B>First Observations of Solar-type Oscillations in a Star Very Different from
the Sun

Text with all links and the photo is available on the ESO Website at URL:


About 30 years ago, astronomers realised that the Sun resonates like a
giant musical instrument with well-defined periods (frequencies). It
forms a sort of large, spherical organ pipe. The energy that excites
these sound waves comes from the turbulent region just below the Sun’s
visible surface.

Observations of the solar sound waves (known as "helioseismology") have
resulted in enormous progress in the exploration of the interior of
the Sun, otherwise hidden from view. As is the case on Earth, seismic
techniques can be applied and the detailed interpretation of the
observed oscillation periods has provided quite accurate information
about the structure and motions inside the Sun, our central star.

It has now also become possible to apply this technique to some solar-type
stars. The first observations concerned the northern star eta Bootis (cf.
ESO PR 16/94). Last year, extensive and much more accurate observations
with the 1.2-m Swiss telescope at the ESO La Silla Observatory proved that
Alpha Centauri, a solar "twin", behaves very much like the Sun (cf. ESO
PR 15/01), and that some of the periods are quite similar to those in the

These new observational data were of a superb quality, and that study
marked a true break-through in the new research field of
"asteroseismology" (seismology of the stars) for solar-type stars. But
what about other types of stars, for instance those that are much larger
than the Sun?

Based on an extremely intensive observing project with the same telescope,
an international group of astronomers [1] has found that the giant star
xi Hya ("xi" is the small greek letter [2]; "Hya" is an abbreviation of
"Hydrae") behaves like a giant sub-ultra-bass instrument. This star is
located in the constellation Hydra (the Water-Monster) at a distance of
130 light-years, it has a radius about 10 times that of the Sun and its
luminosity is about 60 times larger.

The new observations demonstrate that xi Hya oscillates with several
periods of around 3 hours. xi Hya is now approaching the end of its
life — it is about to expand its outer envelope and to become a "red
giant star". It is quite different from stars like the Sun, which are
only halfway through their active life. xi Hya is considerably more
massive than any other star in which solar-like oscillations have so
far been detected.

This observational feat allows to study for the first time with seismic
techniques the interior of such a highly evolved star. It paves the
way for similar studies of different types of stars. A new chapter of
stellar astrophysics is now opening as asteroseismology establishes
itself as an ingenious method that is able to revolutionise our detailed
understanding of stellar interiors and the overall evolution of stars.

PR Photo 13a/02: Oscillation frequencies in the Giant Star xi Hya

PR Photo 13b/02: Non-radial oscillations of xi Hya (computer graphics)

PR Audio Clip 01/02: Listen to the sound of xi Hya (RealMedia and MP3)

The difficult art of asteroseismology

Helioseismology (seismology of the Sun) is based on measurements of the
changing radial velocity of the solar upper atmospheric layers (the
"surface") by means of the well-known Doppler effect, as this surface moves
up and down during acoustic oscillations. The corresponding amplitudes are
very small, with velocities of up to 15 – 20 cm/sec, and the typical period
is around 5 minutes. Therefore the phenomenon was first known as the
"five-minute oscillations".

Intensity measurements have also been tried, but the noise level is larger
than for velocity data due to the presence of "granulation" (moving cells
of hot gas) on the solar surface.

In the case of larger and brighter stars like the giant stars, the
corresponding amplitudes and periods increase. For instance, theoretical
predictions for the giant star xi Hya have indicated that velocity
amplitudes of about 7 m/sec and periods of the order of 3 – 4 hours could
be expected.

Observations of such oscillations are much more difficult, because the
demands on the performance of the spectrograph increase dramatically,
as this timescale is similar to that of variations of conditions in the
Earth’s atmosphere during the observing night.

Spurious instrumental effects, like mechanical flexure, would be detrimental
to such demanding observations. However, the experience from the search for
exoplanets orbiting other stars — by observing the periodic change in
velocity of the parent star due to the weak pull of the orbiting planet
over even longer timescales — has proven to be very useful. Indeed,
asteroseismology has benefitted greatly from the development of accurate
techniques now employed in the search for exoplanets.

