eso9417 — Science Release
First Observations of Solar-Like Oscillations in Another Star (1)
Minute temperature fluctuations detected in Eta Bootis.
23 November 1994
A group of astronomers from the Aarhus University (Denmark) and the European Southern Observatory (2) have for the first time succeeded in detecting solar-type oscillations in another star. They observed the temperature of the bright northern star Eta Bootis during six nights with the 2.5-metre Nordic Optical Telescope at the Roque de los Muchachos observatory on the island of La Palma (Canary Islands) and were able to show that it varies periodically by a few hundredths of a degree. These changes are caused by pressure waves in the star and are directly dependent on its inner structure. A detailed analysis by the astronomers has shown that the observed effects are in good agreement with current stellar models. This is a most important, independent test of stellar theory.
The Sun is an Oscillating Star
About twenty years ago, it was discovered that the nearest star, our Sun, oscillates like the ringing of a bell with a period of about 5 minutes. The same phenomenon is known in the Earth, which begins to
vibrate after earthquakes; in this way seismologists have been able to discern a layered structure in the Earth's interior. The recent impacts of a comet on Jupiter most likely had a similar effect on that planet.
The observed solar oscillations concern the entire gaseous body of the Sun, but we can of course only observe them on its surface. It has been found that each mode moves the surface up and down by less than 25 metres; the combined motion is very complicated, because there are many different, simultaneous modes, each of which has a slightly different period. The exact values of these periods are sensitive to the speed of sound in the Sun's interior, which in turn depends on the density of the material there. Thus, by measuring the periods of solar oscillations, we may probe the internal structure of the Sun, that is otherwise inaccessible to observations.
Why does the Sun oscillate and what is the cause of these oscillations? We do not know yet, but it is thought that the driving force is convective motion in the interior. For a better understanding of this basic phenomenon, we may compare the Sun with a pot of boiling water on the stove. The "bubbles" in the pot, known as solar convection cells, rise upwards towards the surface and jostle the Sun from the inside. This causes it to oscillate, although we still do not know the details of how these oscillations are triggered. We can just be grateful they exist, and by measuring their periods we obtain important information about the inside of the Sun. A great deal of progress has recently been made in this way.
Do Other Stars Oscillate Like the Sun?
The Sun is a normal star. It oscillates, and we would therefore expect that other, similar stars also do.
Indeed, large-amplitude stellar oscillations have been known for centuries to manifest themselves as significant changes in the observed brightness of some "variable stars". However, in most cases, for example in the so-called Cepheid and RR Lyrae variable stars, only one or two periods have been detected. What distinguishes solar-type oscillations is the large number of observed periods, that potentially gives a great deal of information about the stellar interiors, as well as their much smaller amplitudes.
During the past decade, astronomers have been trying to detect this type of oscillations in stars other than the Sun, but with little success. Such oscillations are much more difficult to detect, because the stars are much further away and therefore fainter than the Sun.
Most observational attempts have tried to detect the movement of the stellar surface directly, by measuring velocity (Doppler) shifts in the stellar spectra, cf. the Appendix. Due to the up-down motion of less than 1 metre/sec, the spectral lines should shift backwards and forwards by about 1 part in 300,000,000, or less than 0.00002 nanometres in red light, a minuscule shift that is very difficult to measure with current astronomical instrumentation, even in the very brightest stars.
Faced with this problem, the Danish/ESO group came up with an entirely new method. It relies on the fact that the oscillations are sound waves which deposit energy in the various stellar layers and therefore intermittently heat the star very slightly. For example, each mode changes the temperature on the surface of the Sun by about 0.005 degrees during the oscillation.
But how to measure such small temperature changes? It turns out that this is possible by recording the strengths of the spectral lines, specifically, the absorption lines due to hydrogen. Their strengths change slightly with the changes in temperature (see Appendix). Although this is still a very small effect, it should be easier to measure than the velocity shifts.
Yes, Eta Bootis does!
