I was reading this question: Physical state of Fog And it made me curious about the larger question of how fog exists at all, that is, what causes an aerosol to remain suspended in air?

There's an article in Scientific American that tries to elucidate this but it contains statements that make no sense to me such as "The water and ice particles in the clouds we see are simply too small to feel the effects of gravity". I thought everything feels the effects of gravity.

I don't have a physics background but I understand basic science so could someone please explain in educated-layman's terms what causes water droplets to float in the air instead of plummeting to earth? What role do the characteristics of the air play? If the air was perfectly still (no wind or updrafts) what would happen?

If we had solid spheres of the same density as water (so evaporation wasn't a factor) and we reduced the air pressure or changed the composition or temperature of the air how would it affect their suspension?

  • $\begingroup$ This doesn't apply to the droplets in fog, but it's easy to think that humid air (i.e. air with dissolved water vapour) is heavier than dry air at the same temperature and pressure, and that's not true: water molecules are lighter than air molecules, so humid air is lighter than dry air. $\endgroup$ – Emilio Pisanty Feb 3 '17 at 12:40

The tiny droplets of water indeed tends to fall under the effect of gravity. However, as any other particle falling in the atmosphere, they accelerate downwards until they reach a terminal velocity. For a spherical particle of mass $m$ and radius $R$, falling in a fluid of viscosity $\eta$, the terminal velocity, neglecting buoyancy, is given by $$v_t=\frac{mg}{6\pi\eta R}.$$ For an aerosol droplet of radius $\approx 10^{-6}\, m$ this velocity is very small, typically $10^{-3}\, m/s$. At the same time, the droplets are being constantly pushed upwards by convection currents, since the hot air raises.

  • $\begingroup$ So to summarise, it really is falling, but very slowly but there are enough air currents to counteract that. So if we could create a controlled environment where the air was perfectly still, I assume the droplets would eventually fall to the ground. To test that I assume we would need to use solid spheres of the same density as water so that evaporation would not cool the surrounding air and create convection currents, true? $\endgroup$ – user316117 Jan 23 '17 at 17:37
  • $\begingroup$ @user316117 In a controlled environment they would fall with that small terminal velocity. This was actually used by Millikan in his famous experiment with oil droplets where he found the value of the elementary electrical charge. $\endgroup$ – Diracology Jan 23 '17 at 17:41

So there's two parts to this. The first is that the gaseous vapor in the air does not "want to be" condensed unless it has the appropriate pressures and temperatures around it. The bottom of a typical cumulus cloud is "flat" because the stuff that's falling down beneath this surface is re-evaporating while the stuff that's above the surface is condensing. In other words, clouds are a very non-equilibrium phenomenon; they look fixed and constant but it is very similar to how a candle flame looks fixed and constant: secretly we know that the wax molecules involved in combustion are always new ones from the solid candle-body, and the candle can only be sustained if that body is being consumed. Similarly, clouds are always part of a heat-transfer process from a warmer region to a colder region.

That would not account, however, for the "puffy" tops of these clouds. To understand those, you have to understand that when the vapor condenses into a droplet it releases its heat to the surrounding air, which warms up and therefore, since hot air rises, it feels a buoyant force and heads upwards. It "pulls" the droplet a bit with it, and since the droplet is small it has a high surface-area-to-volume ratio and this "pull" is actually relatively significant. That's what causes the droplets to fly upwards and create these puffy tops of the clouds, compared to their flat bottoms.

A fog is just a cloud that is formed at surface level. The basics are not so different and just require the wind to be blowing the damp air in a direction other than upwards. If the damp air comes into a colder space, even if it blows across the surface or up a mountain, then under the right circumstances it will form those low-lying clouds that obstruct visibility that we call fog. Again, the droplets still, as they form, heat up the surrounding air a little bit and this drags them upwards a bit, so that they seem to defy gravity. Unlike in the atmosphere, there is no surface that they will re-evaporate underneath, so they eventually just fall to the ground: this is why you might get damp when you're out in the fog, as this moisture hits you on its way down.

  • $\begingroup$ Thanks for all that, but I'm not really interested in clouds, per se. Clouds are just little spheres of water so I'm trying to reduce this to the physics of one sphere of the density of water, in air. What keeps it from falling to earth? $\endgroup$ – user316117 Jan 23 '17 at 17:30
  • $\begingroup$ That's what I'm telling you: if it's in a cloud, what stops it from falling to the earth is that it re-evaporates. But if it's in a fog, which is just a low-lying cloud, then it either falls back to the earth or it's blown into a warmer place.There is nothing that stops it from falling to Earth. It happens to be dragged slightly upwards because there is a net upcurrent of air, and it happens to fall at its terminal velocity, which is very low: but there is nothing else stopping it from falling to the Earth. $\endgroup$ – CR Drost Jan 23 '17 at 17:36

"The water and ice particles in the clouds we see are simply too small to feel the effects of gravity". I thought everything feels the effects of gravity.

You are correct and the former statement is indeed wrong.

Fog, however, doesn't really 'float'.

Fog is a suspension of very small water droplets in air. Although the droplets experience gravity like any other 'massive' object, they also experience other forces that kind of stabilise the suspension. The droplets undergo air drag, roughly modelled by Stokes' law, which greatly reduces their tendency to fall.

Furthermore, weak convection currents in the air (due to small temperature differences) also help keeping the droplets airborne, as can other types of weak air currents.

Given enough time, fog will either be dispersed by wind or coalescence will cause droplets to clump together and form rain.

In very static, uniform conditions, fog will eventually precipitate or dissipate, even in the absence the aforementioned effects.


This is the result of Brownian movement.

As you may know, that fog is a colloid called aerosol in which the dispersed phase is liquid and the medium in which the liquid is dispersed in air.

enter image description here

Brownian movement has a stirring effect which counters the force of gravity acting on colloidal particles and hence providing stability to colloidal sols by not allowing them to settle down.

When a large mass of air containing dust particles in cooled down below its dewpoint, the moisture from the air condenses on the surface of these particles forming fine droplets. These droplets being colloidal in nature continue to float in air in form of mist or fog.

  • $\begingroup$ This does not really answer the question. Brownian movement is the microscopic driver of diffusion, but if you only consider gravity and brownian motion, the probability cloud of your particle will still fall to the ground. $\endgroup$ – Emilio Pisanty Feb 3 '17 at 12:08
  • $\begingroup$ Yeah u r right...but Brownian movement surely can slow down the process...and as the particles come down new ones can be formed and join in the colloid... $\endgroup$ – Mitchell Feb 3 '17 at 12:21
  • $\begingroup$ In foggy days grasslands are wet...but still the fog appears... $\endgroup$ – Mitchell Feb 3 '17 at 12:26

protected by Qmechanic Jan 23 '17 at 17:28

Thank you for your interest in this question. Because it has attracted low-quality or spam answers that had to be removed, posting an answer now requires 10 reputation on this site (the association bonus does not count).

Would you like to answer one of these unanswered questions instead?

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