The temperature at which the air first saturates and dew forms. It tracks the actual water in the air: above about 16 °C feels sticky, above about 21 °C oppressive.
The lowest temperature evaporative cooling can reach. When it nears skin temperature (about 35 °C) the body can no longer shed heat.
A percentage of what the air could hold at its current temperature, so it falls when the air warms even though no water has left.
Relative (how full), absolute (grams per cubic metre) and specific (grams per kilogram of air) respond very differently to warming.
Three different questions about the same water. They never share a unit on this page, and each keeps its colour everywhere: in the readouts, the tables and the chart below.
How full the air is. The water present as a percentage of the most the air could hold at its current temperature. It says how close the air is to saturation, not how much water it contains, so it needs a temperature beside it. The number for mould, condensation and static.
RH = 100 · e / e_s(T)Warm the air 10 °C: roughly halves.
Water per cubic metre. The grams of vapour in one cubic metre of air: a density. The natural measure when a volume is what matters, such as a room, a drying cabinet or a museum case. Warm air expands, so the same water spreads over more cubic metres.
AH = 216.7 · e / (T + 273.15)Warm the air 10 °C: falls about 3%.
Water per kilogram of air. The grams of vapour in one kilogram of the whole air, vapour included. Mass does not change when air is heated, cooled or squeezed, so this stays put until the air reaches its dew point and water condenses out. The meteorologist's measure.
q = 622 · e / (P − 0.378 e)Warm the air 10 °C: no change.
Water per kilogram of dry air. Specific humidity's close cousin, and the vertical axis of the chart above. It divides by the dry air alone, so it is always slightly larger than specific humidity. HVAC engineers call it the humidity ratio.
w = 622 · e / (P − e)Warm the air 10 °C: no change.
Both are reported in g/kg, which is exactly why they get muddled — but they divide the vapour by different things. Mixing ratio (w) is grams of vapour per kilogram of the dry air alone. Specific humidity (q) is grams of vapour per kilogram of the whole moist air — dry air plus the vapour itself. Because q's denominator includes that vapour, q is always a shade smaller than w.
How much smaller? Negligible when it's dry, growing with moisture: about 0.5% apart at 5 g/kg, ~2% at 20 g/kg, ~3% at 30 g/kg. So for everyday comfort talk they're interchangeable — but in precise energy balances, or in hot humid and tropical air, the gap is real. Rule of thumb: HVAC engineers use mixing ratio (they call it humidity ratio) because kit adds or removes water from a fixed stream of dry air; meteorologists prefer specific humidity because the total mass of a moving air parcel is what's conserved.
q = w / (1 + w) with both in kg/kg · both ≈ 622 · e / P when humidity is lowWhat a normal thermometer reads — the air temperature, full stop. It's the horizontal axis of the chart and the single biggest lever on how much vapour the air can hold. Every other moisture quantity is "relative to" this number in some way.
Cool the air, holding moisture fixed, and this is the temperature at which it first saturates and dew forms. Because it tracks the actual vapour amount, dew point is a far better "mugginess" gauge than RH: above ~16 °C feels sticky, above ~21 °C feels oppressive. On the chart it's where a horizontal line from your point meets the saturation curve.
What a thermometer reads with a wet wick around it, cooled by evaporation. It's the lowest temperature reachable by evaporative cooling — the limit a swamp cooler or sweating body can hit. When wet-bulb approaches skin temperature (~35 °C), the body can no longer shed heat: a hard survival limit. Always sits between dew point and dry-bulb.
The slice of total air pressure contributed by water vapour alone (e), against its saturation value (e_s) — the most the air could sustain before condensing. Relative humidity is literally these two divided. Everything else on the chart is derived from e.
Total heat content of the air — sensible heat (its temperature) plus the latent heat locked in its vapour. This is what an air-conditioner actually has to remove, and why drying humid air is so energy-hungry: most of the work is condensing water, not lowering temperature.
The psychrometric chart plots dry-bulb (across) against mixing ratio (up). The bright cyan saturation curve is the 100%-RH ceiling — air can't exist above it; a faint dotted copy stays put so the line is always legible even while the glowing one draws in. Dashed lines are constant RH; the green box is a rough thermal-comfort zone.
Pick a city under Climate overlay and its twelve monthly-mean states are plotted and joined into an annual loop, coloured cool-to-warm by month. The shape tells the story: maritime places like London or San Francisco trace a tight, near-vertical loop (mild all year, moisture doing most of the moving); continental places like Minneapolis or Moscow sweep a long diagonal (cold-dry winter to warm-humid summer); monsoon climates like Mumbai hook sharply upward as the rains arrive. Step to a month — or hit Play year — and the calculator loads that month's mean, with a whisker showing the average daily low-to-high range (nearly horizontal, because the moisture barely changes between night and afternoon).
One thing to know about “Annual mean”. It is the year’s air averaged — mean temperature carrying mean vapour pressure — and not the average of the twelve monthly percentages. Those are different numbers, sometimes very different. Beijing’s twelve months run 44–77% and average 57%; its annual mean reads 75%, higher than eleven of the twelve. Two things cause it: relative humidity is a ratio and the saturation curve is exponential, so the mean of the ratios is not the ratio of the means; and the humid summer months hold most of the year’s moisture, which pulls the mean vapour pressure up while the mean temperature sits in the middle. Mean temperature and mean mixing ratio have no such problem — they are straight averages, and the annual point is plotted where it belongs, at the centre of the loop. Only the percentage behaves this way, and it does it in every continental climate: Minneapolis +17 points, Chicago +12, Moscow +11. Maritime and desert cities barely show it.
Air holds water vapour, but only up to a temperature-dependent limit. A psychrometer works out the full moisture state of moist air from a couple of inputs — typically temperature and relative humidity — and derives the rest: dew point (the temperature at which air becomes saturated and condensation forms), wet-bulb temperature, and the actual water content. Type the state into the bar (25c 60%, 72f dew 55), pick which two numbers you know, or drag the point on the chart. The strip above the chart gives the verdict — dew point is the comfort number, wet-bulb the heat-stress number — and the chart reads both off the saturation curve the way a paper chart does: straight left for dew point, down the wet-bulb line for wet-bulb.
Air at 25 °C and 60% relative humidity has a dew point of about 16.7 °C. Cool a surface (a window, a cold drink) below that and water condenses on it.
Raise the temperature without adding moisture and the relative humidity falls, even though the actual water content hasn't changed — which is why heated indoor air in winter feels dry.
Relative humidity is a percentage that changes with temperature; dew point is a temperature that tracks how much water is actually in the air. Dew point is the better comfort guide.
Relative humidity (%) is how close the air is to saturation at its current temperature. Absolute humidity (g/m³) is the grams of water in each cubic metre of air. Specific humidity (g/kg) is the grams of water in each kilogram of the air. Warm the air and the first falls a lot, the second a little, and the third not at all.
Warmer air can hold more vapour, so the same amount of water is a smaller fraction of the maximum — lower relative humidity — even though nothing was removed.
The lowest temperature reachable by evaporative cooling at the current humidity. It matters for comfort, cooling-tower performance, and — at high values — human heat tolerance.
Whenever a surface drops to or below the dew point. That's the principle behind everything from foggy windows to the water that drips from an air conditioner.
Releases in which this page changed, newest last. Derived from the archived copy of every release, not from notes written afterwards — so it reflects what actually shipped. Site-wide passes are left out; they are in the full changelog.