The coninuum part of a galaxy's spectrum is the sum of the billions of stars' black body spectra:
The 400nm (4000 Angstom) break pointed out by the arrow is caused by (i) the absorption of small wavelength (high energy) photons by metals (non-H, He) in the atmospheres of stars and a lack of hot, blue stars. From what we know, the 400nm break should be weak in spiral galaxies (lots of luminous, hot blue stars) and strong in elliptical galaxies (very few blue stars). But we're getting ahead of ourselves...
Lets's look at some actual spectra of galaxies:
From top left to bottom right, here are the optical images of these galaxies:
NGC4889:
NGC 2775
NGC 6181
NGC 4449
The first two are elliptical galaxies, the third a spiral, and the fourth an irregular galaxy. Notice that the elliptical spectra have notcieable absorption lines and the spiral and irregular emission lines. Absorption lines indicate metals in stellar atmospheres or cold gas in the interstellar medium. Emission lines indicate hot gas, likely heated by hot, blue stars. We can conclude that ellipticals have very few hot, blue stars, leaving only an older population of stars. Spirals and irregulars, on the other hand, appear to have many young, blue stars.
Stay tuned...
Hmmm.... interesting.... I'm still a bit confused by how this data is collected. Wouldn't that make a big difference in how the spectra appear? Specifically, I'm wondering about the size of the area "sampled". The elliptical and irregular galaxies are so spread out that it seems to be that you'd get a lot more background noise in the sample, thus skewing your results... even possibly hiding the very thing you are tying to find.
ReplyDeleteThat's a very good point. There will often be background noise, so we have to be able to subtract that out. Background galaxies will give themselves away from their recession velocities, which will put their emission or absorption lines at different wavelengths (e.g. if you see sets of Calcium lines at four different wavelengths, then you're taking the spectra of four different galaxies). Their continuum spectra will also be different (background galaxies will be fainter), so putting all this information together allows you to seperate out all the spectra.
ReplyDeleteA second point to note is that the spectra above are integrated spectra - the spectra of all parts of the galaxy put together. This is what you automatically get if you are looking at distant galaxies which appear close to points of light (i.e. you can't resolve the different regions of the galaxy), but for nearby galaxies we can take the spectra of certain different regions of that galaxy (i.e. for Andromeda, we can take the spectrum of the bulge or the spiral arms).