Tony Onodi's Blog

Putting Solar Panels on Farmland With No Downsides

It's very common in the UK for applications for solar farms on agricultural land to be met with objections regarding food security. "I'm not against solar panels" opponents say, "but we should put them on rooftops, or car parks, or brownfield land". And we should. But would doing that provide enough energy for us to decarbonise?

The UK needs 1,650km2 of solar panels

The Climate Change Committee estimate that, alongside other power sources like wind and nuclear, the UK needs 75-90GW of solar capacity by 2050 in order to reach net zero (for context we currently have about 13.8GW). That works out to about 1,500-1,800km2 of solar panels. I've split that down the middle and gone with 1,650km2.

The squares below represent all of the area we need for solar panels. Each square represents 10km2, slightly counterintuitively you can think of this as a square with 3.16km sides, while the total area needed for solar panels is roughly equal to a square with 40km sides.

We can get 770km2 by putting solar panels on rooves (although this is very expensive)

An EU study from 2019 (before the UK left the EU) found that the UK could technically install 771km2 of solar panels on rooftops. Though it also found that only a small fraction of this would be economically viable. It doesn't give a number for the economically viable area but, for context, it found that though the maximum technical rooftop solar energy production was 43,646GWh/year only 6,517GWh/year of this was economically viable—around 15% of the technical potential, though I'd expect this number to go up as panel costs come down.

Regardless, I'm giving rooftop solar a fairer hearing than it deserves and including all of the potential, so our solar now looks like this:

We can get 260km2 by putting solar panels on brownfield sites

According to a report by the CPRE the UK had 21,566 brownfield sites covering 26,256 hectares in 2021. Let's round that to the nearest 1000ha and call it 263km2.

I strongly suspect that, as with rooftops, much of this land would not be economically viable for solar power, either because it's in a gloomy part of the country, or happens to sit in shade. It also seems like a really bad idea to use all of our brownfield land for solar farms but, for the purposes of this investigation, we're doing it anyway. Here's where we stand now:

Putting solar panels over every surface car park gives us another 200km2

It's pretty expensive to mount solar panels on top of a car park because you have to mount them onto a structure that's tall enough for cars to pass underneath. But you can do it. And that's what we're assuming we'll do here, according to a report commissioned by the Ministry for Housing the UK has 20,000 hectares of surface car park, so with the same caveats as everything else we've looked at, let's add that to our total:

Having now covered every car park, every building, and everything that could one day be a building with solar panels, we've pretty much exhausted our supply of surface area that isn't farmland or wilderness, and yet we're still 420km2 short of our 1,650km2 target. I think this is pretty conclusive. Even though I've been insanely generous, and included a lot of solar panels that would spend their lives sitting in shade, producing almost no power, we're still nowhere near where we need to be.

But the good news is, with one weird trick, we can find the land we need for solar panels and increase the amount of land available for growing food at the same time...

We can get the remaining 420km2 from land currently used for biocrops, and free up an additional 790km2 for farming food!

DEFRA says we use 121,000ha in the UK for growing biocrops. Of this 29,000ha is for biofuel, which is pretty bad; 75,000ha is for biogas, which is also pretty bad; and 10,000ha is biomass (burnt to generate heat or electricity), which... honestly is not something I know a lot about, but there's a pattern here so I have my suspicions.

By displacing them we can finally fill up our solar grid:

And look, free farmland!

It seems a bit absurd to say we can stop growing all of these crops entirely, but bioethanol, and biodiesel are carbon-negative, that is to say the quantity of fossil fuel they displace is smaller than the amount of fossil fuel that goes into sowing, fertilising, harvesting, transporting, and processing them. Photosynthesis is less than 1% efficient, for solar panels that number is 12-20%. Given that we've already set aside a lot of land to produce energy rather than food, it's insane that we've chosen to use a method that is less efficient by 1-2 orders of magnitude (for maize the efficiency number is only 0.25%!).

I haven't even mentioned that the ground is the cheapest place to put solar panels. But it is, by a lot, electricity from the cheapest ground-mount solar is 2.4x cheaper than the cheapest rooftop solar, and more than 5x cheaper than the cheapest residential rooftop solar!

What if we put all our solar on farmland?

Let's now look at the worst case for farmland. What if, for some reason, we decided we wanted to put all of our solar on agricultural land? Well, we have 172,000km2 of it, which means solar panels would take up 1% of that:

Since we can liberate more than half of the space we need from land currently used for biocrops—and unlike biocrops, solar panels don't care about soil quality—we should be able to improve our food security by putting solar panels primarily on lower quality land that's only suitable for grazing.

Overall it seems like it might be a good idea to restrict the number of solar farms we put on our best farmland, but not putting solar on any farmland simply isn't compatible with reducing our carbon emissions to zero. And replacing inefficient biocrops with much more efficient solar panels should leave us with more space to grow crops than we have right now.