Alfred Wegener
Page 38
Bjerknes’s work provided a qualitative treatment, with a sketch of the kind of mathematics required for a quantitative treatment of circulation in both the ocean and the atmosphere. Bjerknes’s focus was on the pressure gradient, and he gave striking illustrations of upwelling and downwelling in ocean basins caused by circulation associated with layers marked by sharp density gradients. This, of course, was impressive and congenial to Wegener, who could see in these “oceanographic elements” the same characteristic distributions governed by sharp surfaces of discontinuity with which he was so concerned in the atmosphere. The heaviness of water compared to air and the relatively narrow range of temperatures for ocean water compared to those for atmospheric air made it logical that isobar surfaces (lines of equal pressure) should govern the ocean in the way that lines of equal temperature governed discontinuity surfaces in the atmosphere.
There were other choices in the monograph literature available to Wegener to update Sprung’s treatment of atmospheric mechanics. Principal among these was the work of Max Margules (1856–1920) in Vienna. Here the emphasis was on the energetics of storm systems, but discontinuity surfaces still played an important part. Much nineteenth-century work on the motion of storms had to do with the vertical circulation of heat and the generation of thermal energy to drive the storms. But as aerological research indicated and Margules’s theoretical treatments elaborated, this was not really the answer. The nineteenth-century notion of a cyclonic storm was that a storm was a self-subsistent entity within a boundary created only by its outer surface with still air. As Margules and others showed, the existence of storms depended on a hydrostatic relationship: they were trapped between the surface of Earth and the tropopause, where, with broad penetrations, generally their convective activity ceased. This was a profound change in the view of storms in air: from full atmospheric circulation to the restricted circulation within the sharply bounded discontinuities between which they lived.
Moreover, Margules was concerned, as were Helmholtz and Köppen before him, with sharp horizontal surfaces of discontinuity, of the kind observed in squall lines of thunderstorms with either sharp drops in temperature or sharp rises in temperature.17 Margules’s work differed from Bjerknes’s in that Margules was interested in changes in kinetic energy which might drive storms, resulting from the interaction of contiguous air masses, whereas Bjerknes was concerned initially with the question of global circulation; the treatments were complementary but not identical in their content and aim.18
Wegener could appreciate the important work being done at the research front by Bjerknes and Margules with attention to the importance of both vertical and horizontal discontinuities in the atmosphere, but the direction his researches were taking had to do with the permanent structural features that constrained the movements of such air and water masses, discontinuities that created the conditions for the existence and behavior of these masses, not the reverse. Air masses in the atmosphere, as well as water masses in the ocean, had no constructive role in establishing discontinuities in the vertical dimension.
Wegener’s Approach to Thermodynamics
As he worked his way more deeply into the subject in the late spring and early summer of 1910, the impetus to write a book on atmospheric thermodynamics grew stronger with his discovery that the literature was filled with errors, and this included many topics in Bezold’s thermodynamic papers published only four years before. Moreover, the treatment of topics seemed chaotic; there wasn’t any order or systematic presentation of matters as fundamentally important as the size of cloud elements (such as water drops and ice crystals), or even a systematic treatment of the ice phase of water vapor in the atmosphere, with its various forms of snow, graupel, hail, and ice crystals.
Wegener published two papers in Meteorologische Zeitschrift in the summer of 1910 which show the interesting character of the work style he was developing.19 As usual, a sort of apologia accompanied the papers. In this case it was not his formulaic apology that the “hypothesis might be premature,” as in some of his work on layering where the preliminary hypothesis he had formulated seemed to find “striking confirmation” in the scant available data. Here the apology was for publishing scientific communications that contained no novelty in their elements but rather took their novelty from the understanding generated by the systematic presentation of material widely distributed in the literature.20
The subjects of these two papers were topics of core significance for meteorology: the constitution and therefore the microphysics of clouds, and a systematic treatment of the different morphologies of ice formation from water vapor in the atmosphere. These were things about which all meteorologists cared and were central to the thermodynamics study of a moist atmosphere. The main line of development was once again Helmholtz, then Hertz, then Aßmann and Bezold, with Bezold making his last significant contribution in 1899. Wegener was again in the position of updating general ideas, based on results from balloon flights and aerological investigations conducted in the first decade of the twentieth century.
