Bug: Wind direction on map differs from wind direction shown at location picker and in timeline

[Ftjwinfs]

I just found a serious inconsistency in the data between the map, the indication shown on the map at the location picker and the timeline shown in the lower part of the screen.

The streamlines on the map show me easterly winds. The direction indication at the location picker and also the wind arrows in the timeline show a northerly wind.

Please see screenshot. The selected model is Meteoblue. You can clearly see the above described inconsistencies

Can you confirm this as a bug?

Also, there seems to be an issue with the color on the map. While for the Meteoblue model 22kn of wind are shown as green, for the UKV model 22kn are shown as orange, which is consistent to the timeline colors (see screenshot below.

[idefix37]

@Ftjwinfs
It is a voluntary bug ? Meteoblue is only available at point forecast. It is a non-gridded model which does not allow to draw a map. So in Windy when Meteoblue is selected in the point forecast for wind, it is the ECMWF that is displayed in the map despite Meteoblue is labeled.
So the colors are consistent since the overlay is the one of ECMWF (not Meteoblue)

[Ftjwinfs]

@idefix37 Ok, understood. But this is grossly misleading and Windy really should show a notification, that the map does not show the selected model but something else.

I just realized, that when I only show the map, Meteoblue is not a model I can select. This supports what you wrote. But as soon as I open the "timeline" at the lower part of the screen, Meteoblue is shown an option. The general behavior of Windy implies, that as soon as I select a model, the data shown on the map and in the timeline comes from the same model. A "fallback" to ECMWF should not be "silent", but the user needs to be made aware of this.

[idefix37]

@Ftjwinfs
By the way, I saw you posted several screenshots with point forecast over the sea.
Note that near the sea shore, for wind and temperature Meteoblue is not so good as explained here:
https://community.windy.com/topic/30040/meteoblue-vs-nems-naming/5?=1720626260780
https://community.windy.com/topic/28594/wildly-incorrect-temperatures-and-rain-forecast/5?=1720626260770

[Ftjwinfs]

@idefix37 Thank you! Actually I do not use Meteoblue for sailing forecasts, I rather go for ICON and ACMWF. But since Meteoblue is always part of the shown forecast in the timeline, I accidentally stumbled upon what I documented above.

[luis.delatorre]

Hi @idefix37

I just want to follow up on this topic, because I'm having the same issue. I'm a paragliding pilot and I use Windy to check for the wind direction. As you can see from the the attached image. That for today (Wednesday 13th at 20:00) the wind direction from the weather models don't reconcile with the particles shown on the map.

The wind direction from the particles indicate a SW wind at 20:00. However, the different models show the following wind directions:

• ECMWF: W Wind

• GFS 22: SE

• METEOBLUE: E Wind

• ICON-D2: E Wind

image url)

Am I ready this correctly or like in the previous case the wind direction marked in Windy takes a differnet model?

Thank you in advance.

[idefix37]

@luis-delatorre
Your screenshots are blurry. It would be better to crop them.
It seems the first screenshot show at 20:00
GFS predicts SE, same in the table
ICON D2 predicts E, same in the table

What does the second screenshot is supposed to show? On which model is it based?

If I do the same exercise, I get consistent wind directions

By the way with Premium you would have hourly forecast in the table.

[luis.delatorre]

@idefix37 said in Bug: Wind direction on map differs from wind direction shown at location picker and in timeline:

If I do the same exercise, I get consistent wind directions

Hi @idefix37 !

Hopefully, these screenshots are better.

The first screenshot shows the different models where wind directions come from either E, SW, or W. The second screenshot shows the wind direction particles from SW, which doesn't reconcile with the models.

[idefix37]

@luis-delatorre
If the second screenshot is from ECMWF, the SW wind direction is not so far from the one shown in the multi-arrows of the first screenshot ? Where ECMWF seems to be WSW