General Circulation Ideas And Controversies s To Early s

To appreciate the momentous changes that took place in general circulation theory between ~ 1940 and 1955, one has only to read Brunt's classic text (Brunt, 1944, Chap. 19), and follow this with a reading of Eady's (1957) contribution 13 years later, "The General Circulation of the Atmosphere and Oceans." From Brunt, the reader is left feeling that a consistent theory of the atmosphere's general circulation is out of reach: "It has been pointed out by many writers that it is impossible to derive a theory of the general circulation based on the known value of the solar constant, the constitution of the atmosphere, and the distribution of land and sea____It is only possible to begin by assuming the known tempera

1 The adjudicators also commended the excellence of the entry "On the dynamics of the general circulation" by Robert Fleagle (1957).

ture distribution, then deriving the corresponding pressure distribution, and finally the corresponding wind circulation" (Brunt, 1944, p. 405).

Eady's discussion, on the other hand, promotes a sense of confidence that the general circulation problem, albeit complicated, was yielding to new theoretical developments in concert with upper air observations. His final paragraph begins "If from this incomplete survey, the reader has gained the impression that general circulation problems are complicated, this is as it should be. The point is that mere complication does not prevent their being solved. Much of the complication shows itself when we attempt to give precise answers instead of vague ones____ To answer problems in any branch of geophysics we need vast quantities of observations but we also need precise, consistent, mathematical theory to make proper use of them" (Eady, 1957, p. 151).

Certainly the 10-year period prior to Phillips's numerical experiment was one of ferment as far as general circulation was concerned. A brief review of the major issues and themes during this period follow.

A. Rossby: Lateral Diffusion

Rossby's interest in the general circulation problem can be traced to his review paper on atmospheric turbulence (Rossby, 1927). In this paper, the work of Austrian meteorologists Wilhelm Schmidt and Albert Defant was highlighted. Defant (1921) had suggested that traveling midlatitude cyclones and anticyclones could be viewed as turbulent elements in a quasi-horizontal process of heat exchange between air masses, and he quantified the process by calculating an austausch or exchange coefficient following Schmidt (1917). Rossby was attracted by this concept (especially in the context of momentum transfer), and he applied it to the gulf stream and tropospheric westerlies (Rossby, 1936, 1937, 1938a,b, respectively).

Rossby summarized his ideas in a wide-ranging review article in Climate and Man (Yearbook of Agriculture), a compendium of meteorology that was shaped by a diverse committee headed by Chief of the Weather Bureau Francis Reichelderfer (Rossby, 1941). Rossby relied on the three-cell model of circulation that emanated from the work of 19th-century scientists William Ferrel and James Coffin (Ferrel, 1859; Coffin, 1875). This conceptual model, as it appeared in Rossby's article, is shown in Fig. 2. Here we see two direct cells: the equatorial cell (called the "Hadley cell") and the polar cell. The indirect cell in the midlatitudes is called the "Ferrel cell."

EQUATOR-

EQUATOR-

Hadley Cell
(A)

Figure 2 (A) Three-cell conceptual model of global circulation (extracted from Fig. 4 of Rossby, 1941). Deep cumulus cloud is indicated in the equatorial zone, clear sky is associated with descending air in the subtropics (~30°N), and precipitation occurs in association with ascent of air over the polar front zone. Westerly/easterly winds are indicated along the meridional circulation circuits by the solid lines/"hatched" symbols. (B) Rossby is shown sitting at his desk in the U. S. Weather Bureau building in Washington, DC (ca. 1940). (Rossby photo courtesy of K. Howard and the Library of Congress.)

Figure 2 (A) Three-cell conceptual model of global circulation (extracted from Fig. 4 of Rossby, 1941). Deep cumulus cloud is indicated in the equatorial zone, clear sky is associated with descending air in the subtropics (~30°N), and precipitation occurs in association with ascent of air over the polar front zone. Westerly/easterly winds are indicated along the meridional circulation circuits by the solid lines/"hatched" symbols. (B) Rossby is shown sitting at his desk in the U. S. Weather Bureau building in Washington, DC (ca. 1940). (Rossby photo courtesy of K. Howard and the Library of Congress.)

Regarding the westerlies, Rossby (1941) argued as follows:

In the two direct circulation cells to the north and to the south, strong westerly winds are continuously being created at high levels. Along their boundaries with the middle cell, these strong westerly winds generate eddies with approximately vertical axes. Through the action of these eddies the momentum of the westerlies in the upper branches of the two direct cells is diffused toward middle latitudes, and the upper air in these regions is dragged along eastward. The westerlies observed in middle latitudes are thus frictionally driven by the surrounding direct cells... the air which sinks in the horse latitudes spreads both polewards and equatorwards. The poleward branch must obviously appear as a west wind____(p. 611)

Rossby modified his ideas by the late 1940s—vorticity becoming the transferable property rather than momentum (Rossby, 1947).