To test their method, the astronomers used the ESO 3.5-metre New Technology Telescope (NTT) with the ESO Multi-Mode Instrument (EMMI) to observe a bright star for a few hours. This was too short to detect actual oscillations, but it did show that the technique works: it was in principle possible to measure the temperature accurately enough.
The target for the real observations was the 2.68-magnitude, naked-eye star Eta Bootis (Greek letter "eta"). It has the common name of Muphrid and is located just north of the celestial equator in Bootes,
one of the oldest constellation names still in use (it was mentioned already in the Odyssey). This particular star is somewhat more evolved and bigger than the Sun and, according to stellar theory, should have
stronger oscillations than the Sun, hence increasing the chance that they could be detected.
The observations were performed with the 2.5-metre Nordic Optical Telescope (NOT) during six, mostly clear nights in April 1994. A careful data analysis has now shown that the temperature of Eta Bootis is indeed changing periodically, around a mean value of about 6000 K. It seems to be oscillating in at least ten different modes simultaneously, with periods around 20 minutes. These periods are longer than those of the Sun, as expected for a star that is larger and heavier than the Sun.
The figure accompanying this Press Release shows these oscillations in the form of a "power spectrum", i.e., the amount of temperature change at different values of the period. Most of the highest peaks correspond to the real oscillations in the star. The changes (fluctuations) of the temperature of Eta Bootis vary with the oscillation mode and, at the time of these observations, were mostly between 0.03 and 0.08 degrees.
This diagramme provides the first strong evidence ever for solar-type oscillations in a star other than the Sun. An article with the detailed results will soon appear in the "Astronomical Journal".
Agreement with Stellar Theory
The measured periods of the main oscillation modes give important information about the interior of Eta Bootis. Theoretical models of the star have now been compared with these observations and the astronomers were pleased to find that the agreement is excellent, implying that current stellar theory is remarkably good.
This shows that we apparently understand stars quite well, but there is of course still much to be learned. Future observations of this kind, with ground-based telescopes and possibly in a more distant future also from space, promise to open up a new and exciting way of studying stars. From now on, we will be able "to look inside" stars in great detail.
Appendix: Spectral Analysis
Dark spectral lines were first seen in the solar spectrum by the German physicist Johann Fraunhofer in 1814. Later, in the mid-nineteenth century, such lines were also seen in the spectra of other stars.
It is now known that they are due to the upper, cooler layers in the solar and stellar atmospheres, whose atoms and molecules absorb the radiation from the hotter, deeper layers at specific wavelengths. These wavelengths serve as "footprints" of these atoms and molecules and allow astronomers to determine which chemical elements are present in the Sun and the stars.
The exact position of a dark line in the spectrum (its wavelength) depends on the velocity along the line of sight of the corresponding atoms or molecules. If they move in our direction, the wavelength of the line becomes slightly shorter; if they move away from us, it will be a little longer. This is referred to as the Doppler effect and is well known also from sound waves, cf. the sound of a passing ambulance.
Moreover, at a given stellar surface temperature, the "strengths" of these lines (a measure of how dark and broad they are) permit to measure directly the quantities present of the individual elements and
hence the chemical composition. Conversely, observed changes in the line-strengths of the spectra of certain peculiar types of stars indicate changes in the composition, or of the ambient temperature.
Until now, all observed temperature changes in stars have been much larger than those caused by solar-type oscillations as now observed in Eta Bootis, and of different nature.
(1) This is a joint Press Release of ESO and the Institute of Physics and Astronomy, Aarhus University, Denmark (IFA).
(2) The group consists of Hans Kjeldsen, Michael Viskum and Soren Frandsen (IFA), Jorgen Christensen-Dalsgaard (IFA; and Theoretical Astrophysics Center, Danish National Research Foundation), and TimBedding (ESO). Hans Kjeldsen was supported by a grant from the Carlsberg Foundation.
Technical information: Unfiltered 16 minutes gunn-r exposure with EMMI. North is up and East is to the left. 1 pixel = 0.265 arcsec; the field measures 51 x 38 arcsec. Observer: O. Hainaut (who also discovered 1994 TG2).