return to table of content

Weather forecasts have become more accurate

jrockway
71 replies
1d1h

People seem to have different opinions on how good forecasts are. I think it likely depends on which model your forecast source of choice pulls from. I notice that the weather on my Apple Watch corresponds exactly to what GFS says. GFS is OK for medium range, but I don't find it too useful for shorter range. NAM is better for a day or two out. HRRR is better for a few hours out.

Rather than letting some aggregator simplify the weather for you, you can just look at the raw data yourself: https://weather.cod.edu/forecast/

For big events, the media briefings by the National Weather Service are good resources. But they often stop the briefings early; a few weeks ago we had a high probability of a large amount of snowfall. The updates stopped at like 9AM, the snow was forecast to start around 1PM. Watching the short term models showed that the probability for snow was decreasing (NYC was just below the snow/rain line), and indeed we got pretty much no snow. (It snowed, but it didn't accumulate and the change to rain happened early.) To be fair, the briefing from the weather service said that the changeover time between snow and rain was very uncertain and that it would be the difference between a little rain and major snow event. But my point is, you can always go get yourself some more data; the closer you get to the event, the more accurate the forecast is.

(I don't know if any of you watch Skip Talbot, but he was looking at helicity swaths on the HRRR a few hours out, found a big one, and where HRRR predicted the strong rotation in the storm is pretty much exactly the path of a major tornado. HRRR is never going to be perfect, but it is right a lot.)

rootusrootus
37 replies
1d1h

That's an interesting point about the Apple weather forecast. That correlates pretty well with my experience. It is exceptionally inaccurate at short range forecasts. It's kind of a running joke at this point.

herpdyderp
33 replies
1d1h

The most humorous part to me is when it says it's currently raining or snowing and it's clear and sunny. How can a system be so wrong that it can't tell the current state of the weather?

counters
15 replies
23h55m

Do you want a snarky answer or a serious answer?

The serious answer is that the way you'd try to figure this out is by combining weather radar, satellite imagery, and a nearby surface observation to try to estimate the current conditions. But there can be a latency of up to a few minutes from these sources, and they could disagree with one another. You have to use them to bootstrap your near-term or nowcast product, but enforcing consistency with recent real-time and the nowcast is quite hard.

It's a surprisingly nuanced technical challenge. Most of the time, it works out just fine (e.g. if there is no weather). But people are awfully good at remembering when these sorts of analyses end up being wrong!

avar
5 replies
20h36m

Apple devices are constantly phoning home every time they see a random AirTag out in the world.

You'd think that if their users are accepting that level of communications with the mothership that they could ship some AI model to hear rainfall in the wild, and thus improve their live weather data.

stavros
0 replies
9h9m

What's wrong with devices phoning home if they don't actually send any usable data there?

inemesitaffia
0 replies
8h27m

I believe they send barometer information home

counters
0 replies
19h37m

Surprisingly low signal-to-noise ratio for most of the common, creative ways people come up with to detect rain. Windshield wipers on cars are another example.

The thing is, even if you did have a super reliable in situ "rain detector", how do you combine it with the existing datasets like weather radar, which is a gridded product? This is actually a really, really difficult sensor fusion problem when you then super-impose product requirements like the general location real-time detection map and the inputs necessary for whatever internal nowcasting system they use.

brookst
0 replies
6h43m

The AirTag comparison is interesting. AirTag tracking is clever and results in Apple not knowing the location or identity of either the AirTag or the phone reporting it. This relies on rotating keys that can be seen as random, or at least not identifiable.

But rain or other objective information? I suppose it works, maybe a bit like “limit IP address tracking” — cloudflare or other edge provider could mediate so Apple gets the data, knows it comes from an iPhone (to prevent bad data attacks), but Apple can’t tell what phone sent which data.

(the privacy concern being documentation of when you were inside/outside/etc).

TylerE
0 replies
15h6m

If you care that much buy a $100 weather station and sit it outside your window?

Actual ground observation weather stations are fairly rare outside of places like airports and major news stations.

“Is it raining here right now?” Is a harder question than you’re giving it credit for. Radar can show rainfall at as low as a couple thousand feet altitude, but if conditions are right/wrong (depending on how you look at it) it never reaches the ground.

bee_rider
4 replies
20h32m

It would be kind of interesting if the app had a “you are wrong” button, which allows you to take a picture of the outdoors. Apple could either use this to improve their models, or even just use it as input data directly if they get enough complaints. Plus, it would allow people to vent, or it could check if there is something wrong with the phone, maybe location is being mis-read or something like that.

counters
2 replies
19h35m

There isn't a vector where after-action reports like this could "improve the model." That data is useful for verification, but these systems generally have no learning component to feed the data back into them to improve them.

bee_rider
1 replies
17h43m

They have the input from all the different sensors and forecasts. Why not, for a given location, keep track of which one gave the best results? Sure, they don’t have a way of keeping track of they now, but it seems like it could be added.

counters
0 replies
16h1m

This is already SOP at most reputable weather data providers; they consume many different numerical forecasts and use statistical post-processing to choose an optimal blend of the available forecasts based on how different forecasts have verified against observations.

But this sort of technique only works for medium range forecasts. Short-range precipitation nowcasts are almost always a single, deterministic run of a model that extrapolates from patterns in recent radar imagery. They aren't bias corrected at all, so you can't use observations in the same way to improve them.

tspike
0 replies
19h50m

It does have that: “report an issue.”

stavros
0 replies
9h9m

Wait, what's the snarky answer?!

krisoft
0 replies
14h43m

if there is no weather

And the ship has been towed beyond the environment.

There is nothing out there, all there is is sea, and birds, and fish. And 20,000 tons of crude oil. And a fire.

Less sarcasticaly speaking I think there is always weather. Maybe what you mean is “no significant change in the weather” neither in time, nor in space.

jrockway
0 replies
23h47m

Yeah. Like you would think you could just look at reflectivity data to determine whether or not it's currently raining, but at most places you are far from a radar site and even the 0.5 degree tilt is scanning a mile above your head. There might be rain there, but is it reaching the ground? All you can really do is guess.

If you're interested in providing on-the-ground condition reports, install mPING: https://mping.nssl.noaa.gov/

I keep this app on my homescreen and try to report when very light rain starts, since it's not always obvious from the reflectivity data. Ultimately the user reports get fed into things like improving the model, and more data is always good.

Fomite
0 replies
19h18m

Yeah - nowcasting turns out to be remarkably difficult at times, especially at very small spatial resolutions.

Etheryte
10 replies
23h59m

I can't even begin to count how many times I've had this conversation with Siri.

"Hey Siri, is it going to rain?"

"It doesn't look like it's going to rain today."

"It's raining right now."

"It isn't raining right now."

throw0101d
5 replies
22h56m

I can't even begin to count how many times I've had this conversation with Siri.

I live in Toronto, Canada, which stretches about 40km east-west, and 20km north-south:

If the west-end (Sherway) gets hit with rain, but the east-end is dry, did it rain "in" Toronto when folks in Scarborough didn't experience it? Was the forecast wrong?

If it snows in North York but is dry at Billy Bishop, was the precipitation forecast "wrong" for one particular group of people?

