Possible display bug in Aviation forecast layers
-
Hello!
To reproduce this error, I'm using the ICON prediction model. Select the same hour for the next day.
In the Aviation section, first select Low Clouds, then Fog, and then Low Clouds again. You'll notice that the clouds have now changed shape and area.
Next, select something like Cloud Base, and then go back to Low Clouds. This time, the clouds appear as they did the first time.Or perhaps I'm doing something wrong.
The question is: which one of these forecasts should we trust?
Thanks! -
@GerSheph Could you please check it once again? Does the selected model change with the change of the layer?
-
@Suty That's not the issue.
Please follow these steps:
Select the "Low Clouds" layer and try taking a picture of what is displayed.
Now select "Fog" and switch back to "Low Clouds." You'll notice that the layer shows something different than before.
Select "Cloud base" then go back to "Low Clouds" once again, and this time, you'll see the same cloud coverage as the first time.
You can repeat this process multiple times.The problem is that the "Low Clouds" layer shows different coverage depending on which layer you were previously viewing.
Like, Low clouds <> fog <> low clouds <> cloud base <> low clouds <> fog <> low clouds
Thank you!
-
@GerSheph I checked the situation on our end and it works okay. What location do you check? Is it the global issue?
-
@Suty In Argentina. Look:
-
Hello GerSheph,
I see the issue in your video. While selecting the "fog" layer you unknowingly change from ECWMF model to the ICON model. Hence when you select "low clouds" afterwards, it's showing you the "low clouds" from ICON. For some reason it again changes back to ECMWF when selecting the "cloudbase" layer and with it you see your original "low cloud" data. Only the ICON models (ICON, ICON-EU and ICON-D2) plus UKV-model have the "fog" layer. So Windy will always change the model when you request the "fog" layer.
Kind regards,
Mark -
@Putje I see the problem now. Thank you very much!
-
@GerSheph We have an explanation, but the change of the model shouldn't be there most probably. I will take a look and forward the issue to colleagues.
