Using for calibration various data, available at this time, I accepted for stellar populations with normal metal content m0 = 0.03 M, for metal-rich populations m0 = 0.001 M, and for metal-poor populations m0 = 0.1 M. Using these lower mass limits we get for old metal-poor halo populations Mi/Li ≈ 3, for extremely metal-rich populations in central regions of galaxies Mi/Li ≈ 100, and for intermediate populations (bulges and disks) Mi/Li ≈ 10, see Fig. 3.4. Using these data I obtained for galaxies M31, M32, and M87 mean total values M/LB = 9, 4, 22, respectively. M/LB of old galactic populations depends also on the total mass of the galaxy, see Fig. 3.5 (Einasto, 1974a).
Fig. 3.4 The evolution of the mass-to-luminosity ratio fB for stellar populations of different metallicity Z and instant star formation. The age t is given in years (Einasto, 1974a).
As I found later, so wide a range of m0 values for various metal content is not needed. Modern dynamical data yield for all populations lower values of M/L, due to more accurate measurements of velocity dispersions in star clusters, and in central regions of galaxies. As suggested in pioneering studies by Faber & Jackson (1976); Faber et al. (1977), the bulge of the Sombrero galaxy has a mass- to-luminosity ratio M/L = 3, and the mean mass-to-luminosity ratios for elliptical galaxies is about 7, close to the ratio for early type spiral galaxies. New data suggested that central velocity dispersions of giant galaxies are by a factor of 2 lower than accepted earlier. This reduces M/L values for metal-rich cores by a factor of up to 4. For extremely metal-poor globular clusters new data suggest a value M/L ≈ 1. Thus in retrospect I can say that my first set of population evolution models is my version of ‘maximum disk’ models.
Fig. 3.5 The dependence of the mass-to-luminosity ratio fB of old galactic populations on the total mass of the galaxy (Einasto, 1974a).
New data solved one discrepancy in my earlier models — if mass-to-luminosity ratios of central regions are taken from early velocity dispersion data, then circular velocities, calculated from mass distribution models, are too high, much higher than suggested from rotation data in central regions of galaxies, see Fig. 4.3 and the discussion below. Earlier I thought that such large deviations are due to the dominance of random motions in central regions, and that rotation velocities cannot be used to find the mass distribution in these regions. According to new velocity dispersion data, M/L values found for central regions are considerably lower, and observed (optical) and model rotation data are in good mutual agreement, as found in our later model of M31 by Tenjes et al. (1994), shown in the right panel of Fig. 4.3. But this suggests that corrections are needed to my previous galaxy evolution models. The lower limit m0 of masses in the Salpeter law must be higher than accepted earlier, for all galactic populations it should be in the range of 0.05 — 0.1 M. This lower limit is close to the limit found from independent data for stars where nuclear burning of hydrogen is possible.
New data increased the discrepancy between earlier classical galaxy models and new ones. In other words, high mass-to-luminosity ratios of groups and clusters of galaxies cannot be explained by known galactic populations.
My results were similar to those of Tinsley (1968) with one important difference. Tinsley used much lower values of m0 ∼ 10−6 M to get high values of M/LB ≃ 250–500 for elliptical galaxies, as suggested by the dynamics of companions of luminous elliptical galaxies. In other words, in Tinsley models of elliptical galaxies most stars were Jupiter-like objects without internal sources of nuclear energy.
Modern calculations suggest that the first generation of metal-free population III stars have large masses (~ 100 M) due to the large Jeans mass during the initial baryonic collapse (for a discussion see Reed et al. (2005) and references therein). Thus population III stars are not suitable to represent a high M/L halo population at the present epoch.
