Hygrometer
The one that actually measures it
The chilled-mirror hygrometer is the only common instrument that measures dew point rather than deriving it, and it does so almost by acting out the definition.
A small polished mirror is cooled, usually by a thermoelectric element, while an optical sensor watches its surface. The instant condensation begins to scatter the light, the mirror's temperature is recorded. That temperature is the dew point — it is the point at which the air touching the mirror gave up its water.
Accuracy is around ±0.2°C, which is why it is the reference standard other instruments are calibrated against. It is also expensive, needs periodic cleaning of the mirror, and is not what is in a weather station.
Everything else computes it
Capacitive humidity sensors — the kind in weather stations, home monitors and industrial probes — measure relative humidity by how much a polymer film's capacitance changes as it absorbs water. Pair that with a thermometer and the dew point follows from the arithmetic.
This is not a compromise. The relationship between temperature, relative humidity and dew point is exact physics; the approximation used to evaluate it is good to a fraction of a degree. What limits the answer is the humidity measurement, not the conversion.
Doing it by hand
Take air at 75°F, which is 23.9°C, at 60% relative humidity.
γ = −0.511 + 1.577 = 1.066
dew point = 243.12 × 1.066 / (17.62 − 1.066) = 15.7°C = 60°F Convert to Celsius first — the coefficients are defined for it.
The dew point calculator runs this same formula, and the chart is the lookup-table version if you would rather find a row and a column.
Where the error comes from
| Method | Typical accuracy | Resulting dew point error |
|---|---|---|
| Chilled mirror | ±0.2°C dew point | ±0.2°C |
| Good capacitive sensor | ±2% RH | about ±0.3°C |
| Consumer sensor | ±5% RH | about ±0.8°C |
| Uncalibrated / drifted | ±10% RH or worse | 1.5°C and up |
Two things follow. First, the formula is never your problem. Second, the error grows as the air gets drier — the same percentage error on a small humidity number is a bigger absolute error in water content, so a dew point computed at 15% humidity is much shakier than one computed at 80%.
If you own a hygrometer, the salt test will tell you which row of that table you are actually in.
The distance problem
For a reading taken from a weather service rather than an instrument in your hand, the dominant error is usually not the sensor at all — it is that the nearest observation may be several miles away, and humidity varies sharply over short distances. A valley, a lake, a car park and a wood can easily differ by several degrees of dew point at the same moment.
That is worth knowing before treating any single figure as exact, and it is covered properly in where humidity data comes from.