate?

david-gpu• 3 years ago

You don't train the AI with the forecasts made by other systems. You train the AI with the actual weather that was measured hours/days later.

jvalencia• 3 years ago

Weather is a complex mix of many systems. The traditional approach is to understand all the systems and add them together. Since we don't understand them all fully, we get a lot of chaos.

The ML algorithm doesn't care about the science, the agendas, the theories, nothing. It just looks for patterns in the data. Instead of an exact calculation it's more akin to numerical analysis. Turns out that looking at the whole in this case, is better than the sum of the parts.

mesoman2• 3 years ago

The people who predict the weather are often damned smart and very experienced.

It's the problem that's hard.

tchvil• 3 years ago

windguru which is in part or fully based on crowd-sourced weather stations is already surprisingly accurate few days in advance, in many regions I tried. For a few hours forecast nothing beats the rain radar. I wonder if they have already or will put some AI in their models.

hackernewds2• 3 years ago

anecdata does not equal data?

hexo• 3 years ago

"AI" aka machine learning

hackitup7• 3 years ago

I've been really impressed at how much better weather forecasting has become already. I remember weather forecasts feeling like a total crapshoot as recently as 15 years ago or so.

patall• 3 years ago

Isn't that highly subjective to where you live? Because I moved to Scandinavia and the forecast here is so incredibly bad, compared to central europe.

tomesco• 3 years ago

Yes, driven by local data collection. More tightly packed ground stations and the availability of atmospheric measurement at various altitudes will improve accuracy.

hutzlibu• 3 years ago

Central europe (minus the alps) is way easier to predict. You can just look at the clouds on a satelite and see how they move, usually west to east, and then extrapolate linearily.

All the fjords and mountains and lakes in norway really make it hard, to precicesly model it. And I think they strongly and chaotically influence the weather in sweden as well.

Also, there are way more people living in central europe, so probably more effort is spend on them.

vodou• 3 years ago

This is partly due to less satellite coverage, something this project is trying to fix: https://www.esa.int/Applications/Observing_the_Earth/Meteoro...

cameronh90• 3 years ago

The accuracy is definitely location dependent, but I anecdotally agree with the GP that the accuracy has improved substantially, at least for the UK where I am.

Ten years ago, the weather forecast was so unreliable that I just assumed anything could happen on a given day, no matter the season. Frequently it would be unable to even tell you whether it was currently raining, and my heuristic for next day forecast instead was to just assume the weather would be the same as today.

Nowadays I find the next day forecasts are nearly always accurate and hourly precipitation forecasts are good enough that I can plan my cycles and walks around them.

eppp2• 3 years ago

It still is. I farm outside of my day job and trying to schedule time to do things like cut hay is sort of a crapshoot. Hay needs a 3-4 day window to dry, rake and roll. This year I got rained on at least twice on days where the NWS showed clear and sunny for 3 days on the spot forecast. 20% or 50% chance of rain is almost useless knowledge. We went for weeks with a 20% chance and it never rained. We still got everything done but it sticks out a lot when you are watching it closely.

TSiege• 3 years ago

Well it's actually not a 20% or 50% chance of rain. It's that it will definitely rain but only for 20-50% of the projected area forecasted

lettergram• 3 years ago

Yup, I had the same issue. Showed 4 days of clear 80-90 degree weather. Twice in that timeframe it rained (!1 inch each time), ruined the cut.

Workaccount2• 3 years ago

If you live in an area with "summer storms" it's basically impossible to forecast anything more than a general area (usually thousands of square miles) that they will appear in.

Its like a shotgun shooting a wall. You can pretty accurately predict the area of the shot, but its incredibly hard to place where exactly each shot in that area will land.

ghaff• 3 years ago

Weird (very local in particular) stuff still happens and tropical weather tracks, for example, can still be pretty unpredictable. But, living in Massachusetts, I still remember how the Blizzard of '78 basically caught everyone by total surprise and left hundreds/thousands(?) of people stranded at work and on highways. Never say never, but it's pretty unlikely you'd see that level of surprise today.

