Chapter 2. Water vapour in the air

A glass of ice cubes set on a table covers itself in drops within a few minutes. The previous chapter showed where that water came from: from the air of the room. But a question remains, and it is the right one: why do the drops appear on the cold glass, and not on the table, nor on the wall, nor on the hand of the person watching?

The answer lies in a property of air that this chapter installs, and on which everything else in the volume depends. A volume of air cannot hold just any quantity of water vapour. It has a limit, and that limit changes with temperature: the warmer the air, the more it can hold. Météo-France puts it in one sentence: "In practice, the warmer the air, the more water vapour it can hold."

An image helps, provided one knows where it stops: air behaves like a sponge whose size depends on temperature. Warm it and it grows and takes up more. Cool it and it shrinks, and what it can no longer hold comes out. It is that last move that makes dew, fog, mist on glass and clouds. The image stops there: air does not absorb water the way a sponge absorbs, the vapour simply mixes with the other gases. But the limit that changes with temperature is exact.

Before speaking of water vapour, one has to know what it keeps company with. According to Météo-France, the atmosphere contains 78 % nitrogen and 21 % oxygen. Those two gases therefore take up almost all of the air on their own, and their proportion hardly moves from one place to another. One detail of reading avoids a false addition here: those two percentages apply to air stripped of its water vapour. The vapour is added to it and pushes the rest aside accordingly; that is why adding 78, 21 and the share of vapour comes to more than a hundred and means nothing.

Water vapour, for its part, takes up a modest and highly variable place: between 0.1 % and about 5 % of the surrounding air, again according to Météo-France. Remember the two words in that sentence: modest and variable. Modest, because water vapour is a minor component of the air. Variable, because it is the only component of the air whose proportion changes from one day to the next, from one region to another, and even from one hour to the next.

That variability is what makes weather possible. Air whose composition never changed would have neither clouds, nor rain, nor fog. It is because water vapour is constantly entering and leaving the atmosphere, by evaporation and by condensation, that there is such a thing as the weather.

When a volume of air holds all the water vapour it can hold at its temperature, it is said to be saturated. Add more, or cool it, and the excess can no longer remain in the gaseous state: it condenses.

Relative humidity measures the distance to that saturation. Météo-France defines it as "the ratio between the quantity of water vapour contained in the air and the maximum quantity possible, expressed as a percentage". At 100 % relative humidity, the air is saturated. At 0 %, it would be perfectly dry, which never happens in nature: in mainland France, humidity very rarely drops below 20 %.

The word relative is the one that misleads, and it is worth pausing on. A humidity of 80 % does not say how much water there is in the air: it says that the air holds 80 % of what it could hold at its temperature. Now that maximum changes with temperature. Air at 80 % on a winter morning at five degrees therefore holds far less water than air at 80 % on a summer afternoon at thirty degrees, even though the figure shown is the same. Twice the same percentage, two very different quantities of water.

That is why winter air, which often shows high humidities, dries out lips and hands once it has come into a heated house: by warming it, one raises its maximum capacity without adding any water to it, and so its relative humidity falls.

This whole chapter comes down to a single condition, and it holds as well for a glass of lemonade as for a cloud several kilometres across: there is condensation when the air reaches its dew point temperature, that is to say when it is cooled to saturation.

There are therefore only two ways of obtaining condensation. The first is to add water vapour to air that is already close to its limit. The second, by far the more frequent in nature, is to cool the air without adding anything to it. Cooled air sees its maximum capacity fall; at some point that capacity drops below what the air is already carrying, and the surplus falls out in the liquid state.

Take up the glass of ice cubes again. The glass itself manufactures nothing: it cools the thin layer of air touching it. That layer of air, cooled below its dew point, can no longer keep all of its vapour, and the surplus settles on the nearest surface, which is the wall of the glass. The table and the wall, for their part, are at room temperature: they cool no air, so nothing settles on them. The question asked at the opening is answered.

It remains to be seen what cools a surface when nobody is cooling it. The answer lies in an observation everyone has made: a clear night is colder than a cloudy one. An object left outdoors constantly loses heat towards the sky; when the sky is clear, nothing sends that heat back to it, and it falls well below the temperature of the air. When a screen covers it, a cloud, a tree, a canopy, a garden table, that screen sends back part of what it loses, and it stays warmer. It is that simple sharing which explains why dew settles on grass in the open and not on sheltered grass.

Note the word that has done the work: cool. Remember it. The chapters that follow will do nothing but decline it: cooling grass during the night gives dew, cooling all the air of a valley gives fog, cooling air by making it rise gives a cloud.

Fog deserves a section of its own because it makes the bridge to the next chapter. Météo-France defines it as "the suspension in the atmosphere of very small water droplets reducing visibility at ground level to less than one kilometre", and adds a sentence that is worth a whole lesson: "fog is in fact a cloud whose base touches the ground".

In other words, there are not fogs on one side and clouds on the other: there is a single object, and the only difference is the height of its base. To walk in fog is to walk in a cloud. A walker climbing towards a layer of low cloud goes through the same thing without changing its name, and the next chapter will give that layer its exact name.

The one-kilometre visibility threshold is not a convenience of language: it is a measurement convention, which lets every station in the world code the same situation in the same way. Below it, one speaks of fog; above it, of mist. A shared figure makes observations comparable, and the chapter on measurement will come back to this.