Hertz had developed, and von Bezold extended, a picture of an atmosphere with essentially four stages or states. The first of these was a dry atmosphere, where there was only a little water vapor. Then came the rain stage, where saturated water vapor and liquid water coexisted. Third, there was the condition of the atmosphere in which hail could develop—near the triple point of water, where water vapor, water, and ice were all present at the same time. Finally, there was the snow stage, where there was only “ice vapor” and crystalline ice. Hertz had the view that an ascending mass of moist air would cool adiabatically as it expanded and that the water content would pass through these four stages successively. Having reached the snow stage, the ascending cloud would begin to descend.21
In twenty-first-century meteorology, the process under discussion is known as the “precipitation staircase.” One begins with air with increasing moisture content. Submicroscopic particles, called cloud condensation nuclei, accompany the air as it ascends. These might be salt crystals from extremely small water droplets, created in the burst of tiny bubbles on the surface of the ocean, or extremely small particles condensing from gas. As the air ascends, it expands; as it expands, it cools. As it cools, it humidifies, eventually reaching saturation, when it can hold no more water vapor. At this point, depending on the altitude and temperature and pressure conditions, it will condense into one or another form of precipitation. As some of the precipitation elements grow at the expense of others, the larger drops begin to fall. Sometimes the whole cloud descends, and this begins the “down staircase” where the descending air is compressed, warmed, and dried, finally with all the water vapor evaporating away.22
Aerological research, as well as laboratory research in the twenty years before 1910, had shown that it was possible for water in the vapor phase to exist in a supercooled form, where it was well below the normal freezing point of water but not yet formed into ice for lack of something on which to nucleate and grow. Bezold had gone into this topic, and the part of his explanation that Wegener was most drawn to was that moment when, in a cloud of supercooled water, the oversaturation was released and precipitation drops began to form.23 Bezold had imagined a mechanism whereby a sudden change in the pressure and temperature regime in a cloud would force the condensation and solidification of supercooled water. The mechanism was complicated, and we will not go into it here, but Wegener found it to be physically impossible.24
Wegener had a different idea for how water might pass from the supercooled phase, in an oversaturated cloud, into ice crystals. He saw that the simultaneous existence of supercooled water and ice particles would establish a variation in the vapor pressure between the water droplets on one hand and the ice particles on the other. Because the saturation vapor pressure is greater in the immediate vicinity of the water droplets than in that of the ice particles, over time, the vapor pressure gradient would draw water molecules away from the supercooled water drops and toward t
he ice crystals. These ice crystals would grow until they became heavy enough to fall from the cloud. Depending on the temperature gradient, they would reach the surface as either snowflakes or raindrops.25
Wegener imagined he was making a physical correction to an accepted process, without significant novelty. However, in this “correction” of Bezold’s work on supercooled water and clouds, Wegener had in fact made one of his most memorable discoveries. The process of migration of water molecules from supercooled droplets to ice crystals in cold clouds is today known as the Wegener–Bergeron–Findeisen (WBF) theory of precipitation in cold clouds. Developed initially in Wegener’s paper and expanded in Thermodynamik der Atmosphäre, it provided inspiration for the Swedish meteorologist Tor Bergeron (1891–1977) and later for the German meteorologist Walter Findeisen (1909–1945). While the modifications made by both of these other scientists were substantial, Bergeron was quite insistent that the inspiration for this development was clearly laid out in Wegener’s work, and indeed one can see it specifically formulated in Wegener’s article published in Meteorologische Zeitschrift in 1910.26
