Popular Science Monthly/Volume 43/July 1893/Recent Science II



DURING the last thirty years the data of meteorology have been accumulated with a very great rapidity, and the chief desideratum of the moment is, to construct with these data such a general theory of the circulation of the atmosphere as would embody the distribution of heat, pressure, moisture, and winds over the surface of the earth, and represent them as consequences of well-established mechanical laws. The old provisory hypothesis of atmospheric circulation, advocated by Hadley in 1735, and further elaborated by Dove in our century, can be held no more, and a new theory has become of absolute necessity.

We all have learned Dove's theory at school, even though we often found it difficult to understand. The air, greatly heated on or near the equator, rises in the same way as it rises in the summer over a sunburned plain. On reaching the higher strata of the atmosphere it flows toward the poles, but, owing to the speed of rotation which it has acquired in the lower latitudes, it is deflected—to consider the northern hemisphere only—to the right, and blows in the upper strata as a current from the southwest. To compensate this flow, air rushes on the earth's surface toward the equator, and as it also is deflected from its course by the same inertia of rotation, it appears in the tropics as a trade wind blowing from the northeast. However, the upper warm current does not flow all the way to the pole in the upper regions; it is gradually cooled down, and in about the thirtieth degree of latitude it begins to descend to the earth's surface, where it meets with the cold polar current. A struggle between the two winds ensues, and it lasts until they make a temporary peace by blowing side by side, or one above the other, the struggle giving origin to storms and to changes of wind which are fully analyzed in Dove's theory. A rope without end rolling over two pulleys, one of which lies horizontally near the equator, and the other stands upright in higher latitudes—such was the simplest expression of Dove's theory given in text-books.[1]

Under this provisory hypothesis meteorology made an immense progress, and some five-and-thirty years ago, Loverrier in France, and Fitzroy in England, ventured for the first time to foretell weather twenty-four hours in advance, or at least to send out warnings as to the coming storms. This bold step brought meteorologists face to face with a quite new problem. From the air pressure, the temperature, the moisture, and the winds observed at a certain hour of the day at various spots and telegraphed to a central station, they had to infer the next probable state of weather. So, leaving aside the great problems of atmospheric circulation, they directed their attention to the changes of weather rather than to the causes of the changes.[2] For this purpose purely empirical laws were of great value. When the meteorologist saw on a weather chart a region of low atmospheric pressure, with winds blowing in spirals round and toward its center, he named it, by analogy with real cyclones, a "cyclonic disturbance" or a "cyclone," giving the name of "anticyclone" to the region of high atmospheric pressure—and he studied the tracks of both disturbances in their advance across the oceans and the continents. He did not inquire for the moment into the causes of the disturbances; he took them as facts, and, following Buys Ballot's law, he said that the wind will blow as a rule from the region of high barometic pressure (the anticyclone) to the region of low pressure (the cyclone), with a certain deflection to the right or to the left. Immense researches were made to study the routes followed by the centers of barometrical minima, and we now have splendid atlases showing the normal tracks of cyclones across the Atlantic Ocean, over Europe and the States, in Japan, in the Indian Ocean, and so on, at various seasons of the year.[3] With these empirical data meteorologists attained such a perfection in their weather forecasts that in five cases out of six their previsions are now correct, while the coming gales are even foretold with a still greater accuracy.

However, the very progress achieved demonstrated the necessity of a more thorough knowledge of the too much neglected upper currents of the atmosphere. In Dove's scheme, the upper equatorial current, after part of it had been sent back to the equator, was entirely abandoned to itself, to make its way as best it could against the opposed polar winds; but the existence of a strong, nearly permanent, and relatively warm upper wind blowing toward the east in our latitudes—which was only probable thirty years ago[4]—became more and more evident, especially since the movements of clouds began to be systematically studied and observatories were erected on high mountains; and this wind remained unexplained in Dove's theory, while in Maury's scheme of atmospheric circulation, which is still in great vogue in our schools, there was even substituted for it a current in an opposite direction, which does not exist, and which Maury himself could not account for.[5] An entire revision of the subject was thus necessary, and this revision has been done by the American meteorologist Ferrel, in a series of elaborate works which are only now beginning to receive from meteorologists the attention they fully deserve.