Etheryte
3 replies
20h54m

Apple Weather uses your precise location if you allow it to, meaning it knows your location down to a meter, network and positioning issues etc notwithstanding. It doesn't have to guess your weather based on "Toronto", it knows your GPS coordinates. There is no technical limitation here, as I outlined in a separate comment thread [0], other apps already give you weather data and predictions with this granularity.

[0] https://news.ycombinator.com/item?id=39683660

sznio
1 replies
8h27m

It doesn't have to guess your weather based on "Toronto", it knows your GPS coordinates.

Meanwhile the Google Weather app constantly insists I live in Frankfurt while I'm in Warsaw.

StefanBatory
0 replies
7h50m

Fur Deutschland...

On a serious note I'm dealing with this as well - I live in a middle-sized city in the centre of the country, not in Kraków!

throw0101d
0 replies
5h50m

There is no technical limitation here, as I outlined in a separate

The technical limitation would be on the weather data size: what is the granularity/resolution of the radar data on where rain is actually falling?

* https://en.wikipedia.org/wiki/Canadian_weather_radar_network

Further: what is the geography of the area, and how does that effect things as well? Toronto specifically has (a) all sort of heat island effects, (b) certain areas are effected by the lake and how weather systems cross it at certain angles, and (c) has enough of an elevation change going north of the lake (e.g., Niagara Escarpment) that there are a few ˚C change in temperature that makes the differences between snow and rain.

viraptor
0 replies
6h2m

This exists on a smaller scale too. I'm in a town perfectly covered by a small hill range from the direction the wind is coming 99% of the time. This means almost every forecast for rain is valid 3km south and north of me, but not in the town itself.

lotsofpulp
3 replies
23h16m

Why would you expect Siri to know if it is raining at your specific location? Surely there exists an edge where on one side it is raining and on the other it is not raining.

So unless you are sitting next to the the weather station that Siri is getting data from, I would not expect it to know 100% of the time.

ceejayoz
1 replies
21h53m

Why would you expect Siri to know if it is raining at your specific location?

I don't, but as a result, I expect it not to guess.

Filligree
0 replies
4h24m

But you’re asking it to guess, and it’s usually right. Surely that’s better than having it refuse to offer a suggestion.

Etheryte
0 replies
23h7m

I live in the Netherlands. The local weather apps tell me when it's going to rain with nearly minute precision, along with cloud maps with scrollable time, graphs of how heavy the rain will be at what time, etc. It's pure nonsense to claim this is a technical limitation when other apps do it with ease. No one is expecting it to be right 100% of the time, but Apple Weather is wrong about rain most of the time, even on a crude scale of say, a city.

bumby
2 replies
1d1h

From the article:

>”These observations are then fed into numerical prediction models to forecast the weather.”

In other words, the forecasts come from models, not necessarily real-time station readings. Those readings are inputs into the model, and the models may not get updated fast enough to reflect current conditions.

Fomite
1 replies
19h15m

Or it might be raining at the station the reading is being made at, but not where you are. Living in the west for example, a lot of weather stations are remarkably broad in the area they're expected to represent.

bumby
0 replies
18h19m

The point is that the forecasts are often not built from real-time weather station data, but models using various initial conditions.

ako
1 replies
1d

Forecast are usually for a larger area, 5x5 kilometers, or 10x10 kilometers. Even within this area, weather will not be the same everywhere, so they'll give a probability for the entire area.

Windy.com lets you compare different models for a specific location, it also includes the size of the area per model: https://www.windy.com/?49.339,5.054,5

GFS is area is 22km, ECMWF 9km, ICON-D2 2.2Km, Arome 1.3Km, and UKV is 2Km. Even in a 1.3x1.3Km area it may not rain everywhere at the same time.

And then there's also the time element, so it's 1.3Kmx1.3Kmx1Hrs (or 3Hrs). So lot's of variation possible.

amarcheschi
0 replies
1d

Yup, a few days ago I made a python script to help me choose whether to get to uni by bike or by moped when it rains (given two coordinates I calculate the angle(bearing?) and checks whether it rains, and the angle from which the wind blows to see if I'll get all wet in the face) and I had a bit of a hard time figuring out why two different providers, windy and openweathermap, gave me 2 different wind results. Eventually, I found out they were using a different model, it took a bit of time tho, because windy only has increments of hours, while the other one was more granular

NikkiA
0 replies
19h26m

Because the 'current' weather isn't, it'll be whatever the last update of your chosen weather station reported. Often people choose a generic station that can be quite far from where they actually are - my default if I allow weather sites to 'guess' tends to be the airport 14 km away and 200m higher than me. A lot of weather can pass me by and still not have gotten to the airport in the 40 minutes since the last update.

wiredfool
0 replies
1d

Apple weather quite often has the "expected radar" function show storms taking a 90 degree turn right around now, so you'll see rain coming from the west, and suddenly when it gets to predictions, it's traveling north. (Note, this is Ireland). Dark sky was a lot better.

I've also noticed that Met.ie will typically predict more rain, and they're usually right. (e.g., last weekend was basically rain/drizzle/wind the whole time, met.ie nailed it, apple weather said that there would be an hour on Sat and all Sunday morning would be wet.

Of course, predicting rain in Ireland is not difficult.

nirav72
0 replies
4h43m

I'm really surprised, considering that they bought darksky. When darksky was active, it was one of most accurate weather apps I had ever used for location specific rain forecast.

blaufuchs
0 replies
22h26m

This one really kills me during Fogust in the Bay Area. I wake up and see the sun is gonna break through at ~1pm, oh no actually 2pm, oh no actually 3pm... oh no it's just another completely overcast day. I can understand missing a day or two, but it's bizarre when it happens day after day for weeks on end. You'd think the priors would get updated at some point.

unsignedint
6 replies
23h10m

I primarily rely on Windy for weather forecasts, which I find exceptionally useful due to its ability to compare multiple models. The variety of overlays available makes it an indispensable tool for all my weather-related needs.

[0]: https://windy.com

uoaei
2 replies
21h36m

Same here! Not to be confused with windy.app!

WillPostForFood
1 replies
19h20m

That is confusing! Does windy.com have an app?

bamboozled
2 replies
20h4m

Windy uses some of the models mentioned including GFS, you can select the model you want to use. So I’m not sure it would be any more accurate than the Apple Watch.

unsignedint
1 replies
19h5m

If you're simply seeking basic weather information, then what you receive from your Apple Watch won't differ much. However, if you prefer to analyze and interpret the raw data yourself, Windy stands out as an excellent resource. It aggregates numerous data sources, offering a comprehensive platform for informed decision-making regarding the weather.

bamboozled
0 replies
13h28m

I see what you mean thanks. Have you ever sen WunderMap? Not quite in the same league just an interesting source of information. It's kind of like windy, but based on near realtime data from personal weather stations around the country. Kind of fun.

https://www.wunderground.com/wundermap

ryandrake
5 replies
19h53m

Our local TV station weatherman has a YouTube channel[1] where he geeks out every morning about the weather, providing a much more detailed forecast than he has time for during the brief windows he has on the TV news. Walks through the HRRR, NAM, GFS, satellite pictures, and other sources of information. It's a nice compromise if you find the raw data to be overwhelming.