B. Jeffreys -Starr -Bjerknes -Priestley -Fultz: Asymmetric Eddies

Tucked away near the end of a paper that explored atmospheric circulation by analogy with tidal theory, Harold Jeffreys argued that asymmetric eddies (cyclones/anticyclones) "...not unlike that described by Bjerknes____" were an essential component of the atmosphere's general circulation (Jeffreys, 1926). Quantitative arguments based on the conservation of angular momentum led him to state that a steady meridional (axially symmetric) circulation could not be maintained. Balance could only be achieved when the frictional torque was balanced by angular momentum transport due to asymmetric eddies. The governing equation for this transport is the integral (around a latitude circle) of the product of horizontal wind components. Quoting Jeffreys (1926, p. 99), "Considering any interchange of air across a parallel of latitude, then uv [the product of horizontal winds] must be negative both for the air moving north and for that moving south. This corresponds to the observed preponderance of south-westerly and north-easterly winds over those in the other two quadrants." (Jeffreys chose a coordinate system where u was directed southward and v eastward. Thus, the sign of uv in Jeffreys's coordinate system is opposite to that found in the more conventional system where u points eastward and v northward.)

Jeffreys came to this conclusion after grappling with the frictional formulation in his theory. The paper conceals this battle, but his reminiscence exposes it:

... the point was that you could solve the [atmospheric] problem when you had adopted the hydrodynamical equations to a compressible fluid... you could solve that for a disturbance of temperature of the right sort, and you could solve it in just the same way as you did for the tides—and it just wouldn't work! At least it worked all right when you didn't put in any friction. When you put friction in, it turned out that the friction in the result would stop the circulation in about a fortnight, and I had to start again, and I found that the only way to do it was to have a strong correlation between the easterly and northerly components of wind. (Jeffreys, 1986, p. 14)

Jeffreys's theory laid dormant for ~ 20 years. It was rejuvenated in the late 1940s by Victor Starr (1948), Bjerknes (1948), and Charles Priestley (1949). In the second paragraph of Starr's paper, he says "In reality, this essay may be construed as a further extension of the approach to the problem initiated by Jeffreys." Starr, who had exhibited his prowess with mathematical physics applied to the geophysical system (see, e.g., Starr, 1939, 1945), displayed another aspect of his skill as a researcher in this essay—namely, a clarity of expression and an expansive research vision. In essence, the essay became the blueprint for Starr's research plan at MIT during the next decade.2 The upper air observations collected in the postwar period made it clear that there was a decidedly NE-SW tilt to the horizontal streamlines, "... so common on meteorological maps, [it] is a necessary automatic adjustment to provide for the poleward transfer of atmospheric angular momentum" (Starr, 1948, p. 41). Dave Fultz's hydro-dynamical laboratory experiments confirmed the tilted streamline patterns and became an independent source of support for Jeffreys's theory. (Photographs from Fultz's experiment are shown in Starr, 1956.)

The initial investigations by Starr and Bjerknes led to independent, long-term efforts (at MIT and UCLA, respectively) to collect and archive upper air data on a global scale. These assidious efforts led to sets of general circulation "statistics"—measures of the temporally and/or spatially averaged terms in the heat and angular momentum budget equations (see the contributions by Starr and White, 1951, and Mintz, 1951, 1975). Priestley's work is notable, however, because his calculations relied on observed winds rather than geostrophic approximations to the wind. Priestley continued his work on these problems until the early 1950s "... before yielding to the greater resources of the two American pairs, Bjerknes-Mintz and Starr-[Robert] White..." (Priestley, 1988, p. 104).

Photographs of the scientists who were instrumental in studying the asymmetric aspects of the general circulation are shown in Fig. 3.

2 Starr was the second recipient of the Ph.D. in meteorology from the University of Chicago (Summer 1946) [The first recipient was Morris Neiberger (Autumn 1945).] Starr accepted a faculty position at MIT in 1947.

Zef Haller

Figure 3 (A) Harold Jeffreys sits in his office at Cambridge (ca. 1928). (B) C. H. B. Priestley (ca. 1980). (C) Jacob Bjerknes (in the foreground) and Dave Fultz at the University of Chicago's Hydrodynamics Laboratory (1953). (D) Victor Starr (ca. 1965). (Courtesy of Lady Jeffreys, Dave Fultz, Constance Priestley, and the MIT archives.)

Figure 3 (A) Harold Jeffreys sits in his office at Cambridge (ca. 1928). (B) C. H. B. Priestley (ca. 1980). (C) Jacob Bjerknes (in the foreground) and Dave Fultz at the University of Chicago's Hydrodynamics Laboratory (1953). (D) Victor Starr (ca. 1965). (Courtesy of Lady Jeffreys, Dave Fultz, Constance Priestley, and the MIT archives.)

