ATMOSPHERE. The term "atmosphere" usually refers to the gaseous envelope covering the surface of the earth. The word is derived from the Greek words smoke or vapour and a4aipa, globe or sphere. The early Greeks were probably the first to study the weather in a regular and systematic way and the wind was defined by Anaximander as "a flowing of the air." Hesiod in his treatise Works and Days discussed the origin of wind, and many observations of physical properties of the air were made by Ctesibus, Hero of Alexandria, and othexs. The ma terial nature of air is clearly recognized in Hero's Pneumatica.
Anaximenes (c. 500 B.c.) regarded the air as the primordial substance from which all matter was condensed. During the time of Socrates meteorology was neglected, but Aristotle revived in terest in the study of the atmosphere and wrote about the winds. He regarded the atmosphere as consisting of three regions; the lowest in which plants and animals exist he supposed to be im movable like the earth; the uppermost region adjoined the fiery heavens and moved with them ; the division intermediate between the other two, he to be exceedingly cold. Meteors were considered by Aristotle to be exhalations from the earth, which became incandescent when they reached the hot upper layer.
Very little progress was made from this time until the early part of the 17th century, although it is said that during the 11th century the Arabs calculated the height of the atmosphere, from the duration of twilight, as 92 kilometres. In 1643, Torricelli, a student of Galileo, found that if a long glass tube sealed at one end was filled with mercury and the open end closed with the finger while the tube was inverted in a vessel containing mer cury, the liquid sank only to a certain level. It thus became possible to measure the pressure of the atmosphere, and the space above the mercury is still referred to as a Torricellian vacuum. This apparatus was called a barometer (q.v.) by Boyle and soon came into general use. Pascal demonstrated the decrease of the pressure of the air with altitude by measuring the height of the mercury column of a barometer at different points up a tower in Paris. In 1650 von Guericke (q.v.) found that he could pump air and was responsible for the famous experiment with the Magde burg hemispheres.
That air consists chiefly of two gases was first recognized by Scheele (1772), but Cavendish (i 7 8 i) was responsible for a large number of analyses of the air and found that zoo volumes contain 20.83 parts by volume of oxygen (q.v.) and 79.17 of ni trogen (q.v.). Similar experiments were carried out by Priestley (who thought the composition variable) and Lavoisier, but it was not until 1846 that it was definitely established by Bunsen that the composition of the atmosphere is not absolutely constant.
Below a height of 20km. (z 2 lm.) the constituents of the at mosphere, with the exception of -Later vapour, are well mixed by winds and by diffusion. Slight changes in composition do occur, however, at the surface of the earth and these depend on latitude and the presence of large quantities of vegetation or sea-water. The permanent constituents of the air are generally present in the following proportions (according to Humphreys in the Scientific Monthly, 1927) : Substance Volume % in dry air Total atmosphere .
Dry air . . 100•00 Nitrogen . . . . . . . . . . 7 8.03 Oxygen . . . . . . . . . . . Argon Water vapour . .
Carbon dioxide . . . . . . . . . 0.03 Hydrogen o.o i Neon . . . . . . . . . 0•00 i 8 Krypton . . . . . . . o•000 i Helium . . . . . . . . . . . 0•0005 Ozone . . . . . . . . . . . 0•00006 Xenon . . . . . . . . . 0.000009 The following table by Hann shows the variation with latitude.
Water Carbon Nitrogen Oxygen Argon vapour dioxide Equator . . . . 75•99 20•44 0.92 2.63 0.02 Latitude 50° N. . . 77.32 20.80 0-94 0.92 0.02 Latitude ;°b° N. . . 77.87 20.94 0.94 0.22 0.03 The composition also varies with altitude, but not to any very appreciable extent at heights at which respiration is still possible. The amount of water vapour present in the air is usually about z • 2 % by volume, but in very cold weather this quantity falls almost to zero. At other times it may be as high as 5 ` .
The ozone (q.v.) of the atmosphere is produced by electrical discharges and is found over the sea and mountains. Probably it is never present in quantities greater than one part in ten mil lions and the amount varies with the seasons, being greatest in winter, and averaging 2.5 volumes per million of air. Large quan tities of carbon dioxide, steam, nitrogen and hydrogen are con stantly liberated from inside the earth by volcanoes ; and carbon dioxide is the product of respiration of animals and plants. It is probable that when the earth was in the liquid state most of the nitrogen would remain free and would be retained by gravity rather than by chemical combination. The source of the oxygen presents difficulties: it may have been formed by the decomposi tion of volcanic carbon dioxide by plants; but this can only be effected by green plants in the presence of light, and, as most prim itive plants are not green, it is more probable that there was an excess of oxygen in the first place and that it has been retained through the gravitational effect ; however the balance of the quantity of oxygen in the air is maintained by the decomposition of carbon dioxide by plants. Lighter gases such as hydrogen and helium would tend to escape while the earth was still liquid, but would be retained when a solid crust had developed.
