This paper concerns the history of magnetic-flux transport and not so much the technical details and consequences of the model, which Marc DeRosa will discuss in a companion paper. I thought it would be interesting to review the history and to write it down before it is forgotten, but I did not want to punish the reader (or myself) by discussing all of the results that we have learned along the way. The interested reader can track them down fairly quickly beginning with the references that are listed here.
In that respect, the referees of this paper provided additional references, some of which I was aware and some of which I was not. The former category includes a review by Rabin et al. (1991), of which Rick DeVore and I are co-authors, but which I had forgotten. The latter category includes papers by Csada (1949, 1955), who presented a mathematical theory of differential rotation and meridional flow in rotating magnetic stars.
A referee also mentioned papers by Plaskett (1959) and Ward (1965). I don’t remember if I was aware of the former paper in the 1963 – 1965 era when we were beginning our work on flux transport, but I was aware of the latter paper. In 1965, Fred Ward came to Caltech to discuss his idea that the Greenwich sunspot observations show evidence of Rossby waves on the Sun. Leighton was so opposed to Ward’s argument that he wrote a rebuttal (Leighton, 1965b), in which he concluded:
‘To summarize, we have shown that the fluctuations and systematic variations in spot group areas would, a priori, be expected to produce effects of the kinds and approximate magnitudes reported by Ward. While the roughness of our estimates does not yet permit us to be certain that Ward’s results are entirely spurious, the degree of agreement surely suggests that his result be treated with the utmost caution. If the possible sources of difficulty discussed here are in fact less important than estimated, they should be proved so as result of suitable measurements rather than simply by assumption.’
The manuscript is dated May 15, 1965 and contains an inscription indicating that the paper was sent to the Astrophysical Journal. But to my knowledge, it was never published.
One can gain a further sense of the meridional-flow issue by going back a few years earlier to Babcock’s (1961) famous paper, ‘The Topology of the Sun’s Magnetic Field and the 22-Year Cycle’. Here, Babcock cited a previous Babcock and Babcock (1955) paper, in which he and his father had speculated that the poleward migration of the trailing parts of bipolar regions was responsible for the cancellation and reversal of the polar fields, and that the leading parts of the bipolar regions must therefore either be neutralized at the equator or contribute to a quadrupole moment of the Sun. He went on to say that there is no persistent quadrupole, so there must be migration toward the equator at low latitudes and toward the poles at moderate latitudes. However, the cause of this migration was obscure and Babcock could do little more than appeal to supporting observations by Richardson and Schwarzschild (1953), who had reported that individual sunspots drifted slowly away from the 20°latitude circles. For those of us who were skeptical of meridional flow speeds obtained by tracking sunspots, the combination of random-walk diffusion and systematically tilted sunspot groups (Joy’s law) seemed to be a much more convincing way of moving trailing-polarity flux toward the poles and leading-polarity flux toward the equator.
One of the referees wondered if the informal solar lunches and frequent seminars that took place in Pasadena when Zirin, Leighton, and Howard were reinvigorating solar research at Caltech and Mount Wilson played a role in stimulating the development of the flux-transport model. That is an interesting historical question, but the answer is ‘no’. As indicated at the beginning of this paper, Leighton got the idea in September 1963 while he was in Germany preparing his IAU talk on supergranulation. Hal Zirin had not yet arrived at Caltech and the solar lunches were at least a few years in the future. By the time of the solar lunches, I had moved to the Kitt Peak Observatory in Tucson and was less familiar with what was happening at Caltech. However, I think that Leighton had pretty much turned his attention to infrared and millimeter wave astronomy, and that Howard and Zirin were working with a new group of students and post-docs at Mount Wilson and the Big Bear Solar Observatory.
One of them was K. A. Marsh who suggested that the interaction between ephemeral regions and the previously existing network fields might contribute to the large-scale diffusion of flux (Marsh, 1978). The Marsh mechanism was probably an exciting topic of conversation at Caltech lunches when people were trying to reconcile the relatively high diffusion rates 800 km2 s–1 which Leighton used to reverse the polar fields, and the much slower rates of 100 – 20 km2 s–1 found by tracking small-scale flux elements. However, the Marsh mechanism did not resolve this discrepancy because the eruption of ephemeral regions with random orientations is not a diffusive process and, in particular, it does not change the large-scale distribution of flux on the Sun (Wang and Sheeley Jr, 1991). More recently, Schrijver et al. (1996) offered a different explanation for the discrepancy, arguing that the traditional methods of tracking flux give smaller diffusion rates because they emphasize the larger, more sluggish, flux elements and underestimate the significant contribution of the many smaller, more mobile, concentrations.
Work on the flux-transport model has waxed and waned over the years. There were relatively few numerical studies from the mid-1960s when Leighton proposed the model until the 1980s when the NRL group started its work. During this interval, observational studies, performed mainly at the Big Bear, San Fernando, Lockheed, and Mount Wilson Observatories, gave results (like small diffusion rates and reports of meridional flow) that were inconsistent with the model as Leighton had originally proposed it. But eventually flow was added to the model and the numerical simulations provided a number of new results and explanations. Additional groups appreciated the usefulness of the model and developed their own flux-transport codes. Now, we are in a new era of extending and improving the flux-transport model (van Ballegooijen et al., 1998; Schrijver, 2001; McCloughan and Durrant, 2002; Mackay et al., 2002a,b; Baumann et al., 2004), and applying it to interesting subjects like the Sun’s long-term behavior and the magnetic fields of other stars (Schrijver and Title, 2001; Mackay et al., 2004).
© Max Planck Society and the author(s)