Lecture 11. Structure of our Galaxy and other galaxies

What is the nature of the Universe in which we live - thats the big question. At the beginning of the 20th century it was a matter for intese debate. The standard picture was that the Universe consisted just of the stars in our own galaxy, but what were the 'spiral nebulae' ? - were they associated with our Galaxy or were they external star systems, separate Galaxies. As we saw in the last lecture, this was also the time at which the structure and size of our galaxy had been dramatically expanded by Shapley. The two questions (nature of the spiral nebulae and structure of our galaxy) got tied up together (if our galaxy was much larger than had previously been thought then perhaps the 'spiral nebulae' were within our own galaxy). This led to the so-called great debate between Shapley and Curtis with Shapley saying (correctly) that our galaxy was much larger than previously thought and (incorrectly) that the spiral nebulae were just bits within it while Curtis held (incorrectly) to a much smaller galaxy but argued (correctly) that the 'spiral nebulae' were external star systems - galaxies in their own right and so the universe consisted of not just our Galaxy. Scientific debate on current research topics is often like this - several groups with different ideas, each with a bit of the puzzle.

The questions could only be resolved by getting the size of our galaxy correctly and getting the distance to the 'spiral nebulae'. So how do we get distances ? trigonometic parallax is the most direct but very very hard to do, so we are left with spectoscopic parallax. Get the star's spectrum, and so get its spectral type, then the hertzprung-russel diagram tells us how intrinsically luminous it should be. We then use the inverse square law with its observed brightness to get distance (though remember to be careful about DUST). At the start of the 20th century a class of stars called Cepheids had been identified. These are evolved stars, and are highly luminous (10,000x more luminous than the sun) so can be seen at large distances. But their key property is that are unstable and pulse, and the period of pulsation gives a very good measure of the absolute luminosity. There is a nice javascript animation of a cepheid expanding and contracting. Again this is a spectroscopic distance measure, and again uses the apparent brightness and the inverse square law (ie assuming space is empty) to get its distance. Shapley was able to get distances to the globular clusters from Cepheids which was centred on a point 27,000 light years away from the Sun. More on location of the sun . Hubble in ~1925 finally resolved Cepheid variables in a few of the brightest 'spiral nebulae', and got a distance from the Cepheid period-luminosity relation which was much larger than anything proposed for the size of our own Galaxy - millions of light years away.

Knowing the distance, the diameters of the 'spiral nebulae' can be found from their apparent size on the sky - they turned out to be similar in size to our own galaxy - 100,000 light years across the stellar disk. Again, knowing the distance, we can use the apparent brightness of the galaxy to work out its intrinsic luminosity from the inverse square law. These turned out to be approximately 10¹¹x brighter than the luminosity of the sun. Thus if they are made up of stars like the sun then these galaxies contain approximately 10¹¹ stars!!! The 'spiral nebulae' are then separate star systems comparable in size to our own galaxy i.e. they are galaxies in their own right. The Universe consists not just of our own galaxy, the Milky Way, but of many many other galaxies too. See an introduction to galaxies.

Galaxies come in all sorts of shapes and sizes. We can classify these into several types, ellipticals, spirals (and barred spirals), and irregular galaxies. The nearest galaxy of similar size to the Milky Way is the Andromeda Galaxy - it is 2.2 million light years away, so the light we see from it left the stars in that galaxy 2.2 million years ago!! We are looking at it, not as it is now, but as it was 2.2 million years ago - we are looking back in time! And this is the NEAREST galaxy like our own.

We can currently (just!!) see Cepheids with the Hubble Space Telescope out to ~100 million light years (more details here - click on the sidebar 'individual images' to see the Cepheids). But what can we use after that ? We need a something where we know its absolute luminosity (called a 'standard candle'), so we can combine this with the apparent brightness via the inverse square law to get the distance. But we want this 'standard candle' to be much brighter than the Cepheids so we can use it out to larger distances. Using the galaxies where the distances are already known via Cepheids gives that the brightest globular clusters all have much the same absolute luminosity. Alternatively, the very brightest stars in each galaxy have much the same absolute luminosity. ASSUMING that the faroff galaxies are similar to the nearby ones, then we can use these to determine the distance.

But there are more faint galaxies which seem to be at still further distances. How can we get a good distance estimate to these ? We could assume that the Galaxy was like the Milky Way, and has the same absolute luminosity - this is a bit of a wild guess as we already know that galaxies come in all sorts of sizes. Ditto if we assumed that the diameter of the faint galaxies was the same as that for the Milky Way. We need something better to go further. If we take the spectra of galaxies as well then the amount of doppler broadening of the spectral lines tells us how fast the stars orbit in the galaxy, which is proportional to its mass and so to its luminosity. These can be calibrated separately to get the absolute luminosity for spiral galaxies (Tully-Fisher relation) and for elliptical galaxies (Faber-Jackson relation). More on using these to get distances to galaxies.

But theres something far more obvious in the spectra of galaxies than the breadth of the absorption lines and thats their position! In the 1920's it was already known that most galaxies had absorption lines which were doppler redshifted. Plotting the redshift versus the distance of the galaxy (found by Cepheids or any of the other methods above) gave a linear relation - the redshift that we measure in the spectrum is directly proportional to the distance. This is Hubbles Law. We can easily take a spectrum and then use the redshift to get the distance. The current record holder for the furthest galaxy whose distance can be determined by this technique is ~15.5 thousand million light years away!

But what does this all mean ? The Universe must be expanding. Is our galaxy at the center of this expansion ? At first glance it seems so since we see most of the galaxies rushing away from us ? But its not true. Suppose we can see a galaxy receeding from us - from that galaxy an observer would see US receeding from THEM with exactly the same velocity. If we were sitting on any galaxy we would see all the other galaxies rushing away from us. All galaxies would see the same Hubble law. But the Universe must be expanding! And if we run the expanion back then at some point the Universe must have been much smaller and denser than it is now!!!

So we've seen many ways to get distances. And a lot of them build on each other forming a distance ladder (1 parsec = 3 light years) out to measure the Universe.