[HN Gopher] Project Aardvark: reimagining AI weather prediction
       ___________________________________________________________________
        
       Project Aardvark: reimagining AI weather prediction
        
       Author : bentobean
       Score  : 282 points
       Date   : 2025-03-23 23:33 UTC (23 hours ago)
        
 (HTM) web link (www.turing.ac.uk)
 (TXT) w3m dump (www.turing.ac.uk)
        
       | kgwgk wrote:
       | Paper available here:
       | https://www.nature.com/articles/s41586-025-08897-0
        
         | abdullahkhalids wrote:
         | Arxiv version here https://arxiv.org/pdf/2404.00411
        
           | foofoo55 wrote:
           | Why are the Arxiv and Nature versions so different, even the
           | text?
        
             | abdullahkhalids wrote:
             | Its quite common to revise papers. For example, they might
             | have uploaded to arxiv in order to submit to a conference.
             | Later, they revised and submitted to Nature.
        
             | kkylin wrote:
             | As another comment mentioned, papers get revised during
             | review, usually in response to reviewer comments. Also,
             | some journals (not sure about Nature) do not allow authors
             | to "backport" revisions made in response to reviewer
             | comments to preprints; I guess they view the review process
             | as part of their "value add".
        
             | jamala1 wrote:
             | Arxiv is mostly meant for preprints for peer review.
             | 
             | In a Nature paper in particular, the final layout is
             | typically done by the journal's professional production
             | team, not the authors.
             | 
             | Not all publishers grant permission for authors to upload
             | the peer-reviewed and layouted postprints elsewhere.
        
           | benob wrote:
           | According to arxiv paper, code will be made available here:
           | https://github.com/annavaughan/aardvark-weather-public (from
           | 8 months ago)
        
             | sunshinesnacks wrote:
             | The Nature preprint references this zenodo archive with
             | data and code in a big 13GB .tar file:
             | https://doi.org/10.5281/zenodo.13158382
             | 
             | I haven't downloaded it to see what's in it.
        
       | dboreham wrote:
       | Although based on their bios and UK population statistics, it
       | seems likely they're both English, the article omits this
       | information. They do work at a British institution, located in
       | England..
        
         | _joel wrote:
         | The name Turing and turing.ac.uk should maybe give it away. I
         | don't think the authors' nationalities are of importance,
         | however.
        
       | abdullahkhalids wrote:
       | How robust will ML models be in the world of rapid climate
       | change, where the past no longer predicts the future?
        
         | zdragnar wrote:
         | Climate is not weather.
         | 
         | Weather predictions are for specific events and areas, made on
         | the order of days- typically no more than 2-3 weeks into the
         | future.
         | 
         | Climate models predict future averages over large regions on
         | the order of decades.
         | 
         | "rapid climate change" is on the order of "within this
         | century". Whether the climate changes or not doesn't really
         | impact the weather models at all, because the climate's not
         | changing on the same time scale.
        
           | labster wrote:
           | While I really want to agree with your sound argument, I have
           | this suspicion that climate change significantly affected
           | forecast skill on Hurricane Otis.
        
             | melagonster wrote:
             | Climate change: more typhoons in the future.
             | 
             | Weather prediction: which path typhoons will follow?
        
             | zdragnar wrote:
             | The climate didn't change in the time from Otis forming to
             | landing.
             | 
             | Weather models look at current conditions and extrapolate
             | what the next set of conditions will be.
             | 
             | Another way to look at it is that climate change models
             | reflect what we will experience. Weather models reflect the
             | mechanics of humidity, temperature and air flow- given a
             | current state, what is the next likely state. Climate
             | change doesn't change how those mechanics interact. It only
             | guesses what "current state" will look like in the far
             | future.
        
               | avianlyric wrote:
               | Climate change doesn't impact the physics of weather, but
               | our understanding of the inputs to a weather system is
               | still very incomplete. Climate change does impact the
               | relative importance of different physical inputs into a
               | weather system, and inherently pushes weather systems
               | away from the most studied and well understood aspects of
               | weather physics. That in turn degrades the accuracy of
               | our models, as it takes time to research and understand
               | whats changed in a weather system, and what previously
               | unimportant inputs have become much more important.
        
               | zdragnar wrote:
               | Let's say I live in the hypothetical state of Midwestia.
               | It gets up to 100 degrees F in the summer and down to -40
               | in the winter months.
               | 
               | Over the course of ten years, the average daily
               | temperature goes up by 1 degree F (which is many times
               | faster than reality).
               | 
               | That means that there's basically one day in the entire
               | year that doesn't fit within the existing model, or maybe
               | a handful if some other part of the year is cooler than
               | normal. We see more variation between normal years than
               | we expect to see average increase in both the short and
               | medium terms.
               | 
               | Our weather models predict no more than two or three
               | weeks into the future.
               | 
               | There's literally nothing about climate change that will
               | change how weather systems interact that will occur
               | faster than our weather models will adapt to. Our weather
               | patterns may change drastically, but that's a
               | fundamentally different problem.
        