3.2.6 Models of galaxies of the local group and M87; mass paradox in galaxies
In the early 1970’s I collected from all possible sources data on our Galaxy, M31, M32, Fornax and Sculptor spheroidal dwarf galaxies, and the giant elliptical galaxy M87 in the Virgo cluster. These data were used to calculate population models of these galaxies. The most detailed model was found for M31, where parameters for the nucleus (a massive single object at the center, now we know that it is a massive black hole), core, bulge, disk, halo, and young disk were found. Dwarf spheroidal galaxies were represented only by the halo, for other galaxies at least three populations were used. In these models I used results of my calculations of the evolution of populations. These population models as well evolution models were presented as part of my Doctor of Sciences thesis (Einasto, 1972b).
For M31 and our Galaxy I calculated hydrodynamical models for all main galactic populations. These models included densities, density gradients, velocity dispersions, and a number of other parameters (Einasto & Rümmel, 1970b; Einasto, 1972b, 1974a). As an example, for one flat population of our Galaxy these functions are shown in Fig. 3.6.
In the modelling of M311 encountered a serious problem. If rotation data were taken at face value, then it was impossible to represent the rotational velocity with the sum of known stellar populations. The local value of M/L increases towards the periphery of M31 very rapidly, if the mass distribution is calculated directly from the rotation velocity, see Fig. 4.2 in the next Chapter. All known old metal-poor halo-type stellar populations have a low M/L ≈ 1 − 3; in contrast on the basis of rotation data I got M/L > 1000 on the periphery of the galaxy near the last point with measured rotational velocity.
Fig. 3.6 Various description functions of a flat population of the Galaxy with the effective radius ao = 7 4 kpc, axial ratio ε = 0 05, and structural parameter N = 1. In the upper panel the z-dependence of the density is shown in units of the central density of the population. Additionally the vertical density gradient l = −δlog /δz, log, and the velocity dispersion (σz)0 are given for the distance from the center of the Galaxy R = 10 kpc. In the middle panel the R-dependence of the density , of the radial density gradient m = −δlog /δR, and of the three velocity dispersions σR, σθ, σz are shown, at the Galactic plane z= 0. In the bottom panel one line of the constant density is shown, as well as lines of constant velocity dispersion, σz, for the meridional plane of the Galaxy; velocities are expressed in km/s (Einasto, 1974a).
There were two possibilities to solve this discrepancy: to accept the presence of a new population with very uncommon properties, or to assume that on the periphery of galaxies there exist large non-circular motions. I considered both possibilities. My conclusion was that if there exist only stellar populations with known properties, then the first alternative has several serious difficulties.
If the hypothetical population is of stellar origin, it must be formed much earlier than all known populations, because known stellar populations form a continuous sequence of kinematical and physical properties (Rootsmäe, 1961; Einasto, 1974b), and there is no place to put this new population into this sequence: the estimated velocity dispersion of stars of the new population is too high for a conventional stellar population. In other words, the new population must be well separated from all known populations, both in velocity space and spatially.
Secondly, the star formation rate is proportional to the square of the local density (Schmidt, 1959; Einasto, 1972c), thus stars of this new population should have been formed during the contraction phase of the formation of the population near its central denser regions, and later expanded to the present distance. The only source of energy for expansion is the contraction of other stellar populations. The estimated total mass of the new population exceeded the summed mass of all previously known populations. Estimates of the energy needed for the expansion demonstrated that the mass of the new population is so large that even the contraction of all other stellar populations to zero radius would not be sufficient to expand the new population to its present size. This means that the new population must be formed prior to the formation of known stellar populations, before the collapse of the gas fro
m which all known stellar populations formed.
And, finally, it is known that star formation is not an efficient process: usually in a contracting gas cloud only about 1% of the mass is converted to stars. Thus we have a problem of how to convert, in an early stage of the evolution of the Universe, a high fraction of primordial gas into this population of first generation stars.
I also found it psychologically difficult to accept the first alternative. From rotation data it follows that the mass-to-luminosity ratio of this hypothetical population must be very high, thus it should be essentially a dark population. Our studies of the local density of matter near the Sun by Kuzmin and his students have shown that there is no room for the presence of large amounts of dark matter in the plane of the Galaxy. Thus I was not ready to accept the presence of a dark population elsewhere in galaxies.
