Read the announcement:
https://www.noaa.gov/news-release/noaa-deploys-new-generation-of-ai-driven-global-weather-models
Environment friendly : "using up to 99.7% less computing resources"
Grib-files available here:
https://nomads.ncep.noaa.gov/pub/data/nccf/com/aigfs/prod/
Despite the NOAA budget cuts - or perhaps because of them...
Maybe one day on Windy, what do you think? (I'm asking for trouble 😓)

@Gkikas-LGPZ Thank you. You summarized the meaning of title of this post: there are other of these (solid ground) "Madis" stations floating in Windy.

An anticipated sign of the climate change effects, perhaps? With a bit of searching, one can find, for example:

Ferrières-sur-Sichon (pop. 552) has set the sails and is now called Ferrières-sur-Mer:

Not dramatic but a batch run by Windy to block this noise would we welcome.

Hi, awesome! But I am missing the numerical value of the air pressure.

Lisez :
Jean-Yves Bernot: Météo & Stratégie, Croisière et course au large
Chapitre 23 La Météorologie de la Méditerranée
concernant la difficulté avec les phénomènes locaux pour les modèles numériques. Conclusion ch23: ailleurs, on calcule, en Méditerranée, on réagit !

Welcome to the Mediterranean ! While you can certainly find numerical models with higher density from other paying services such as PredictWind, I personally prefer to use the above book and my own judgment when planning a crossing : preparing for the worst and hoping for the best! Windy gives some loose hints for that.

I noticed in my navigation area measured values not corresponding to observations from
Madis station
GW3376 Litchfield
S39°2'28" W89°41'32"

According this page, it is located in Litchfield, IL, US 39° 11' 0" S 89° 39' 0" W elev 690m.

Is this a glitch in the source of data or a persisting parsing error of it? Or a bug in the batch file?

Absolutely great, congratulations to the team! Thank you so much.

@idefix37 said in Isolines of pressure:

Therefore, it is necessary to re-enable them when switching to these two layers.

Actually, the isolines setting is memorized (v49 browser+android) and there is no need to re-enable it when switching back to, say the wind layer view from radar layer view. Thanks, Windy!

I know that somebody is working on this (?) - on a stormy day at sea one can easily find more of these lost landlubber "Mad"iis stations, being probably terrified:

"

@Tsei28 Hi, I would say that your Storm and Heavy rain mobile alerts are enough (or they should, but if your work, etc., depends on it, I would trust more Ilmatieteen laitos).
Your conditions for your own alert look too complicated to me, and I presume that all the conditions will never be met at the same time (it is an AND-operator), and nothing will come out. Also, your requirement of south-east approaching storm is not necessarily assured by selecting the wind direction "all" 5-20 m/s (the front is not following the ground wind, anyway)
I have noticed that creating one simple alert per phenomenon works the best for me, allowing me to adjust the criteria for each alert more accurately by experience: like one alert for the wind from certain direction(s) and one alert for the rain. Forget the cloudiness - IMO.
Stay safe!

@idefix37 @daniel-smith-860f3f08-a85e-4077-b8fd-46b34b41bbc0
IPMA's own application and web page is better for this purpose than Windy's radar view, or the aforementioned IPAM's own radar view, here : Mapa dinâmico Continente.

This product's radar layer is available in 5 minute resolution, from https://www.ipma.pt/resources.www/transf/radar/imgs-radar.json
and https://www.ipma.pt/resources.www/transf/radar/por/<< path-from-json >>

I don't know if Windy is using the same source/frequency and/or their own filters, but the view is different :

The aforementioned IPAM radar image product is not updated as frequently, only once every 30 minutes. Is Windy using interpolation in between less frequently fetched images in order to limit the bandwidth? (I am just guessing, of course) :

note: the above is a JPG-image generated every 30 minutes, unlike the dynamic map, which is using IPAM's own API.

@andrea-vistarini Buongiorno : https://community.windy.com/post/226741 Sail Router plug-in, aka Passage Planner:

Olá! Weather models are not created by Windy but by governmental organizations, then presented by Windy and others. Indeed, none but 50km low-resolution models are available in your region (too many motorboats there! 😊). Anyway, I wouldn't trust any computerized model prediction about a sudden local phenomenon: I was using a radar (my own) over there to take a urgent reef (not an option for you, I know)...
If you are north of Salvador, Windy is displaying Brazilian weather radars (see below). In Salvador, and in wonderful Baía de Todos os Santos, there is no working radar that I know of.
For those interested, a good story (pt-br) about Pirajá here.

