Clouds are gatherings of water droplets floating in the air. Water droplets above a given size fall down as rain. But what about the water droplets forming the cloud---aren't they also heavier than air?

My own guess is that clouds are held aloft on uprising air currents. I know this is true of some clouds. My question is whether this is always true. Can you have a cloud that stays aloft in completely still air?

I suppose that a way for a cloud to stay aloft in completely still air is if each droplet is falling down, but the ones at the bottom of the cloud evaporate, while further ones are condensing at the top. In that case it is continually 'raining' inside the cloud, but the whole looks still from a distance. So then a follow-up question: is this case common?

  • $\begingroup$ Good question, and it's also curious how clouds can keep a more or less constant shape for quite a long time... $\endgroup$ Mar 14, 2021 at 18:16
  • 2
    $\begingroup$ Yes clouds are heavier than air. They don't fall down because warm rising air keeps them up in the sky. $\endgroup$
    – user291777
    Mar 14, 2021 at 18:24
  • 1
    $\begingroup$ @AlistairBain I think with some more elaboration and perhaps some sort of source, this should be an answer. $\endgroup$
    – jng224
    Mar 14, 2021 at 18:38
  • $\begingroup$ naturemuseum.org/2017/11/how-do-clouds-float/…. scientificamerican.com/article/why-do-clouds-float-when Two good explanations. $\endgroup$
    – user291777
    Mar 14, 2021 at 18:45
  • 1
    $\begingroup$ @annav thanks---but don't forget, clouds are not made of water vapour (which is invisible); they are made of droplets of liquid water. Each droplet is very much denser than air, but the terminal velocity is quite low for small enough drops. $\endgroup$ Dec 14, 2021 at 13:19

1 Answer 1


Haha, this question is more evil than it seems.

In general, there is an updraft, that keeps the cloud particles going up, so that they either cancel out the terminal velocity, or they fall very very slowly, like in the scale of 20 m an hour.

This book says :

Cloud vertical velocities are important because the updrafts control the time scale $T_P$ and determine the cloud’s ability to suspend precipitation particles.

But, what happens though, when the water droplet starts falling? Before the water started falling, there was a moment where the vertical velocity was zero, of both the water (may be it was in vapor form) and the air parcel carrying it. Now, if you let the water fall, then it has a downward momentum. Does the air get an upward momentum?

This question remains poorly understood, as whether the momentum conservation should apply or not remains debated - given that there is a phase change - and the released thermal energy could be converted to kinetic energy and perhaps to a force. The momentum equations in the moments of phase change is a lesser known unsettled boundary of physics. In fact the same book does speak about it.

Now, when the droplet has just formed, it is still very very small, like a few microns. That is the so call "Aitken regime" of particle size. In this size, Brownian motion becomes more dominant than gravity. As such, momentum is not going to remain conserved any more. But the spaghetti is going to be more twisted.

Brownian motion in this regime will also kick of few atoms in the falling (or even rising) droplet. This is an effect akin to evaporation. This book gives a good overview.

The particles will vaporize fully, and if you manage to vaporize a particle in a parcel of air, then the parcel will transfer the heat (enthalpie) of its air into the water content. This will end up decreasing the density of the parcel, and it will then start moving up, until it reaches a height where it will cool again due to lapse rate and the process will begin again. That means, there is sort of a convection inside the cloud itself. Here is an overview inside a hurricane. Here is another.

Such a system might explain stability of clouds in a downdraft. But as I said, this spaghetti has not yet conclusively been combed out.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.