In addition to this clarifying and surprisingly novel work on the ice phase of water vapor in the atmosphere, Wegener also reorganized thinking on the size of the particles making up clouds. William Thompson, Lord Kelvin (1824–1907), had long since argued that given a certain partial pressure of vapor, a water drop must be of a particular radius: drops that are slightly larger will grow by condensation, and drops that are slightly smaller will evaporate. This equilibrium between vapor and small drops is not simple and actually involves drops of different sizes and concentrations. Kelvin had been followed by Joseph Perntner (1848–1908), Julius Hann (1839–1921), and more contemporaneously Victor Conrad (1876–1962). Wegener pointed out, in his parallel paper entitled “Die Größe der Wolkenelemente” (The sizes of cloud elements), that all of these celebrated attempts to study the effect of surface tension on the equilibrium radius of a drop at a relative humidity of 100 percent had failed because they had not considered supercooling. In supercooled clouds, the relative humidity was regularly higher than 100 percent, sometimes 110 or 120 percent, and (rarely) even as much as 400 percent. Under such conditions, the radius of the droplets would be substantially different, and the equilibrium, as the water moved from the supercooled aqueous phase into the ice crystals, would be very unstable. If the volume of a raindrop at 100 percent humidity were 4.2 × 10−9 cubic centimeters, at 400 percent humidity it might be as small as 1.7 × 10−21 cubic centimeters.27
The issue of the size of the cloud elements in the supercooled cloud is important because the very small droplet sizes in the exaggerated relative humidity of a supercooled cloud give conditions under which “hygroscopic kernels” (as Wegener described them) would have a significant effect on formation of ice crystals. It would allow not just the very small salt crystals that his friend Lüdeling had observed in sea air to function in this way as condensation nuclei, but even individual (nitrous) gas molecules. This is the so-called ionic theory of nucleation, and it plays an important role in meteorology even today.28
The question of condensation nuclei had been in meteorology for a long time, and his mentor and supervisor at Lindenberg, Aßmann, had made some famous observations in 1884 on the mountain outside Berlin called the Brocken, to determine (a) whether the microstructure of clouds and mists was bubbles or drops, and (b) whether or not these drops formed around minute dust particles called “Aitken nuclei,” after the Scottish meteorologist John Aitken (1839–1919), who had done some experiments with condensation of steam in dusty and clean air. Aßmann walked up and down the mountain slope with a 400-power microscope, capturing droplets on microscope slides and examining them with a micrometer under oblique illumination to determine their diameters and composition.29 Aßmann did not observe nuclei, even though he could have seen particles down to 0.0005 millimeters, and therefore assumed they played no role. Wegener’s work indicated that actual “Aitken nuclei” might have diameters of only a few millionths of a meter. Thus, Wegener’s work on droplet size and “hygroscopic kernels” extended and corrected Aßmann as well as Bezold, Kelvin, Conrad, and Perntner.
Wegener’s invention and novelty were not limited to these two efforts, of course. If one reviews Wegener’s twenty or so scientific papers published from 1908 through 1910, other than results of the Danmark Expedition, they are all on some aspect of atmospheric physics. In the latter topic he was consciously working to establish himself as the leading theorist of the “Berlin school” of meteorology and atmospheric physics. To do so, he made a special point to employ and feature data collected by the Berlin group, though supplemented with other data sets. He also made an effort to update and correct papers written by both Aßmann and Bezold going back a number of years. He was extremely polite and cautious in these corrections, never saying that “Bezold was wrong” or that “Aßmann got the wrong answer.” Rather, his approach was always to say that new data provided by aerology had made thus and such a conclusion no longer tenable, or that such and such an idea needed to be modified.
In addition to his corrections of those senior academics involved in his training and supervision at Berlin, and thus the advancement of their work as well as his own, he was very much involved in bringing together and synthesizing the conclusions of papers by a variety of different investigators, written in the time when he was away in Greenland. Here he disclaimed any novelty in the uncovering of basic empirical facts, casting his contributions as possible avenues of theoretical advance, falling out of the work of keeping some subject matter up to date. Even in his most exhilarating papers on the chemical differentiation of the atmosphere and the new fundamental layer boundary, his stance was always that this material was available in the literature for anyone to see, though no one (save himself) had seen it yet.
