Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

Windy is great. The map is appealing, but be sure to check out the forecast view for a particular location. Also make sure you try different locations. I'm in the Salish Sea on Orcas Island in the San Juans and because we have a lot of topography here mixed with ocean, we have a lot of local effects, contour winds and so on and there are big differences between locations.

Also note that Windy can get it wrong. I grew up in Cape Town and forecasting there is easy compared to here because it's the tip of Africa surrounded by Atlantic and Indian Ocean. Here it's very mixed with land, sea, big 11,000ft mountain ranges like the Olympics and so on and this region is hard to forecast. For where we are, the forecasts - and Windy's map specifically - is wrong fairly often.

A trick that a lot of folks don't know about is using ATIS, AWOS or ASOS at a local airport or airfield. If you want to know what the weather is at a given location, find a nearby airport, get their ATIS (or AWOS or ASOS) phone number and you can call and get a real-time report that is extremely accurate. I do this for KORS, our local airfield all the time. You can get this data off Foreflight although I'm sure there are plenty of free alternatives. Obviously it's current weather, not forecast, but it's often helpful.



> , find a nearby airport, get their ATIS (or AWOS or ASOS) phone number and you can call and get a real-time report. Obviously it's current weather, not forecast, but it's often helpful.

The US Government makes both METARs (current report) and TAFs (area forecast) available online. You can also get PIREPs (pilot reports) if you are interested in the conditions in the air.


Along with free APIs! (XML, though.)

https://www.aviationweather.gov/dataserver


that’s neat! thanks for mentioning it.


I use an Android weather widget called Meteogram Pro. It supports pulling in and displaying METAR data, among a thousand other options.

You can make plots of things like thunderstorm probability, thickness and altitude of different cloud layers, azimuth and elevation for all planets in the solar system, pollen index, tide height, you name it.


Can you let me know which developer it is? There are a few with the almost exact same names. Thanks!


https://play.google.com/store/apps/details?id=com.cloud3squa...

There's also a non pro version, but I'm not clear on the differences because I still have a decade or more worth of the old system credits bought to support what's always been a great app/widget.


When you say Windy can get it wrong, isn't it really the underlying forecast models that are wrong?

It's unclear to me if Windy is doing much more than creating a great presentation of existing data. Their website does list an open ML role, though..

I wonder how much rolling your own weather forecast is tantamount to rolling your own crypto??


By default, Windy uses the ECMWF weather model. You can also change it to GFS or ICON, which many american forecast websites use.

ECMWF, GFS and ICON are made by national/international forecasting agencies. IBM also has its own proprietary weather forecasting service, notably used by The Weather Channel. Other apps mostly use one of these models or aggregate predictions from different services (e.g. AccuWeather claims to aggregate many different models, including those from national agencies around the world)


for people who live in europe, try AROME, it's very limited compared to the other models, but it does (IMO) give the best forecast.

in my experience AROME tends to forecast on the drier side of the weather. Since I'm in the UK I use AROME first and flick between ECMWF. (ECMWF tends to forecast on the wetter side!!)

for the next hour(s) forcast, I tend to use rain radar - windy has this, but it's not the best I use netweather.tv (there's many others though)


I'll add in a vote for the UK Met Office here too (particularly for the UK). AROME is a high resolution 'nested' model, where the region of interest is simulated at a higher resolution that the (often global) driving model. Similar to ICON-D2 (Germany [0])and UKV (UK [1]). These will typically produce a better downscale of the forecast to a local area.

Their utility depends a bit on what is causing the uncertainty in the forecast though. For the mid-latitutdes, a lot of the uncertainty comes from timing the arrival of a weather system. In these cases, the high resolution simulation doesn't help you, as it just incorporates the uncertainties of the lower resolution driving model.

The ensemble simulations that the weather servies perform will help you there, ECMWF has a really nice meteogram for individual locations [2].

[0] - https://www.dwd.de/EN/ourservices/nwp_forecast_data/nwp_fore...

[1] - https://www.metoffice.gov.uk/research/approach/modelling-sys...

[2] - https://apps.ecmwf.int/webapps/opencharts/products/openchart...


I would love to see the UK Met Office model in Windy

I don't know why it's not there - I'm assuming it costs money


> IBM also has its own proprietary weather forecasting service, notably used by The Weather Channel.

Is it not the case that IBM bought the weather channel to acquire the technology with no interest in the WC business?


IBM bought The Weather Company, which was the holding company of The Weather Channel. The forecasting technology is now owned by IBM, which licenses the weather data to The Weather Channel.


Rolling your own weather forecast is more akin to rolling your own election forecast (ahem).

Equally dangerous, but not designed to be private and a huge different set of risks.


Political motive to change weather forecasts, and weather motive to change political forecasts.

Very dangerous.


depends what you mean by 'rolling your own weather forecast'. If by that you mean you're going to do some statistical modelling on ensemble data, you're going to get a much better idea of what is likely to happen than you can from the average datasets you get on sites like windy


"If you want to know what the weather is at a given location, find a nearby airport, get their ATIS (or AWOS or ASOS) phone number and you can call and get a real-time report that is extremely accurate."

