Can everything run on renewable energy? An outline of the negative case.

Ted Trainer.


The potential and limits to renewable energy are hotly debated, and far from settled. Many people take it for granted that it can meet all our energy needs.  However there are reasons for thinking that it can’t and an indication of the case as it now stands is given below.  (This is an update of previous attempts to understand and summarise the situation; for the detail see Trainer, 2014a.) 

The reviews find that in general wind is probably confined to providing 30% of electricity needed. Higher levels are possible but at escalating cost due to the problems set by integrating such an intermittent source into the supply system. Countries that achieve higher levels today, notably Denmark, have favourable conditions such as big neighbours capable of taking surpluses when the wind is strong, and providing energy when there is no wind.

Similarly, even on a very sunny day photovoltaic panels can deliver no energy for about 16 hours and this probably limits it to a 15%-20% contribution, no matter how low its cost falls.  If the proportion was to be higher than this most or all of the other sources would have to be idled during the day, and large amounts of PV electricity would have to be stored or dumped.  

So let’s proceed on the assumption that wind plus PV might contribute about 50% of the electricity needed (and vary the assumption later.) But that is only about 10% of all energy needed, because electricity makes up only 20% of rich world energy consumption. 

So where are we going to find the other about 50% of electricity? There are only two major possibilities, solar thermal and biomass. Hydroelectricity potential is not estimated to be capable of more than doubling, to provide around 5% of world energy.  The reviews generally agree that at the global level tidal power, waves and geothermal are not capable of contributing much compared with sun and wind.  

When we consider the amount of solar thermal plant that would be needed to meet demand in winter it is evident that the capital cost for a big enough system would be very high.  Here’s some of the arithmetic.  (See Trainer, 201.) NREL says 100 MW plant located at Blythe Riverside in the US would theoretically deliver an average 28 MW in the winter Direct Normal Irradiation of 5.2 W/m2, and would cost $658 million. That’s a capital cost of $23.5 per Watt generated. (The wind cost is $2,250/kWp.) If we subtract 20% of its output to pay for the energy it took to produce the plant plus the energy lost in power transmission from distant deserts, the cost of sufficient capacity to deliver one kWh in winter would be $29,375. If solar thermal was to provide 45% of the present approximately 2,215 million kW world electricity demand, the cost of the power stations would be 2,215 million x $29,375 = $65 trillion dollars, or $2+ trillion per year assuming a plant life of 30 years. But that is about 7 times what the world presently invests each year in total electricity generation, even though the sum does not include the perhaps additional 30% capacity that has to be built to cope with peak demand, the dollar and energy costs of operations and management, and the long distance transmission lines from deserts ... or the PV, wind, hydro and biomass components of the system. (The sum also assumes only 6 hour solar thermal storage.  The capital cost of the Gemasolar plant with 15 hour storage is about twice as high as it would be for the Blythe Riverside plant.)

But that calculation does not take into account the most awkward problem set by renewable energy sources. Because wind and sun are so variable at certain times many and maybe all sites are producing little energy.  Therefore proposals for 100% renewable supply have to assume a great deal of excess or redundant plant.  The best known proposals have to assume about four times as much capacity as would suffice in the form of coal-fired plant. (See for instance Hart and Jacobson 2011, Elliston, Diesendorf and MacGill, 2012, Budischak et al., 2012, Weisback et al., 2013.)

And even that is not sufficient because there will occasionally be some very difficult times when the amount of redundant plant proposed will still be insufficient and some kind of back up source will have to be used.  The Elliston, Diesendorf and MacGill proposal for Australia has to assume so much back up capacity in the form of biomass-gas-electricity generation as would be sufficient to meet the total 23 GW average demand.

Thus it can be seen that the capital cost of sufficient plant to maintain supply through gaps in the availability of wind and sun would probably be very high. Nevertheless it might be possible for countries with a lot of biomass potential, such as Australia, to provide all its electricity from wind and sun plus biomass backup. Europe is in a much worse situation, having very little biomass potential and winters with long, cold and stable periods, along with high energy demand.

But there are three big problems with the use of biomass. Firstly it can be argued that we should not try to use a lot of biomass energy.  We are now heading into a major holocaust of species loss mainly because we are taking so much of nature, especially the habitats that plants and animals need. We should be returning large areas to natural vegetation, not taking more land from nature to plant trees to harvest.

Even if we ignore that point, the recent IPCC Report (2014) says that global biomass potential (plantations plus wastes) might be at most 270 EJ/y, sufficient to produce only about one quarter of present world final energy consumption, e.g. 90 EJ/y as power or as liquid fuel.