The observations of the giant star xi Hya

An international team of astronomers [1] observed xi Hya with the Swiss
1.2-m Euler telescope at the ESO La Silla Observatory (Chile). They used
the CORALIE spectrograph, which is well known for numerous discoveries of
exoplanets (cf. PR 07/01), and recently for the detection of 7-min acoustic
oscillations in the solar-twin star Alpha Centauri A (cf. PR 15/01).

The same technique that delivered superb observations of Alpha Centauri A
was employed to investigate the oscillations of xi Hya. The sound waves
make the surface of the star oscillate periodically in and out, and the
CORALIE spectrograph measures the velocities of the up-down motion.

As xi Hya is a giant, these waves need more time to propagate through the
stellar interior up to the stellar surface than they do in a solar-like
star. Thus, the generated oscillations of the surface are slower.

An observing campaign lasting no less than one full month, taking about two
measurements every hour was necessary to detect the tiny movements of the
surface of xi Hya.

The detected oscillations have periods of about 3 hours, and have speeds of
only up to 2 metres per second. This is somewhat smaller than expected, but
the predictions for these amplitudes were very uncertain as the conditions
in xi Hya are so very different from those in the Sun.

First results for xi Hya

ESO PR Photo 13a/02

Caption: PR Photo 13a/02 shows the "frequency spectrum" of the giant star
xi Hya, as deduced on the basis of extensive velocity measurements with
the 1.2-m Leonhard Euler telescope at the ESO La Silla Observatory
(Chile). The abscissa unit is microHertz; 100 microHz corresponds to a
period of 10,000 seconds (2.78 hours).

PR Audio Clip 01/02

Listen to the sound of xi Hya! This 15-sec audio clip was produced by
mixing the 16 strongest frequencies in the observed sound spectrum (PR
Photo 13a/02) with the correct, relative amplitudes. In order to render
the signal audible, all frequencies were multiplied by a factor of one
million. Note that quality loudspeakers are required to fully appreciate
this rich and complex signal, especially the underlying bass tones.
Several beat frequencies are obviously present. Available in RealMedia
(requires RealPlayer software) and MP3 (264k) formats.

PR Photo 13a/02 shows the frequency spectrum of xi Hya, based on these
extensive observations. The "power peaks" indicate the frequencies of the
oscillation of the stellar atmosphere. The broad distribution means that
several different sound waves are clearly present. This is the first time
such a spectrum has ever been obtained for a giant star.

A first analysis showed the presence of about one dozen significant
frequencies and correspondingly, periods. Among those, four have amplitudes
above 1 metre per second. In addition to these twelve frequencies, others
appear to have been detected as well, but with less certainty and their
reality must be confirmed by a subsequent, more detailed study.

The "sound of xi Hya" has been synthesized in PR Audio Clip 01/02.

Stellar models

A good model of the star is necessary before the observed oscillation
frequencies (periods) can be properly interpreted. Current models of the
Sun are accurate and represent a typical main-sequence star at midlife,
and the oscillations are well understood. The sound spectrum corresponding
to the full disk — i.e., what we would observe if the Sun were as distant
as other stars and we would therefore see it as a light point in the sky —
shows a regular pattern in which the observed frequencies are separated
by two different and constant intervals, the "large" and the "small"

It is much more difficult to "model" the interior of a giant star as the
core has changed a lot during the evolution of the star. The nuclear fuel
has been exhausted, the stellar core has contracted and the envelope has
expanded substantially [3]. The resulting sound spectrum has therefore also
changed considerably. Now there is only a small group of oscillating modes
that display the same regular pattern as seen in the Sun. They are the
radial modes, pressure modes that correspond to a radial expansion and
contraction of the star (up and down motion of the surface).

The modes in the Sun are sound waves for which most of the oscillation
energy is concentrated in the outer parts of the Sun. In stars as highly
evolved as xi Hya, they partly take on the character of gravity modes in
the interior of the star.