               | avianlyric wrote:
               | How quickly do you think our weather models adapt? Our
               | current models are built on top of decades of detailed
               | weather research by tens of thousands of researchers
               | around the world, and has involved the launch of some the
               | largest satellites in space. All to help us try and
               | understand how weather works.
               | 
               | Those models don't just "adapt", they're carefully and
               | manually refined over time, slowly incorporating better
               | physical models as they're developed, and better methods
               | of collecting data as it become available. Climate change
               | absolutely changes how weather systems interact faster
               | than our models can adapt, primarily because we can only
               | adapt models to changes _after_ their accuracy degrades,
               | and it becomes possible to start identifying potential
               | weaknesses in the existing approaches.
               | 
               | You're example of seasonal temperature changes vs global
               | average temperature increases, vastly underestimates how
               | complicated weather models are, and how much impact
               | changes in global climate systems (like the North
               | Atlantic Current) can drastically impact local weather
               | conditions. Weather models, like all models, will make
               | assumptions about the behaviours of extremely large
               | systems which aren't expected to change very much, or are
               | very hard to measure in real-time. As those large systems
               | change due to climate change, the assumptions made by
               | weather models will grow increasingly incorrect. But that
               | hard part is figuring our _which_ assumption is now
               | incorrect, not always easy to identity exactly what
               | assumptions have been made, or accurately measure the
               | difference between the assumption and reality.
        
           | danaris wrote:
           | But when climate _changes_ even by what appear, to us, to be
           | very small amounts, that can have huge effects on how weather
           | systems behave.
           | 
           | Thus, if you want your AI model that predicts _weather_ ,
           | which was trained on data before that change, to predict
           | weather _after_ that change, you may very well find that it
           | rapidly loses accuracy.
        
             | jvanderbot wrote:
             | I think the naive answer is that physics doesn't change
             | when the climate does, just the initial conditions for the
             | day's simulation of the next week's weather.
             | 
             | And every year over year change is within error bars of the
             | prior years results, once you account for those initial
             | conditions.
        
               | danaris wrote:
               | And if your model is based on physics, then that's fine.
               | 
               | But if your model is an LLM (or close cousin) purely
               | based on previous weather patterns, then it may struggle
               | to accurately predict once the underlying conditions
               | change.
        
               | geysersam wrote:
               | The less naive answer is that a typical weather model has
               | hundreds of parameters that have been tuned manually over
               | decades of use.
        
           | abdullahkhalids wrote:
           | One of the characteristics of the present climate change on
           | Earth is that weather events (storms, heat waves, maximum and
           | minimum temperatures, start of seasons) display a lot more
           | variability, and consequently also a lot more extremes than
           | in the recent past. This is already happening.
           | 
           | If you train a weather model ML on past weather data, which
           | has reduced variability, and use it to predict future
           | weather, it is possible that it under predicts variability of
           | weather variables (amount of expected rainfall, maximum
           | temperature tomorrow etc). Variance not mean.
           | 
           | I say "it is possible" because only rigorous modeling can
           | tell us the truth. Which is why I asked the question
           | originally. I don't think it can be denied by abstract
           | arguments.
        
         | labster wrote:
         | AI is not just pattern matching. Assuming they have some idea
         | of initial and boundary conditions, they should be fine.
        
         | crm9125 wrote:
         | "Tomorrow will be partly cloudy, with a chance of hurricane."
        
         | KennyBlanken wrote:
         | I wish HNers would ask themselves "If I, someone with nothing
         | more than layman's knowledge, thought of this - would someone
         | who has spent years studying in their field, possibly the head
         | of their lab, the people they collaborated with, the people who
         | approved their grant, and the people who reviewed the paper for
         | publication - think of this?"
         | 
         | I can hear the keyboard mashing already. YES, _I am aware there
         | are problems in both research and publishing_.
         | 
         | That does not mean that a HN'ers "did they think of X?" is any
         | more valid. It's like saying that because bridges collapse, we
         | should be allowed people with no relevant experience,
         | knowledge, or training to look at new bridges and say "that bit
         | there doesn't look strong enough to me!"
        
           | throwup238 wrote:
           | I wish HNers would stop and ask themselves "Do HNers need to
           | read yet another misguided Dunning-Kruger rant from someone
           | who clearly misread the parent comment?"
           | 
           | The GP asked _ _how_ _ robust the ML models will be, which is
           | a perfectly good question to ask. Maybe a climate scientist
           | specializing in ML can answer that question.
        
           | RicoElectrico wrote:
           | If you know about a particular field and read HN comments
           | it's even worse.
        
             | genewitch wrote:
             | Murray Gell-Mann says "et tu?"
        
           | graemep wrote:
           | On the other hands experts do need to be questions and
           | provide explanations.
           | 
           | "It's like saying that because bridges collapse, we should be
           | allowed people with no relevant experience, knowledge, or
           | training to look at new bridges and say "that bit there
           | doesn't look strong enough to me!""
           | 
           | Not a valid comparison. Civil engineering is a field with a
           | lot of known answers. bridge designs rarely rely on cutting
           | edge research.
        
         | interludead wrote:
         | Honestly one of the biggest open questions in applying ML to
         | weather and climate
        
       | lenerdenator wrote:
       | Hmmmmmmm.
       | 
       | I have a challenge for the model:
       | 
       | Accurate (within 3deg F) weather predictions for the Kansas City
       | metro more than two days out. As of 2024 these were rarely
       | accurate.[0]
       | 
       | [0]https://www.washingtonpost.com/climate-
       | environment/interacti...
        
         | lytedev wrote:
         | Indeed we have it tough out here. I like the excitement,
         | though, I must say! Funny to find the sentiment here on HN,
         | though!
        