Taking into account all these difficulties I accepted the second alternative — the presence of non-circular motions (Einasto, 1969b; Einasto & Rümmel, 1970c). A similar decision was made by many other astronomers, as discussed during the Third European Astronomy Meeting by Materne & Tammann (1976).
As I soon realised, this was a wrong decision.
One more episode from this period. In the late 1960’s a young astronomer from the Leningrad University published two papers in theArmenian JournalAstrofizika, where he developed a model of M31 (Sizikov, 1968, 1969). He used both rotation and photometric data, while the density was expressed by one spheroidal component of variable flatness as in the Kuzmin (1956a) model. These papers were the basis of his PhD (candidate) thesis, and I was appointed as the reviewer (opponent). Our main discussion during the defence was devoted to the mass distribution on the outskirts of M31. At this time I already had preliminary results of my galaxy evolution model calculations, and knew that it is very difficult to assume the presence of a large amont of very faint dwarf stars in outer regions of galaxies, as explicitly assumed by Sizikov. However, my results were only preliminary and not yet published, thus it was difficult for me to show any error in the Sizikov model. I gave the thesis high opinion, and we agreed that further study was needed to understand the difference between our models.
3.3 Tartu Observatory in the 1960’s
3.3.1 New observatory
The first scientific conference in the new observatory was a cosmology summer school in July 1962, where we met for the first time Yakov Zeldovich. He had just returned from military to civil science, and started to study the physics of the most powerful explosion — the Big Bang. At this time we had no idea that in years to come our group would have a very close collaboration with him.
The new observatory was opened with a conference in 1964 under the motto “Science is carried by search for truth that is as sincere and honest as the Nature itself “ (T. Rootsmäe). A number of leading astronomers attended, among them the director of the Pulkovo Observatory, academician A. Mikhailov, the director of the Byurakan Observatory, academician V. Ambartsumian, the chairman of the Astronomical Council of the USSR Academy of Sciences, academician E. Mustel, and also a great-granddaughter of F.G.W. Struve. After the conference trees were planted by many participants. The trees grew well and now there is a beautiful park around the main building.
After defending my PhD thesis I spent several years teaching at the University and working in the station to observe artificial Earth satellites. The observation of satellites (or sputniks) gave the students a certain experience in astronomical observations and thus popularised studying astronomy at university. Among other things I found a way to construct a four-axis mount for satellite tracing telescopes (Liigant & Einasto, 1960), together with my student M. Liigant we got an author’s certificate for this telescope and a medal at the All-Union Exhibition. A similar mount was invented by the Carl Zeiss factory in Jena, and a series of satellite-tracking telescopes were built. Zeiss representatives invited me to Jena to inspect their telescope. This visit was in summer 1962. We did not patent our version, so Zeiss was not obliged to pay us any royalties. This was my short excursion away from galactic modelling.
Fig. 3.7 The cosmology school in the new observatory, July 1962. In the center of the second row with beret is Yakov Zeldovich, in the first row in front of Zeldovich are Alla Massevich and Bruno Pontecorvo (author’s archive).
After moving to Tõravere, together with our students and young collaborators I spent nearly half a year exploring the astronomical literature and studying new trends of development in astronomy. We had a series of astronomy seminars where we discussed the new directions in astronomy. According to earlier plans, it was intended to buy a Schmidt camera as the main telescope in order to continue stellar- statistical observations. As a result of discussion we came to the conclusion that in our climatic conditions it would be rational to pay more attention to stellar physics, particularly to spectral observations of stars, where much more useful information can be obtained in relatively short periods of clear weather.