The Rare Gases.—During his experiments on the combination of nitrogen and oxygen by means of an electric spark, Cavendish, in 1785, observed that a small bubble of gas always remained after the absorption of the nitrogen oxides by potash and of the oxygen by "liver of sulphur." This was "certainly not more than .o of the bulk of the phlogisticated air (nitrogen) let up into the tube." Cavendish's observation was overlooked until in 1894 the late Lord Rayleigh noticed that atmospheric nitrogen is slightly denser than nitrogen prepared by chemical means (Atmospheric nitrogen, z. 2 5 7 z 8; Chemical nitrogen, 1.25107). The difference could only be explained by the presence of an unknown gas in the air, and Lord Rayleigh and Sir William Ramsay used two methods to isolate this, one of which was a repetition of Cavendish's experi ment on a large scale, while the other consisted in absorbing the nitrogen by heated magnesium. The new gas was found to be chemically inactive and hence was called argon (Gr. (q.v.). During the course of the separation of argon from liquid air by fractional distillation, Ramsay and Travers (1898) dis covered four more inert gases, which received the names helium, neon, krypton and xenon (qq.v.).
These five inactive gases are best characterized by their spectra, which are quite distinctive. They are all monatomic, i.e., each molecule only contains one atom. Certain springs evolve these gases, and helium is found in some natural gas and also occluded in certain minerals, e.g., broggerite.
Impurities.—Besides the normal constituents air always contains other substances which we can call "impurities." Organic and inorganic particles are always present to a certain extent, but are more plentiful over towns than elsewhere. The organic matter consists chiefly of plant spores and micro-organisms, and decreases in quantity when the temperature falls. Over the open sea the number of such particles is about one per cubic metre, while in crowded places it may rise to several thousand per cubic metre. Air may be sterilized by treatment with ozone if dry, by passing it through a hot tube, or partially by filtration through cotton wool. Inorganic dust is introduced into the atmosphere by the dis integration of meteors, volcanic eruptions, the combustion of fuel, and from the earth's surface by wind. The minute crystals of sodium chloride (common salt) found in the air owe their origin to ocean spray. The larger dust particles are visible as "motes" in a beam of light, but by far the greater number cannot be seen with the naked eye. In town air there are about zoo,000 per cu. cm. but over the sea this number falls to some hundreds. The particles can be made to settle out by washing and scrubbing. Atmospheric dust is the chief cause of haze in dry weather. Very slight traces of radioactive substances are also found in the air.
A large number of gaseous impurities are present such as ammonia, oxides and acids of nitrogen, small quantities of hydro carbons, sulphuretted hydrogen, carbon monoxide, sulphur diox ide and sulphurous and sulphuric acids, chlorine and hydrochloric acid. Nitrogen compounds (see NITROGEN) which are produced by electrical discharges, e.g., during thunder-storms, and carried down by rain play an important part in the fertilization of the soil. Ammonia (q.v.) is also introduced by the decay of organic matter. Carbon monoxide is contained in the exhaust gases of petrol engines and is found in railway tunnels. Other impurities are released from various chemical works.
Height of the Atmosphere.—The height to which the atmos phere extends cannot be definitely stated, although at an altitude of 5om. the air cannot exert any measurable pressure. Three methods are available for the estimation of the height : (z) obser vation of meteors, (2) measurement of the duration of twilight, (3) observation of auroral displays. The first method gives results ranging from 150 to 3ookm., while the duration of twilight indi cates a value of about 64km. at lat. 45°. It is difficult to make reliable calculations from auroral displays, but it is claimed that these occur up to a height of 5ookm. If the density of the atmos phere remained uniform throughout with the same value as at the earth's surface, the air would form a layer only 8km. thick and this is sometimes called the "height of the homogeneous atmosphere." Half of the air is below a height of 5.8km. At low levels temperature is usually considered to decrease 0.56°C per metres increase in altitude, but the rate is extremely variable. Above 2km. the temperature is on an average below o°C and con tinues to fall up to 1 okm. (6m.) when it is about —55° C. At 37km. the temperature is practically the same as at iokm. The lower region of the atmosphere is known as the troposphere and extends up to i okm., beyond which clouds are not generally found, except in tropical latitudes.
The Outer Atmosphere.—The upper region of the atmos phere, above iokm., is called the stratosphere and is separated from the troposphere or lower region, by a boundary region known as the tropopause. In the stratosphere the temperature gradient runs parallel to the earth's surface, whereas at lower alti tudes it is vertical, i.e., in the former case the air is arranged in columns each with a given temperature, while in the troposphere there are layers of air at different temperatures. Knowledge of the upper atmosphere, its constitution and physical properties, is by no means complete, although sounding balloons have been used up to about 25km. Lindemann and Dobson have deduced from observations on meteors, that the stratosphere does not extend beyond about 6okm. and they also conclude that above this level the temperature rises to about 3o° C. This high temperature region they believe to extend up to i 5okm. at which height meteors become luminous. Evidence for such a warm region has also been brought forward by F. J. Whipple from a study of the abnormal audibility of explosions, but its existence has been questioned by Sparrow.