Ferrel's theory is based upon considerations as to the laws of motion of liquids and gases of different densities. If the whole atmosphere were equally heated in all its parts, and at full rest, the air would be disposed in horizontal layers, of greater density at the bottom, and of decreasing density toward the top. Considering some part only of the atmosphere, from pole to equator, and neglecting the curved surface of the earth, we should thus have something analogous to a trough filled with layers of different liquids. If one end of the trough were now warmed, and the other end were cooled, the layers would be horizontal no more. They would be inclined, but in two different ways: the lower ones would be inclined toward the warm part, while in the upper layers the inclination would be the reverse. A full circuit of the lighter liquids flowing one way on the surface, and of heavier liquids flowing the other way on the bottom, would thus be established. The same would happen in our atmosphere with the lighter warm currents and the heavier cold currents if the earth had no rotation on its axis. But it rotates—the solid globe as well as its gaseous envelope—and this modifies the whole circulation. The air which flows from the equator to the poles maintains, not its velocity of rotation, as has been hitherto taught, but its energy of rotation, which means that it obeys the law of preservation of areas; therefore, when it is transported from the equator to a higher latitude it is endowed (in the northern hemisphere) with a much greater easterly velocity than if it simply maintained its speed of rotation. On the other side, the air which is flowing from the higher latitudes toward the equator also obeys the same law and acquires a westward velocity, but much smaller than the eastward velocity of the former; this is why the west winds have such a preponderance in our latitudes.[6] Moreover, in virtue of the centrifugal force, all masses of air moving in any direction—not only north or south, but also due west or east—are also deflected to the right in the northern hemisphere, and to the left in the southern hemisphere,[7] Consequently the air flows in great spirals toward the poles, both in the upper strata of the atmosphere and on the earth's surface beyond the thirtieth degree of latitude; while the return current blows at nearly right angles to the above spirals, in the middle strata as also on the earth's surface, in a zone comprised between the parallels 30° north and 30° south.[8]

Such are, very briefly stated, the leading features of the theory which Ferrol laboriously worked out during the last thirty years, submitting all its parts to the test of both observation and mathematical analysis. By the end of his life (he died in 1891) he embodied his theory in a well-written and suggestive popular work, which. fully deserves being widely known. All taken, his views so well agree with the facts relative to the movements of the atmosphere, and they give such a sound method for further investigation, that they are sure to become for some years to come the leading theory of meteorology. They already have given a strong impulse to theoretical research, and have created a whole literature in Austria and Germany.[9]

Another theory of the general circulation of the atmosphere which is also awakening a good deal of interest among physical geographers was propounded in 1886 by Werner Siemens, and further developed by him in 1890.[10] Siemens did not consider that air might flow down the density surfaces, as supposed by Ferrel and Helmholtz, and admitted by many meteorologists, and he maintained that the source of the energy required for all disturbances of equilibrium in the atmosphere must be looked for in the unequal heating of its different strata by the sun, and in the unequal loss of heat through radiation in space. From these considerations he inferred the existence of an ascending current in the equatorial belt, an upper warm current, and a cold polar current. As to the eastward and westward directions of these currents, he made the very just remark that the energy of rotation of the whole atmosphere must remain constant and unchanged, even though masses of air move from one latitude to another. The velocity of rotation of the atmosphere in tropical latitudes must therefore lag behind the rotation of the earth, and it must outstrip it in higher latitudes, mathematical calculation proving that the thirty-fifth parallel is, in both hemispheres, the line of division between the two. The general system of air circulation deduced from these principles is very similar in its results to the system of Ferrel; but the interest and importance of Siemens's views lie elsewhere. His memoirs were an appeal and an attempt to apply the principles of thermodynamics to the aërial currents, and they have opened the way for a series of important researches, which, however, are not yet sufficiently advanced to be discussed in these pages.

