Atmosphere
atm — non-SI unit · quantity: pressure · 101 325 Pa exactly
Florentine well-diggers complained that no pump would lift water higher than ten and a half metres, however hard one tried. Galileo joked that nature apparently runs out of its fear of the void. His student Torricelli decided to test it in earnest and replaced water with mercury — fourteen times heavier, so the column would come out shorter and fit on a table. The mercury in the sealed tube stopped at seventy-six centimetres, and above it remained a void that was not supposed to exist. The pump, it turns out, sucks nothing: air presses the water up the pipe, and presses with exactly the force this column can balance. We live, Torricelli wrote, submerged at the bottom of an ocean of air.
01 · Definition
The atmosphere is a non-SI unit of pressure equal to 101 325 pascals exactly. The number is not measured but assigned: it is the pressure of a 760 mm mercury column at zero degrees under standard gravity, computed once and fixed as the definition since 1954.
Note what happened to the logic. At first the unit was an object: a column of mercury one could pour, measure with a ruler and see with one's eyes. Then the object was computed through the density of mercury and the force of gravity, a five-digit number came out — and the number itself was declared the unit. After that the mercury became unnecessary: if tomorrow its density turns out to be known slightly better, the atmosphere will not budge, because it is no longer about mercury.
The only deceptive thing about this unit is the second half of its name. The «standard» atmosphere says nothing about what is going on outside the window: real surface pressure wanders from 870 hectopascals in the eye of a typhoon to 1085 over winter Siberia, and exactly one atmosphere occurs, strictly speaking, by accident. It is a reference point, not a measurement: gas volumes are reduced to it, boiling temperatures are counted from it, pressure gauges are calibrated by it — and out of habit it is called «normal pressure», though no normal pressure exists in nature.
From the fixed number follows all the arithmetic of this sheet: the atmosphere is exactly 760 times the millimetre of mercury, 1.013 25 times the bar and 1.033 times the technical atmosphere — that very «at», the kilogram-force per square centimetre, still confused with the real one. The difference between them is three per cent: enough to ruin a calculation, and too little to notice the error on a gauge.
One atmosphere is ten tonnes three hundred and thirty kilograms of air standing on every square metre: about six hundred kilograms on an adult's shoulders, a hundred and fifty on a palm. It does not crush us because it presses both from outside and from inside — but remove half from one side, and the difference becomes a very tangible force. The suction cup, the drinking straw, the vacuum lifter for glass and the chest at inhalation all work on the same trick: slightly lower the pressure on one side and let the atmosphere do the work itself.
02 · Conversion
Enter a value — the sheet will convert it into the rest
Three families: the pressure scale itself, familiar objects under this pressure, and the standard conditions set by the atmosphere.
The main trap is not in the coefficients but in what one counts from. A household gauge — tyre, compressor, any with a needle — shows gauge pressure, that is, the difference with the surrounding air; «two and a half in the tyre» means three and a half absolute. Gas-law formulas, boiling temperatures and solubilities demand the absolute one. Hence the two letters in English-language catalogues, psig and psia, and hence half of all ruined calculations: to confuse them is to be off by exactly one atmosphere, which at low pressures is more than the measured value itself.
| Notation | Notation | Value | Where it is used |
|---|---|---|---|
| atm | standard atmosphere | 1.0000 | chemistry, diving |
| Pa | pascal, the SI unit | 101325 Pa | standards and calculations |
| kPa | kilopascal | 101.325 | tyres, SI engineering |
| bar | bar | 1.01325 | hydraulics, diving |
| hPa | hectopascal, alias millibar | 1013.3 | weather reports |
| mmHg | torr | 760.00 | medicine, vacuum technology |
| at | technical, kgf/cm² | 1.0332 | legacy documentation |
| psi | pound per square inch | 14.696 | Anglo-Saxon engineering |
| mmHg | Blood pressure | 101325 Pa | 120 mm — systolic in a healthy adult |
| bar | Car tyre | 101325 Pa | 2.5 bar on the gauge |
| atm | Scuba cylinder | 101325 Pa | 200 atmospheres in a steel one |
| hPa | Weather report | 101325 Pa | 1013 hPa at the surface |
| atm | IUPAC standard conditions | 101325 Pa | 1 bar since 1982, before that the atmosphere |
| kPa | Normal conditions | 101325 Pa | 101.325 kPa and 0 °C |
| mmHg | Barometric calibration | 101325 Pa | 760 mm at 0 °C and g₀ |
The conversions are exact: all these units are defined through the pascal by whole coefficients, and none of them is subject to refinement by experiment. The only living number in the table is the hectopascal: it coincides with the millibar, and so meteorology changed its unit without retraining observers or redrawing maps.
03 · Orders of magnitude
decimal logarithm of pressure, atmospheresThe habitable range and everyday machinery
04 · Measuring instruments
What catches the atmosphere
A metre-long glass tube, sealed at one end, filled with mercury and upturned into a cup: the mercury sinks until it stops at the equilibrium height, and above it remains the Torricellian void — the first vacuum made by human hands. The instrument is impeccably accurate and utterly inconvenient: it is heavy, fears tilting and demands corrections for the mercury's temperature and the local gravity. In return it needs no calibration — the ruler is the scale.
