The debate over the relative levels of carbon emissions from the various energy generation options has been quite long and tortuous. It seems clear that fossil fuels are bad news, but there is uncertainly about the relative merits of nuclear and renewables – most see the carbon footprint of renewables as very low, but what about nuclear?
A recent study by researchers at Potsdam Institute for Climate Impact Research, Germany, published in Nature Energy, estimates the 2050 full lifecycle greenhouse gas emissions of a range of sources of electricity in a 2 °C scenario. It claims that the carbon footprints of solar, wind and also nuclear power are many times lower than coal or gas with carbon capture and storage (CCS). This remains true after accounting for emissions during manufacture, construction and fuel supply, i.e. the so-called embodied energy.
Wind, photovoltaics (PV) and nuclear all do well, in the range 3.5–12 g CO2 equivalent/kWh, with solar at a mid-range 6 g CO2 eq/kWh, and wind and nuclear at 4 g CO2 eq/kWh. Fossil CCS is high, nearing 100 g. There are still upstream emissions from mining and not all the plants’ CO2 can be captured. Biomass and hydro are also high, around 100 g, but “highly uncertain”. Bio-energy with carbon capture and storage (BECCS) has the advantage of negative emissions – see my next post – but with hydro in some locations there can be methane production from trapped biomass.
In terms of the percentage of the lifetime energy produced by each plant, the energy needed for building wind turbines comes out by far the lowest, PV next best, with Energy Return on Energy Invested ratios (in effect the inverse of embodied energy) of 44:1 and 26:1, respectively. Nuclear comes out with an EROEI of 20:1. That seems high (it’s usually put at 15:1 for pressurized water reactors (PWRs)), given the energy/materials intensive nature of the complex plants and, crucially, the energy used in uranium mining and processing. What’s more, although new extraction techniques may help reduce the energy used for fuel production initially, longer term it may well rise, and EROEIs fall, since lower grade ores, harder to get and process, will have to be used in future as uranium reserves deplete. Certainly, an earlier meta review of multiple studies by Ben Sovacool suggested that the figure for nuclear was much lower.
…is there any point in trying to live on a polluted, radioactive planet with the ecosystem undermined in other ways – and draconian political regimes reinforcing social inequalities in response?
However, in the Potsdam study, all the rest are very much worse, including fossil fuels with CCS, but also hydro – surprisingly given its long operational life. But then it does involve a lot of concrete. Biomass also comes out poorly, including BECCS, but it’s a low energy-density fuel and you need more energy to collect it and get less energy out per unit of input than from fossil fuels.
The Potsdam paper’s estimates of full lifecycle greenhouse gas emissions for the different sources of electricity assume a 2 °C world in 2050, when global electricity supplies have been largely decarbonized. So the indirect emissions due to the electricity used for manufacturing systems are lower than now, for example, for a solar cell fabrication factory or for some parts of nuclear fuel production, though not presumably for uranium strip mining, if that is still used – it will need a lot of diesel diggers and trucks, unless they are all battery or syngas powered by then.
Overall the report concludes: “The indirect greenhouse gas emissions induced by upscaling wind, solar and nuclear power are small compared with other emissions sources, and thus do not impede the transformation towards climate-friendly power supply.” That is certainly interesting. For example, in a review of the report Carbon Brief notes that some earlier studies had suggested the opposite – renewables would need a lot of energy to build, and more than nuclear.
Be warned though, this sort of analysis is tricky. There are lots of unknowns and assumptions about assessment boundaries. What’s more, although the Potsdam researchers say they have taken these into account in the ranges they offer, it’s hard looking to 2050. The technology is changing fast and market and regulatory pressures may shift priorities and valuations in unexpected ways. If by then we are running a basically low-carbon energy system, the main issues may be different – less to do with carbon, more to do with relative environmental and social impacts, land-use, water use and so on. For example, a recent report from University College London (UCL) has proposed new, wider metrics for sustainable development.
Clean energy is vital – but still not enough
There are several candidates beyond just eco-impacts and costs. For example, employment creation is potentially a positive socio-economic impact, which some see as part of the case for adopting sustainable energy. There are now national campaigns for “One Million Climate Jobs”, e.g. in South Africa. And the EU has adopted job creation as a key metric in its approach to development aid. So we are moving beyond just carbon counting.
That is implicit in many recent studies that include wider social and economic costs and benefits: carbon saving is only one factor. Some interventions may deal with both carbon and other issues, for example, the adoption of renewables will also reduce air pollution. Some may deal with carbon but introduce other problems – for example, with nuclear power, the risk of incidental and accidental radiation release. Nuclear radiation hazards are, of course, not included in the Potsdam greenhouse gas analysis, nor in the UK government’s carbon accounting approach. It’s not easy to do that, given the uncertainties involved – for example, we are still debating the human cost of Chernobyl with, worryingly, a new UN report raising the thyroid cancer rate significantly.
Even leaving that aside, given the range of issues and options, strategically, in some cases, there may be conflicts over which focus should have priority. As the UCL paper says, “there are complex trade-offs between the natural resource dependencies of energy, food and water systems, and environmental threats including biodiversity loss, climate change and localized air and water pollution”. It concluded: “These synergies and trade-offs will manifest differently in different settings, and the impacts for different social groups will need to be understood and accommodated. Considerations of rights, justice and equity must be integrated into the exploration of solutions for these complex energy dilemmas to ensure we leave no one behind.”
Those worried mainly about climate change may object that everything else must be secondary, arguing that all other issues are irrelevant if the climate system is seriously disrupted. But equally, is there any point in trying to live on a polluted, radioactive planet with the ecosystem undermined in other ways – and draconian political regimes reinforcing social inequalities in response? Hopefully, extremes like this can be avoided, but as UCL says, “decision-makers can no longer think in silos, and will need to find ways of widening participation, creating collective ownership and building consensus.”
The trade-off issues can clearly be complex and that shows up when we look at specific sets of options in relation to carbon reduction, as I will explore in the next few posts.