> Most our energy comes from fossil fuels. Even fertiliser for our food. So we can't just replace it all with solar power.

But we can with nuclear plants, and I'm thinking garden variety fission plants, not the always-30-years-from-now fusion ones. There are various areas where we use fossil fuels only because its more convenient - e.g. ground transportation. We can go electric with most of it. Fertilizers - I'm not sure if petroleum is a necessary chemical component or just energy delivery vector for the production process. Plastics are I think the only area where we can't easily sidestep fossil fuels. But with enough electricity we can still synthesize the compounds we need for those processes.

I have thought about this problem and came to the same conclusion - we need to move to nuclear. But then if most of our energy needs are from nuclear, now we're going to be back in the same situation in 50 years running out of uranium.

Or we could sit in top of our coal power plants slurping oil, wait for the magical thorium reactor and/or solar power to become another 10 times more economical.

Maybe not in 50 years, because it seems that we have a lot of fissile material available[0] that should last us for a long time - long enough to expand past this planet if we continue the growth curve, or at least to advance and deploy renewable solutions without a hurry. But yeah, with continued growth we would hit an energy ceiling in the future too. That ain't bad if reasonably managed - we could fix all the damage done to this planet and expand to the Solar System. But then again, humanity isn't exactly known for reasonable management.

[0] - http://www.withouthotair.com/c24/page_162.shtml

http://www.withouthotair.com/c24/page_162.shtml:

>> "A once-through one-gigawatt nuclear power station uses 162 tons per year of uranium"

>> "World total (conventional reserves in the ground) 4.7[million tons]"

https://en.wikipedia.org/wiki/World_energy_consumption

>> "The IEA estimates that, in 2013, total world energy consumption was 13,541 Mtoe , or 5.67 × 1020 joules, equal to an average power consumption of 18.0 terawatts.[3]"

One gigawatt nuclear plant uses 162 tons per year.

The world uses 18000 gigawatts per year. (40% oil, 10% coal, 15% natural gas, 35% other).

If all the fossil fuel-based energy were replaced with nuclear power, the nuclear power plants need to provide 18000 gigawatts times 65% = 11700 gigawatts.

11700 gigawatts times 162 tons of uranium is 1.9 million tons of uranium, per year.

The world's conventional reserves of 4.7 million tons will last for three years.

The 22 million tons in Phosphate deposits will last for another 10 years. (if and when methods to mine them economically are discovered)

Perhaps the seawater uranium will come in time, but as yet it is not an economical method of uranium extraction. The cost needs to be reduced by 5 to 10 times.

> Most our energy comes from fossil fuels. Even fertiliser for our food. So we can't just replace it all with solar power.

That doesn't follow at all. Why not?

There's plenty of sunlight in the areas with most of the people, and the areas with most of the farmland. We have storage solutions--they're still a little expensive, but they're certainly getting close.

If our current energy came from hydroelectric or geothermal could we replace it with solar? Your statement seems like a complete non sequitur.

This statement may be wrong but for different reasons - it boils down to just how much electricity vs. petroleum we need to use to make fertilizer. I don't know much about fertilizer manufacturing, but from a quick look at Wikipedia I think that for nitrogen ones we could conceivably replace natural gas with some other hydrogen source and run the whole Haber-Bosch stack off electricity.