@Filip_K Thank you very much for the quick fix!

Hello,

I've noticed that the madis-F4072 PWS, located in Chisinau, Moldova, often reports temperatures that are too high. The problem appears during daytime hours only and is most severe in sunny weather, when the temperatures can be up to 3-7 degrees Celsius higher compared to other PWSes in the city and official stations. This hints to inadequate sun shielding as being the root cause of the overestimated temperatures.

This station uploads data to Weather Underground as well, under the ID ICHIINU9 (I've compared the readings at the same time moments - they are identical). Using the WU archive, I've made a plot of daily maximum temperatures at different stations in Chisinau during July 2024, when anomalously hot and sunny weather was prevalent. The overestimation is clearly seen.

I wonder if it is possible to contact the owner of the station in order to fix the shielding or stop reporting temperature (and likely dewpoint) data, for example?

I can provide more comparisons if it's required.

A quick example - the data is about 22 minutes old on this screenshot (14-minute update delay + 8 minutes after the latest update).

Dear @Filip_K, thank you very much for the quick integration!

Dear @Filip_K, thank you very much for the quick integration!

Radar data freshness isn't explicitly shown anywhere on Windy, so anyone not familiar with the peculiarities of Romanian data can mistakenly use 20-25-minute-old data (15 minutes of delay + up to 10 minutes of waiting for the next update) as fresh ones.

In addition to that, the combined effects of high delays and slow updates result in substantial still periods when animating Romanian radar data, as can be seen on the video below about 0:05-0:09 time marks.

A quick example - the data is about 22 minutes old on this screenshot (14-minute update delay + 8 minutes after the latest update).

Dear Windy developers,

I wanted to bring to your attention that data from Romanian weather radars is now available through an open data server: https://opendata.meteoromania.ro/radar/.

This new data source offers significant advantages over the current one, which (as far as I know) relies on image overlays from https://www.meteoromania.ro/radarm/radar.index.php. Specifically:

• Lower update delay: about 4-5 minutes, compared to ~15-minute delay of the current source.
• Faster individual radar updates: If data from individual radars is used instead of the composite, the update interval is 5 minutes rather than 10 minutes.

In my opinion, integrating this source could improve real-time weather tracking on Windy.

Hello,

I've noticed that the madis-F4072 PWS, located in Chisinau, Moldova, often reports temperatures that are too high. The problem appears during daytime hours only and is most severe in sunny weather, when the temperatures can be up to 3-7 degrees Celsius higher compared to other PWSes in the city and official stations. This hints to inadequate sun shielding as being the root cause of the overestimated temperatures.

This station uploads data to Weather Underground as well, under the ID ICHIINU9 (I've compared the readings at the same time moments - they are identical). Using the WU archive, I've made a plot of daily maximum temperatures at different stations in Chisinau during July 2024, when anomalously hot and sunny weather was prevalent. The overestimation is clearly seen.

I wonder if it is possible to contact the owner of the station in order to fix the shielding or stop reporting temperature (and likely dewpoint) data, for example?

I can provide more comparisons if it's required.

@Filip_K Thank you very much for the quick fix!

@Filip_K Yes, these dBZ values are correct.

There's an article about the nowcasting system in Romania where the scale can be found: https://www.todaysoftmag.ro/ampArticle/3631/avertizarile-meteorologice-de-tipul-nowcasting.

Dear Windy developers,

There's an issue with the Romanian radar mosaic as it's presented in Windy, and it persists for multiple months as of now. The reflectivity values are significantly lower than in the original maps available at https://www.meteoromania.ro/radarm/radar.index.php.

An example from today: 26 dBZ on Windy and 35-40 dBZ in the original image.

The color scale for the original image starts at 5 dBZ and proceeds in 5 dBZ increments.

The issue has already led to dangerous precedents like the one below, when an extremely severe hail-bearing supercell had reflectivities of mere 45 dBZ.