Wind speed differences between maps and wind tap

Hello,
Does anyone knows why for a same spot,time,model and altitude has different forecast?
For example on the first picture we have 28 km/h for icon eu model at 750 m which is the altitude for this village. At 15h
Then at the second picture the icon eu model says 22kmh for the same point and same hour.
Do you know why?.
Thanks a lot
Please check the pictures

@aug1234
I was unable to reproduce this problem
Sometime 1 kt difference. 
@aug1234
On the first image, wind is at 750m above ground.
On the second, wind is on the surface. 
@GkikasLGPZ Hello, but the village is at 750 m above the sea level, should be the same, right? Thanks a lot.

@idefix37 thanks a lot

@aug1234 Please read this post (7 years ago) https://community.windy.com/topic/5014/windspeedanddirectiondifferentheights
and
https://community.windy.com/post/20810 
@GkikasLGPZ is right. I had not seen that your setting was not at surface. The comparisons I made were with altitude slider set “at surface” as it should be for all forecasts at a location.

@GkikasLGPZ I really appreciate your comments thanks so much. Could you help me once more?
I’ve read your old post, and I understood : as the wind is “measured” at levels of pressure, then will be the same to compare the wind at surface in a spot which is 3000 m than set the altitude to 3000 m and select a point in the map which is at 3000 m. Is it right?But then my results are not the same and I’m very confused. Thanks a lot
Please see this example
In picture one the wind is 11km/h for 16 h on Saturday
On the second picture is 21 km/h for the same point and same hour.
With the explanation from your post I’ve expected to have the same result as the two pictures are at the same level of pressure (hpa)Thanks again

@aug1234
No, wind speed is not the same because wx. models are smoothing out steep peaks.
e.g. ECMWF has a grid resolution of 9 km. So, when you point at Pico de Mulhacen (elevation 3460m/11350ft), the model takes into account the mean elevation of an area 9x9km, wich is lower.
If you scroll the airgram to the right, and click on Sounding
you'll see that the model's elevation is only 2224m.The model's wind for elevation 13366ft (Pico de Mulhacen) is from 40 degrees (NE) at 40 kts (74 km/hr).

.... For Sat 13 Apr / 1600 .....
wind is forecast ENE 30 kt.
For differences in the wind in both sides of a mountain,
please read https://www.thoughtco.com/leewardandwindwardsidesofmountain3444015 
@GkikasLGPZ
I would like to start by thanking you for the time and dedication you have invested in trying to help me solve my doubts. It is greatly appreciated and very helpful. I am a beginner paragliding pilot and I am starting to study meteorology.From your last post, I have understood:

WX models approximate the real terrain relief, and this approximation is clearly seen in the sounding forecast in the model elevation section. So, I understand that the greater the difference between the real altitude and the model prediction, the more difficult it is to have an accurate prediction.

ECMWF (9km) means that this model has a resolution of 9 x 9 km, and the prediction is made at a mean altitude of the selected point, so predictions will be better in flat areas than in peaks, as the model altitude will be more similar to the real terrain altitude.

I have also learned to interpret wind direction in degrees, with North being 0º (simple, but I had never thought about what 40 degrees means). I understand it now.

From your last comment, I understand that the best way to know the wind prediction at a point is by using sounding and selecting the desired time and altitude.

I also found the link you shared very interesting.
Surely, by this point, I should have understood all your explanations, but I still have doubts. Could you tell me in simple words what the difference is between these two predictions? Same point, same time; one is surface wind at a point at 750m and the other is the prediction for a point at 750m where the point is at 750m altitude. I understand that selecting surface at this point is incorrect.
On the other hand, I now have three places on Windy where I can compare the wind for the same point and time:
 The basic time window, where the prediction is 26km/h.
 The general map where I can select the prediction altitude: surface (23 km/h) or at 750m (34 km/h).
 The sounding, which with an elevation model of 586m predicts 32km/h.
Which of all these predictions should I assume will be on the slope for takeoff? Around 23 km/h or closer to 34 km/h?
Thank you very much for your help." 

@aug1234
Concerning your points 1 to 4, they are well understood by you.For takeoff forecast (wind), the easy answer is "closer to 34km/h".
For your question "what the difference is between these two predictions? "
I want to point out that the surface wind (predicted 26 km/h by the model) in your location (elevation 750m=2460ft)
is not the wind at 750 m !!!
"Surface", for the model, is at 586m !!! as this is the "model elevation".
In general, "surface" reffers to the model's lowest atmospheric layer.
As the model's elevation is 586m, the "surface" wind of 26 km/h is a fictionary wind speed.
.
Better rely on the wind from the sounding (from140 degrees (SE) at 16 kts=30 km/h)
.

@GkikasLGPZ
Thanks so much for your answer.
I’ve just realized I was not predicting the correct wind speed after 2 years hahaha. I was always using the “search tool” entering the place and seeing the “basic model” just like this. I always thought that the search tool knew the point in the map with the correct altitude. And giving us the “surface wind” at that altitude.
Know I understand that maybe it works like this if it is in a really flat area, if not… probably the prediction for this point is some hundred meters below… so you need to change the altitude on the scroll bar or in the sounding forecast.
Thanks so much for all your comments! I really appreciate them!

@aug1234
May I have a feedback?
After all, how much was the wind at the point?