One recalls that in his probationary lecture at Marburg the previous year he had spoken of the alternating periods in meteorology between the collection of data and creation of new instruments on the one hand and the theoretical integration in advance of the subject on the other. He had characterized the first decade of the century, following the lead of Köppen, as a time of the collection of new facts—and most of these facts concerned, of course, aerology.
Yet in 1910, as he was increasingly aware, theoretical forward motion of atmospheric physics was everywhere evident—except at Berlin and Lindenberg. Aßmann had used up an enormous amount of money building the station at Lindenberg, with little to show for it.30 He had tried (unsuccessfully) to get Wegener to join him again at Lindenberg and produce theoretical results that Lindenberg could claim as its own.
In donning the theoretician’s mantle, Wegener was aware that there were others simultaneously vying for the prize. The work of Ficker, as well as that of Margules at Vienna, was built on the same scale as his own: global atmospheric structure and atmospheric motion. Wegener’s historical sense placed him self-consciously in that generational group that would carry out the next set of theoretical advances. His choice of thermodynamics kept him close to his own empirical work and to the greatest scientific successes of the Berlin school.
Wegener was positioning himself in his published work and in his lectures to build a career leading to a professorship at the earliest possible moment. His reasons for wishing this were as much financial as anything else; he was virtually penniless. His lecture fees were almost nonexistent, and his parents could provide almost nothing toward his support. With the help of Richarz, he made a successful appeal in April, May, and June to the Ministry of Education for a supplemental stipend of 1,500 marks per year; there was some humiliation involved, as he had to submit his father’s tax returns to show that his parents were unable to support him.31 In return for this stipend, Wegener took on some “additional duties” in the supervision of Marburg’s Astronomical Observatory atop the Physical Institute, with its modest refracting telescope.
The stipend he arranged in June 1910 from t
he Ministry of Education was a great help, even though it was less than half of the salary that Aßmann had offered to entice him to Lindenberg. He still needed more money, as evidenced by the correspondence back and forth with the Danmark Committee requesting the use of lantern slides and permission to give paid public lectures, for which he had to obtain permission each time under the stringent conditions set down by the now-deceased Mylius-Erichsen. He was giving so many of these lectures in June and July that he asked if he might hold on to the slides rather than mail them back each time.32 He fretted at the loss of time these lectures entailed, but he needed the money.
Along with his stipend and his fees for public lectures, he wrote a few popular articles for which he picked up honoraria.33 His final source of income, as he tried to stitch together a living to support his life as a researcher, was the remainder of his Danmark Expedition salary, now down to a trickle, and constantly in need of extension. In order to keep the money flowing, he had to make continuous progress on the expedition reports. There was the remainder of the Danmark data to publish, and in the summer of 1910 he worked continuously at the next full volume on the station observations from the terminal at Danmarkshavn, on which he had to use extreme care because they were the reference values for all his other comparative measurements and would be the data set most eagerly sought by other investigators, other than the aerology. This work was dreary and time-consuming, very much like his dissertation at Berlin—the reduction of values taken under very different conditions, and requiring absolute vigilance in making sure, as the fragments or sections of this manuscript traveled back and forth to the press in Denmark, that the values were set in type correctly.
Wegener could not concentrate wholeheartedly on the thermodynamics of the atmosphere in summer 1910 because he was constantly pulled away from this work by moneymaking schemes. He obtained some relief from this with a series of balloon flights in association with his newfound friend, the physicist Karl Stuchtey, who had been sitting in on his lectures on atmospheric physics. They began to sketch out a plan to build an instrument to be used aloft to measure the albedo (the percentage of sunlight reflected rather than absorbed by the surface) of Earth and of cloud surfaces, as a way of computing the thermal regime in the middle of the troposphere. They built a prototype that summer and used it on several flights. The instrument allowed them to observe which wavelengths of light were being reflected or radiated back from a cloud and thus make inferences as to the composition of the cloud elements. In aerology, as in astronomy, much work was done by indirect means: the shape of a shadow, the color of light; all these optical phenomena and many others were employed to squeeze out every bit of information concerning the physics of the clouds.34