The same principle taken to metrology: the cup at the bottom is closed with a leather pouch, and each time a screw brings its level to the zero of the scale — an ivory point that must barely touch its reflection in the mercury. Only then is the top of the column read with a vernier to a tenth of a millimetre. Such barometers set the pressure on calibration benches until the middle of the 20th century, and it was from their readings, reduced to zero degrees and standard gravity, that the number 101 325 grew.
Instead of mercury — a flat corrugated capsule with the air pumped out, which flattens slightly when pressure rises and straightens when it falls. Its wall travels tenths of a millimetre, and the whole cunning of the instrument is in the levers that magnify this motion to the sweep of a needle. An aneroid can be carried, dropped and hung in a hallway, but it lies with time: the spring tires, and it has to be checked against the mercury one. Thus one and the same quantity got a «proper» instrument and a «handy» one.
Inside a phone lives a silicon membrane the size of a grain of sand: it flexes under pressure, and with the flex changes the capacitance or the resistance of the strain gauges at its edge. The sensitivity is about ten pascals — less than a metre of altitude, which is why the navigator can tell the first floor from the third. There are now more barometers in the world than thermometers, and their readings, gathered from millions of phones, are already fed into weather-forecast models.
05 · Writing rules
Three lower-case letters and one caveat
The unit is not named after a person, so the symbol is lower-case; prefixes are not attached to it.
A capital «A» turns up in catalogues of imported equipment and always means carelessness, not another unit. The bar differs from the atmosphere by 1.3 per cent and was introduced precisely to have a round number of pascals; meteorology abandoned the atmosphere in its favour back in the thirties, and reports go in hectopascals numerically equal to millibars. The technical «at» is the kilogram-force per square centimetre, three per cent smaller than the real one, and in old domestic documentation it occurs more often than atm. The last line on the right is about what is missing: without the note «absolute» or «gauge» a pressure figure is incomplete, and that is the one caveat that cannot be dropped.
07 · Historical section
How the void became a unit
The experiment took a minute, and the argument about it lasted twenty years: above the mercury in the tube there was a void which, by common conviction, could not be. Torricelli explained everything at once: the column is held by the weight of the air, and its height is not constant — it changed from day to day. An instrument invented for a philosophical dispute instantly turned out to be a weather forecaster, and this is the only case of meteorology being born out of vacuum.
Blaise Pascal reasoned that if the column is held by air, then on a mountain there is less air and the column is bound to sag. He himself was too frail, and up the Puy de Dôme went his brother-in-law Florin Périer with two tubes: one he left at the foot in a monk's care, the other he carried up. At the summit the mercury dropped by three inches and a line. The dispute was over: air has weight, and less of it the higher one climbs.
Otto von Guericke, burgomaster of Magdeburg, put two copper hemispheres together, pumped out the air with his own pump and invited teams of horses to tear them apart. Eight horses on each side failed; on a half-metre diameter the atmosphere held the sphere with a force of about two tonnes. The calculation is clear to anyone who can multiply area by pressure, but the spectacle explained it better than any derivation — and entered the textbooks for four centuries.
By the middle of the 20th century an inconvenient thing had emerged: the definition through 760 millimetres of mercury dragged along the density of mercury and the local gravity, and both quantities kept being refined. The Tenth General Conference on Weights and Measures cut the knot by declaring the atmosphere equal to 101 325 pascals by definition. The unit lost its tie to a substance and became simply a multiplier — the same move by which the metre, and later the kilogram, were untied from objects.
The unit that was abolished and could not be removed
The atmosphere is not part of the SI and is not recommended for use: pressure has the pascal, and for large values — prefixes to it. Nevertheless the unit holds on wherever what matters is not the number but the comparison with the air outside the window. A chemist writes «reaction at 3 atm» because what matters to him is how many times above atmospheric; a diver counts depth in atmospheres because every ten metres of water add exactly one, and that is the handiest arithmetic in his trade; a tyre gauge is marked in bars and atm, since the difference between them is smaller than the instrument's error. The metrological paradox here is that after 1954 the atmosphere ceased to be a measurement and became an exact multiplier — that is, exactly what is convenient to use and exactly what cannot be verified by experiment. One can verify a barometer, not a unit; and a barometer calibrated in pascals will show one atmosphere only by rare coincidence. A similar fate befell the rest of its kin: the millimetre of mercury is likewise defined through the pascal and likewise alive — in medicine, where blood pressure is measured in it, and no recommendation has changed that.
Catalogue · units of measurement
A passport for every quantity
Seven SI base units, twenty-two derived ones with names of their own, and non-SI units without which neither engineering nor daily life gets by. Each is a separate sheet: definition, conversion, instruments, writing rules. In 36 languages.
Pressure: the units and their reasons
one quantity, a dozen scalesSI base units
kilogram, metre, secondPassport language
36 languages. The symbol atm is international, but the familiar scales are not: in translation the tables are brought to local practice.