         | metaphor wrote:
         | > _Accurate (within 3deg F) weather predictions_
         | 
         | Sniff test: ASTM E230 standard tolerances for the venerable
         | Type K thermocouple is +/- 2.2degC or +/- 0.75%, whichever is
         | greater.
         | 
         | Expectations are in need of recalibration if anyone thinks a
         | single number is going to meaningfully achieve that level of
         | accuracy across a volume representing any metro area at any
         | given point in time, let alone two days out.
        
           | genewitch wrote:
           | I think you misunderstood the assignment. The temp
           | predictions in my area wildly swing around, until the day of,
           | and even then I have to adjust for my actual location. And
           | I'm not in Kansas City. The only place that has ~3deg
           | accuracy is San Diego. And maybe Antarctica.
        
           | IshKebab wrote:
           | The bigger the volume the _more_ accurate temperature you
           | should be able to get.
           | 
           | Do you think 2C global warming is irrelevant because a single
           | thermocouple couldn't accurately measure it?
        
           | kwertzzz wrote:
           | I am not sure if type K thermocouple are used for
           | meteorological air temperature measurements.
           | 
           | These sensors (based on thermal resistance) for example have
           | an accuracy of 0.2 degC under typical conditions [1].
           | 
           | [1] https://www.ti.com/lit/ds/symlink/tmp1826.pdf?ts=17427997
           | 638...
        
           | zipy124 wrote:
           | That is the accuracy of one sensor. The law of large numbers
           | comes into play when using forecasts such as this, since
           | accross any reasonably sized metro area you will have
           | hundreds if not thousands or even tens of thousands of
           | weather stations, accross which you can average to bring down
           | error thresholds.
        
         | baq wrote:
         | What are your use cases for such level of accuracy on such long
         | time frames?
        
           | genewitch wrote:
           | Is it safe to put my seedlings outside or is there going to
           | be a soft freeze that kills them all in two days.
        
             | bongodongobob wrote:
             | Then use a thermometer. A city is a large area and can have
             | variances of +/-10 degrees depending on where you are in
             | the city.
        
               | genewitch wrote:
               | a thermometer will tell me if it's going to freeze in 2
               | days?
               | 
               | here's the scenario. I have started a bunch of seeds
               | inside my house. I am waiting for the "last frost" as per
               | the instructions on the seed packets. Now, how do i use a
               | thermometer to tell if the last frost has passed? You
               | need prediction models that can be accurate out at least
               | a few days with temperature, and a 10% error "at
               | freezing" means my plants either live or die, based on
               | that error.
               | 
               | there's no counterargument, here. "oh just cover the new
               | plants" or "just wait longer" don't work, especially with
               | larger gardens or "farms" or "homesteads". Most of us
               | home-gamers just use environmental clues - number of
               | bugs, buds on dogwood or pecan trees (native ones), when
               | other trees flower. But this year i got a lesson that
               | pecan trees get it wrong, too. They are in the process of
               | leafing out and there was an _unpredicted_ , non-forecast
               | cold snap that took the temperature so close to 32.0F
               | (0.00006C) that i think any plant not equipped for that
               | would have died that night. Now, _now_ i 'm fairly
               | certain there's no more frost chance, but it's just a
               | guess.
               | 
               | Now imagine i lived anywhere outside of the subtropics,
               | like nebraska or montana and needed to plant food for
               | livestock or whatever.
        
               | bongodongobob wrote:
               | Forecasting isn't fine grained enough to get your
               | backyard temp within 2 degrees. What your asking for is
               | silly. The temp across your city is going to vary by more
               | than 2 degrees every single day.
        
               | genewitch wrote:
               | you're still not listening. Once more, the forecast can't
               | tell me _with any certainty_ whether or not there is
               | going to be a frost, specifically in my area, and
               | apparently in Kansas City, as well.
               | 
               | What i am asking for is what is promised by weather
               | forecasters and the models. If it can't say with any
               | certainty if it's going to freeze, it's completely
               | worthless _for this common circumstance_. Like most
               | people, i don 't care if the forecast is 75F and it's
               | actually 70F or 80F (or 68F), but what i do care about is
               | a forecast for lows of 50F and it ends up being 32.15F.
               | If you were a roofer and you were off by 18 degrees,
               | you'd still be in prison.
        
         | Joker_vD wrote:
         | Well, take the actual weather observations for the past year.
         | Take the actual weather observations for the last week. Overlay
         | and slide this week of observations on top of the observations
         | of the past year, until you find the window that matches "the
         | best" -- then take the day right after this window and predict
         | that the weather will be just like that (or maybe try to tweak
         | the values a bit).
         | 
         | I wonder how poorly this thing operates, and whether taking
         | several years of history to look at would help much.
        
           | counters wrote:
           | Extremely poorly, because forecasting the weather is all
           | about forecasting the deviations from the expected seasonal
           | patterns - the "eddies" in the atmospheric flow which give
           | rise to storm systems and interesting, impactful weather.
        
         | bongodongobob wrote:
         | Where in the city?
        
         | kubav027 wrote:
         | According to paper model grid resolution is 1.5 degrees. I do
         | not think it can predict accurate weather in any location. It
         | show global weather trends.
        