@Tsei28 Hi, I would say that your Storm and Heavy rain mobile alerts are enough (or they should, but if your work, etc., depends on it, I would trust more Ilmatieteen laitos).
Your conditions for your own alert look too complicated to me, and I presume that all the conditions will never be met at the same time (it is an AND-operator), and nothing will come out. Also, your requirement of south-east approaching storm is not necessarily assured by selecting the wind direction "all" 5-20 m/s (the front is not following the ground wind, anyway)
I have noticed that creating one simple alert per phenomenon works the best for me, allowing me to adjust the criteria for each alert more accurately by experience: like one alert for the wind from certain direction(s) and one alert for the rain. Forget the cloudiness - IMO.
Stay safe!

@idefix37 @daniel-smith-860f3f08-a85e-4077-b8fd-46b34b41bbc0
IPMA's own application and web page is better for this purpose than Windy's radar view, or the aforementioned IPAM's own radar view, here : Mapa dinâmico Continente.

This product's radar layer is available in 5 minute resolution, from https://www.ipma.pt/resources.www/transf/radar/imgs-radar.json
and https://www.ipma.pt/resources.www/transf/radar/por/<< path-from-json >>

I don't know if Windy is using the same source/frequency and/or their own filters, but the view is different :

The aforementioned IPAM radar image product is not updated as frequently, only once every 30 minutes. Is Windy using interpolation in between less frequently fetched images in order to limit the bandwidth? (I am just guessing, of course) :

note: the above is a JPG-image generated every 30 minutes, unlike the dynamic map, which is using IPAM's own API.

@eric-ludinart Correct, il en faut un API de Navily pour vous développer un plugin. J'ai bien peur que Navily ait déjà pris l'initiative de vendre l’équivalent de votre idée comme un produit payant :

[https://www.navily.com/navily-premium](link url)
Access 72 hours of weather forecasts and our personalised protection score on all anchorages.

Ils achètent leurs données météo d'ici :
[https://www.meteomatics.com/en/customer-stories/weather-data-helps-navilys-users-choose-the-best-anchorages/](link url)

Bien sûr, cela n'empêche pas Windy d'entrer en discussion avec Navily.

Si vous êtes un développeur en recherche d'un API, peut-être le STW, dont les données ont un peu stagné, pourra vous ouvrir leur interface :
[https://stw.fr/fr/map?active=filter-marinas](link url) - y ajoutant les mouillages.

Comme ça, on aurait un peu de concurrence Open Source pour le Navily sur un bon plateau alternatif ! 😉

@idefix37 said in Isolines of pressure:

Therefore, it is necessary to re-enable them when switching to these two layers.

Actually, the isolines setting is memorized (v49 browser+android) and there is no need to re-enable it when switching back to, say the wind layer view from radar layer view. Thanks, Windy!

Thanks, @Suty, that's what I thought also. Maybe @Eloy-Herrero-Gimeno can confirm us how his problem was resolved (?) and/or UNCAP the alarming title... 😊

@Eloy-Herrero-Gimeno I have had this behavior occasionally after computer was put in power save mode. Reloading the page restores. Search how to do that for your browser. Maybe cmd+R.

Read the announcement:
https://www.noaa.gov/news-release/noaa-deploys-new-generation-of-ai-driven-global-weather-models
Environment friendly : "using up to 99.7% less computing resources"
Grib-files available here:
https://nomads.ncep.noaa.gov/pub/data/nccf/com/aigfs/prod/
Despite the NOAA budget cuts - or perhaps because of them...
Maybe one day on Windy, what do you think? (I'm asking for trouble 😓)

In Atlantic North now: no problem with Starlink, with or without VPN, both browser and App OK. Regarding the IP above (better not publish your IP in public, BTW), it looks like the hosting server is in Sofia, Bulgaria and that it is building up a bad spam reputation :
https://cleantalk.org/blacklists/as14593/217.142.20.0/22
In October 2025, 180 spamming IP addresses in that network today 95 spam active IPs, spam rate 9%... Not everybody is using Windy only, unfortunately.
Cannot speak anything about Windy's hosting policies, and are they automatically blacklisting that network (and they will not tell you that in public), but I have doubts that they would whitelist your IP alone, since it can change.
I would advise using a VPN, or sailing away until Starlink changes the ground station.