Solar farms, energy yields and indirect land use change

In the past month, there has been a ripple of interest through the UK media about the loss of agricultural land to solar farms (large installations of photovoltaic cells), for instance this article from the Western Evening News. Sir Jim Paice, the Member of Parliament for South East Cambridgeshire and a former minister at the Department for the Environment, Food and Rural Affairs, has been reported as saying that, ‘“Solar farms are also leading to a loss of prime agricultural land, with the DECC aiming for solar power to supply about 15 per cent of green electricity needs by 2020.” The micro-furore has been largely triggered by a report by the Cambridge Institute for Sustainability Leadership (CISL). Now, this report doesn’t actually mention solar cells at all, but it seems like as good an excuse as any to talk briefly about a related issue that has been cropping up from time to time for a little while now – what is the indirect land use change (iLUC) implication of losing prime agricultural land to solar photovoltaics?

Now, the first thing to say is that, especially when looked at globally, there are extensive areas in the world that have limited agricultural potential that could make great sites for solar farming. Still, without a grand centralized land planning framework some solar developers are eschewing the Sahara for Devon, and turning ‘green fields blue’. An obvious question to anyone who’s been watching the debate on iLUC and biofuels is whether the iLUC impact of losing farmland will undermine the carbon savings from these installations, and therefore whether the UK Government (and indeed other countries) should act to protect farmland from the spread of renewable energy. 

In the analysis of the indirect effects of biofuels, estimates of iLUC emissions tend to vary by crop, and there is also variation between studies. However, it would be fair to say that most estimates fall within a range from about 10 to 100 gram of carbon dioxide equivalent released per megajoule of energy (gCO2e/MJ). Can we extrapolate from these numbers to get a sense of the likely magnitude of iLUC emissions from a solar farm? Well, according to the UK’s National Farmers’ Union (NFU), solar farms are typically installed at something like one megawatt of capacity for every two hectares. UK Government statistics from DECC suggest that every MW of installed capacity in the UK delivers about 70 kW of actual average power. At that rate, each ha of solar panels should deliver about one million megajoules of energy per year. Compare this to a biofuel crop like wheat, which delivers about 40,000 megajoules of energy per hectare. The solar farm is much more efficient in terms of energy per unit area – as you would expect, because it is a much more expensive exercise to install a field of solar panels than to sow a field of wheat. Biofuels do produce co-products, which increases the effective efficiency – but even if we make a fairly generous allowance and assume that the effective energy per hectare is doubled, a solar farm is still over 10 times more energy efficient per unit area than wheat for ethanol. This suggests that if biofuel iLUC sits somewhere in the range 10-100 gCO2e/MJ, then the iLUC from solar panels sits in the range from 1-10 gCO2e/MJ – which is to say that it’s not a very major term in the carbon footprint. Similarly, while loss of farmland to solar panels will have some impact on food prices, that impact will be only a fraction per unit of energy of the impact caused by biofuels.  

There is guidance from the UK Government and the NFU that farmers should aim to install solar cells on less productive fields by preference, and indeed that brownfield and other non-agricultural sites should be favored over agricultural land in general. As far as I’m concerned there’s no question that it makes sense to use less productive areas for solar panels wherever possible. Still, while there may be valid arguments to be made against deploying solar panels in some agricultural areas, iLUC emissions and impacts on food security are not really among them. And, of course, whatever amount of land is being turned over to solar PV, it is dwarfed by the land in the UK and elsewhere needed to produce crop-based biofuels (back in 2009, UK demand for biofuels required 1.4 million hectares of land according to the Joint Nature Conservation Committee). 

The NFU says it well: “If 10 GW of solar power were ground-mounted (half the national ambition for 2020 set by DECC), this would occupy at most 25,000 hectares – just 0.14% of total UK agricultural area (18 million ha) with a negligible impact on national food security.”