To get the phone number:

     curl -Ls https://www.airnav.com/airports/get?s="$1"|sed -n '/<TR>.*WX/{s/.*=right>//;s/&nbsp;.*top>//;s/<.*//p;}'"
 
For example, Orcas Island

     curl -Ls https://www.airnav.com/airports/get?s=orcas|sed -n '/<TR>.*WX/{s/.*=right>//;s/&nbsp;.*top>//;s/<.*//p;}' 
or, without the redirect

     curl -s https://www.airnav.com/airports/KORS|sed -n '/<TR.*WX/{s/.*=right>//;s/&nbsp;.*top>//;s/<.*//p;}'


You live in a beautiful area. I grew up scouting around there...on one trip we were invited to embark & leave Eastsound earlier than anticipated because a couple of our members decided to "casually" lift some cigars from a store there (IIRC). xD Thanks for the ATIS info too. I wonder if it's the same type of message I hear on non-noaa VHF from nearby.


Thanks - I'd forgotten about the marine VHF reports. Rarely listen to them, and I should. LOL! Trust me Orcas has plenty of scandal that's worse than that. I'm sure less than half the island has all your social security numbers memorized.


Another option is the point forecast from weather.gov. For example, here's the forecast for Eastsound: https://forecast.weather.gov/MapClick.php?lat=48.6968&lon=-1...


If you're in the US, you can get METARs from aviationweather.gov:

https://www.aviationweather.gov/metar/data?ids=KSEA&format=d...


Another thing to consider is the difference between a global scale forecast mode (GFS, ECMWF) and a mesoscale model (HRRR, NAM, RAP). The later uses a much smaller grid size and can take into account terrain. In a place the the San Juan's (and the PNW in general), a lot of the weather patterns are coastal terrain driven so these models can be much more accurate. They catch is, they don't see out of far due to their higher computational complexity.

Here is a great resource to read up on the various models (there are far more than Windy offers): https://luckgrib.com/models/


1000% agree! I sail and Windy is great for an idea of what is going to happen, but no substitute for what is actually happening on the water, i.e., the fine-tuned weather that is necessary to sail a boat. If you want to know a hurricane is coming or the potential for a "weather event," then Windy is truly your friend. Otherwise, what the OP wrote is totally necessary. For sailing, you want to look at lighthouse weather data (e.g., NOAA) and the numerous buoy systems (e.g., https://buoybay.noaa.gov/) really are your friend.


I live less than 3 miles from an airport, but due to microclimates the weather is completely different. The NOAA does a reasonable job of forecasting my specific area, but most weather apps will take the current conditions from the airport, which can be off by over 10F in temperature alone.


"Also note that Windy can get it wrong."

I kind of lost trust in it, when I was in the middle of a thunderstorm, yet windy showed me all sunny.

Granted, it was in the alps and forcasting there is hard, but it was the current state of things they got wrong.


Convective ("heat") thunderstorms are pretty much impossible to forecast precisely. On the day of, I'd recommend to use the radar, satellite and wind measurement, to get a picture what's happening, not the forecast anymore.


"Use the radar, satellite and wind measurement"

I did. And they showed me no clouds, rain or wind worth mentioning.


If you were in the storm.and the radar didn't show anhthing maybe the radar was out of service in that area, which is very unusual.


It's not realtime, some of the underlying models only get updated every 6 hours.


The forecasts for the US at forecast.weather.gov are really good too. I constantly refer to the hourly precipitation charts to figure out how much rain gear I should hike with, and their temperature predictions are useful too. Plus if they have any warnings or watches it’s helpful. It’s much better than any of the paid services out there.


Also worth mentioning that Windy provides several different forecast models that you can choose between. There are high resolution models like NAM, and lower resolution models like GFS - toggling between them often gives me a better sense of what to expect.


Agreed. It was interesting to see the deltas between the models for Hurricane IRA. Was significant 3 days out.


You can also get the METAR (observations updated hourly) from the Aviation Weather Center: https://aviationweather.gov/metar/data?ids=KORS&format=decod...


If you're in the Pacific Northwest, specifically Washington/Western Washington, then UW's weather models reach much higher spatial resolution than Windy appears to do.

https://a.atmos.washington.edu/mm5rt/

I recommend starting with the 4km or 1.3km WRF-GFS model and looking at something like 3hr Precip.


Windy also has HRRR (among others) for hi res near term domestic US incorporating radar.

https://windy.app/features/hrrr-weather-forecast-model.html

Other models:

https://windy.app/guide/windy-app-weather-forecast-models.ht...


I live a couple islands over from you. As a frequent sailor, I use Windy regularly before a race, but it's never completely accurate (PredictWind tends to be better). Great for looking at general trends and visualizing patterns. This area is tricky — lots of microclimates for the reasons you mentioned and it's almost impossible for a service like Windy to be completely accurate.


The forecast that worked best for me (at least in summer in Europe) was local sailplane text forecast from the respective national service. It was an interpretation of the model by a very experienced human with a big picture introduction ("this front" or that "high pressure area") and also expressed uncertainties. They are sometimes behind soft paywalls, though (need to register, but its free. There are some weird rules of the EU about some weather data - cannot be freely available. I don't get it)


> Windy is great.

I don't like the zooming ratio




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: