@pi3-141592
Thank you for the quick reply. I have read the content of the page to which you sent me a link. I think I was misunderstood. The attached figure shows that the cloud cover (3-5 km high) is adequately reflected in the pictograms at the top of the graphic only in two cases. The case with question marks is not understandable to me.

Hello everyone,

there is a layer in Windy.com that displays the cumulative rainfall over a selected time period. In an analogous way, it would be possible to develop a layer that would return the average temperature over a given time interval. Such a layer would have a very high potential in terms of practical applications! Based on this data, administrators of housing cooperatives and owners of single-family houses would be able to determine very precisely, for example: the beginning and end of the heating season. An example condition could be formulated as e.g. “the heating season begins when the average temperature over a period of 3 days falls below 15 C”. In the transitional period, when some houses are heated with fireplaces, the knowledge of the average daily temperature allows very precise determination of the amount of firewood - especially when, for example, at the weekend the wood is chopped for the next week. Having such data, power plants would be able to predict the demand for electricity by heat pump owners. I don't want to prolong it any longer, but the applications are many. Such data is published by the Polish IMGW (attached photo) but not very precisely due to the small number of locations. The average temperature is a key parameter in medium and long-term forecasts, and it would be great how it would be available in your weather service.

Best regards
Pi3.141592

@pi3-141592 The bug has been fixed. The alerts are now working properly. Thanks a lot!

@pi3-141592
Thank you for the quick reply. I have read the content of the page to which you sent me a link. I think I was misunderstood. The attached figure shows that the cloud cover (3-5 km high) is adequately reflected in the pictograms at the top of the graphic only in two cases. The case with question marks is not understandable to me.

I would like to add that for the icon model, the symbolic notation (cloud, sun) of cloudiness is not consistent with the cloud cover plot (in the attached example, we observe a cloudless sky).

@korina I selected a clear sky in the alert settings.

I received an alert where you can see full cloud cover.

I am quite surprised.

Cloud setting in the alert does not seem to be working properly. Setting full cloudiness or no cloudiness produces the same result.