           | scellus wrote:
           | It's lower than many other medium-range AI forecasts, but
           | note that those other models get state-of-the-art with pretty
           | coarse grids, 0.5deg or so. The point is that upper
           | atmosphere and broad patterns are smooth, so with ML/AI they
           | don't require high resolution (while simulating them with
           | physical models does require). And at the forecast lag of say
           | 5-10 days, all local detail is lost anyway, so what skill
           | remains comes from broad patterns, in all models. (Some extra
           | skill can be gained by running local models initialized with
           | the broad patterns, for there are clear cases like mountains
           | where fine resolution is useful.)
        
           | counters wrote:
           | 1.5 degrees is perfectly fine for predicting large-scale
           | (synoptic) weather patterns. They're not just "global
           | trends." But yes, typical global NWP models and their MLWP
           | competitors are run at 0.25 degrees or finer. All forecasts
           | are statistically post-processed and biased-corrected to
           | create local forecasts.
        
       | MostlyStable wrote:
       | I'm curious if some future, hypothetical AGI agent, which had
       | been trained to have these kinds of abilities, would be akin to
       | how most humans see a ball in flight and just instinctively know
       | (within reason) where the ball is going to go? We don't
       | understand, consciously and in the moment, how our brain is doing
       | these calculations (although obviously we can get to similar
       | results with other methods), but we still trust the outputs.
       | 
       | Would some hypothetical future AI just "know" that tomorrow it's
       | going to be 79 with 7 mph winds, without understanding exactly
       | how that knowledge was arrived at?
        
         | defrost wrote:
         | > would be akin to how most humans see a ball in flight and
         | just instinctively know (within reason) where the ball is going
         | to go?
         | 
         | Up to a point .. and that point is more or less the same as the
         | point where humans can no longer catch a spinning tennis
         | raquet.
         | 
         | We understand the gravitional rainbow arc of the centre of
         | mass, we fail at predicting the low order chaotic spin of
         | tennis raquet mass distributions.
         | 
         | Other butterflies are more unpredictable, and the ones that
         | land on a camels back breaking a dam of precariously balanced
         | rocks are a particular problem.
         | 
         | * https://en.wikipedia.org/wiki/Tennis_racket_theorem
         | 
         | * Dzhanibekov effect demonstration in microgravity:
         | https://www.youtube.com/watch?v=1x5UiwEEvpQ
         | 
         | * https://en.wikipedia.org/wiki/Horseshoe_map
        
           | MostlyStable wrote:
           | Yes, humans are obviously limited in the things we can
           | instinctively, intuitively predict. That's not really the
           | point. The point is whether something that has been trained
           | to do more complicated predictions will have the a similar
           | feeling when doing those predictions (of being intuitive and
           | natural), or if it will feel more explicit, like when a human
           | is doing the calculus necessary to predict where the same
           | ball is going to go.
        
             | defrost wrote:
             | Humans have both intuition and explicit calculation,
             | predictive calculation can be stable or inherently
             | unstable.
             | 
             | The point is whether a LLM has any _feelings_ ...
        
               | MostlyStable wrote:
               | My phrase "future, hypothetical" was trying very
               | specifically to avoid any discussions about whether
               | current AI have qualia or internal experiences. I was
               | trying to think about whether something which did have
               | some kind of coherent internal experience (assumed for
               | the sake of the idle thought), but which had far greater
               | predictive abilities than humans, would have the same
               | intuitive feeling when making those much more complicated
               | predictions.
               | 
               | It was an idle thought that was only barely tangentially
               | related to the article in question, and was not meant to
               | be a comment at all on the model in the article or on any
               | current (or even very near future) AI model.
               | 
               | I don't expect anyone would have an answer, given the
               | extreme degree of hypothetical-ness.
        
               | criddell wrote:
               | When you say "The point is whether a LLM has any
               | feelings..." are you thinking specifically about an LLM,
               | or AI in general? I've seen nothing to indicate that
               | Project Aardvark is using an LLM for weather prediction.
        
         | avianlyric wrote:
         | I think "chain-of-thought" LLMs, with access to other tools
         | like Python, already demonstrate two types of "thinking".
         | 
         | We can query an LLM a simple question like "how many 'r' are in
         | the word strawberry", and an LLM with know access to tools will
         | quite confidently, and likely incorrectly, give you an answer.
         | There's no actual counting happening, and any kind of
         | understanding of the problem, the LLM will just guess an answer
         | based on its training data. But that answer tends to be wrong,
         | because those types of queries don't make up a large portion of
         | its training set, and if they do, there's a large body of
         | similar queries with vastly different answers, which ultimately
         | results in confidentiality incorrect outputs.
         | 
         | But provide an LLM tools like Python, and a "chain-of-thought"
         | prompt that allows it to recursively re-prompt itself, while
         | also executing external code and viewing the outputs, and an
         | LLM can easily get the correct answer to query "how many 'r'
         | are in the word strawberry". By simply writing and executing
         | some Python to compute the answer.
         | 
         | Those two approaches to problem solving are strikingly similar
         | to intuitive vs analytical thinking in humans. One approach is
         | driven entirely by pattern matching, and breaks down when
         | dealing with problems that require close attention to specific
         | details, the other is much more accurate, but also slower
         | because directed computation is required.
         | 
         | As for your hypothetical "weather AI", I think it's pretty easy
         | to imagine an AGI capable of confidently predicting the weather
         | tomorrow, not be capable of understanding how it computed the
         | prediction, beyond a high level hand wavy explanation. Again,
         | that's basically what LLM do today, confidently make
         | predictions of the future, with zero understanding of how or
         | why they made those predictions. But you can happily ask an LLM
         | how and why it made a prediction, and it'll give you a very
         | convincing answer, that will also be a complete and total
         | deception.
        
