I've followed most of this thread and have found it interesting. I
think Mike Borgelt's analysis is a good accounting of what's going on.
I get to deal with underground mine ventilation and it has some
interesting analogies to this situation, water vapor and thermals. In
some situations moist air underground will be lighter than air in
other parts of a mine or tunnel and the difference in the "wieght"
will be sufficient to change the air flow - just because of humidity!
Most people, and pilots too - are surprised that dry air is heavier
than air with some moisture in it. A parcel at a given temperature
will be the lightest at 100% RH. I've been in thermals that I'm sure
originated in stock tanks or isolated ponds out west where the air is
"relatively" dry. It doesn't occurr in New England though, I'd guess
because there isn't enough of a humidity differential between the two
air parcels ( over the water, and adjacent ) and thier relative
humidities. Out west where the air is drier this does occurr and could
very well trigger a thermal. And then probably in conditions where
land forms or the area is conducive to preventing mixing with
surounding air by wind - a sheltered area or no air movement to allow
a parcel to build above a water source. And then the bouyant
difference in the air masses due to the humidity diferences has to be
enough to overcome the air cooling near the surface of the water ( the
swamp cooler thing! ) by the evaporation taking place into the air
above the water. Clear as mud - Mark Guay