1: https://www.youtube.com/@markfinanweather

jayknight
2 replies
17h37m

There's no substitute for a local meteorologist who knows how weather patterns work in your region and knows how to interpret the models and is good at communicating that to regular people.

dylan604
1 replies
15h53m

Absolutely. It is something that I find amusing about people moving from California, LA/SoCal in particular, where their weather is basically just a nice segue from celebrity talk into more celebrity talk that reads some report the producers pulled. Then, they move to town and are amazed at how much people actually pay attention to the weather and comment on how many people actively have open tabs with the live radar. I usually reply with when you need to know if your ride to Oz is coming or not, you pay a lot closer attention.

ghaff
0 replies
4h50m

With snow in particular, 20 miles can be the difference between sort of a nothingburger and you really don't want to leave the house if you don't need to.

IG_Semmelweiss
4 replies
16h38m

I find that more important than the model used, is that you get actionable intel. To me, actionable intel boils down to 2 P's . Precision, and Probability.

So, I don't care if tomorrow there's a 50% chance of rain. I care that at a precise time of the day , say 9am, has a 10% precipitation and at noon is 90% , because i commute at 9am, not at noon. Wind is also an important factor if its raining. Temp as well. I need all this info presented as a mosaic.

For this purpose, I find the NOAA forecast local by hour is unrivaled. https://www.weather.gov/okx/ . Enter ZIP and then in enter local forecast by hour.

I have this URL bookmarked in my browser. I haven't looked back since. Example:

https://forecast.weather.gov/MapClick.php?lat=33.797&lon=-11...

I'd love to know if there is an android app that gives this level of detail, preferably, without spying into my microphone...

vel0city
0 replies
5h58m

I use the myRadar app on Android. It has a per hour graph of temperature and precipitation that makes it easy to quickly understand the next few hours. The paid version also gives easy access to more radar types like velocity

sklargh
0 replies
5h25m

Just seconding this. Weather.gov’s hourly charting is the best weather tool in the USA.

Sn0wCoder
0 replies
14h3m

I second weather dot gov mapclick. This makes me want to finally fix my PWA that used the Dark Sky API that quit working in 2023. This PWA also had a feature to click an icon and feed the long/lat into the mapclick URL which is the feature I used most often. I had a bookmark for a free replacement API for Dark Sky (that I got from HN) but have since re-formatted my computer. I know slightly off topic, but does anyone have a recommendation for a good free (couple users a day) weather API? Any suggestions would be appreciated as this has given me the motivation to hook it back up. Cheers.

dylan604
2 replies
15h57m

I think it likely depends...my Apple Watch corresponds exactly to what GFS says. GFS is OK for medium range

which is interesting, as i'm noticing the "within 15 minutes" level of notice on rain starting/stopping to have been close enough. the daily forecast last week said no rain even though the conditions really looked like it could at any moment. my iDevices pinged with rain starting soon even though the same apps forecast still did not suggest rain. it started raining with in "good enough" range of the app's notifications.

the update to the native weather app have all been very good over the past 2 OS updates. maybe they have integrated whatever company they purchased for good, but for my local area on the globe, it has been pretty good. i haven't traveled in a good while, so maybe my market is in the sweet spot of getting a lot of attention??? BigD in case you're wondering

perfectritone
1 replies
14h49m

Apple acquired DarkSky and sunsetted their API in the past few years. It always seemed more accurate to me than any other weather service. Sad to see it go.

dylan604
0 replies
14h12m

to me it didn't go, it just got rolled into Weather.app. I never used the original app, so I have not direct comparsion. however, if they are using the darksky data/tech/etc into weather.app, then it's better for me. not really sure where/how/why the new app updates are better under the hood, but they just are which makes it all seem like worthy upgrades

LeoPanthera
2 replies
22h47m

Ultra-short-range weather on Apple devices uses what used to be called "Dark Sky", before Apple bought it. It's how you get those alerts that say things like "Light rain in 17 minutes".

swores
1 replies
22h38m

Dark Sky was just the name of a weather app that included that feature earlier than Apple's weather app.

But things like "rain in X mins" is a feature multiple providers & apps have (including Apple once they bought Dark Sky), it's not specifically what Dark Sky was nor is it exclusive to them/Apple. (And actually, Dark Sky was probably the best weather app all round, yet Apple despite buying them and using some of their tech still produce one of the worst weather apps in my experience.)

tomaskafka
0 replies
5h29m

Hi, if you miss the Dark Sky, I am running a privacy-conscious indie weather app that can be configured to look just like the Dark Sky app: https://weathergraph.app

It's subscription based though.

AtlasBarfed
1 replies
19h4m

It depends if you're over the age of probably about... 35

Three day forecast in the 80s and probably early 90s are about where, crap, 15 days out is, actually the 15day is probably better.

Modern forecasting long term identifies the front movements, really well long term, even if they might be off a day and 5degrees.

The old forecasts would be completely off.

Source 50something

dylan604
0 replies
15h45m

Wow, you're being generous. To me, Nostradamus could have predicted the 10-day forecast as accurately as some that I've seen from the 80s/90s.

It's June in Texas, so we'll just say the 10-day is going to be sunny, hot, no rain. It's California in June, so we'll just say warm with June gloom burning off in the afternoon; warmer to hot inland. I didn't use any science data, and my forecast will probably have just as good of a chance as one that did.

Nothing like the old Aggie weather station consisting of a chain suspending a rock. If rock is wet, it's raining. If rock is moving, it's windy.

y-c-o-m-b
0 replies
1h39m

GFS = Great Faulty System as I like to call it. For 10-day regional weather, ECMWF is king. The problem will always be resolution though. For example here in the Pacific Northwest, the ECMWF can't distinguish our dynamic terrain very well (surface level vs 500-1000ft hilltops for example).

This website allows you to select which weather model you want to use: https://www.pivotalweather.com/model.php?fh=loop&dpdt=&mc=&r...

willmadden
0 replies
22h31m

Where you live matters more. If you live near a mountain range, good luck getting accurate weather predictions.

monkburger
0 replies
18h54m

The NAM overestimates southern jetstream energy. And the 3km is just awful for anything beyond 24hr.

HRRR is OK, but usually <10 hours for better numbers.

You should relllllly look at the soundings, though.

everdrive
0 replies
6h31m

I use weather.gov the same way. If you look at the hourly forecast for your area, you get very verbose, useful, and accurate information. My mother in law always reports the weather, and it's always wrong because she's "asked google" or has seen it on TV. The un-aggregated information is excellent and almost always more accurate than what is reported from other sources.

dan-robertson
0 replies
17h32m

Mostly unrelated: are there any good weather apps that get and can visualise an ensemble forecast?

SiempreViernes
0 replies
1d1h

Also useful to keep in mind is that predictions can become more accurate without necessarily improving in precision.

jrmg
15 replies
21h29m

I remember reading in The Signal and The Noise* that people _think_ that forecasts are bad if it rains, but the chance of rain was reported as below 50%. Getting rain when the forecast told you there would probably not be rain is annoying; getting a sunny day when the forecast predicted likely rain is a pleasant surprise.

To get what people judge to be a ‘good forecast’, the chance of rain has to be adjusted to be wildly too high - so that’s what consumer-focused forecasters do.

