I don't understand why people are screaming at each other to do the math, or referring to others who have hypothetically done the math, or accusing each other of lying, or claiming that solar panels are made out of
rare earth elements (!!), instead of just doing the math themselves here in the comments. It's almost as if they're more interested in being vicious to one another than in finding out the truth.
World marketed energy consumption is 18 terawatts. Terrestrial insolation is 127 petawatts (below the atmosphere. Also my earlier figure here of 41 was incorrect.) Common PV solar panels are 16–24% efficient, so the total PV resource is 20–30 petawatts, 1000–1500 times larger. Utility-scale PV capacity factors (relative to the nominal 1000 W/m² peak insolation) in polar countries like Germany and the Netherlands are around 10%, which is abysmal, but in California they average 29%, and in more equatorial countries with less clouds they're presumably higher (I'd be grateful for more data here but I've only found what look like trustworthy, transparent operational reports containing this data from the US, Germany, and the Netherlands).
Let's look at three cases here: a worst case, a best case, and a plausible case.
In the worst case, all our solar panels are in terrible places like Germany with a 10% capacity factor, and they're all cheap 16%-efficient panels, and also the world marketed energy consumption doubles to 36 terawatts. In that case they take up 2.3 million square kilometers, a circle of 850 km radius. This is about 13% of the area of Russia or 17% of the area of Siberia.
You have: double 18 TW / 10% / (1000 W/m^2) / 16%
You want: km^2
* 2250000
/ 4.4444444e-07
You have: double 18 TW / 10% / (1000 W/m^2) / 16%
You want: circlearea
846284.38 m
You have: double 18 TW / 10% / (1000 W/m^2) / 16%
You want: 17125191 km^2
* 0.1313854
/ 7.611196
In the best case, world marketed energy consumption goes down slightly to 16 terawatts, all the panels are using expensive 24%-efficient monocrystalline solar cells, and they're all located in places like the Atacama, the Gobi, and the Sahara, so there are no clouds, it virtually never rains, and the sun is nearly directly overhead, so their capacity factor is even better than California's. Say, 35%. In that case we only need 190 thousand km², a 246-km-radius circle. This is a little bigger than Tunisia or about 36% of the area of Yemen:
You have: 16 TW / 35% / (1000 W/m^2) / 24%
You want: km^2
* 190476.19
/ 5.25e-06
You have: 16 TW / 35% / (1000 W/m^2) / 24%
You want: circlearea
246232.52 m
You have: 16 TW / 35% / (1000 W/m^2) / 24%
You want: 527968 km^2
* 0.36077223
/ 2.771832
For the plausible case, let's figure on a nominal capacity factor of 25% (like Arizona), 20% growth in usage to 22 terawatts, and low-cost 16%-efficient panels, even though PERC is starting to see mass adoption. This is 550 thousand km², a circle of 420 km radius (groovy, mang!), about 80% of the area of Texas:
You have: 22 TW / 25% / (1000 W/m^2) / 16%
You want: km^2
* 550000
/ 1.8181818e-06
You have: 22 TW / 25% / (1000 W/m^2) / 16%
You want: circlearea
418414.19 m
You have: 22 TW / 25% / (1000 W/m^2) / 16%
You want: 695662 km^2
* 0.79061383
/ 1.26484
This 18 TW figure isn't just electricity; it includes human energy use in forms like jet fuel, diesel for trucks, bunker fuel, and oil for heating, but not firewood, livestock feed, or human food. And it's
world marketed energy consumption, not just the US: that Texas-sized area would be powering Australia, China, Brazil, and Madagascar too.
So, no, land availability is not a bottleneck on this.
I'd like to challenge you all collectively to step up your epistemological game a bit from the profoundly disappointing "I'm sorry but" level you seem to be stuck at. It's just not that hard. There's a lot of concrete, verifiable information out there, and it's easy to do the calculations. Then if you're wrong you'll change your mind to agree with the people who were right, and if you were right the other people who will wrong will change their minds to agree with you. Either way you won't have anything to fight about. Unless they're just looking for an excuse to be vicious.
You should be ashamed of yourselves. You do not become a hacker by attacking people and repeating talking points you don't understand. You become a hacker by figuring things out.
Calculemus.