         | klabb3 wrote:
         | > would be akin to how most humans see a ball in flight and
         | just instinctively know (within reason) where the ball is going
         | to go?
         | 
         | Generally no. If I show you a puddle of water, can you tell me
         | what shape was the ice sculpture that it was melted from?
         | 
         | One is Newtonian motion and the other is a complex chaotic
         | system with sparse measurements of ground truth. You can
         | minimize error propagation but it's a diminishing returns
         | problem, (except in rare cases like for natural disasters where
         | a 6h warning can make a difference).
        
           | mnky9800n wrote:
           | While generally correct there has been evidence that machine
           | learning models can predict multiple Lyapunov times past
           | traditional models.
           | 
           | [1] https://link.aps.org/doi/10.1103/PhysRevLett.120.024102
           | 
           | [2] https://link.aps.org/doi/10.1103/PhysRevResearch.5.043252
        
         | sebastiennight wrote:
         | I remember learning that humans trying to catch a ball are not
         | actually able to predict where the ball will land, but rather,
         | will move in a way that maintains the angle of movement
         | constant.
         | 
         | As a result a human running to catch the ball over some
         | distance (eg during a baseball game) runs along a curved path,
         | not linearly to the point where the ball will drop (which would
         | be evidence of having an intuition of the ball's destination).
        
           | 01HNNWZ0MV43FF wrote:
           | Sounds a bit like proportional navigation from missile
           | guidance
           | https://en.wikipedia.org/wiki/Proportional_navigation
        
           | eszed wrote:
           | This hypothesis could be tested, now that major league
           | baseball tracks the positions of players in games. In the MLB
           | app they show animations of good outfield plays with "catch
           | difficulty" scores assigned, based (in part) on the straight-
           | line distance from the fielder's initial position to the
           | position of the catch. The "routes" on the best catches are
           | always nearly-straight lines, which suggests that high-level
           | players have developed exactly this intuitive sense.
           | 
           | Certainly what I was coached to do, what outfielders say they
           | do, and what I see watching the game, is to "read" the ball,
           | run towards where you think the ball is going, and _then_
           | track the ball on the way. I was and am a shitty outfielder,
           | in part because I never developed a fast-enough intuitive
           | sense of where the ball is going (and because, well, I 'm
           | damn slow), but watch the most famous Catch[1] caught on
           | film, and it sure looks like Mays knew right away that ball
           | was hit over his head.
           | 
           | [1] https://m.youtube.com/watch?v=7bLt2xKaNH0
        
           | spiderfarmer wrote:
           | Humans are definitely able to predict where a ball wil land.
           | https://www.youtube.com/watch?v=aoScYO2osb0
        
             | pests wrote:
             | Agreed. Reminds me of juggling, while learning I noticed
             | that as long as I could see each ball for at least a split
             | second on its upwards trajectory I could "tell" if it would
             | be a good throw or not. In order to keep both hands/paths
             | in my view I would stare basically straight forward and not
             | look at the top of the arc and could do it at any height.
             | Now I can do it much more with feel and the motion is
             | muscle memory but the visual cues were my main teacher.
        
             | aredox wrote:
             | It makes sense that there are several heuristics. After
             | all, "Thinking: Fast and Slow" already makes the point that
             | human brains have several layers of processing with
             | different advantages and drawbacks depending on situations.
        
           | jampekka wrote:
           | There are a few that kind of theories. You are probably
           | referring to the Optical Acceleration Cancellation theory[1].
           | There are some similar later so called "direct perception"
           | theories too.
           | 
           | The problem with these is that they don't really work, often
           | even in theory. People do seem to predict at least some
           | aspects of the trajectory, although not necessarily the whole
           | trajectory [2].
           | 
           | [1] https://pubs.aip.org/aapt/ajp/article-
           | abstract/36/10/868/104...
           | 
           | [2]
           | https://royalsocietypublishing.org/doi/10.1098/rsos.241291
        
         | nsm wrote:
         | To quote Iain M. Banks, probably not :)
         | 
         | > "Sma," the ship said finally, with a hint of what might have
         | been frustration in its voice, "I'm the smartest thing for a
         | hundred light years radius, and by a factor of about a million
         | ... but even I can't predict where a snooker ball's going to
         | end up after more than six collisions." [GCU Arbitrary in "The
         | State of the Art"]
        
           | Vecr wrote:
           | 6? That can't be right. I don't know how big a GCU is, so the
           | scale could be up to 1 OOM off, but a full redirection of all
           | simulation capacity should let it integrate out further than
           | that.
        
             | Fuzzwah wrote:
             | The target precision wasn't specified.
        
             | myrmidon wrote:
             | For ball-to-ball collisions, 6 is already a _highly_
             | conservative estimate-- this is basically a chaotic system
             | (outcome after a few iterations, while deterministic, is
             | _extremely_ sensitive to exact starting conditions).
             | 
             | The error scales up exponentially with the number of (ball-
             | to-ball) collisions.
             | 
             | So if the initial ball position is off by "half a pixel"
             | (=> always non-zero) this gets amplified extremely quickly.
             | 
             | Your intuition about the problem is probably distorted by
             | considering/having experienced (less sensitive) ball/wall
             | collisions.
             | 
             | See: https://www.lesswrong.com/posts/JehyrC6W3YTtdxw6S/a-pr
             | imer-o...
        
