Studying the universe.

[epicstruggle]

Im just curious, is it better* to study the universe from begining to today, or work backward from what we know today and work backwards to the begining?

epic

 

[Neeyik]

Not wishing to sound obtuse but what do you mean by "better" and by "study"? Are you talking about from the point of view as a student trying to gain a knowledge of modern cosmology or as somebody trying to build a model to explain observations of the universe?

 

[epicstruggle]

Not wishing to sound obtuse but what do you mean by "better" and by "study"? Are you talking about from the point of view as a student trying to gain a knowledge of modern cosmology or as somebody trying to build a model to explain observations of the universe?

From a scientists point of view, trying to model the universe's timeline, would you start from the beginning to today, or from what we know today and go backwards?

epic

 

[nutball]

If you're talking about simulations and models, generally speaking the idea is to construct a model assuming some physics and some initial conditions, run the model, then compare the results with current observations. If there's a difference between the model prediction and the observations it tells you that the model and/or the initial conditions are incorrect, so you modify them, re-run the model, redo the comparison to observation. Rinse and repeat until you can't distinguish between the model output and reality(*).

The problem with trying to go backwards in time is that you don't have any observational data with which to compare your models.

Note though that depending on what you're doing you don't always have to start from the Big Bang. If you're studying, eg. planet formation, you start from a point where the host star and proto-planetary disc have already formed. You don't break out your favourite cosmology code and start from first principles -- that's not feasible at this time and won't be for many decades to come.

(*) Even such a match doesn't tell you that your model + initial conditions are "true" or "correct". Basically it tells you that you need some more detailed observations!

 

[epicstruggle]

http://www.nature.com/news/2006/060619/full/060619-6.html

Hertog argues that the same must be true of the path through time that took the Universe into its current state. We must regard it as a sum over all possible histories.

Take it from the top

He and Hawking call their theory 'top-down' cosmology, because instead of looking for some fundamental set of initial physical laws under which our Universe unfolded, it starts 'at the top', with what we see today, and works backwards to see what the initial set of possibilities might have been. In effect, says Hertog, the present 'selects' the past.

In other words, some of these alternative histories have left their imprint behind. This is why Hertog and Hawking insist that their 'top-down' cosmology is testable. Hertog says that the theory predicts the pattern of the variations in intensity of microwave background radiation, the afterglow of the Big Bang now imprinted on the sky, which reveal fluctuations in the fireball of the nascent Universe. These variations are minute, but space-based detectors have measured them ever more accurately over the past several years.

 

[nutball]

Good luck to them, I hope it increases the sales of his book (Hawking that is).

 

[KimB]

Im just curious, is it better* to study the universe from begining to today, or work backward from what we know today and work backwards to the begining?

epic

Well, that really depends upon what you're attempting to do. If you're just looking at the expansion of the Universe, for instance, the equations that govern that are symmetric in time, so going from the present back or going from the past to the present are completely equivalent. So the usual thing becomes to go from the present back in time, so that you're only looking as far back as you have to.

With other phenomena, like the formation of structure (galaxies, galaxy clusters, etc.) in the universe, the evolution has a distinct arrow of time, and thus you basically have to start with some initial conditions and work forward.