Pekka
You wrote (regarding Graeff’s experiments) “It’s very easy to disturb the system enough to initiate convection.”
Now, Graeff acknowledged that he has had very little formal training in physics and, frankly, that is obvious. For a start he incorrectly multiplied by the degrees of freedom and was thus expecting steeper thermal gradients than really would be the case.
But he also went to a lot of unnecessary trouble to prevent convection. You also see convection as a problem But it’s not, because convection just “follows” the thermal gradient which evolves spontaneously as a corollary to the Second Law of Thermodynamics.
This is what happens in the real world in calm conditions: the thermal gradient forms as the state of thermodynamics equilibrium. Then, if new thermal energy is absorbed, convection takes place in all accessible directions away from that source. For example, when the Sun warms Earth’s surface, convection will be only upwards away from the surface, because the air cannot move down through the surface, of course. But when new energy is absorbed at dawn in the upper troposphere of Venus, some of that energy moves downwards to what are warmer regions, but the energy flow follows the gradient and is helping to form a new state of thermodynamic equilibrium.
There! I just about given away most of the content of my book “Why it’s not carbon dioxide after all” which is now in the production stage with the text finalised. It should be available late April through Amazon and Barnes & Noble, so I’ll leave it at that here.
D C