The Slow Problem
Ladies and gentlemen, we have a small problem.
I say a ‘small’ problem – actually it’s arguably quite a large one. It’s called stellar evolution . You see, from the moment the original protostar ignited, our Sun has – in a sense – been dying. Now, before everyone looks too shocked, this isn’t actually news. Also, it’s not dying particularly fast. It’s been going for ~5 billion years, and it should run for another ~5 billion years. However…
…this is a bit of an extreme assertion, you may well say. Fair point. So, why do I assert that the Sun is dying?
Well, what I mean is that the Sun only has a finite supply of hydrogen available at its core. It shines by fusing this hydrogen into helium. However, the helium nucleus is slightly lighter than the sum of the masses of the four hydrogen atoms that get crushed together to make it. That difference of mass is converted into energy – this is where the Sun’s light and heat come from. (Odd little statistic: per kilo of mass, the Sun actually emits less power than your body. However, the Sun has vastly, vastly more kilograms of energy-emitting mass then you do, so it still wins.) Obviously, the problem with there being a finite supply of hydrogen is that eventually, it will run out. This is the ultimate fact behind red giant expansion, a red giant being the ultimate destination of all intermediate-mass stars .
The Sun won’t leave the Main Sequence for another 5,000,000,000 years or so. However, we will have a problem long before that. When the young Sun first lit up, it was something like only about ~67% as bright as it is today. This follows the general pattern for Main Sequence stars – as they age, they brighten. In 5*10^9 AD, the Sun will be about twice as bright as it is now. (And then, shortly after, it will get much, much, much brighter as it ascends to the red giant branch.)
This is a perfectly-normal piece of stellar evolution. However, it’s not good for the Earth . Doubling the Sun’s luminosity will obviously make things rather hot down here – too hot, in fact. On a rigorous analysis, it emerges that the Sun’s evolution will trigger a runaway greenhouse effect here on Earth somewhere between 500 million AD and 2.5 billion AD. (The specifics of when exactly depends on non-solar inputs like weather, plate tectonics and green plants. If the atmosphere is reasonably free of greenhouse gas, and the planet is unusually shiny, then it can hang on for longer. If it’s clogged with carbon dioxide and methane and the surface is darker, then it will go faster.) Some estimates have even claimed that problems will begin as early as 50m AD (although I saw that in a pop-sci article which didn’t quote any citations … so I don’t personally credit that one).
Anyway, precisely when is hair-splitting. The point is, it will happen sooner or later. The physics of nuclear burning demands it. And this means that, sooner or later, we’ll have to do something about it.
Yes, that is a very bold statement, isn’t it?
However, just for a laugh, I thought I’d get my optimist’s hat out and take it for a trip around the block. So let’s proceed on the assumption that we’ll still be here long enough into the future for this to be an issue. This isn’t necessarily that absurd – the crocodile has existed for 200 million years, supposedly, and has survived a string of extreme natural events in that time. And we’re a lot clever-er than crocodiles. (Just look at who ends up as the handbags – us or them? I rest my case.) Anyway, the point I’m making is not that I think we necessarily *will* survive into geological time, but that it is *possible* that we might.
So, imagine you’re in the government in the year 200 Million. The Sun’s getting brighter, things are getting uncomfortably-warm down on Earth, the Pacific Warm Pool‘s reached a rolling boil and, God forbid, the voters have started noticing… Clearly, you’re going to have to do something – but what?
The first approach would be the ‘medicate the symptoms’ approach.
You could put something on the Lagrange One point, between Earth and the Sun. A big disc, perhaps. A mirror, essentially, that blots out some of the sunlight. Reduce the input, reduce the heat. But, gradually, you’ll have to keep replacing the mirror with a larger and larger one. This could be somewhat inconvenient. Also, it will stop being very useful when the Sun finally expands into a red giant. In fact, on that day it will probably melt.
The second approach is to move the Earth.
This is actually less insane, and possibly cheaper, than it sounds. You see, the Earth is relatively easy to move if you don’t mind doing it slowly. Luckily for us, the Sun’s evolution is also a slow process, so we have a lot of time to work with.
There are, roughly speaking, two ways to move a planet. One involves adding momentum to it by dropping asteroids onto it – technically you can change its orbit this way, an impact at a time, but this is obviously a bit sub-optimal for the people living on the surface! The pub pool table looks a lot less fun when you’re living on the white ball, after all.
Don’t despair. There is a clever way to do this as well.
The clever way involves sending asteroids past, but not into, the Earth. This takes advantage of something we tend to forget. You see, asteroids have mass, and thus some gravity of their own. When an asteroid passes the Earth, the Earth pulls on it but the asteroid also pulls very slightly back on the Earth. So, send an asteroid whizzing past at high speed, and it will pull the Earth slightly after it. Basically, the asteroid gets a bit slower and the Earth gets a bit faster. And when it speeds up, it will move out from the Sun – just a consequence of basic orbital mechanics there.
Now, obviously, the change in momentum will be tiny in any one encounter – the Earth is a lot heavier then an asteroid, after all. However, if you keep doing it, then the cumulative effect can be surprising. (Remember – there’s no friction in space.) In fact, there was one estimate that I saw that suggested we could keep pace with the Sun’s gradual expansion over the next 5 billion years with as little as one planet-asteroid interaction per century. Now, I don’t have their figures to hand, so I can’t double-check the specifics, but the physics is reasonable.
But here’s the sting. We have a fixed timetable to work with. I have trouble imagining anyone influencing the Sun’s internal processes in any significant way, in any really-believable future. So, whether we’re still here or not, the Sun will leave the Main Sequence in 5,000,000,000 years’ time. But, the later that we leave the asteroid-flybys, the more of them and the more energetic they will have to be to work. And obviously, the higher the energies involved and the more that happen in any given unit of time, well, the greater the chances of an ‘accident’.
So it would actually make sense to take the long view, and start now…
 An important terminology note here. In astronomy, ‘evolution’ can refer to the change of a single object over time, rather than populations. This is quite distinct from biological evolution – the two processes should not be confused!
 Okay, strictly, the final end-state is a white dwarf, which is formed from the old stellar core, after the red giant envelope is shed into space. This ‘white’ dwarf will cool slowly over billions of years, and eventually will become barely warmer than the space around it. However, even the oldest ‘white’ dwarfs in our galaxy are still hotter than ~3700 degrees Celsius (so maybe more ‘pink’ dwarfs than white). We don’t see any fully-evolved white dwarfs. The universe just isn’t old enough yet…
 To forestall the obvious question, no! This process has nothing to do with anthropogenic climate change. The solar warming is not significant – or even measurable – on timescales as short as decades or centuries. And if you wish to question this, I can provide the maths in the comment thread…