C. PALMEN AND RlEHL: JET STREAMS

The existence of the strong and narrow band of upper level westerlies, labeled the jet stream, was established by forecasters in Germany (late 1930s) and the United States (early 1940s) (see Seilkopf, 1939, and Flohn, 1992; Riehl et al., 1954; and Plumley, 1994, respectively). Following World War II, Rossby obtained funding from the Office of Naval Research (ONR) for a comprehensive study of atmospheric general circulation (including the dynamics of the jet stream). He invited Erik Palmen to assume a leadership role in this research. Palmen had spent his early career at Finland's Institute for Marine Research, and was named director of the institute in October 1939, just 2 months before Russia invaded

Figure 3 (Continued)

Figure 3 (Continued)

Erik Herbert Palm

Figure 3 (Continued)

Figure 3 (Continued)

Finland. Throughout the remainder of WWII, Palmen's scientific work was severely curtailed. "He [Palmen] was born again in the setting of the general circulation project at the U of C [University of Chicago]" (C. Newton, personal communication, 1990). He remained at Chicago for 2 years (1946-1948), returning to Finland in late 1948 as chair professor of meteorology at the University of Helsinki. His frequent long-term visits to Chicago during the next decade, however, made him a fixture at the U of C's Institute of Meteorology.

In June 1947, the expansive report on the ONR project appeared under the authorship of staff members of the Department of Meteorology (Staff Members, 1947). Salient features of the jet stream were enumerated in the Summary section of the paper. Notable were the following: (1) The jet is located in or just south of a zone in which a large fraction of the middle and upper troposphere temperature contrast between polar and equatorial regions is concentrated; and (2) below the jet stream, it is possible to identify a well-defined frontal zone, intersecting the ground south of the jet stream.

Palmén became convinced that the concept of a single circumpolar jet was questionable, and he proposed the existence of a second jet, which he called the subtropical jet. "He [Palmén] thought that the great mass of air convected to the upper troposphere in the tropics could not all then descend in the subtropics. As evidence kept mounting, one began to speak of the "subtropical jet stream" found mainly above 500 mb and not undergoing the many violent north-south oscillations of the northern, soon called "polar jet stream" (Riehl, 1988).

Following Palmén's return to Finland in 1948, Herbert Riehl became the scientific leader of the jet stream project. Through the continued sponsorship of ONR, research flights across the circumpolar jet stream were initiated in 1953 (Riehl, personal communication, 1994).

D. Controversies

Amid such rapid advancement in meteorology, along with the slate of competing ideas, there is little wonder that this period had its share of controversies. A considerable amount of heated debate occurred at the daily map briefings at University of Chicago in the late 1940s. George Cressman offered daily discussions and forecasts with all the available maps (from mid-Pacific Ocean to the Ural Mountains in Russia—240° of longitude in the Northern Hemisphere). There was no end to the arguments about general and cyclone circulations that followed Cressman's briefings. The "reverse cell" of midlatitudes created fuel for the verbal exchanges. The abrupt transition from equatorward westerlies at high level in this middle cell to the neighboring easterlies in the equatorward or Hadley cell was conceptually difficult to understand (see Palmén and Newton, 1969, Chap. 1, for a summary of research that established the existence of the upper level easterlies). In Riehl's words, "... [why should] the equatorward westerlies, virtually friction-free in high atmosphere, ... quickly diminish and go over into easterlies, just where the maximum west wind is observed" (Riehl, 1988, p. 4).

One of the most celebrated scientific exchanges occurred in the Correspondence section of the Journal of Meteorology. Starr and Rossby (1949) wrote a short article reconciling their differences on the role of angular momentum conservation in the atmosphere's general circulation. Their "differences" were minor, essentially related to the interpretation of terms in the equation of angular momentum conservation. One of the statements in the article, however, created an uproar. This cardinal statement reads: "Most of the classic theories for the general circulation were based upon the assumption that it is this effect of meridional circulations which maintains the angular momentum of the zonal motions in the atmosphere. It is this assumption that both of us call into question for reasons enumerated by Rossby [1941]." They go on to say that, in their opinion, it is the advective transport of relative angular momentum—the uv term in Jeffreys's formulation—that is of prime importance in the mechanics of the general circulation.

Four months after the appearance of the Rossby-Starr article, Palmén wrote a letter to the editor that adamantly questioned the conclusion stated above (Palmén, 1949). He argued that the mean meridional circulation term could not be discounted; furthermore, Palmén made order of magnitude estimates of the meridional transport and found them comparable to the eddy transport term. The verbiage was strong and it elicited an ordered yet acerbic response from Starr (1949). Quoting Starr, p. 430 "Apparently Palmén suspects me of highest heresy lest I suggest that the energy production process may also be accomplished without the aid of meridional circulations. This I have indeed proposed... the hypothesis that meridional cells are of small importance seems to be bearing fruit. Indeed if such are the fruits of heresy, then I say let us have more heresy."

Although more stimulating than controversial, the general circulation statistics generated by the research teams at UCLA and MIT were demanding explanation. For example, the work of Bjerknes (and Mintz) at UCLA showed that the poleward eddy heat flux had its maximum at 50° latitude and was strongest near the ground. On the other hand, the poleward eddy angular momentum flux had its maximum near 30° and was strongest near the tropopause (Bjerknes, 1955).

Thus, by the mid-1950s, major questions related to the atmosphere's general circulation begged for answers. Among the issues were the respective roles of the mean meridional circulation and transient eddies in the momentum and energy budgets, mechanism for the maintenance of the westerlies (jet streams), and the dynamical basis for alternating wind regimes at the surface.

Was this article helpful?

+3 0

Post a comment