I’ve noticed this problem for years, It always seemed to me that a gust of wind picked up by my weather station and WD, as listed by testtags, “fell” between the minute updates to the log file. Sometimes as much as 5 of 6 MPH greater. When warranted I would correct the log file to preserve a potential daily/monthly record. Not a huge deal but it would be nice to not have to correct these gusts. There may be a great deal of difficulty to fix this given the transient nature of wind gusts compared to other weather parameters as measurements such as temperature.
I had a similar issue (I’m using km/h). The plot did not contain a peak gust but it might either miss it or reported a lower value. Few years ago Brian explained that WD does a routine check every NN minutes (1,2,10?) for the max gust in that period and then corrects the records, if the WD’s value is different from console’s by NN km/h or knots. This might explain the difference of few knots you see.