[TassyJim]

I strongly recommend to not do this. The wind gust should be sampled at a most shorter interval than 12 seconds. You will lost readings of real wind gusts, and most important the readings of average wind speed will also be affect by your change (if you don't believe it, just try with default setting and then with the changed one, then compare the graphics). Resuming,  your wind readings will be much flat.

Many weather station (FO and many others), which send RF data with longer intervals, usually send the highest wind gust recorded within the period between each data packet. WeatherDuino don't work this way, it works more like the Davis Weather Stations (which send wind data at each 2.5s - WeatherDuino sends each 3s).

There is even other implications... if one data packet with wind data is lost (by any reason) your readings at the receiver will have a 24 seconds interval.

I thought about this for a long time.
My other system uses the Davis protocol but is connected to my Cumulus software with hard wiring so RF is not a problem.

One packet takes 380mS and sending that every 3 seconds was too high a duty cycle for me. I have other RF data links also on 433MHz and I am worried about interference. I had to accept the loss in wind data.

I will set up a second instance of Cumulus and connect it to the WeatherDuino and do a comparison.
Not much use today, we have very little wind.

Jim