* https://en.wikipedia.org/wiki/The_Signal_and_the_Noise

HankB99
5 replies
21h3m

the chance of rain was reported as below 50%.

I've come to think of that as "it is going to rain 50% of the time. I don't know if that's what really is meant by "50% chance of rain," but it seems to fit.

And overall I tend to believe that the forecast is astonishingly accurate. This is in the Midwest (Chicago market) where weather has to cross large portions of the country or Canada before it gets to us. I suppose there are areas on the coast where weather is more volatile and harder to predict.

kxrm
1 replies
20h40m

Chance of rain is defined by NWS as:

"The probability of precipitation (POP), is defined as the likelihood of occurrence (expressed as a percent) of a measurable amount of liquid precipitation (or the water equivalent of frozen precipitation) during a specified period of time at any given point in the forecast area. Measurable precipitation is equal to or greater than 0.01 inches. Unless specified otherwise, the time period is normally 12 hours. NWS forecasts use such categorical terms as occasional, intermittent, or periods of to describe a precipitation event that has a high probability of occurrence (80%+), but is expected to be of an "on and off" nature."

Source: https://www.weather.gov/bgm/forecast_terms

lathiat
0 replies
15h24m

One of the realisations I only really had in my early 30s is that whether or not it's raining is much more localised than I thought. It can be raining here, and not at the shops a suburb over.

The percentage chance of rain includes whether or not it might rain in your specific dot of a given forecast area, which might be a suburb or entire city, as spelt out in your quote "at any given point in the forecast area".

The first time I drove over the Nullarbor in Australia, which has an entirely flat and straight 100+ KM section of road, I got to see rain far in the distance and experience driving into it having been able to clearly see it's edge from far out. That was an experience I had never had in the costal city I live in (Perth). That also led into similar realisations as the above.

It sounds so simple in theory but was not obvious to me for a long time :)

nektro
0 replies
20h47m

"50% chance of rain" means that there's 100% chance of rain for 50% of the area

luplex
0 replies
20h55m

I think it means "for any given point in the specified area, and for any given point in time, p(rain)=50%

So it of course won't rain for exactly 50% of the time on a given day, but over the long run, it will.

bombcar
0 replies
20h49m

My experience is that about 50% is where “rain will happen somewhere nearby, it may affect me”.

avar
4 replies
20h42m

Here in the Netherlands everyone mostly uses short-term live radar tracking of rain clouds and precipitation over actual predictive weather forecasting.

In an urbanized area most "is it going to rain?" questions are short-term, e.g. is now or 30 minutes later a good time to bike home?

Perhaps this wouldn't be as useful in other areas. The Netherlands gets very spotty rain. So even if you've got a 100% chance today it's probably 1-2 hours spread throughout the day, and sometimes very heavy rain followed by a dry spell.

The only time I've seen it to be incorrect is if a moving rain cloud just barely misses you due to changes in wind patterns.

moritzwarhier
3 replies
20h36m

Same here from Germany.

Never really thought about it, but I've opened the "Rain radar" more frequently than any weather app including the native one during the last couple of years, too.

diordiderot
2 replies
15h6m

Same in UK

Though I don't know anyone else who does this that doesn't cycle.

ericholscher
1 replies
4h49m

Same in Portland, and when cycling. So checks out here :)

dockd
0 replies
3h30m

Unfortunately the current Apple weather is terrible with their blurry blobs of clouds. Maybe they're trying to hedge when there is any chance of showers.

bee_rider
1 replies
20h37m

I sometimes wonder if people think forecasts are bad because they think of it in terms of: there are two possibilities, the forecast will be wrong, or it won’t. Therefore, the weatherman should be right at least half the time.

Of course, there are countless ways for the for the forecast to be wrong, and only a couple ways for the forecast to be right!

bobbylarrybobby
0 replies
13h29m

I had a high school teacher (not math!) who insisted that if you guessed on a multiple choice question, your chance of getting it right was 50%, regardless of how many choices there were.

kirrent
0 replies
20h26m

Amusingly he was unwittingly writing about his own future. People still make fun of Silver for Trump's win in 2016 because 538's final prediction of about 30% likelihood for Trump was 'wrong'.

WarOnPrivacy
0 replies
5h37m

getting a sunny day when the forecast predicted likely rain is a pleasant surprise.

Less so in FL where drought begins 4 minutes after the last rainfall. The 13th month of summer can have us begging for days w/o the migraine-making cancer ball.

bcardarella
12 replies
1d1h

Not in Boston, they've become far far worse. I presume because of climate change but still has become frustrating in recent years.

ejb999
8 replies
1d1h

why would you think that short term weather forecasts are somehow affected by climate change?

bcardarella
4 replies
22h55m

Climate models are based upon historical data. Recent climate change has changed weather patterns where historical data being used is making predictions less reliable.

counters
3 replies
16h0m

No, climate models are the same sorts of physics-based simulations as weather models.

hackerlight
1 replies
12h32m

Isn't GraphCast now state of the art for weather forecasting? It outperforms physics-based simulations over the relevant time horizon.

counters
0 replies
4h6m

"Outperforms" is a stretch; its skill is on par or slightly exceeds the state-of-the-art NWP models currently run today, but you as an end user would not notice the difference between GraphCast and the ECMWF HRES. Furthermore, raw model guidance is almost always passed through a blending and statistical post-processing process alongside competitor models to fine tune the forecast, and there are no rigorous benchmarks of whether or not the inclusion of GraphCast improves this process because very few groups are running it operationally.

Also, it's worth noting that GraphCast's outputs are a tiny subset of what we would traditionally forecast as weather parameters. You can't out-perform a competitor on a task that you aren't solving!

hiddencost
0 replies
14h51m

lol

dingnuts
1 replies
1d1h

You gotta keep up with the narrative bud, it's not "weather isn't climate" anymore; as long as the weather seems unusual to adults, it's evidence of climate change.

But don't forget! That doesn't mean nice weather is counter-evidence of climate change. Nice weather is /also/ evidence of climate change, because it's merely the lull before the weird weather.

Got it yet?

bcardarella
0 replies
22h51m

Ok smoothbrain

Y-bar
0 replies
1d1h

I don't think OP i correct that the forecasts there are less reliable, but I would gather that:

More warming == more energy in the system.

More energy in the system -> more volatile weather.

More volatile weather -> harder to predict weather.

Zamiel_Snawley
1 replies
1d1h

Which forecasts have become worse in Boston? 24 hour, 5 day, all of the above? This is surprising to me, because even when traveling across the USA, I've found the predictions to be very useful.

bcardarella
0 replies
23h0m

All forecasts. I race sailboat around Boston and they've been absolutely horrible. Not just for extreme weather events but even regular weather wind direction is off by nearly 180 degrees in direction regularly, wind strength is regularly wrong too. The predictions overemphasize rain events in the 10 day forecast during the Summer that nearly always completely go away and they fail to predict rain and lightning events. That's overing multiple weather models at various resolutions.

jghn
0 replies
21h27m

they've become far far worse

Where are your data to back this claim up? And over what time horizon?

elil17
10 replies
1d1h

Like many other people are commenting, I have subjectively felt that rain forecasts have gotten worse. I can think of two theories that could explain this. I'd be curious to hear from someone more knowledgeable if any of them are right or plausible.