The behaviour of long wave-length electromagnetic radiation (wireless waves) points to the existence of a conducting layer of ionized gas (the Kennelly-Heaviside layer) at a level of 4okm. 5okm. during the day and rising to about 9okm. at night. In the daytime this ionization could be caused by the ultra-violet (short wave-length) radiation from the sun, but its existence at night can only be explained if it is assumed that some substance is present which is capable of dissociation in the dark. It is believed that this substance is ozone. Various workers (Fabry and Buisson, 1921; Harrison and Dobson, 1925) have shown by studies of the absorption of solar radiation that there is a considerable quantity of ozone in the upper atmosphere and that it is probably formed by radiation of wave-lengths shorter than X = 2,000 A.U. (2/Io' cm.). The region in which this ozone occurs would be expected to have an abnormally high temperature and electrical conductivity, because it absorbs strongly radiation of certain wave-lengths. This region may therefore be considered to be identical with that where meteors become luminous and wireless waves are reflected back to the earth's surface. Its upper boundary is probably at a level of about i 5okm. and since it contains ozone, oxygen must also be present. Moreover, as oxygen is less dense than ozone, it will tend to rise to even greater heights. At a height of 3,2ookm. according to Jeans, there can only be about 300,00o gas molecules per cubic centimetre.
Much information concerning the upper atmosphere may be derived from studies of spectrum photographs obtained from the displays of the Aurc:a Borealis. These displays take place at levels varying from about 8okm. to 5ookm., but are most fre quent at Io6km. The auroral spectrum always contains a well defined, strong, green line of wave-length X =5,577.35 A.U., the origin of which remained a mystery for a long time. In 1925 M'Lennan and Shrum examined the radiation emitted by a mix ture of oxygen and either helium or neon in excess, under the influence of an electrical discharge, and found a green line, A =5,577.35 A.U., which was also shown with pure oxygen under low pressure. That nitrogen, in the same form as we know it, must exist at these great altitudes was shown by spectroscopic work carried out by Lord Rayleigh in 1921. Ozone, hydrogen and helium are also inferred. (See also AURORA POLARIS.) Absorption of Radiation by the Atmosphere.—The blue colour of the sky is due to the fact that the air is not perfectly transparent and its particles reflect and scatter light, that from the blue end of the spectrum being most widely scattered. This effect also obscures the light of the stars. Very little of the sun's thermal radiation is absorbed by the air, which derives most of its heat from the earth by conduction and convection. A layer of air one metre thick absorbs about 0.007% of the radiant heat passing through it. Of the radiation incident on the outer atmosphere about 37% is lost by reflection and scattering. The fraction of the radiant energy from the sun which reaches the earth is termed the coefficient of transparency of the atmosphere. The absorption is chiefly dependent on the amount of water vapour, carbon dioxide and solid impurities present and consequently is much greater in the neighbourhood of towns. The following coefficients of transparency are given by Wild for one metre of air:—Dry, dust-free air, 0.99718, Dry air, containing dust, from a room, 0.995 20, Dust-free air saturated with water vapour 0.993 28. The ozone, which appears to be present at very high altitudes, is responsible for the removal of practically all the ultraviolet radiation of wave-length shorter than X= 2,885 A.U.
Since the temperature of the upper atmosphere is practically constant and no convection or condensation takes place there, it is important to consider what would be the effect of dust par ticles which might be forced into the stratosphere by volcanic eruption. After certain eruptions, e.g., Krakatoa 1883, Mont Pele and Santa Maria 190'2, Katmai 1912, a reddish halo was observed round the sun owing to the dust ejected to very great altitudes, and it was possible to calculate the size of the particles. It has been estimated that a quantity of dust of volume less than 7 cu. km. distributed in the upper layers of the air, would reduce the intensity of solar radiation by 20%. It is possible to explain the occurrence of ice ages in this way. (See also SPECTROSCOPY; METEOROLOGY ; CLIMATE.) BIBLIOGRAPHY.-Sir Napier Shaw, The Air and Its Ways (1923) ; Bibliography.-Sir Napier Shaw, The Air and Its Ways (1923) ; Sir Napier Shaw, Manual of Meteorology (1926) ; W. J. Humphreys, Physics of the Air (1920) ; W. J. Humphreys, "The Atmosphere: Origin and Composition," Scientific Monthly (1927) ; "The Atmos phere," Scientific American (1928) ; F. H. Bigelow, Atmospheric Circulation and Radiation (1915) ; Geddes, Meteorology (1921) ; Clara M. Taylor, The Discovery of the Nature of Air (1923) ; J. C. M'Lennan, "The Spectrum of the Aurora," Proceedings of the Royal Institution (1926) . (J. R. P.)