If we assume that we can get that 270 EJ/y of biomass and put it into backup for wind and solar electricity generation plus liquid fuel production, these three sources would only account for about 45% of today’s world energy use. We would still have to explain how the remaining 55% of world final energy demand is going to come from renewable sources? 

Australia has greater biomass potential than most countries (possibly a 96 million tonne yield p.a.), and we could probably back up a solar and wind power supply system with it, but we would certainly not have enough to also make much of a dint in the liquid fuel for transport demand.

But perhaps the biggest drawback is that to use biomass would be to greatly worsen the greenhouse problem, because all that carbon would be released into the atmosphere and not be taken out again for many decades until the trees harvested had been regrown.

If electricity could be stored in very large volume renewable sources might be able to meet a large fraction of demand – but that is very difficult and costly. The best option is pumping water up into dams when there is a lot of wind or sun. But because there is extreme fluctuation in wind, and strong sunlight for only maybe six hours on an average good day, we would need a very large pumping capacity to harvest and store in those short periods when wind or solar surpluses are available.  A study of Ireland, possibly the best of all wind sites, found that to store enough wind energy to meet almost all Irish power demand they would need so many pumps that they would be using energy at six times the rate corresponding to total Irish electricity demand. (Connelly et al., 2012.)

Energy can be stored in compressed air, but it is not likely that enough sites can be found. Lithium batteries are a possibility, but global lithium resources seem to rule this out as capable of providing grid level storage for the world as a whole. Hydrogen is a viable storage option, but it is very energy inefficient and costly; to deliver one kWh to the car wheels requires 4 kWh to be generated.  It also, requires much expensive plant to produce, compress, pump and store the gas.

So it is not at all easy to see how the storage problem could be solved.

Conservation effort and technical advance will reduce the size of the energy supply task, but the demand problem will grow enormously and far outweigh their effect.  If 9 billion people were to rise to the per capita energy consumption people in rich countries have now, world supply would have to be 8 times as big as it is now. 

And be careful about assuming large demand reductions through conservation and efficiency effort. Big claims are made for electric vehicles and the German Passivhaus, but these refer only to running costs and when the energy cost of construction is included some studies find that these use more energy in their lifetimes than normal houses or petrol driven cars.


As I see it all this adds up to a strong case that it will not be possible for all people to live on renewable energy sources at anything like present rich world levels of energy use. The picture is not altered greatly if we double the above wind limit assumption.

Energy is only one of our many problems.  We have to face up to the fact that there are savage limits to growth, that we have gone through them and are now rapidly depleting the life-support systems of the planet, and that tech-fixes such as renewable energy cannot make this grossly unsustainable affluent-consumer-capitalist society sustainable.

Footprint and other measures show that the required reductions in present rich world per capita levels of production and consumption are enormous, probably to one-tenth.  This cannot be done unless we make a huge and radical transition to The Simpler Way (detailed in Trainer, 2014b. See also The Simplicity Institute.)

This has not been an argument against transition to renewable energy. We must make that transition as soon as possible, and we could live well on renewables in a Simpler Way, but not in anything like the present energy-intensive consumer-capitalist society.

Budischak, C., Sewell D., Thompson H., Mach, L., Vernon, D. E., Kempton W., (2012), “Cost minimised combinations of wind power, solar power and electrochemical storage, powering the grid up to 99.9% of the time”, Journal of Power Sources, 225, 60 – 74.

Connolly, D., H., Lund, B. V. Matheison, E. Pican and M. Leahy, 2012. The technical and economic implications of integrating fluctuating renewable energy using energy storage, Renewable Energy, 43, 47 – 60.

Elliston B., Diesendorf, M.,and I. MacGill, (2012). “Simulations of scenarios with 100% renewable electricity in the Australian National Electricity Market”, Energy Policy, 45, 606 – 613.

Hart, E. K., and M. Z. Jacobson, 2011.  A Monte Carlo approach to generator portfolio planning and carbon emissions assessments of systems with large penetrations of variable renewable, Renewable Energy, 36, 2278 – 2286.

Trainer, T., 2013. Limits to solar thermal energy set by intermittency and low DNI: Implications from meteorological data. Energy Policy, 63, 910 -917.

Trainer, T., 2014a, Can the world run on renewable energy? The negative case.   August 2014 update of an ongoing analysis.

Trainer, 2014b, The Alternative Society.

The Simplicity Institute,

Weisback, D., G. Ruprecht, A. Huke, K. Cserski, S. Gottlleib and A. Hussein, (2013), “Energy intensities,  EROIs and energy payback times of electricity generating power plants”, Energy, 52, 210 - 221.