1. High frequency 5G has thrown off rain forecasts in urban areas. Average prediction accuracy has still improved because rural/suburban areas don't have high frequency 5G.

2. The weather app now shows rain forecasts in time blocks as small as 15 minutes, even though predictions this granular are still inaccurate. This has inflated our expectations for forecast accuracy.

Zamiel_Snawley
1 replies
1d1h

How does mm wave 5G effect forecasts? Interfering with weather radar?

elil17
0 replies
1d1h

23.8-gigahertz 5G signals can look like water vapor to the instruments on weather satellites.

Workaccount2
1 replies
23h23m

Learning to read radar is phenomenally helpful in determining whether or not it will rain. It's not very difficult either.

A few years ago I was able to stop my friend's outdoor wedding (on the terrace as opposed to the hall, the venue had both ready) from getting rained out by reading the radar and catching a small pocket storm that had formed and coming right towards us. Sure enough it down poured, but everyone was inside for the ceremony. Reading just the weather report, there wasn't even rain forecasted.

vaughnegut
0 replies
15h9m

Where can one learn to read radar, sounds fun and useful!

lathiat
0 replies
15h22m

Another possibility from another top-level comment

"I remember reading in The Signal and The Noise* that people _think_ that forecasts are bad if it rains, but the chance of rain was reported as below 50%. Getting rain when the forecast told you there would probably not be rain is annoying; getting a sunny day when the forecast predicted likely rain is a pleasant surprise. To get what people judge to be a ‘good forecast’, the chance of rain has to be adjusted to be wildly too high - so that’s what consumer-focused forecasters do."

https://news.ycombinator.com/item?id=39684844

jerf
0 replies
1d1h

A theory I've been entertaining lately is that the raw engineering of the weather forecasts has indeed gotten better, but it has been offset by the clickbait-driven need for weather forecasts to declare everything to be the Worst Thing Ever, Click Here To Not Die. Snow storms that would have in my youth been a medium experience hardly worthy of note get their own names and days of breathless pre-coverage from the weather channels nowadays.

The net is the improved raw accuracy of the weather forecast is offset by the difficulty of reversing the clickbait layer slathered on top.

counters
0 replies
23h50m

(2) is a big ol' bingo. There was a race towards the bottom line of higher spatial and time resolution over the past 5 years (claims along the lines of "higher resolution means higher accuracy!"), which led to an awful lot of products that are nothing more than naive interpolations of coarser data. So couple the perception of "better"/"more accurate" products with a wholly insufficient technical approach to realizing this and you have a perfect storm for end users to feel that weather forecasts are getting worse. They just over-promised and under-delivered because many people who entered the field from outside of it completely underestimated how hard it is to push weather forecasting technology forward.

(1) is irrelevant for weather forecasting.

cogman10
0 replies
1d1h

An area has to be extremely densly trafficked before high frequency 5G is deployed. And even then, the whole point is to minimize broadcast range to avoid interference.

Further, it's only the upper range of high frequency spectrum that's being used (not sure who owns it) so it's not even every carrier that could interfere.

Finally, the most powerful radars are transmitting in the kilowatts range of output. It's hard for me to imagine that the microwatt output of cellphones are often the cause of radar interference.

brewdad
0 replies
23h26m

Short-term forecasts (1-2 days) seem more accurate than ever. However, weather as a business has meant a race at both ends of the forecast spectrum. Apps now offer minute by minute forecasts on the one hand or 10 and 15 and even 90 day forecasts on the other. Neither of those forecasting models are anywhere near ready for prime time but there is a market demand for them, so they get put out there anyway.

SiempreViernes
0 replies
1d1h

I would guess it is mostly down to option 2, coupled with the fact that aspects such as precipitation onset are possibly not something that actually has improved much: the examples given are hurricane tracks and atmospheric pressure which don't obviously couple tightly to when it starts raining.

dataflow
9 replies
1d1h

I'm sure they have, but I've also been drenched while reading a weather report that refused to admit it was raining in my city right now. It just told me it was cloudy, despite clear and rather heavy rain for 30+ mins straight over the whole city. To this day I haven't figured out how that's even possible.

NordSteve
3 replies
1d1h

One factor is your distance from the nearest weather radar, and nearest airport with automated weather observation. This sort of prediction is heavily dependent on whether the precip is detected by sensing.

I've seen similar things in our area (Minnesota) where you drive through a snowstorm, but the radar shows nothing in theare.

cogman10
1 replies
1d1h

I've seen variations in weather literally 10 miles apart. With torrential rain at work and nothing at home.

I can't see how any weather predictor could be correct in that situation.

importantbrian
0 replies
1d

Living in Florida I've driven down the road and had it be raining on one side of the road and not raining on the other while the sun is shining.

Fomite
0 replies
19h14m

Yeah. There's a place nearby where I live that's used for a lot of outdoor recreation like fishing and rafting, and the two weather stations that serve that area are a long way away.

n_plus_1_acc
2 replies
1d1h

Depending in which sources they used, they simply interpolate on a very rough grid

dataflow
1 replies
1d1h

I mean, I guess, but how rough of a grid are we talking? This wasn't a tiny city or something... it was a pretty populated city spanning a few miles across in a very populated and well known region. Granted I didn't walk around to check the whole city for rain, but the sky didn't make it look like the clouds were only above my head...

InSteady
0 replies
1d

microbursts are a thing. It's entirely possible much of your city was dry despite the cloud cover. Also possible they just done goofed.

hazbot
0 replies
20h49m

This is probably because of either poor sensor coverage, or a stale (old) forecast. Many weather services do not issue 'nowcasts' that constantly update with the latest weather observations (it's a hard and interesting problem), but rather a single forecast say 4x a day as the latest Numerical Weather Prediction model run comes in.

Fwiw, I agree with your bemusement and scorn - it's not good enough! (I say this as someone who has had roles where I issued these 'always stale' forecasts)

bumby
0 replies
1d1h

Was it saying there was 0% chance of rain or did it just not update to 100% even though you were experiencing rain?

The latter is somewhat common because the models (AFAIK) use probabilistic estimates, where different initial conditions generate potentially distant outcomes. The number of “rainy outcomes” defines the probability of rain, and doesn’t necessarily get updated with real conditions.

Zamiel_Snawley
7 replies
1d1h

This article mostly discusses longer-term forecasts, but I have also been impressed with the quality and reliability of imminent-storm alerts. They have saved me from getting drenched in a rain storm or allowed me to pull off the road for a break before a downpour.

It doesn't get a ton of press, but as this article highlights, progress has been steady and significant.

This article asserts that improving forecasts in low-income countries is underrated--does anyone know of studies that predict the impact better forecasts would have? Helping the poor with tech seems like the kind of project that many philanthropists could get excited about, and hopefully more effective than gravity lights and the like.

macintux
6 replies
1d1h

As someone who drives much of the summer with no top on his Jeep, Dark Sky was a revelation. I also managed to find a route between two bands of heavy thunderstorms (with a tornado watch to boot) one night far from home with no top and no doors using the radar.

Modern technology is amazing.

darknavi
5 replies
23h27m

Dark Sky was a revelation

Any replacement for it on iOS? Maybe I am crazy but Apple's weather alerts just don't seem like the same sauce.