Gravity modes are oscillations that move matter up and down in the gravity
field, under the influence of buoyancy, with only small changes of the
pressure. This is the same effect that makes an air-filled ball pop to the
surface when released under water. Gravity modes are normally trapped in
the stable interior inside the upper (convective) envelope of a star.

So far gravity modes have not been detected in the Sun. In a giant star,
however, there is a chance to see some, because some of the oscillations
have a mixed character: they behave like gravity modes in the interior
and like sound waves in the envelope.

The nature of the oscillations observed in xi Hya

ESO PR Photo 13b/02

Caption: PR Photo 13b/02 is a computer-generated illustration of one
possible non-radial oscillation mode in the giant star xi Hya. The blue
parts contain particles in the upper stellar atmosphere moving away
from the stellar centre, hence they cause a "blue-shift" (towards
shorter wavelengths) in the spectrum for the observer. At the same time,
particles in the red parts move towards the stellar centre and cause a
"red-shift" (towards longer wavelengths). Particles in the white regions
do not move during the oscillation cycle. Half an oscillation cycle
later, the red parts will have become blue and vice versa.

The high-resolution spectra of xi Hya were also used to determine improved
values of the fundamental parameters of this star: its temperature is
4950 +/- 100 K, the mass is 3.31 +/- 0.17 times that of the Sun, and the
age is 276 +/- 21 million years [3]. These values may be refined in a
subsequent, more extensive analysis.

With this improved model for xi Hya, the astronomers calculated the
frequencies of all oscillations likely to be observed. As in the Sun, the
radial modes are expected to be the dominating ones. In fact, three out of
the four modes actually observed in xi Hya coincide within the errors with
the predicted radial modes. The fourth mode seems not to be radial, but
agrees with a non-radial mode with 2 or 3 wave peaks and valleys over the
surface. PR Photo 13b/02 provides a graphical illustration of this in the
case of a star seen almost equator-on.

Some of the observed lower-amplitude modes must be mixed non-radial modes,
since more modes are detected than can be accounted for by the radial
modes of the models alone.

Future plans

Moving directly from stars of about one solar mass to the giant star xi Hya
is a rather great leap. With the CORALIE and HARPS instruments (the latter
soon to be installed on the ESO 3.6-m telescope at La Silla), an entire
sequence of stars at different evolutionary stages will be observed next:
from newly born to middle-aged stars like the Sun, and also old ones that
are near retirement.

The new observations of xi Hya show that this is now technically feasible.
Once more stars have been observed, changes in the interior structure and
composition can be followed and current theories of the internal stellar
structure can be verified and improved. Clearly, asteroseismology is bound
to have a major impact on the understanding of stellar evolution.

The detection of oscillations in the giant star xi Hya also has implications
for the target selection of several space missions aiming at seismic
measurements: the Canadian MOST mission, the French-led European COROT
mission (with launch expected in 2005), and some that are still under
consideration, as the Danish Rømer mission (now in the detailed design
phase) and the ESA Eddington mission. The present observations have proven
that these space missions will be able to observe oscillations in a wide
range of stars, and thus will constitute a major new source of detailed
information about the interior of stars, not accessible from the ground.

More information

The results described in this Press Release are about to be submitted to
the research journal Astronomy & Astrophysics (Letters) by the present


[1]: The team consists of Conny Aerts and Thomas Maas (Dept. of
Physics and Astronomy, Catholic University of Leuven, Belgium), Fabien
Carrier, Michel Burnet, Jose de Medeiros and Francois Bouchy (Geneva
Observatory, Switzerland), Soeren Frandsen, Dennis Stello, Hans
Kjeldsen, Teresa C. Teixeira, Frank Pijpers, Joergen
Christensen-Dalsgaard and Hans Bruntt (Dept. of Physics and
Astronomy, Aarhus University; and Theoretical Astrophysics Center,
Aarhus University, Denmark).

[2]: Some HTML-browsers support character entities for greek letters —
"xi" is then represented by "ξ".

[3]: In astrophysical terms, xi Hya is currently in the core-He-burning
phase, having left the main sequence some time ago and now near the
sub-giant/giant border.