Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.1 6/24/83; site alice.UUCP
Path: utzoo!watmath!clyde!burl!ulysses!allegra!alice!td
From: td@alice.UUCP (Tom Duff)
Newsgroups: net.astro
Subject: Re: Measurement of Light Years and Distance
Message-ID: <3049@alice.UUCP>
Date: Fri, 19-Oct-84 11:24:02 EDT
Article-I.D.: alice.3049
Posted: Fri Oct 19 11:24:02 1984
Date-Received: Sun, 21-Oct-84 11:35:42 EDT
References: <872@ihuxp.UUCP>
Organization: AT&T Bell Laboratories, Murray Hill
Lines: 59

The problem of measuring astronomical distance always puzzled me as well.
It turns out that it's all VERY indirect.  The distance to nearby (less than
a few parsecs) objects can be measured by parallax.  Because there is a slight
perspective shift when observing nearby stars from opposite ends of the earths
orbit, we can effectively triangulate their positions by measuring their angular
shift relative to the more distant (and therefore relatively stationary) stars.
The use of parsecs as a unit of distance is an artifact of this means of
measurement.  One parsec is the altitude of an isosceles triangle whose
base is the diameter of the earth's orbit and whose opposite-the-base angle
is one second of arc.  This is the distance to a star whose stellar parallax is
one second.

Now that I think of it,  I'm not sure how the diameter of the earth's orbit
would be measured.  I guess you could use the parallax from simultaneous distant
observations of planets to fix their orbits and then use Kepler's laws to
compute the size of earth's orbit.  (But, how did Kepler do the measurements
to establish his laws in the first place?)

Anyway, the next step is to notice that the period of certain variable stars
(called Cepheid variables after the constellation in which the first one was
noticed) whose distance we can measure by parallax
is a function of their absolute brightness (i.e. the luminous flux at their
surface.)  Since we can measure the period and apparent brightness of Cepheid
variables, we can use the fact that apparent brightness decreases by an inverse square law to
compute the distance to many stars whose parallax is too small to measure.
In fact, distances to nearby galaxies can be measured by observation of
particularly bright Cepheids within them.  This is how it was first established
that the spiral nebulae were much farther away than any star in the galaxy,
and that there were in fact other galaxies.  (Imagine being the first person to
recognize that the universe isn't just thousands of light-years wide, but millions!)

The next step (due to Edwin Hubble?) is to notice that all the galaxies whose
distance we can measure seem to be moving away from us with a velocity that
varies directly with their distance (i.e. the universe is expanding.)  We
can measure their velocities by looking at the amount of Doppler shift of certain
characteristic lines of their spectra.  This is how the distances of the most
distant objects in the universe are measured.

As I said, this is an extremely indirect way to measure distance.  Each step
depends on lines of reasoning that may have flaws.  For example, the diminution
of stellar brightness might not be governed by an inverse square law.  If
space is filled, however tenuously, with small dust particles that absorb
light, the attenuation will have an exponential term that will certainly
dominate at stellar distances.
Furthermore, those Cepheids that seem to be in other galaxies may
not be related to them at all and may just but superimposed from our
point of view (but see below.)  Also, it was not known for many years that
there are two different kinds (or `populations') of Cepheid variables, which
obey different period/luminance laws.

The best `plausible reasoning' argument I know for all of this is as follows:
Stars and galaxies are not distributed uniformly across the night sky.
The seem to be organized into recursive clusters (i.e. clusters of clusters of
clusters of ...)  Unless this clustering is some sort of cosmic practical
joke on the human race, the measured distances to objects within these clusters
should also cluster.  And they do, so well that when astronomers notice an object
within a cluster has a measured distance very different from its neighbours,
it is taken as a sign that something very strange and worthy of detailed
examination is afoot.