Larrikin
3 replies
22h49m

I think they were doing some magic with the Android phone sensors, large amounts of user reports, as well as the actual forecast models.

Before Apple bought them, my Android phone was its own party trick at the bar. I'd be able to tell people down to the minute when it would start and stop raining. It was amazing for bar hopping on bad weather days.

counters
2 replies
22h29m

Nope. Simple computer vision / optical flow applied to radar image sequences.

declaredapple
1 replies
21h32m

What the actual crap did Apple do to mess it up so bad then?

Switching between providers on Carrot, Apple Weather often doesn't predict any amount of rain for the entire week, meanwhile I'm soaked in water in a thunderstorm, and NOAA and others predicted rain the entire week (which it did).

counters
0 replies
19h40m

No clue. They have strong folks on their weather team, too. Not obvious what's gone wrong over there.

declaredapple
0 replies
21h24m

Dark sky used to be accurate almost to the minute for me.

Apple Weather will tell me it won't rain today or all week.

Meanwhile NOAA will tell me I'm currently in a thunderstorm and that it will rain all week - And it was right.

Carrot is nice because you can switch between several providers.

callalex
5 replies
1d

Dark Sky brought a lot of this powerful forecasting to a hyper-local level. It’s such a shame that Apple bought it up and just…threw it all away. What a waste.

counters
2 replies
23h50m

Dark Sky didn't have "powerful forecasting." They literally just had a simple computer vision app which used optical flow to track blobs and weather radar, and then they extrapolated those blobs forward.

brewdad
1 replies
23h18m

It was a tool that was either very accurate or inaccurate depending on your perspective. If DS said rain would be starting in 8 minutes, it almost always rained at my house very soon thereafter. Very accurate. However, sometimes that rain came 4 minutes later or perhaps 12 minutes later. Now the forecast was off by a factor of 50%. Could be no big deal or a thing that ruins your morning depending on whether you got caught out in it and expected to by dry or not.

counters
0 replies
19h39m

Well, tracking a rain blob on radar that is 8 minutes from your house is an extraordinarily linear problem, so not surprising they'd have absurdly high P/R for that forecast :)

LeoPanthera
1 replies
22h42m

If by "threw it all away" you mean "integrated into the built-in weather app on every Apple platform" then, sure. I guess.

jdminhbg
0 replies
20h46m

Apple even opened it up as an API that's cheaper than lots of the others! I don't know where people get this idea that DS died, it's like they just took what happens to lots of other startup acquisitions and extrapolate it like it's a blob of precipitation moving towards their current location.

goatkey
4 replies
1d1h

Anyone who lives in a hurricane-prone area like myself (Florida) knows that while forecasts have gotten a _lot_ better, there is still so much room for improvement.

I am not affiliated, but I recommend checking out https://www.forecastadvisor.com/ to see what forecasts are best for your city. I totally changed weather providers and it seems much better now.

'The Secret World of Weather: How to Read Signs in Every Cloud, Breeze, Hill, Street, Plant, Animal, and Dewdrop' by Gooley is a fun read for anyone interested in figuring out weather without a forecast (or to supplement).

jrockway
0 replies
23h35m

It also keeps tigers away. I don't see any tigers around there, anyway.

artdigital
0 replies
13h32m

Sadly something like this doesn't exist internationally from what I can tell. I live in Japan and have no idea which sources are good and which are bad.

The local apps pull data from the Japan Meteorological Agency, so does Apple Weather, and so does Carrot Weather since a recent update (though those 2 still give me different results). Outside of Japan, when I travel, I have no idea so I just leave the Carrot Weather source on Apple Weather, because that at least pulls data from local weather services if available (https://developer.apple.com/weatherkit/data-source-attributi...)

Alifatisk
0 replies
5h41m

I am not affiliated, but I recommend checking out https://www.forecastadvisor.com/ to see what forecasts are best for your city.

What a great recommendation, it’s sadly US only. I have previously used an app called Climendo which claimed to digest over 15k forecasts and use the most accurate one in my city.

JoeAltmaier
4 replies
21h6m

Twelve year old won the San Jose Mercury News weather prediction contest one year, by predicting each day that the weather would be the same as the weather the day before.

Consumer weather prediction isn't about being right. It's about pleasing the customer by appearing to be helpful. Which often means exaggerating the chances of abnormal weather, so if it happens you can be a hero.

Real prediction is boring.

TheCowboy
1 replies
18h54m

Was he competing against other humans vs weather models? I'd hesitate to draw a conclusion from this anecdote.

I also couldn't find a link to this, and if you have one I'm interested in reading more.

JoeAltmaier
0 replies
2h29m

It was back when I lived in San Jose, during the 1980's. It was in the paper. They had a contest every year, with amateurs trying to beat the weatherman.

hazbot
0 replies
20h56m

Depends, the "optimal" forecast can be very sensitive to the scoring metrics used.

E.g. Darwin in Australia's tropics - persistence forecasting (as you describe above, just predicting the weather the day before) does very well on a metric like 'mean absolute error'. But has no practical skill at forecasting a severe tropical cyclone (aka hurricane/typhoon)! Many are willing to accept some level of false positives and a higher mean absolute error, because the cost of a surprise cyclone is so devestating.

ImaCake
0 replies
20h41m

This would work in San Jose because it’s a hot Mediterranean climate. Such climates have very predictable hot dry summers and cool wet winters. In Perth, similar climate, we often go month’s without rain in summer but will have several consecutive days of rain in winter.

I imagine using the previous day would have a much lower skill score in more variable climes.

LgWoodenBadger
3 replies
1d1h

Maybe coincidentally related that Simpson's paradox article from earlier, but if they've gotten more accurate overall, I certainly do not see it in the 24-hour forecasts. Of course, I'm also probably only paying attention to when it's wrong, not when it's right.

I plan my motorcycling based on rain, and the number of times I've gotten caught in rain when it wasn't supposed to rain at all that day is non-zero just this year.

VHRanger
2 replies
1d1h

Daily reminder that data isn't the multiple of anecdote.

Izkata
1 replies
22h44m

Except it literally is.

That statement is pretty much only used as a thought-terminating cliche that means "you're not allowed to have an opinion".

not2b
0 replies
22h10m

No, it literally isn't, at least, the plural of anecdotes isn't useful data. To be useful, data need to be collected in a uniform and systematic way. Anecdotes are memory, and it seems we are wired to remember the unusual and the unexpected. So you remember wrong forecasts (especially if you were caught outside unprepared), don't remember correct forecasts. Collecting everyone's anecdotes would not give you any insight about how good or how bad forecasts are.

Pointing this out is not an attempt to silence you.

kylehotchkiss
2 replies
1d1h

One thing I came to appreciate about growing up on the east coast was how much more accurate the forecasts were than they are in Southern California. The winds pushing east, the expansive radar coverage over USA that's publicly available, and the commercial airlines collecting weather data means the storms and weather systems are well understood as they're coming over. Plus the storms make nice straight lines from north to south that push through. In Southern California the rain forecasts always seem off. Even right now, it's raining and the forecast told me it was just supposed to be a little cloudy.

chasd00
1 replies
1d

the forecast for Southern California can just be a standing Sunny and 85F. The days it's wrong are infrequent enough to be tolerable. I didn't think that area even had meteorologists.

antod
0 replies
20h56m

The movie LA Story comes to mind where the TV forecaster prerecords their weather reports.

