SETI bioastro: Re: [ASTRO] The Ages of Stars

From: Larry Klaes (
Date: Fri Apr 21 2000 - 13:39:57 PDT

From: Ka Chun Yu <kachun@CASA.COLORADO.EDU>
Subject: Re: [ASTRO] The Ages of Stars
To: (astrolist)
Date: Fri, 21 Apr 2000 14:01:36 -0600 (MDT)
X-Mailer: ELM [version 2.5 PL2]
Reply-To: Ka Chun Yu <kachun@CASA.COLORADO.EDU>

Amalendra Anandaraj wrote:

> > Although you can compare
> > a key set of lines for any particular spectral type of star to known
> > calibrated line strengths for an abundance, in the end, the way
> > these
> > are calibrated are via detailed stellar atmosphere models that take
> > into account abundances, temperatures, and gravities at the stellar
> > surface.
> How are these models tested or validated? I may be misreading the above,
> but it reads as if to 'calibrate' abundance, you need abundance as an
> input. In any case, there seems to be a large number of independent
> variables.

I am not an expert on this so my explanations probably could have
been a little better. Let me try again in more detail.

Basically if what spectral lines you see in a star depend on many
different factors, of which abundance is only one. Temperature is of
utmost importance since this determines the ionization level of any
particular atom. The reason why iron lines are so prevalent in the
Sun's photosphere is that the Sun's surface is cool enough that iron
tends to be in the form of ions with a few electrons stripped off.
Hydrogen has much weaker spectral lines since the single electron of
hydrogen is much more tightly bound than the loose outer electrons in
iron and other "metals" such as calcium and potassium. Thus although
these other elements make up a tiny fraction of the Sun's
composition, they contain lines that are the strongest and the most
easiest to see. Thus what lines you will see what heavily depend on
the temperature at the surface of your star.

When you take a spectrum of a star, you are integrating emission
originating from the photosphere but which subsequently passes
through the stellar atmosphere. You thus have to take into account
the interactions between the photons with each gradually cooler layer
of atmosphere. (This is also the reason why most stellar lines are
seen in absorption. The cooler layers above the photosphere tend to
absorb emission at each particular spectral line.)

There are thus lots of different variables that one has to account
for. However there are hundreds if not thousands of spectral lines
that one can use to compare the modelling with. Thus stellar
atmosphere codes have some of the best explanatory power in all of
astrophysics. People have been working on them for many decades, and
they not only explain the presence of the lines that you see, but
also how deep or bright these lines appear, the exact shape of the
line to as fine a precision as the highest resolution modern
spectrometers can measure, and so on and so forth. (The modelling of
line shape is even more arduous and intricate; here the effect of the
surface gravity of the star as well as bulk motions in the gas such as
rotation and stellar winds all play a role.)

Over the past century, the observers have built up an archive of
countless spectra of different stars, which the modellers then try to
explain. The theorists have been able to finetune their computer
codes over the past several decades by making comparisons with not
just a few elements and a handful of lines, but of all the observed
elements at all observed lines. The success of these codes have been
proven when theorists make predictions as to the existence of lines
that were not previously observable--say in the ultraviolet or the
infrared, because of the available technology at the time--but are then
found and confirmed.

> >For instance, Hydrogen emission and absorption lines from
> > the Sun are actually quite weak, whereas most of the absorption
> > lines
> > are dominated by metals like Iron and Calcium. Without detailed
> > codes to model the behavior of the emitting gases, a simple
> > interpretation would be that the Sun consists mostly of Iron and
> > Calcium.
> This seems to justify my doubts about the effects of "mixing."

I am not sure what you mean by "mixing". (I must have missed it in
one of the previous posts.) However since the light that one
observes is coming from the stellar surface and from the stellar
atmospheres, you are not observing elements that have been processed
in the core of the star. However we know that stars produce the
heavier "metals" from nuclear burning of hydrogen and its
by-products. Generally these tend to stay in the core (the only
place hot and dense enough for nuclear fusion to occur). However for
stars that have substantial convective currents, contamination by
heavier elements via upwelling can occur which will affect the
abundances at the stellar surface. There are models that are a
combination of stellar atmosphere radiative models and of stellar
interior models, so that the affect of interior convection is handled
in a consistent way. From a talk I heard several years ago, these
codes are not as advanced as the atmosphere-only or interior-only

--kachun +** Center for Astrophysics and Space Astronomy, CB 389 **+
          +** University of Colorado, Boulder, CO 80309 **+
          +** Email: **+
          +** **+

This archive was generated by hypermail 2b30 : Wed Mar 28 2001 - 16:07:53 PST