And, finally, a third new point of view has been introduced into the same discussions by Helmholtz. Sitting one day by the seaside, and observing how wind blows on the surface of the sea by sudden gushes, how it originates waves, and how they grow when wind blows with an increasing force, Helmholtz came to consider what would happen with two air currents blowing one above the other in different directions. A system of air waves, he concluded, must arise in this case, in the same way as they are formed on the sea. The upper current, if it is inclined toward the earth's surface (as is often the case), must originate in the lower current immense aërial waves rolling at a great speed. We do not generally see them, but when the lower current is so much saturated with moisture that clouds are formed in it, we do see a system of wavelike parallel clouds, which often extend over wide parts of the sky. To calculate the sizes. of the waves in different cases is extremely difficult, if not impossible; but by taking some simpler cases Helmholtz and Oberbeck showed that when the waves on the sea attain lengths of from sixteen to thirty-three feet, the air waves must attain lengths of from ten to twenty miles, and a proportional depth. Such waves would make the wind blow on the earth's surface in rhythmical gushes, which we all know, and they also would more thoroughly mix together the superposed strata, dissipating the energy stored in strong currents. These views are so correct that they undoubtedly will throw some new light, as they already begin to do, upon the theory of cyclones.[11]

At the same time, Bezold is now endeavoring to reconstruct meteorology from the point of view of thermodynamics;[12] and the well-known Austrian meteorologist, J. Hann, whose work is exciting just now a great deal of interest, has openly broken with the old theory as regards the origin of cyclones and anticyclones.[13] From observations made for several years in succession on the top of the Sonnblick—a peak twelve thousand feet high, of the Tyrolese Alps—as well as from observations made on several high-level stations, he has concluded that a cyclone can not be due to a local heating of the earth's surface and to an ascending current of warm air provoked by this cause, just as an anticyclone can not be due to a local cooling of the earth's surface, and to a consequent condensation of the air. Contrary to the previsions of the meteorologists, the ascending column of air within a cyclone, up to a height of some ten thousand feet, is not warmer than the surrounding air; it is cooler within the cyclone, and its upward motion thus can not be due to its temperature. So also in an anticyclone the descending current of air is warmer than it is under normal conditions, and its downward motion must be due to some other cause than an increase of density resulting from a lowering of its temperature. The decrease of pressure in the one case, and its increase in the other, thus can not be caused by differences of heating or cooling of the lower strata; and both cyclones and anticyclones must be considered as parts of the general circulation of the atmosphere, such as it was conceived by Ferrel.[14]

Such a deep modification of the current views, though supported to a great extent by weighty evidence, will obviously not be accepted without opposition; but it is already making its way, and certainly will exercise a deep influence on the further development of meteorology.

Abandoning now the domain of theoretical investigation, I must mention a work—also a life's work—which may safely be placed side by side with the best achievements in theory. I mean the beautiful charts of Mr. Buchan, representing the distribution of pressure, temperature, and winds over the surface of the globe, embodied in the last volume of the Challenger Expedition Reports. When Mr. Buchan published twenty-three years ago his first maps of monthly isobars and prevailing winds, they were quite a revelation, even though the data upon which they were based were very incomplete at that time.[15] But better data have been collected since, and in the hands of Mr. Buchan they have undergone such a careful and able analysis that the Challenger Reports charts may be taken as the best reliable representation of the winds, the temperatures, and the pressure in the lowest strata of the atmosphere, as well as the surest basis for further generalizations.[16] The theories which have been mentioned in the preceding pages give the grand lines of atmospheric circulation; on Buchan's maps we see how the grand lines are modified in the lowest strata by the distribution of land and sea, and the unequal heating or cooling of continents and oceans. The leading features indicated by theory are still maintained, and they become even still more apparent if we consult isobars traced for a certain height, like those of Teisserenc de Bort; but the immense plateaus of East Asia and North America act in winter as colossal refrigerators, where cold and heavy air accumulates, to flow down in all directions toward the lowlands. We see also how in July the air is heated in the lower lands of northwest India, in the corner between the Afghanistan and the Thibet plateau, how pressure is lowered there by the ascending current, and how winds blow toward this region of lowered pressure. We see more than that: on looking on the maps it strikes the eye how the moisture or the dryness of the climate is dependent upon the distribution of pressure, and how the dry anticyclonic winds make barren deserts of parts of North and South America, of Africa, and central Asia, and how they will continue to dry the lakes and the rivers of these regions and occasion total failures of crops so long as that distribution of pressure lasts on the globe, and man has not yet learned to eschew its effects by getting water from the depths of the earth. The life of the globe during the present period is written on these splendid charts.—Nineteenth Century.