Vagantem
2 replies
20h59m

Interesting! As a contrast, I'm using historic weather data to predict future weather - couples use my free wedding weather predictor to find the perfect date for their wedding: https://dropory.com/

hazbot
1 replies
20h37m

Cool! If you end up wanting to expand beyond just the nearest weather station (forgive me if I've misunderstood your process), you could look into ERA5 - free Numerical Weather Prediction 'reanalysis' of past weather on a regular grid. openmeteo has some open source tools for extracting time series data from it.

But, although you get good spatial coverage, the drawback is 'the map is not the territory' - the model's representarion of reality doesn't perfectly mesh with the weather on the ground.

Vagantem
0 replies
19h23m

Very cool - will check out!

redavni
1 replies
1d1h

There is a gap between the title of the article and the contents. Starts out with weather forecasting is improved, but spends most of the article talking about how poor people and countries have other things to spend their money on than forecasting weather.

pornel
0 replies
15h43m

They also spend more on forecasts as a share of GDP. It's a pretty bleak picture — it's a bigger cost for them, and they still get worse results.

open-meteo
1 replies
9h42m

Creator of the open-source weather API open-meteo.com here.

The future of weather forecasting is likely to rely heavily on AI models. The article discusses Pangu Weather and HN comments mention GraphCast as examples. Interestingly, on the first of March, the European weather forecast center ECMWF released their new AI weather model AIFS as open data. This model is not only more accurate than their existing numerical model, but also requires significantly less computing power to run. They've published comparisons showing AIFS outperforms other models in terms of forecast precision: https://www.ecmwf.int/en/about/media-centre/aifs-blog/2024/f...

timthorn
0 replies
7h59m

Richard Turner gave an overview of AI weather forecasting to the Cambridge Philosophical Society at the end of last year. Recording available here: https://www.youtube.com/watch?v=JGn18WH0d6s

nipperkinfeet
1 replies
17h6m

Ever since IBM destroyed TWC and Wunderground, things have become worse. Wunderground personal weather stations were the most accurate weather forecasts.

hedora
0 replies
16h29m

The Dark Sky (now Apple Weather) acquisition somehow went worse than those IBM ones. Not only is it less accurate than Dark Sky used to be, it’s somehow managed to be less accurate than pre-acquisition Apple Weather.

https://weather.gov/ still works (type your zip code into the box on the top left), thankfully.

I’m worried weather.gov is only one election cycle away from being decommissioned.

Lobbyists from commercial weather sites nearly got the US weather service killed under Trump.

If/when they finally kill NOAA, global shipping will probably collapse, which is why killing it was blocked last time.

glitchinc
1 replies
1d

I could not disagree more.

I paid far less attention to weather forecasts 30 years ago than I do now, but I have numerous anecdotal examples of how weather forecasting models and information provided by publicly available weather services have trended towards uselessness.

There is no publicly accessible weather information service that can accurately forecast weather at my house. One of the first purchases I made when I moved in to the house was an Ambient Weather Station resulting from pure curiosity that has evolved into an interest in keeping a historical record of "actual weather". Daily hi/low temperatures generally have positive correlation with forecasted temperatures, but the spread between forecasted temperatures and actual temperatures is generally ten degrees less than forecasted.

Long term qualitative temperature trends ("above average for the winter" and similar) are positively correlated.

But ...

- Forecasted storm intensities are wildly inaccurate. Forecasted high-intensity rain storms end up being all-day drizzle events or on and off rain showers, and visa versa. A forecast of “a passing afternoon shower” ends up being an all-day wash-out.

- Precipitation forecasts are wildly inaccurate, without correlation. Actual precipitation can be far less than forecasted or far more than forecasted, even when compared to short term forecasts--to include same day and intrahour forecasts. Just this past weekend we had accumulating whiteout snow squalls on an off all day long on Sunday, yet there was never any mention of any possibility of snow by any local meteorologists or by any weather forecasting service I routinely check.

Dark Sky was the best app I ever used for weather forecasting. Its short and long term forecasts were more than sufficient for planning purposes, but where the app to this day has had no equal was in its intrahour local forecasts and precipitation forecasts. If Dark Sky alerted me that there was going to be tornado in my area within the next 15 minutes, I saw a funnel cloud 15 minutes later. If Dark Sky alerted me that it was going to stop snowing in 15 minutes, the snow stopped 15 minutes later. Sadly, Apple lobotomized the service when they claimed to have integrated Dark Sky functionality in to Apple Weather. Even though I fairly regularly report weather accuracy issues to Apple via the Weather app, the reporting and forecasting provided by Apple Weather has never improved.

- Seasonal precipitation forecasts are wildly inaccurate without correlation. Modeling (from NOAA, local meteorologists, etc.) suggested we were to have "above average snowfall" this winter, with the official average winter snowfall being 48 inches. We have received 20 inches so far this winter. Either winter will go out with a bang in the next few weeks (which would be nice, IMO), or modeling will have predicted more than 140% of the actual snowfall. This is an altogether unfair comparison, but why not: if the executives of a publicly traded company forecasted 140% more revenue to shareholders than the company they preside over realized, they would all be immediately fired, sued, jailed, etc.

If society collectively will not tolerate 140% inaccuracy in financial matters (stock price manipulation, value destruction, and so forth), should we be content with weather forecasting and modeling that is just as inaccurate? After all, weather is treated as (only) a financial matter by insurance companies. On an individual level, viewing weather's impact through financial optics still makes sense--from lost days of work and lost wages, to insurance premiums, to food prices, to transportation costs, to taxes, to paying for the ability to get your money back for a concert ticket you bought months ago if the weather is too bad.

Climate change is certainly wreaking havoc on weather modeling, but it has been doing so for a significant period of time and the models do not appear (to me) to be getting better at adequately accounting for the effects of climate change. If current weather forecasting models cannot be adapted to accurately account for the effects of climate change, it may be time to either fundamentally change the way weather modeling and forecasting is done, or not do it at all. Taking out my broad brush and bucket of paint: are there any companies relying on AI to develop a more accurate weather forecasting service?

And if anyone has a weather service to recommend that will not “Night at the Roxbury” me with ads and that has accurate 3-day-or-less weather forecasts, I am all ears. Please post them here.

counters
0 replies
23h22m

Climate change has no impact on weather modeling. The vast majority of weather forecasts derive from physically-based simulations of the atmosphere; the physics of the atmosphere don't suddenly change because the climate is warming. However, we rely equally heavily on statistically post-processing these physically-based simulations to correct systematic biases and better contextualize their outputs. Drift in the distribution of weather conditions - even small - can contaminate some of these types of applications. But not really in a way that you can honestly claim "climate change is making weather forecasts less accurate."

are there any companies relying on AI to develop a more accurate weather forecasting service?

