cosmolog

ding dong merrily on high

Today was going to be the day I visited the telsecope but the sky is overcast so it'll have to wait.

The project to create the sound of the CMB (documented over the last few months) has jumped the rails. I've been thinking that my idea of using an IFFT on the CMB power spectrum was misconceived and various conversations today seem to confirm that. I never really understood what a power spectrum was and thought I could use it as a plot of frequency and amplitude but I now realise that though this is possible, a number of different sounds could give rise to the same power spectrum. So - it's necessary to look at the data in a different way, to look at it before its made into a power spectrum. Pedro has ideas about this which I don't fully understand yet.

Talking about this I learned of a past visitor to the department, David Whittle, who had worked on the same idea, although from the basis of actually understanding the problem in a much more rigourous way. I tracked him down on the www and emailed him.
His reply suggests that the way to proceed is indeed complicated and involves starting by simulating the early universe.
I look forward to hearing his results of which he says: "I now have sounds not just for the CMB as observed, but also the evolution of the sound spanning the first 1 million years of the life of the universe --- starting about 50 years after the Big Bang, up to recombination (about 380,000 years) and then on past recombination when the normal matter, once freed from being trapped by the radiation, is free to follow the structure of the dark matter."

I've finally got hold of a book I was shown back in October, "An Introduction to Modern Astrophysics", which weighs in at half a ton. Deep within lies a description of the subtle pulsation of some stars and their modes of oscillation. Again and again reading about such chiming stars, their ringing is likened to that of a bell. I'd like to hear what a sky full sound like all dinging away. Of course their oscillations are too low to hear even if our ears where to be in the right place (which is a ridiculous notion, the bodies to which they are attached would have long been reduced into a fiery heap of cinders). Transposed up into our audible range and simulated on a computer though they may be safely heard.

Posted by Jem Finer at March 17, 2004 4:33 PM