How this instrument works
Dew point is the temperature air would have to cool to, at the same pressure and with no moisture added or removed, before the water vapor already inside it starts condensing into liquid. It is a genuine physical threshold, not a percentage: chill a parcel of humid air past that point and dew beads on grass, fog forms in a valley, or a cold glass starts to sweat, all from the same mechanism. The formula reaches it by inverting the Magnus curve that describes how much vapor pressure saturated air can support at a given temperature — instead of asking how much this air can hold, it asks at what temperature the vapor already present would be exactly enough to saturate it.
That inversion is why a logarithm shows up. The Magnus-Bolton saturation curve is exponential in temperature, so solving it backwards for temperature pulls a natural log of the humidity ratio out of the exponent: γ = ln(RH ⁄ 100) + 17.67T ⁄ (T + 243.5) folds the actual vapor pressure into one dimensionless term, and Td = 243.5·γ ⁄ (17.67 − γ) rearranges the same curve to read off temperature instead of pressure. The constants 17.67 and 243.5 are David Bolton's 1980 least-squares fit to laboratory vapor-pressure data, the same pair this site's absolute-humidity instrument uses for the forward direction.
Two limits are worth knowing. Dew point can never exceed air temperature — the two are equal only at 100% relative humidity, the instant fog or dew actually forms — so a result above the entered temperature signals a bad input, not a new state of matter. And this formula reports the dew point of air sitting still at its measured pressure; it is not the figure a meteorologist means by lifting condensation level, which asks where a rising, compressing parcel would first saturate and needs the dry and moist adiabatic lapse rates as well, not just a single surface reading.
- Air temperature takes the current reading, in °C by default, though its unit menu also offers °F and K.
- Set Relative humidity, % to the hygrometer's percentage, anywhere between 0 and 100.
- Read Dew point: the temperature this air would need to cool to before condensation begins.
- Compare Dew point against Air temperature — the narrower that gap, the closer the air sits to fog, frost, or a sweating window.
- Switch Dew point's unit menu to °F or K to match a thermostat, forecast, or spec sheet quoted differently.
Worked example — a mild room at half saturation
With Air temperature set to 25 and Relative humidity, % set to 50, the instrument first builds γ: ln(50 ⁄ 100) = −0.69315, and 17.67 × 25 ⁄ (25 + 243.5) = 1.64525, so γ = −0.69315 + 1.64525 = 0.95210. Feeding that into Td = 243.5 × 0.95210 ⁄ (17.67 − 0.95210) gives 231.837 ⁄ 16.718 = 13.87 °C — the dew point this instrument reports for that room.
Thirteen-point-nine degrees sitting more than eleven degrees below the 25 °C air means this room is comfortably dry: nothing sweats or fogs until either the air or a cold surface inside it drops to 13.87 °C. That same eleven-degree spread is what a pilot uses to gauge cumulus cloud base — a widely used rule of thumb multiplies the spread by roughly 125 metres per degree Celsius, putting cloud base about 1,390 metres, or 4,560 feet, above the ground on a day shaped like this one.
Questions
Can dew point ever be higher than the air temperature?
No. Physically the two can only be equal, and that happens exactly at 100% relative humidity, the instant condensation actually starts. Td comes from RH and T through the same saturation curve that caps humidity at 100%, so any result above the entered air temperature means the relative-humidity figure fed in was itself above 100% — a sensor glitch, not a new physical state worth trusting.
Why does a muggy day feel worse than a dry, hotter one?
Because skin cools by evaporating sweat, and that only works as fast as the surrounding air can still absorb water — which is exactly what dew point measures and relative humidity does not. Forecasters treat a dew point under about 16 °C as comfortable, 16–19 °C as noticeably sticky, and above 24 °C as oppressive, regardless of the thermometer reading; a 30 °C day with a 12 °C dew point evaporates sweat easily, while 25 °C with a 22 °C dew point barely does.
How is dew point different from relative humidity?
Relative humidity is a ratio to the current saturation ceiling, so the same 50% reading describes far less actual water on a cold morning than on a warm afternoon. Dew point tracks the water itself: it only moves when moisture is added to or removed from the air, staying put while temperature swings the RH percentage up and down. That is why weather reports increasingly lead with dew point for comfort and fog forecasts rather than relative humidity alone.
Can this be used to estimate cloud base while flying?
Yes, approximately. Subtract the entered Dew point from Air temperature and multiply the spread by roughly 125 metres per degree Celsius, about 400 feet per degree Fahrenheit, to estimate the height of the lifting condensation level — the base of cumulus cloud on a day with that surface spread. It is a rule of thumb built from the dry and moist adiabatic lapse rates, not a substitute for this instrument's dew point, which describes air sitting still rather than air being lifted.
Why do cold drinks and window panes start to sweat?
Because their surface temperature has dropped below the surrounding air's dew point. The thin layer of air touching that cold surface cools with it, can no longer hold as much vapor once it falls under the dew point, and sheds the excess as a liquid film — the identical mechanism that lays dew on grass overnight. Warming the surface above the room's dew point, or lowering the room's dew point with a dehumidifier, stops it.
Does the calculation need barometric pressure or altitude?
No. Dew point comes from the ratio of actual vapor pressure to saturation vapor pressure, and Bolton's fit builds both sides from temperature alone, so surface pressure cancels out of the formula entirely. What it does not capture is a parcel that is being lifted and compressed — that scenario needs the lifting condensation level instead, a related but distinct calculation built on adiabatic lapse rates rather than a single surface reading.