         | namaria wrote:
         | That's assuming we are actually tracking all the relevant
         | indicators.
        
         | mmazing wrote:
         | > Would some hypothetical future AI just "know" that tomorrow
         | it's going to be 79 with 7 mph winds, without understanding
         | exactly how that knowledge was arrived at?
         | 
         | I think a consciousness with access to a stream of information
         | tends to drown out the noise to see signal, so in those terms,
         | being able to "experience" real-time climate data and
         | "instinctively know" what variable is headed in what direction
         | by filtering out the noise would come naturally.
         | 
         | So, personally, I think the answer is yes. :)
         | 
         | To elaborate a little more - when you think of a typical LLM
         | the answer is definitely no. But, if an AGI is likely comprised
         | of something akin to "many component LLMs", then one part might
         | very well likely have no idea how the information it is
         | receiving was actually determined.
         | 
         | Our brains have MANY substructures in between neuron -> "I",
         | and I think we're going to start seeing/studying a lot of
         | similarities with how our brains are structured at a higher
         | level and where we get real value out of multiple LLM systems
         | working in concert.
        
         | interludead wrote:
         | I think that's actually a pretty good way to frame how these
         | deep learning-based forecasting models might evolve
        
       | IncreasePosts wrote:
       | Why would we need weather models on our desktop, instead of just
       | consuming the output of a far more powerful model?
       | 
       | Unless people start interacting with weather forecast differently
       | in the future it doesn't seem like something that needs
       | personalization
        
         | rafram wrote:
         | Read the article. The point is that it can run on sub-
         | supercomputer hardware, not specifically that it can run on
         | consumer desktops.
        
         | avianlyric wrote:
         | The title has been editorialised. The paper is about building
         | AI weather models on par with current state-of-the-art weather
         | models that require super computers to run. But without the
         | need for the super computers, just a normal desktop computer is
         | enough.
         | 
         | That represents a huge step-change in need compute for accurate
         | weather models, and opens up the possibility of even more
         | accurate models, if the accuracy of these techniques scales
         | with available compute. If you can get state-of-the-art
         | accuracy with a desktop computer, something that normally
         | requires a super-computer, what happens if you run those
         | techniques on a super-computer?
        
           | benob wrote:
           | Didn't read the paper, but it would be cool to apply
           | distillation to these big physics-driven models, in order to
           | simulate their outcome on much smaller models.
        
         | brigandish wrote:
         | Additionally (to the other replies to your question), asking
         | for the weather is just another data point in the huge amount
         | of surveillance occurring. If I can get the weather for my
         | location without accessing a remote server where someone else
         | gets to see my query and probably location, all the better.
         | This provides that possibility.
        
       | rkagerer wrote:
       | Are all the inputs (from buoys, weather balloons, stations, etc)
       | from decades of history stored, as well as past daily forecasts
       | of existing weather models, so that this AI algorithm (and any
       | future new ones) can be run across historic data and compared in
       | terms of performance to existing methods?
       | 
       | Is there a big Clearinghouse for this data?
       | 
       | Kind of like how fintech algos can be run against historic stock
       | market data to evaluate them.
        
         | lgeorget wrote:
         | The World Meteorological Organization has data that national
         | weather services exchange to make forecast. Apart from that,
         | its on a country-by-country basis.
         | 
         | In France for instance Meteo-France has released all of its
         | historical data in January 2024:
         | https://www.data.gouv.fr/fr/organizations/meteo-france/#/dat...
        
           | graemep wrote:
           | But national models go beyond there borders and usually have
           | some modelling outside so the exchanges will cover quite a
           | lot of the world.
           | 
           | For example, the UK's metoffice has a low resolution medium
           | term global model:
           | https://www.metoffice.gov.uk/research/approach/modelling-
           | sys...
        
           | cship2 wrote:
           | How does windy do it? Did they just scrap all the sites or
           | api. Would be great to have global dump every 1 minute of all
           | the countries. Or have it have in a radio broadcast.
        
             | jcd000 wrote:
             | Windy uses openmeteo, which in turn aggregates many weather
             | providers under one roof/api
        
         | mschuster91 wrote:
         | > Is there a big Clearinghouse for this data?
         | 
         | Many commercial and even third party governments rely on the
         | data from NOAA and its archives, on top of that the EU runs its
         | own EUMETSAT fleet of data, plus a ton of national services -
         | unfortunately, the result is there's a looooot of datasets.
         | 
         | NOAA's dataset is public domain [1], EUMETSAT only requires
         | attribution for most of its data [2]. On top of that you got
         | the EU's Climate Data store [3], ECMWF [4], and ECA&D [5].
         | 
         | The service that many private weather services provide is to
         | aggregate and weigh all of the publicly available datasets, and
         | some also add in data from their own ground stations,
         | commercially licensed "realtime" data from governmental
         | services, and their own models as well.
         | 
         | The interesting question is what DOGE will do regarding NOAA -
         | it is increasingly possible that NOAA will shut down, either to
         | be turned into a pay-for-play model, replaced by private
         | services, or just carelessly dropped in its entirety.
         | 
         | [1] https://www.ncei.noaa.gov/archive
         | 
         | [2] https://user.eumetsat.int/resources/user-guides/data-
         | registr...
         | 
         | [3] https://cds.climate.copernicus.eu/
         | 
         | [4] https://www.ecmwf.int/en/forecasts/datasets
         | 
         | [5] https://www.ecad.eu/dailydata/
        
           | Loughla wrote:
           | NOAA saves lives. Cutting that organization would be an
           | absolute slap in the face to rural trump voters who rely on
           | that system for weather alerts.
        
