Posted: Jun 24, 2019 1:55 pm
by hackenslash
Blip wrote:I'm hoping for some help from one or more of the physicists here.

Some of you know that I'm doing a course on astrophysics with Oxford ContEd. I don't have a maths or physics background, but the course is designed to be accessible to a wide range of people, and so it is. Mostly.

However, I've come across something which I don't understand and with which our tutor's answer hasn't really helped me. Looking at results from Hubble, it seems that the most distant, i.e. oldest galaxies 'look strange – smaller, irregular, lacking clearly defined shapes.'

Nearer, i.e. more recent galactic views show more ordered galaxies; as that site explains '[c]loser in, we see numerous galaxy interactions and collisions as galaxies come together and merge, growing in the process. And nearer still, we see versions of the large, stately galaxies we know today. '

Help! How does this square with the entropy of the universe increasing?


Because gravity is an ordering force that increases entropy. The more ordered galaxies actually represent an increase in entropy, not a decrease.

This is the problem with equating entropy and disorder. Disorder is a good description of high entropy only in very specific circumstances, because it's a measure of the probability of a given configuration. Thus, because a disordered system is more probable than an ordered one, the disordered system is the one with higher entropy. This does not mean, however, that disorder and entropy are the same thing (this is another conflation of map with terrain, such as humans seem to be fond of).

Properly, entropy is the tendency of energy gradients downwards, and a system whose entropy is higher is one which is closer to its lowest energy state. Entropy is really a measure of how much energy in a system is unavailable for performing work or, in the terms I've presented here, how much energy is further down the gradient than it would need to be to perform work (where work is the equalisation of differentials).

More here:

Order! Order!

And especially here:

All Downhill From Here

As for the deck of cards thing, in reality, an ordered deck has only slightly lower entropy than a shuffled one. Such a system has greater entropy if the deck is scattered around the room. The reason for this is that, because it requires work to put the deck in order, entropy has been brought lower locally at the expense of an increase in entropy less locally, because energy has been expended in ordering it, and is thus unavailable for performing work.