Hi, sorry for the late reply. I only briefly read the second link, but I don't think it's what I was talking about. I read the Wikipedia article on thermals a week ago, and it seems a bit brief. I was a bit quick with my first reply as well. This is a pretty big subject.
Thermals can be triggered by any perturbation in the atmosphere - a gust of wind, a truck on the highway or a temperature differential caused by the sun. But they are _sustained_ by vertical temperature differences in the atmosphere: As an airmass rises, it expands due to the lower atmospheric pressure. This expansion causes the airmass to cool adiabatically, approximately one degree Celcius per 100 meters of gained altitude.
However, meteorological conditions can cause the temperature distribution of the _surrounding_ air to be different. For instance, after a clear night with no clouds, perhaps the air from the ground up to 2000 meters altitude actually decreases in temperature 1.1 degrees Celcius per 100 meters of altitude gained. If an airmass on the ground starts moving upwards in these conditions, it will still cool adiabatically - but slower than the rest of the atmosphere cools due to the meteorological conditions. So a rising airmass will actually accellerate, because the airmass grows hotter relative to the surrounding air as it rises. This is a thermal. It will keep moving until the temperature of the atmosphere starts rising with altitude. Condensation (cloud formation, cumulus clouds are caused by thermals) can increase its vertical speed.
But to answer your initial question: Yes, they can and do. But when the atmosphere is so _unstable_ (unstable atmosphere == temperature falls very quickly with altitude) that thermals are spontaneously triggered without help from the sun, it will usually mean bad weather or thunderstorms later in the day. So if the air seems calm at the surface, it will pretty much be guaranteed to be calm at 10 meters altitude. You can detect thermals at ground level by rapidly changing wind direction, often on nice summer days. In the air at low-ish altitudes (less than 2000 meters), thermal activity is indicated by turbulence.
Thermals can be triggered by any perturbation in the atmosphere - a gust of wind, a truck on the highway or a temperature differential caused by the sun. But they are _sustained_ by vertical temperature differences in the atmosphere: As an airmass rises, it expands due to the lower atmospheric pressure. This expansion causes the airmass to cool adiabatically, approximately one degree Celcius per 100 meters of gained altitude.
However, meteorological conditions can cause the temperature distribution of the _surrounding_ air to be different. For instance, after a clear night with no clouds, perhaps the air from the ground up to 2000 meters altitude actually decreases in temperature 1.1 degrees Celcius per 100 meters of altitude gained. If an airmass on the ground starts moving upwards in these conditions, it will still cool adiabatically - but slower than the rest of the atmosphere cools due to the meteorological conditions. So a rising airmass will actually accellerate, because the airmass grows hotter relative to the surrounding air as it rises. This is a thermal. It will keep moving until the temperature of the atmosphere starts rising with altitude. Condensation (cloud formation, cumulus clouds are caused by thermals) can increase its vertical speed.
But to answer your initial question: Yes, they can and do. But when the atmosphere is so _unstable_ (unstable atmosphere == temperature falls very quickly with altitude) that thermals are spontaneously triggered without help from the sun, it will usually mean bad weather or thunderstorms later in the day. So if the air seems calm at the surface, it will pretty much be guaranteed to be calm at 10 meters altitude. You can detect thermals at ground level by rapidly changing wind direction, often on nice summer days. In the air at low-ish altitudes (less than 2000 meters), thermal activity is indicated by turbulence.