(A friend of mine who moved to the Boston area about ten years after the event once told me that she had never seen a northern city in which so many people headed home from work if they saw so much as a snowflake.)

asdff• 3 years ago

If you live in an area with a lot of microclimates within one city, weather forecasting is honestly no better than astrology.

runemadsen• 3 years ago

I was just telling my wife this after looking up the "no rain" weather report and getting absolutely showered 5 minutes later in an hour-long rain storm. Weather reports suck so much.

drcongo2• 3 years ago

Is there a chance that it just made something up and got lucky like ChatGPT?

Kuinox• 3 years ago

Is there a chance you made something up and got lucky ?

drcongo2• 3 years ago

A huge chance!

alberth• 3 years ago

Is this really an "AI" story?

Aren't existing weather forecasting models, already a form of "AI"?

I'm no AI/ML expert, but isn't the real story here is that a new model (like GPT-4.0) is better than the previous/existing model (GPT-3.5).

It's just grabs way more attention calling the new model "AI" (vs not referring to the old as such).

mdpye• 3 years ago

It's an ML story. The article specifies that the current (now previous?) state of the art models are numerical, crunching vast equations representing atmospheric physics.

surfmike• 3 years ago

No, existing models use more numerical methods. This is using a completely different approach.

GraphCast utilizes what researchers call a "graph neural network" machine-learning architecture, trained on over four decades of ECMWF's historical weather data. It processes the current and six-hour-old global atmospheric states, generating a 10-day forecast in about a minute on a Google TPU v4 cloud computer. Google's machine learning method contrasts with conventional numerical weather prediction methods that rely on supercomputers to process equations based on atmospheric physics, consuming significantly more time and energy.

ramkumarkoppu• 3 years ago

How to translate the graphcast model output to usual weather variables like temperature, rain, wind, etc if I have to build a weather dashboard?

dauertewigkeit• 3 years ago

The multimesh is interesting. Still, I bet the Fourier Neural Operator approach will prove superior. Members of the same team (Sanchez-Gonzales, Battaglia) have already published multiple variations of this model, applied to other physical scenarios and lots of them proved to be dead ends. My money is on the FNO approach, anyway, which for some reason is only given a brief reference. To their credit DeepMind usually publishes extensive comparisons with previously published models. This time such a comparison is conspicuously missing.

Full disclosure: I think DeepMind often publish these bombastic headlines about their models which often don't live up to their hype, or at least that was my personal experience. They have a good PR team, anyway.

kleiba• 3 years ago

How is that a disclosure?

counters12• 3 years ago

Pragmatically speaking, it doesn't really matter if one is better than the other, at least until there is a massive jump in forecast quality (e.g. advancing the Day 5 accuracy up to Day 3). In the real world, we would never take raw model guidance from _any_ source - the best forecasts invariably come from consensus systems that look across many different models. So it's good to have a diverse lineage of forecasting systems, as uncorrelated errors boost the performance of these consensus systems.

thriftwy• 3 years ago

Yandex claims to be using AI-based weather forecasting for a good part of a decade and claims it as a success. It is quite good.

https://meteum.ai/

counters12• 3 years ago

My understanding is that they just use an AI-based precipitation nowcast (see [1]). Very different forecast/modeling problem than GraphCast.

[1] https://arxiv.org/abs/1905.09932

devit• 3 years ago

Seems like it would be much better to do conventional weather forecasting and then feed the predictions along with input data and other relevant information to a machine learning system.

rldjbpin• 3 years ago

GraphCast makes forecasts at the high resolution of 0.25 degrees longitude/latitude (28km x 28km at the equator).

the resolution, while seemingly impressive, is very imprecise compared to the SOTA in the theoretical modelling side.

this discredits the computational claims made by the paper for me. i understand that the current simulations can go down to meter scale, but i wonder what the compuational requirements are when you calculate for this resolution.

GraphCast: AI model for weather forecasting

Related Stories