M. Thoraddsen, in the narrative of his travels in Iceland, observes a peculiar feature of the oases at the foot of Mount Hecla. These oases are subject to constant displacement by the violent sandstorms which are common. On the windward side all vegetation is gradually destroyed, while on the other side grass takes root, and in a wonderfully short time the level and sterile surfaces are converted into good pasture lands.

  1. E. E. Schmid, Lehrbuch der Meteorologie, Leipsic, 1860, p. 568.
  2. See W. Bezold's short sketch of meteorological progress in Sitzungsberichte der Berliner Akademie der Wissenschaften, 1890, ii, 1295, sq.
  3. Besides the earlier works of Ley (Laws of the Winds prevailing in Western Europe, Part I, 1872) and Köppen (Wissenschaftliche Ergebnisse aus der monatlichen Uebersichten des Wetters, 1873-'78), we have now the splendid work of W. J. Van Bebber, which embodies the tracks of all cyclones in Europe for the last fifteen years (Die Zugstrassen der barometrischen Minima, für 1875-'90), the researches of Blanford, S. E. Hill, and Elliot in the Indian Meteorological Meirioirs and Cyclone Memoirs, Part IV (published by the Meteorological Department of India), the work of E. Knipping for Japan, in Annual Meteorological Report for 1890, Part II, Appendix, and several excellent works for Russia.
  4. Observations in Siberia—namely, at the graphite works on Mount Alibert, at a height of eight thousand feet (52° north latitude)—were especially conclusive. Alibert's observations, buried in the Russian Trudy of the Siberian expedition, proved the existence of a nearly permanent west and west-northwest wind on the top of the peak, and they showed at the same time that the average yearly temperature on the top of the peak was by some fourteen to eighteen Fahrenheit degrees higher than it otherwise ought to be. When I visited the then abandoned mine in 1864, and saw the peak dominating all surrounding mountains, and could judge of the force of the west wind from the immense works accomplished to protect the road which was traced on the western side of the peak, I could not refrain from explaining the extraordinary great height of the snow-line in east Siberia by the existence of a relatively warm equatorial current blowing with a great force at a height of from eight to ten thousand feet in the latitude of 52° north. Later on the observations which I brought from the Voznesensk mine (60° north, altitude twenty-six hundred and twenty feet) induced my friend Ferd. Müller, who calculated those observations, to conclude that in higher latitudes the same current descends still lower to the earth's surface, and still maintains some of its initial warmth.
  5. See James Thomson's paper On the Grand Currents of the Atmosphere, in Philosophical Transactions, A. 1892, p. 071.
  6. Full tables giving the eastward (or westward) velocities for each latitude, under the two different hypotheses, have been calculated for the Meteorologische Zeitung, 1890, pp. 399 and 420.
  7. Ferrel seems not to have been aware that the same had been demonstrated by R. Lenz for rivers (about the year 1870), in a discussion of Baer's law, applied to the Amu Kiver, in the Mémoires of the St. Petersburg Academy.
  8. William Ferrel, A Popular Treatise on Winds, comprising the General Motion of the Atmosphere, Monsoons, Cyclones, Tornadoes, Waterspouts, Hailstorms, etc. New York: Wiley, 1889. See also analysis of it by W. M. Davis (in Science, xv, p. 142; translated in Meteorologische Zeitung, 1800; Literaturbericht, p. 41), who gave the best diagram of circulation according to Ferrel's theory, and by H. F. Blanford in Nature, xli, 124. A full bibliography of Ferrel's works was given after his death in the American Meteorological Journal, October, 1891.
