Spectral Analysis of Stars

[Sunshine]

Hello! I’m in the middle of a great book called “The Human Cosmos, Civilization and the Stars” one chapter is dedicated to the pioneering work of Joseph Von Fraunhofer who invented the science of spectral analysis. While reading about his findings when observing Sirius I began to wonder how one would single in on the light of Sirius itself. One would assume that Asside from our own stars light enveloping our solar system and whatever other background stars light is mixed in with the light of Sirius that it would be difficult to single out the particular spectrum from one particular star. Unless there is something I am missing, which I’m sure there is.

[saac]

I think in the same way as when you view Sirius at the eyepiece, you could argue that some of that light is background light from other sources. However this light, compared to the light from Sirius itself, is simply background noise. The main signal from Sirius swamps the background signal. So in a similar way, when the light from Sirius is collected by the telescope and directed through the spectroscope to produce a spectrum - any background signal is simply dwarfed by the main signal from Sirius. Sure,  that background signal will be there but too low level for the instrument response to resolve it.   The trick is to ensure that the star under observation is positioned such that the incident light falls on the spectroscope slit. The more professional setups doing spectroscopic analysis will use a fibre optic system to allow for a more precision approach to positioning and collecting the target starlight so avoiding other extraneous light.  

Jim 

Edited February 22, 2023 by saac

[andrew s]

Actually it is easier to remove the sky background with a slit spectrograph. 

You define an areas either side of the star spectrum and assume they average to the contamination in the star spectrum. By subtracting it from you star spectrum you remove the contamination very effectively even from urban sites.

With a fibre you can't do this.

Regards Andrew 

[saac]

26 minutes ago, andrew s said:

Actually it is easier to remove the sky background with a slit spectrograph. 

You define an areas either side of the star spectrum and assume they average to the contamination in the star spectrum. By subtracting it from you star spectrum you remove the contamination very effectively even from urban sites.

With a fibre you can't do this.

Regards Andrew 

Ah, that makes sense Andrew and quite a neat solution. 

Jim 

[robin_astro]

19 hours ago, Sunshine said:

One would assume that Asside from our own stars light enveloping our solar system and whatever other background stars light is mixed in with the light of Sirius that it would be difficult to single out the particular spectrum from one particular star. Unless there is something I am missing, which I’m sure there is.

A slit can be used to separate out the object of interest and the subtraction of the sky background  measured above and below the star can be subtracted but even if you just image the sky through a diffraction grating (or prism) each star produces its own spectrum against the sky background as here for example using my Star Analyser grating

https://britastro.org/observations/observation.php?id=20201216_234948_8cabda965bfe692f

and the sky background above and below the star can be subtracted in the same way

https://britastro.org/observations/observation.php?id=20210406_144443_9e1c6a4cf219d14d

Cheers

Robin

 

[robin_astro]

23 minutes ago, robin_astro said:

even if you just image the sky through a diffraction grating (or prism) each star produces its own spectrum against the sky background

 

The first stellar spectral classification systems were developed in the late 19th, early 20th century using spectra recorded using this technique by  the women of the  "Harvard Computer" team

https://en.wikipedia.org/wiki/Harvard_Computers

and you can see a modern example of an objective prism spectra here by Mike Harlow

https://britastro.org/observations/observation.php?id=20221003_080406_1603b0503cc5bcca

Cheers

Robin

Edited February 23, 2023 by robin_astro
clarified date