I would call that a minor hurdle. There's actually no shortage of unused land. Or roofs. And agrivoltaics (combining solar with farming) is a thing. And wind turbines and farms are a common combination as well. And we have off shore wind, and floating offshore wind. Which you can combine with solar. Floating solar to limit evaporation in hydro basins is also a thing.
As an electrical engineer I have to sadly inform you that electricity transport is not free. Transporting electricity from those places to the places with populations is overall inefficient.
So the best place for a solar panel is right next to were the energy is needed (provided there is at least some sun).
Solar installations in deserts could still be a thing if you are willing to think it differently (e.g. use the energy for desalination and to split water into hydrogen and oxygen, transport that hydrogen via container ship etc).
So the solar panels will have dust and extremely harsh temperature cycles. I don't have any source on that, but I don't think the gained efficiency from the hot sun is going to outweigh the pain (and efficiency loss) from maintaining that efficiency in such an environment.
typically desert solar farms have higher capacity factors, like 25% to 29%, than non-desert solar farms, which are typically more like 20%, or 10% in very polar countries like the uk, germany, or the netherlands; possibly faster degradation will eventually reverse the relationship
but, for the time being, the gained efficiency from the hot sun does seem to outweigh the efficiency loss
as an ai language model, i cannot feel pain, so i do not know what outweighs it
Their data source suggests that rooftop solar, onshore and offshore wind and nuclear are basically tied on that metric so not sure what problem you're seeing?
Is it the ground mounted grid solar? I'm sure some countries will happily trade extra land use for much cheaper energy.
no because even though pv uses orders of magnitude more land per megawatt (not per megawatt hour; please get your units straight) it still uses two orders of magnitude less land than is available
please do the math instead of posting random bullshit without any regard to whether or not it is true
> A crude calculation (earthradius_equatorial^2 * pi * (1000 W/m^2) * 1 year in units(1) --- gosh, Unix is great!) suggests that the total solar energy falling on the earth is about 40000 * 10^20 joules per year.
... that's not a good approximation. There's someone who actually has done less crude math for maximum possible solar energy, at least for the UK: https://www.withouthotair.com/c6/page_38.shtml (though the HTML version is somewhat annoying because it's still paginated as if it were a book). Spoiler alert: it's roughly enough energy to cover total transportation energy demand, nowhere near total energy demand in the UK.
you would look much less foolish if you read more than just the introduction to my notes; i did a great deal more math than that
while i appreciate mackay's calculations a great deal, his estimates for very polar countries such as the uk are not generally applicable, and even for the uk are probably conditioned on overly pessimistic assumptions; he would undoubtedly agree if he were alive today
in particular, he assumed (reasonably) that 10%-efficient solar panels would be much cheaper than 20%-efficient ones, which would remain impractically expensive. but in fact most solar farms are being built with 21%-efficient panels because they're nearly as cheap as the 16%-efficient kind, and the 10%-efficient kind has been competed out of the market. so mackay's excellent calculations are all too low by more than a factor of 2, because one of his reasonable assumptions turned out to be wrong
but i do think it's plausible that without wind the uk would have to continue importing energy from abroad, as it has done since the 19th century, unless it goes nuclear. because mackay was aware his estimates for very polar countries such as the uk were not generally applicable, importing solar energy from abroad was in fact what he recommended in the chapter you linked but evidently didn't bother to read
> you would look much less foolish if you read more than just the introduction to my notes; i did a great deal more math than that
Perhaps, but I see nothing in your post that actually corrects for effective solar irradiation on Earth, or for the fact that half the Earth's surface is by definition not receiving any sunlight at any given time, or for the fact that most of the Earth is water and not land (although I suppose you dropping the 4 from the multiplier for the surface area is meant to account for that). In other words, at no point did I see anything that took into account the error I pointed out.
you didn't point out any errors; maybe you noticed one and forgot to mention it. please let me know if so
if you're interested in taking the capacity factor into account, which accounts for things like night, clouds, and oblique illumination, a number of my other notes in https://dercuano.github.io/topics/solar.html (linked from the bottom of my above-linked note) do that; for example in https://dercuano.github.io/notes/japan-energy-autarky.html i calculated that energy autarky for japan, if purely solar, would require 5% of its land area and about 1.7 trillion euros of solar modules, taking into account all of those factors as well as panel efficiency. since then, the price has dropped by more than a factor of 2, but the land area required remains about the same, or slightly increased
(floating that 5% of their land area on solar barges off the coast, instead of occupying precious land area, is also clearly feasible; it just isn't economically competitive, much like nuclear power)
of course, the real-life solution also involves wind and grid-scale storage
perhaps it goes without saying that very few places are as densely populated or as heavily industrialized as japan, so much smaller fractions of their land area would suffice
i suggest learning the basics of the field, so you can do a modicum of critical thinking, instead of parroting talking points from thought leaders, without any idea of what it would mean for them to be true or false
you can and should use 'per megawatt produced' rather than 'per megawatt nameplate capacity' when comparing things like solar to things like nuclear
but intermittency is irrelevant to basic incommensurability of units; neither nuclear nor solar uses more and more land over time to produce a constant amount of power, which is what 'land per megawatt hour' implies
https://ourworldindata.org/land-use-per-energy-source