6
$\begingroup$

On a partly cloudy day, many individual clouds can be observed in the sky. I would expect that the moisture in the clouds would want to diffuse out into the other less dense non-clouded areas.

What holds the clouds together and prevents this? Is it a form of surface tension?

$\endgroup$

1 Answer 1

4
$\begingroup$

I think the answer in this reddit discussion is basically right.

But first, I should point out that pure diffusion is excidingly slow. A quick calculation shows that it would take 500 years for an average water molecule to diffuse 1 meter. Of course, in most cases real particles diffuse much faster thanks to small turbulent flows.

This already hints at one of the keys to understanding clouds: flow is much more important than diffusion. In the atmosphere clouds are essentially formed by convective flow (hot air raising, cooling, and falling down). In particular, a cloud forms when the air parcel is at the top of this cyclic flow (when the air is coldest).

Still, in this convective flow, one may expect the cloud to get deformed or something, as it goes with the flow. However, the second key point is that clouds are not a particular parcel of air, but a cold region in the atmosphere. Therefore, the right picture is a cloud being like a little ripple in a stream, which has a constant shape, even if the water that forms it is constantly changing. The cold region (the cloud) is still moving, due to wind, but wind is more or less homogeneous over much longer length scales than clouds, and doesn't deform them much. A mountain can help "lock" these cold regions, though, as seen here. This is a rather common pattern in nature, and self-organization. In fact I think that not only the stability of clouds, but the particular length-scale of clouds is just a manifestation of the famous patterns that self-organize in Rayleigh-Bernard convection, such as cells, and rolls. See here for a demo.

I think these are the key reasons why clouds appear as rather localized things. However, some turbulent diffusion may still be expected to diffuse a non-negligible number of water droplets around. However, another key idea weakens this effect: clouds can be seen because they are dense, this is just because their white color is due to light scattering. This means that even if there are some cloud droplets floating around in the "clear" areas of the sky will still appear clear, and we will only see cloud where the concentration is high, which is where the droplets are being produced from condensation, i.e. in the cold regions discussed above.

$\endgroup$
2
  • $\begingroup$ This is a wonderful answer. I see your main expertise on this site seems to be statistical mech, and it shows in your eloquent description of several things I have never thought about / known. It's a shame we don't see more of you here. $\endgroup$ Commented Jul 10, 2016 at 2:35
  • $\begingroup$ Thanks! StatPhys is indeed one of my passions. I guess I don't often have much time, but I'd like to come here more often yeah. $\endgroup$
    – guillefix
    Commented Jul 10, 2016 at 2:54

Your Answer

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

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