PDA

View Full Version : how is the distance to stars calculated?



ern258
2001-Dec-17, 10:43 AM
I was just wondering, and if any of you could answer this for me that would be great.

2001-Dec-17, 10:59 AM
<pre>
Earth posit days 9 8 7 6 5 4 3 2 1 0
? : : ^
O e:ectrons gama 8 min
prOtons : g
O : g
O : g
O : g
O : g
this message sent 9 lines at a time 4:32 A.M. and YEST the PRE endings missing temporally

ToSeek
2001-Dec-17, 01:18 PM
On 2001-12-17 05:43, ern258 wrote:
I was just wondering, and if any of you could answer this for me that would be great.



It's an overlapping chain of different techniques, depending on how far out you want to go. Here's one set of explanations:

http://hyperphysics.phy-astr.gsu.edu/hbase/astro/distance.html

David Hall
2001-Dec-17, 03:32 PM
That's a pretty good site ToSeek, but a little sparse on the details.

I was wondering, Cephid Variables are considered standard candles, and I understand the concept, but I'd like to know just how dependable they are. Do all Cephids of the same period show exactly the same change in brightness, or is there some margin of variablility that needs to be factored in? How strong is the evidence that this relationship is true in all instances, other galaxies for example?

Also, what is the relationship of Cephids to the HR diagram (http://library.thinkquest.org/12713/noframes/hrdiagrm.html)? They look to be mostly giants in the F and G range. Does the period depend on the spectral class? What I mean is, would the periods of 2 Cephids be different because one is K3 and the other is G3 for example, or does some other mechanism control the period and the two can be the same?

And as long as I'm asking, what causes the variations in the first place? It's been a long time since I studied this, and I don't remember at all. Hope someone can answer. /phpBB/images/smiles/icon_smile.gif

_________________
David Hall
"Dave... my mind is going... I can feel it... I can feel it." (http://www.occn.zaq.ne.jp/cuaea503/whatnots/2001_feel_it.wav)

<font size=-1>(Ok, so this board doesn't let you set link targets to a new window.)</font>

<font size=-1>[ This Message was edited by: David Hall on 2001-12-17 10:42 ]</font>

lpetrich
2001-Dec-17, 08:32 PM
The mechanism of Cepheid variation is an ionization instability in the stars' interiors; this means that their interiors cannot be static. What one sees is a limit cycle; a stable state of oscillation. Its period is approximately related to the star's average density;

(oscillation rate) ~ sqrt(G*(density))

as can easily be worked out the form of the law of gravity. In turn, the density is related to the star's mass and size, and the size is related to the overall luminosity; these factors combine to yield a period-luminosity relationship.

Kaptain K
2001-Dec-17, 09:03 PM
If I might add.
The relationship is between luminosity and period. The range of luminosity is not part of the equation and varies considerably from star to star. In fact, it can change over time for a given star. Polaris is a Cephied variable whose variability used to be fairly noticeable. Currently, its luminosity is almost constant, but the period still shows up as a Doppler shift as it expands and contracts.

lpetrich
2001-Dec-17, 10:20 PM
The distance methods used for distant stars, star clusters, and galaxies depend on what one's trying to look at; they've traditionally been described as a ladder because their ranges of usefulness overlap, allowing distances to be inferred out to cosmological distances.

The most accurate is parallax; as the Earth orbits the Sun, a star makes a small mirror-image ellipse whose angular size is (size of Earth's orbit)/(distance to star) Even the biggest ones are very small; the first stellar parallax had been measured in 1831. And until recent decades, it was difficult to see very far with parallaxes; however, the Hipparcos astrometric satellite has significantly outdone ground-based measurements, and there are more astrometric satellites to come.

One long-used method has been the moving-cluster method; the Hyades cluster's stars are moving in nearly the same direction -- and their radial velocities can be measured spectroscopically. Putting these two pieces of the puzzle together yields a distance. However, this distance was found to be about 10% off when compared to the Hipparcos parallaxes, which suggests that other recalibrations may eventually have to be done.

Comparison of near and far star clusters allows for estimates of relative distance, with the near clusters' distance being measured by parallax or with the moving-cluster method. The farther ones may contain Cepheid variables, which are useful for seeing to much greater distances, such as to nearby galaxies.

And for galaxies, one can use star clusters such as globular clusters to estimate distance; these are naturally calibrated with the help of nearby ones.

Galaxies, like stars, are often in clusters, and one can find the distance of distant galaxies by checking on the brightness of their brightest members -- and comparing to the brightest members of nearby clusters whose distance can be determined by other means.

But there is one kind of star that is useful for finding distances across the size of the entire accessible Universe: the supernova, especially the Type I kind. However, these are very transient, so one has to continuously observe large numbers of galaxies in order to see them.

As is evident, the distances to more distant parts of the Universe can be difficult to measure; and since these distances are inputs to estimates of the Universe's expansion rate, that means that the Universe's age is difficult to measure.

However, coming astrometric satellites like FAME and SIM and GAIA will be able to find precise parallaxes for most of the visible parts of our Galaxy, thus enabling better calibration of Cepheid and bright-star distances -- and those distances that depend on these.

ern258
2001-Dec-18, 08:51 AM
what is the farthest we can accurately measure?

ToSeek
2001-Dec-18, 12:54 PM
On 2001-12-18 03:51, ern258 wrote:
what is the farthest we can accurately measure?


This depends a great deal on how you define "accurately." I'd confine that to measurements by stellar parallax, and the best results from Hipparcos (http://astro.estec.esa.nl/SA-general/Projects/Hipparcos/hipparcos.html) go out to about 1000 parsecs (3300 light years).

Mr. X
2001-Dec-19, 03:34 AM
Distance to stars is not measured. They observe a star, the astronomer takes out his calculator and uses the Randint() (Random integer) function to get a number between 0 and 10000000. The returned value is the number of lightyears it is away. /phpBB/images/smiles/icon_biggrin.gif

DStahl
2001-Dec-19, 07:16 AM
And when the RADINT function returns 0 for a red supergiant, we're all in trouble.

Mr. X
2001-Dec-19, 02:55 PM
Never happened up to now. /phpBB/images/smiles/icon_smile.gif

Let's hope it doesn't! /phpBB/images/smiles/icon_razz.gif

ljbrs
2001-Dec-28, 12:38 AM
Never happened up to now.

Let's hope it doesn't!


Mr. X:

Think of the magnificent astronomical and/or astrophysical experience!

ljbrs /phpBB/images/smiles/icon_biggrin.gif /phpBB/images/smiles/icon_smile.gif /phpBB/images/smiles/icon_wink.gif

ToSeek
2001-Dec-28, 01:33 PM
On 2001-12-27 19:38, ljbrs wrote:

Never happened up to now.

Let's hope it doesn't!


Mr. X:

Think of the magnificent astronomical and/or astrophysical experience!

ljbrs /phpBB/images/smiles/icon_biggrin.gif /phpBB/images/smiles/icon_smile.gif /phpBB/images/smiles/icon_wink.gif


However brief it might be. /phpBB/images/smiles/icon_wink.gif

Manchurian Taikonaut
2004-Mar-07, 08:46 AM
when they start to measure distances to things like quasar, then its almost like a guess, but still good.