           | sunshinesnacks wrote:
           | > The service that many private weather services provide is
           | to aggregate and weigh all of the publicly available datasets
           | 
           | Yes, and more people need to understand this. Too many people
           | seem to think that commercial services won't be impacted if
           | NOAA stops doing what they do.
        
           | K0balt wrote:
           | NOAA marine forecasts and FAA integration is critical
           | infrastructure for marine transportation and aviation. If
           | they shut that down it will result in direct losses much
           | larger than NOAAs budget, and thousands of lives lost through
           | disaster and workplace accidents.
           | 
           | I sure hope those "smart" people are capable of understanding
           | that.
        
             | mschuster91 wrote:
             | > I sure hope those "smart" people are capable of
             | understanding that.
             | 
             | Well, NOAA can be privatized, sold off to the highest
             | bidder and be fed money from the annual government to
             | provide said critical infrastructure.
             | 
             | In the end, it's always one giant ass grift.
        
         | _joel wrote:
         | > Kind of like how fintech algos can be run against historic
         | stock market data to evaluate them.
         | 
         | Backtesting, that's called.
        
           | axismundi wrote:
           | In weather it's called hindcast
        
             | trillic wrote:
             | afttest
        
         | NitpickLawyer wrote:
         | > Are all the inputs (from buoys, weather balloons, stations,
         | etc) from decades of history stored
         | 
         | I was also thinking about smartphones. They have barometric
         | data, and while it might vary from phone to phone, I'm sure
         | something like a kalman filter + historic data could do
         | something there.
         | 
         | Think about gathering all the data from "stationary" phones,
         | correlate that with weather sat data, and with real "ground
         | truth" weather stations, and then go back 30 - 60 min / a day,
         | and see what comes out.
        
           | Havoc wrote:
           | >smartphones. They have barometric data
           | 
           | For anyone else having a TIL moment: It's apparently for
           | vertical position and supposedly sub meter accurate. o_O
        
             | p_l wrote:
             | Depends on smartphone (and smartwatches). Not all have it,
             | at times it disappeared from brands that had it earlier.
             | 
             | The smartwatch series I use explicitly include it because
             | it's essentially the "all in one" version in a series that
             | had smartwatches designed for aircraft use (including
             | military versions where they serve as _backup cabin
             | pressure warning_ , apparently)
        
               | usrusr wrote:
               | Smartwatches include barometers because they enable
               | elevation measurement far more accurate than than GNSS
               | (GPS and GPS-alikes, they tend to be noticeably less
               | accurate in the vertical than in the horizontal). In
               | particular when both are combined: barometry for high
               | frequency changes (going up a hill) continuously
               | calibrated against a long term average of the GNSS
               | elevation to filter out low frequency changes in
               | barometry (weather influence).
               | 
               | But back to device could weather observation, this would
               | require continuous GNSS, devices don't do that (network
               | location would not be good enough).
        
           | orion138 wrote:
           | Dark Sky used that data for hyper local forecasting...
           | 
           | https://news.ycombinator.com/item?id=22740466
           | 
           | https://www.theverge.com/2015/6/22/8822767/dark-sky-
           | weather-...
        
             | xgulfie wrote:
             | I'm still grumpy about apple buying and shuttering darksky
        
             | thatcat wrote:
             | They stopped using it during later years.
        
             | counters wrote:
             | Not really; it was a gimmick. They used standard forecast
             | post-processing techniques to bias correct global/regional
             | weather models. There is virtually no evidence they
             | actually used device data in this process.
        
           | phillipseamore wrote:
           | "Collecting and processing of barometric data from
           | smartphones for potential use in numerical weather prediction
           | data assimilation"
           | 
           | https://rmets.onlinelibrary.wiley.com/doi/10.1002/met.1805
           | 
           | https://github.com/dmidk/smaps
        
         | TheJoeMan wrote:
         | It would also be nice to have a historical store of the weather
         | predictions and not just instantaneous parameters. But I'm not
         | aware of such record, perhaps because weathermen don't want a
         | record of their (mis)predictions...
        
           | ano-ther wrote:
           | Oh they absolutely do track their predictions and how the
           | different models perform. That's how they keep improving.
           | 
           | This is a lay-person overview which cites some of the in-
           | depth studies: https://ourworldindata.org/weather-forecasts
        
         | interludead wrote:
         | Getting high-quality, harmonized data at global scale can still
         | be a challenge, especially when it comes to observational
         | coverage in developing regions
        
         | wewxjfq wrote:
         | There's climate reanalysis, which combines historical
         | observations with weather models to get clean data of past
         | weather conditions, which is then used by researchers for
         | various purposes. Most notably is ERA5 by ECMWF.
         | 
         | [0] https://en.wikipedia.org/wiki/ECMWF_re-analysis
        
         | serialdev wrote:
         | Is there an equivalent modelling approach for earthquake
         | prediction? A data repository for it widely shared on the same
         | line as the top comment would work
        
         | roelschroeven wrote:
         | What data is used to train this AI? The article doesn't say
         | anything about that (tough I have to admit I didn't read it
         | super carefully). My first thought would be exactly all this
         | historical data, but then you can't use that same data to test
         | the AI's performance. Are different subsets of the available
         | historical data used for training vs testing?
        