After these discussions we prepared a new program of the observatory’s development, in which we planned to construct a reflecting telescope with a mirror of diameter 1.5 meters. This plan received the support of the Astronomical Council and became the basis of our future development. In the mid-sixties I worked with the design of the telescope and its dome for nearly five years. Among other things I studied the thermal regime of the telescope dome to avoid micro-turbulence of air during observations. Results of this study were published (Einasto & Laigo, 1973) and used in the construction of the dome of our 1.5-m telescope, as well as the dome of the large Soviet 6-m telescope in North Caucasus.
Fig. 3.8 The opening ceremony of the new Tartu Observatory in 1964. In the first row sitting from left are: Mrs. Naan, G. Naan, A. Mikhailov, E. Mustel, V. Ambartsumian, A. Kipper, K. Ogorodnikov, Mrs. Kuzmin, G. Kuzmin (author’s archive).
The 1.5-m telescope was completed in 1975. At first we tried to use it for spectral observations of galaxies too, but the climatic conditions were unstable, so the new telescope remained mainly for spectral observations of stars, as originally planned.
The sixties were the years of rapid growth for Tartu Observatory. Many talented young people had received their initial training at the satellite station of the Tartu University. The satellite station provided the students with good observation practice which was highly useful in the new observatory. In accordance with our development plans, most of the students moved to stellar physics.
3.3.2 Philosophical seminars and New Year parties
In the new observatory we continued philosophical seminars. The foundation for this kind of seminar has already been laid in the fifties by professor Aksel Kipper. The invited speakers were distinguished scientists from other fields, as well as cultural and political figures. The atmosphere at the seminars was quite free. For example, one of the most interesting presentations was professor Uku Masing’s “Religion in the History of Mankind”. In the presentation Uku Masing developed the thesis that all major religions have taken the form of a political movement, and vice versa, all political movements acquire the characteristics of a religion. As an example, Uku Masing brought up Hitler’s Germany. Everyone of course realised which country he really meant. According to Masing’s thesis, the main task of a religion is to establish a system of conventions and beliefs, without which no society can exist. If such a system of beliefs and customs is destroyed, the society would become very unstable. Once again it was clear which society the talk was of.
Fig. 3.9 Grigori Kuzmin and Viktor Ambartsumian planting a tree at the opening ceremony of the new Tartu Observatory in 1964 (author’s archive).
Fig. 3.10 The 1.5-m Telescope dome of the new Tartu Observatory in 2011. In the foreground are Laurits Leedjarv and Tim de Zeeuw (author’s photo).
I once appeared at the seminar myself, on the topic “The Effectivity of Scientific Research”. Studies had recently begun on this topic so I could rely on some earlier work. But most of the analysis had to be done myself without using examples. A
mong other things, I looked at how many articles appeared in different publications in the field of astronomy, and how often they were cited. Thus it was possible to determine the impact factor of a magazine or a publication series. I did not know that the impact factor has already been used. It turned out that at that time (1965) nearly half of all astronomical papers were published in the observatories’ own publications, but there were almost no references to those papers. The journals “Nature” and “Annual Review of Astronomy and Astrophysics” had the highest impact factors, followed by “Astrophysical Journal”. The effectivity of scientific research also varied; sometimes some astronomers had many publications but they were not cited. Citations of the publications of Tartu Observatory were very rare, as were that of most journals in Russian.
I drew my own conclusions and thereafter refrained from publishing my papers in our own Observatory publications. I published a few papers in Russian-language journals, but soon I started sending my papers to English-language journals. Another reason to avoid Russian was because, when translated from Russian to English, my name became distorted (it had the shape Ya. E. Ejnasto). For a while I pondered over the best way to write my address, because I did not want to use “Estonian SSR”. Finally I decided to use as my address “Tartu Observatory, Estonia, USSR”. There was a hidden message in this address — Estonia is still alive, but occupied by the USSR; curiously enough, all our papers were censored, but the censors did not notice the message. Later, when the time was right, the “USSR” could be omitted; the rest of the address was already correct.
Dark Matter and Cosmic Web Story Page 10
