I found discrepancies between wind numeric value and its representation


The wind vector shows 5 kt less than the numeric value when the speed is divisible by 5 kt.
If it is due to unit conversion and rounded value, e.g. 64 km/h = ~34.56 kt, rounded value = 35 kt, I suggest to round the vector to the closest value instead of rounding down, in the example vector of 35 kt instead of 30 kt.

Moreover, it is possible to show the wind vector for 5 kt with the short line at the second position like the third picture


source: https://upload.wikimedia.org/wikipedia/commons/c/c6/Wind_barbs.gif

Alexandre
Based at LSGL (Lausanne - La Blécherette, in Switzerland)

I guess the barbs represent the closest value (in steps of 5 kts) that is equal to or larger than the actual windspeed. So 51 kts would be represented by a pennant (50 kts) and a half-barb (5 kts) even though just a pennant would be closer to the value. This must be to give the pilot a firm number he can rely on and be sure that the wind will not be stronger than it. There probably is a set definition of how to convert wind speed to barbs. Tim, do you know?

-- Chris.

But that is the opposite of how SD does it.

Is there an operational reason why it would be more relevant for the pilot to have the wind speed rounded down to the next lowest multiple of 5 instead of rounded to the closest multiple of 5?

Usually, you would prefer the "nearest" rounding as it would minimise roundoff errors in calculations. The only real exception is when it is important to always round in a "safe" direction, e.g. when checking operating minima.