           | axismundi wrote:
           | It's most likely ERA5:
           | https://cds.climate.copernicus.eu/datasets/reanalysis-
           | era5-s...
        
             | scellus wrote:
             | No, they say end-to-end, meaning they use raw obsevations.
             | Most or all other medium-range models start with ERA5.
             | 
             | There's a paper from Norway that tried end-to-end, but
             | their results were not spectacular. That's the aim of many
             | though, including ECMWF. Note that ECMWF already has their
             | AIFS in production, so AI weather prediction is pretty
             | mainstream nowadays.
             | 
             | Google has a local nowcast model that uses raw
             | observations, in production, but that's a different genre
             | of forecasting than the medium-range models of Aardvark.
        
               | counters wrote:
               | > Google has a local nowcast model that uses raw
               | observations, in production, but that's a different genre
               | of forecasting than the medium-range models of Aardvark.
               | 
               | It's very clear from the MetNet announcement blog[1] that
               | they require HRRR or other NWP output at runtime.
               | 
               | [1]: https://research.google/blog/metnet-3-a-state-of-
               | the-art-neu...
        
               | wafngar wrote:
               | They train with ERA5 and observations.
        
         | guhidalg wrote:
         | There are several places you can get this data, both historical
         | and in near-realtime.
         | 
         | I work on one of them:
         | https://planetarycomputer.microsoft.com/catalog?filter=weath...
        
         | counters wrote:
         | > Is there a big Clearinghouse for this data?
         | 
         | The short answer is, "no." There are some projects like the
         | "NNJA-AI" project at Brightband[1] which is attempting to
         | create such a clearing house in order to focus research efforts
         | across the community.
         | 
         | [1]: https://www.brightband.com/data/nnja-ai
        
       | westoque wrote:
       | not to hijack this thread but my dad did extensive research in
       | sea breeze and rainfall modeling and he would have loved to see
       | these AI and machine learning advancements in weather prediction.
       | 
       | [0]:
       | https://www.revistascca.unam.mx/atm/index.php/atm/article/do...
       | 
       | [1]: https://wmo.int/about-wmo/awards/international-
       | meteorologica...
        
       | thenthenthen wrote:
       | Another challenge I would add to the list is that weather data is
       | strategic data.
        
       | politelemon wrote:
       | > replaces all of the steps
       | 
       | Unable to tell if this is an exaggeration or if I'm just missing
       | nuance, how would the model replace the data gathering which is
       | listed as step 1.
        
         | sunshinesnacks wrote:
         | I think it's replacing data assimilation. Ingesting observation
         | data from lots of sources. Not replacing observations
         | themselves.
        
       | jedberg wrote:
       | It's a shame they just cut funding for launching the weather
       | balloons and other equipment needed to collect this data.
       | 
       | https://apnews.com/article/weather-forecasts-worsen-doge-tru...
        
         | HumblyTossed wrote:
         | Let's name "they". It was Trump via DOGE.
        
       | echelon wrote:
       | Super smart to name it "Aardvark".
       | 
       | It'll sort first in any alphabetized list. Before GFS, etc.
       | 
       | By the same measure, it's probably smart to name your company or
       | app A-something rather than Z-something.
        
         | desas wrote:
         | People were doing this in the days of printed phone books.
         | There would be an "AAA Plumbers", "ABC Taxis" and so on.
         | 
         | Reputedly, "Apple Computers" coming ahead of "Atari" (Jobs' ex-
         | employer) in the phonebook was one of the reasons for Apple
         | being Apple.
        
       | Onavo wrote:
       | How does it compare to Google's
       | https://www.nature.com/articles/s41586-024-08252-9
        
         | scellus wrote:
         | Google's is initialized with a gridded dataset, ERA5, from
         | ECMWF. Using ERA5 is the current standard here, and ECMWF
         | themselves build on that mostly now. Meanwhile, Aardvark tries
         | to do the same directly from observations.
        
       | bazzargh wrote:
       | When I saw this I thought... "The Turing Institute? Does that
       | still exist?"
       | 
       | https://en.wikipedia.org/wiki/Turing_Institute
       | 
       | There was a previous Turing Institute in Glasgow doing AI
       | research (meaning, back then rules-based systems, but IIRC my
       | professor was doing some work with them on neural networks),
       | which hit the end of the road in 1994. There was some interesting
       | stuff spun out of there, but it's a whole different institute.
        
       | interludead wrote:
       | Curious to see how it handles edge cases though, like hurricanes
       | or rare extreme weather events that aren't well-represented in
       | historical data
        
       | dsr_ wrote:
       | I note that they don't actually say "matches the accuracy of
       | current best models".
       | 
       | Or even "almost as accurate as the current best models".
        
       | herodoturtle wrote:
       | For a second I thought this was related to that old school
       | FogCreek internship programme with the same name.
       | 
       | Fun documentary at the time. Something about 12 weeks with geeks,
       | and jumping out of windows comes to mind.
        
       | noiv wrote:
       | Model data here:
       | https://zenodo.org/records/13158382?token=eyJhbGciOiJIUzUxMi...
       | 
       | ~14GB, needs login, works with some federated accounts.
        
       ___________________________________________________________________
       (page generated 2025-03-24 23:01 UTC)