It's not photosynthesis that's .00001% efficient or whatever, it's the whole process, because it's also fuelling the entire lifetime of those organisms
before they become fuel in the ground, and all those organisms that
don't have just the right conditions for becoming fuel in the ground, and all the fuel in the ground that
didn't stay there over those millions of years because it didn't have the right conditions.
And in any case, the problem we need to solve is not at all replicating the amount of energy stored by fuel in the ground. The problem we need to solve is generating the amount of energy we're actually using. And that has the advantage that we know that number quite precisely, instead of having to derive it based on ridiculously imprecise guesses.
And TFA makes a projection for required area using exactly this, and arrives at
> 300 TW of solar generation capacity, occupying about 5% of Earth’s land surface area, and split between roof top installations in cities and dedicated plants on nearby less developed land. For comparison, agriculture uses 18% of Earth’s land surface area, and largely uninhabited deserts are 33%.
And that completely ignores the additional power generation capacity of wind power.