Sure there are. But AI isn't a silver bullet, and existing weather forecasting technologies are _really freaking good_. For all of the hullabaloo over AI-NWP systems like Google's GraphCast and Huawei's PanguWeather, these state-of-the-art systems are about _on par_ with the best-in-class existing numerical weather models; they offer incremental improvements in tuned forecast accuracy, but these improvements are statistical descriptions of a very, very large number of forecasts - end users really wouldn't see any practical difference in forecast quality if they relied on these forecasts. But to my point above - even AI-NWP outputs would be filtered through statistical post-processing to boost their accuracy/utility.

There are a lot of companies that _claim_ they use AI at different parts of the weather value chain to improve forecasts. A lot of them stretch the truth as to what extent they really use AI or ML. The simple reality is that the weather community has used ML since the 1970's to improve weather forecasts.

vanilla_nut
0 replies
22h52m

I find that 4 to 7 day forecasts tend to be 80% accurate. So probably a little bit better than they were when I was a kid.

Unfortunately, the most important part of any forecast IMO is intensity. I don't care if we're going to get snow flurries all day, but if we're going to get a foot of snow, I would like to know -- and not just when the winter storm warning goes into effect!

Similarly, I don't care if we're going to get scattered showers all day. But if we're going to get a downpour in the afternoon, I'd like to know so I can avoid getting caught in a flash flood on a trail or on the road.

Same thing applies with temperature: if it's going to be cold all day, good to know. But if a rainstorm is going to remain active during a deep freeze and create a layer of ice on every exposed surface, I need to be prepared for walking, biking, or driving.

Fortunately there's a somewhat local weather station near me that provides an RSS feed of longform weather forecasts. But I notice that more and more people wind up surprised by slightly-abnormal weather events as they rely more and more on smartphone weather apps. Weather apps that utterly lack the nuance that a paragraph of text can provide.

ufocia
0 replies
20h50m

A more catchy title would've been "Weather has become more predictable".

toolslive
0 replies
23h20m

But it's still a chaotic system and Lyapunov would claim we're quite vain to even try.

timetraveller26
0 replies
1d1h

okay, nice, but what about software delivery forecasts?

throw0101d
0 replies
1d1h

Recommend the book The Weather Machine by Andrew Blume (also wrote Tubes) on some history of forecasting and what happens in the background nowadays:

In The Weather Machine, Andrew Blum takes readers on a fascinating journey through an everyday miracle. In a quest to understand how the forecast works, he visits old weather stations and watches new satellites blast off. He follows the dogged efforts of scientists to create a supercomputer model of the atmosphere and traces the surprising history of the algorithms that power their work. He discovers that we have quietly entered a golden age of meteorology—our tools allow us to predict weather more accurately than ever, and yet we haven’t learned to trust them, nor can we guarantee the fragile international alliances that allow our modern weather machine to exist.

* https://www.andrewblum.net/the-weather-machine-2

* https://www.goodreads.com/en/book/show/42079139

For the very early history of meteorology, see perhaps The Invention of Clouds about Luke Howard:

* https://www.goodreads.com/book/show/1148768.The_Invention_of...

* https://en.wikipedia.org/wiki/Luke_Howard

sklargh
0 replies
5h26m

Just a reminder that weather.gov’s local hourly forecast is exceptionally good and a masterclass in synthesizing information well. (weather.gov/“zip”)>go to right hand rail > hourly weather forecast.

I haven’t found an app or tool outside of building a grafana dash that beats it.

qwertox
0 replies
22h16m

I once listened to a podcast [0] with interviews of a couple of scientists at the ECMWF (European Center for Medium-Range Weather Forecasts).

I think it was in that episode where one said that every 10 years we improve the forecast by 1 day.

It was recorded in 2019, so AI wasn't really that much of a topic as it is today, considering that Google published an AI weather model in November of last year [1].

[0] https://omegataupodcast.net/326-weather-forecasting-at-the-e...

[1] https://deepmind.google/discover/blog/graphcast-ai-model-for...

paxys
0 replies
1d1h

All the stuff mentioned in the article is accurate – better raw data, faster computers, smaller grids, better predictive algorithms etc. all result in vastly better weather info in general today. This also means though that you have to put in more effort to get a better result for yourself. What algorithm is the app using? Does it localize all the way to your neighborhood, or your street? How frequently does it update? Is your GPS accurate? People generally don't think about this stuff, but some fine tuning can result in vastly better results.

okdood64
0 replies
1d1h

Anecdotally, they have been very frustrating in the Bay Area with this El Niño rain season. Not as reliable. Many ruined plans, but I have learnt my lesson.

nvahalik
0 replies
19h44m

This might be true for some areas but maybe not for others. Where we live is on the very edge of the NWS coverage. Our forecasts and messaging is generally a lot more “loose” than folks closer to the main NWS “office”.

I am not sure if this has to do with radar capability but all the old time hams seem to corroborate this.

mikeortman
0 replies
22h22m

This speaks on long term outlooks at synoptic scale, we really should put some energy on researching mesoscale long term outlooks, or even 4 hour short term. It's a difficult problem to solve as the variables are quite complex, but the reward can be substantial -- on-land severe weather impacts less people but often is deadlier and can cause huge financial loss in areas that may not expect it.

egl2021
0 replies
20h12m

Geezer alert: forecasts look amazingly good to me. When I was a kid in the Pacific Northwest, it was routine to miss major storms until they hit land. We didn't have satellites, oceanic buoys, etc., and I remember the TV weather guy saying things like "we've had a report from a ship at sea..." and proceeding to make wild guesses.

cameronh90
0 replies
19h40m

The weather forecasts in the UK are definitely much better than they were a decade ago, especially in the 3 to 14 day range. However, I still find my stupid heuristic works quite well for predicting tomorrow's weather: the weather tomorrow will be the same as what it was today. The UK often gets sticky ("blocking") weather patterns, so it works surprisingly well.

bilsbie
0 replies
21h54m

I can’t believe they killed dark sky. That was accurate to the minute. Incredible.

belorn
0 replies
17h41m

I wonder if energy market speculation is playing a role in making weather forecasts more accurate. I would not be surprised if some stock markets also get influenced by weather, especially during periods of high fluctuations in energy costs.

beanjuiceII
0 replies
4h58m

The question is WHEN are they more accurate, my weather stations and apps tell me as weather changes what it is, which I can usually do too by stepping outside. But weather forecast (a prediction) I have not noticed it being more accurate and often wonder what changed, is it too much reliance on this real time data?

Workaccount2
0 replies
23h19m

Keep in mind, weather forecasting is something that is exceedingly difficult and that concerns just about everyone. That means it is particularly ripe for confirmation bias.

We objectively know that weather forecasts are more accurate than ever. We subjectively know that they are bad/gotten worse, because last Thursday I brought my umbrella to work for nothing.

TheCowboy
0 replies
18h6m

I took meteorology in high school and our teacher made us do a daily prediction exercise that I think would benefit all people, especially those who put too much weight into their own anecdata.

We just had to make a prediction for the next day's weather. Then compare our prediction to the forecasted prediction. It didn't matter how accurate we were for the grade. Just that we systematically performed this exercise.

It really made one appreciate the quality of forecasts, and that no, the "weather man" is not always wrong at all. A lot of huffing and puffing is from people who lack any rigor in their observations. And if you're trying to contest the accuracy of weather forecasts, or any form of forecasting really, then you really should provide some hard evidence.