Why are we adding speeds for HW and/or gust, but with Tailwind we had nothing if we have a steady wind (no gusts of course)?
(a) the presumed friction layer used for certification is a power law (the 1/7th relationship) which describes a commonly observed reduction in steady wind velocity as the height above ground reduces in the absence of obstructions. It is presumed that the steady wind will be reasonably predictable in its behaviour. This relationship typically is the basis for low level wind profiles incorporated into Flight Manual data.
(b) for an headwind, the wind speed is expected to reduce somewhat as the aircraft approaches the runway resulting in an undershoot shear .. hence we carry a margin to give some fat for this. If a tailwind, we would be at double jeopardy if we were to carry a margin as the effect now becomes an overshoot shear situation.
(c) the usual Boeing margin (half the steady headwind) is an attempt to provide some protection against undershoot shear while not exposing the landing to an overly excessive speed entering the flare.
(d) for the gust situation, we presume a randomness about the gust time history so it is more appropriate to allow for all the gust value on the basis that this may appear at any time during the landing.
(e) one recalls that the landing factor is 1.67 dry and this provides a reasonable margin to accommodate a small variation in speeds in the flare. The 20kt limit reflects a reasonable maximum speed delta to fit in with the 1.67 if circumstances on the day conspire to find the aircraft entering the flare with the full additive still present. This has been a long used value both in US and UK certification practices.
An
older thread looked at these things in some detail and is probably worth a read in the context of this present thread.