  9. Roth has already abandoned the mathematical objections he had raised against Ferrel's theory in the Wochenschrift für Astronomie, 1888. The objections raised by Teisserenc du Bort and Supan against the "density surfaces" have been answered by Prof. Davis in Science, and are not shared by the most prominent meteorologists. And the mathematical analysis of Prof. Waldo, Sprung (the author of the well-known Treatise of Meteorology), M. Möller, and Pemter has further confirmed the accuracy of the theory. So also Hildebrandsson's observations of upper clouds (Annuaire de la Société météorologique de France, xxxix, 338), Teisserenc du Bort's high-level isobar?, and Guaran de Trommelin's researches relative to coast winds. The transport of the Krakatoa dust and Abercromby's observations of clouds having rendered the existence of an upper east current very probable on the equator, Pernter has mathematically deduced from Ferrel's theory-the existence of such a current in a belt 4º 45' wide on both sides of the equator, and he therefore has withdrawn the restrictions he had previously made in a lecture (published in Nature, 1892, xlv, 593) in favor of Siemens's views. It must be added that the idea of three superposed currents blowing in spirals may have been suggested to Ferrel by a communication of James Thomson to the British Association in 1857. Such was, at least, the claim raised and developed at some length by the Glasgow professor before the Royal Society in a Bakerian lecture, now published in the Transactions (A. 1892, pp. 653-685). Though Thomson's paper was never published, and only given in a very short abstract without a diagram (the diagram in the Transactions is now published for the first time), the few lines in which his theory was stated (British Association Reports, Dublin, 1857, pp. 38, 39) contained the idea clearly expressed. It is certainly a matter of great regret that James Thomson has not returned to this subject.
  10. Ueber die Erhaltung der Kraft im Luftmeere, in Sitzungsberichte der Berliner Akademie der Wissenschaften, March, 1886, p. 261; Ueber das allgemeine Windsystem der Erde, in same publication, 1890, ii, p. 629.
  11. H. Helmholtz, Zur Theorie von Wind und Wetter, and Die Energie der Wogen und des Windes, in the Sitzungsberichte of the Berlin Academy, 1889, ii, and 1890, ii. Oberbeck's calculations of the waves are given in the Meteorologische Zeitung, 1890, p. 81.
  12. Zur Thermodynamik der Atmosphäre, in Sitzungsberichte of the Berlin Academy of Sciences, 1888, p. 485; same year, p. 1189; 1890, p. 355; and 1892, p. 279.
  13. Das Luftdruckmaximum vom November 1889, in Denkschrift der Wiener Akademie dor Wissenschaften, 1890, Bd. Ivii, p. 401. Bemerkungen über die Temperatur der Cyclonen uud Anticyclonen, in Meteorologische Zeitscbrift, 1890, p. 328.
  14. See the discussion of this subject between Hazen and J. Hann in Science, 1890, xv, 382-384, and Meteorologische Zeitschrift, 1890, p. 328.
  15. To trace the isobars, or lines of equal atmospheric pressure, reduced to the sea-level, the real altitude of each meteorological observatory must be known from direct geometrical levelings; but in 1869 the altitude of not one single station in Siberia, central Asia, or even the Urals was known. A leveling across Siberia, as far as Lake Baikal, has been made since, Mr. Buchan's isobars having been one of our best arguments to press the necessity of the leveling. But Mr. Buchan may not be aware that the leveling beyond the ninetieth degree of longitude is now considered by Russian geodesists as utterly unreliable; it is supposed to contain some substantial error, so that a new leveling between Krasnoyarsk and Lake Baikal is insisted upon. The incertitude in the isobars on an immense space in northeast Asia resulting from this cause may attain as much as one or, perhaps, even three tenths of an inch.
  16. An excellent résumé of the whole work and its results in a popular form has been published by Buchan himself in the Proceedings of the Geographical Society, March, 1891.