The Sao Paulo radar is broken

jesperrhode

Hi
The Sao Paulo / Brazil radar has been broken for some time now.
The rain "dissapears" into the line on the image.
Other radar services shows the weather correctly.

best, Jesper

Screenshot 2024-10-11 200452.png

Rodolfoar07

@Filip_K This error persists to this day on Windy.

I live in São Paulo and it is very difficult to use the weather radar here.

This error can simply be corrected by attaching the image from this link in its entirety, covering the area of São Paulo, including the area of the red box, which currently does not happen. I ask that you pay attention to this if possible.

https://www.simepar.br/simepar/radar_msc

Several official agencies in Brazil (Civil Defense) use this application and this change can save lives here. In addition, Windy is widely used to cover sporting events such as Formula 1, and this correction will be very useful. I ask that you pay attention to this if possible. Thank you.

Filip_K

Hi @Rodolfoar07, yes, we are aware that some Brazilian radars are missing from our data feed. There are many sources of radar data for Brazil and it is high on my TO-DO list for upcoming weeks to consolidate them and bring best possible radar data for Brazil to windy. Thank You for Your patience.

Rodolfoar07

@Filip_K The update of Brazilian radars was great. Thank you very much!

Filip_K

@Rodolfoar07
You're welcome. I gathered all the sources I could and now we should have most complete and up-to-date data possible.

Rodolfoar07

@Filip_K Hello. I think there's a problem with Windy. Simepar images haven't been loading in the São Paulo and eastern areas of southern Brazil since Thursday morning.

Filip_K

@Rodolfoar07
The SIMEPAR images have been loaded and used at the time You mention (28-02-2025T03:50Z). Maybe Your custom colorscale for radar doesn't show it?

Also, our color->value mapping from SIMEPAR image might be wrong, since I wasn't able to find, what each color means exactly. So now the colors from the legend in the lower right corner map linearly from 10 dBZ to 55 dBZ. This way it matched most with values from neighboring sources that we know values of.