Why the world can't rely on renewable energy if we want to remain affluent

Can sun and wind provide base-load power? The answer is, of course they can! But that’s the wrong question. The right question is, can they provide enough and the answer to that question is, no they can’t.

There are several impressive studies and reports proving that the world could indeed run entirely on renewable energy sources.  As this is what everyone wishes to believe, it is not surprising that there has been almost no examination of the possible limits to renewable energy. For some years I have been attempting to clarify the situation and I believe there is a strong case that our world cannot run on renewable energy. 

The main problem for renewables is to do with the variability of the two major sources, sun and wind. For years Mark Diesendorf and many others have argued that this does not prevent renewables from providing all the energy that energy-intensive societies will demand. Following is a brief indication of the reasons for thinking that this conclusion is mistaken.

First, the obvious point that even on a sunny day PV panels can provide no energy for about 16 hours of that day. Similarly there are times when there is close to no wind blowing in your region, and these times can last for many days. Weather comes across in very large synoptic patterns and these can leave the entire continent of Europe under conditions of intense calm, cloud and cold for a week at a time. Lenzen’s review of renewable energy (Current State of Electricity Generating Technologies 2009) includes a plot for the whole of Germany showing hardly any wind input for several days in a row.

Germany's not in a good wind region but several studies show that the same problem applies to the U.K, probably the world’s best inhabited wind region.

Coppin and Davey (2003) make the same point for Australia, indicating that for 20 per cent of the time a wind system integrated across 1500 km from Adelaide to Brisbane would be operating at under 8 per cent of peak capacity. Mackay (2008) found that data from Ireland between October 2006 and February 2007 had a 15-day lull over the whole country. For five days output from wind turbines was 5 per cent of capacity and fell to 2 per cent on one day.

What’s more at these times of low renewable energy demand can peak. Most renewable energy enthusiasts make the mistake of discussing the issues only in terms of averages. What matters are: minima in available renewable sources (the solar radiation over a whole mid winter month for a particular year and place can be 40 per cent below the average level for that month and place, and lower than that on specific days (NASA, 2010); and maxima/peaks in demand. What matters even more is the fact that the two can coincide in time, for example, Victorian demand peaks in stable winter cold snaps. On these occasions you might need more than twice the generating capacity that would meet annual average demand, and you might be able to get none of it from wind and PV. That means that on these occasions you will have to meet most of demand from other sources.

All the renewable-optimistic reports I have read, including those by Stern (2006), The World Wide Fund for Nature (2010), Greenpeace, Zero Carbon Britain and Zero Carbon Australia make the same fundamental and fatal mistake. They fail to recognise the need for massive redundancy in generating capacity, caused by the fact that often one or more component systems will not be contributing much if anything. When the solar energy is low you will need enough wind or some other capacity to make up that deficiency. Stern for instance proposes wind will provide 8 per cent of annual demand. He then proceeds as if we will only have to build enough wind plant to generate 8 per cent of annual demand. This fails to recognise that there will be times when all that wind capacity is contributing almost nothing and will have to sit idle while PV or some other source fills the gap. Similarly there will be times when there is no sun and you will need to have enough windmills etc., to meet all the demand. So we might have to build enough wind capacity to meet 100 per cent of demand. When there is no sun, and we might also need to build enough solar capacity to meet 100 per cent of demand. When there is no wind, it means total system capital cost might be several times what we thought it would be. 

This exposes the common fallacy expressed as “...but the wind is always blowing somewhere.”  Sometimes there is hardly any wind anywhere you can tap, but more importantly if it is blowing strongly today in region ‘A’ and Stern is going to provide his wind quota from that region today, then he will have to build in that region enough capacity to provide it all. And what if tomorrow the wind is only blowing well in region ‘B’? Obviously we will need to build sufficient capacity to meet the wind quota in every region where the wind might be blowing well on a particular day. We will have to build far more windmills than would contribute that 8 per cent of total demand.

"We'll store it"

This problem of intermittency and redundancy would not exist if electricity could be stored in very large quantities. But this can’t be done and it is not foreseen. Pumping water up into high dams is the best option. Mackay (2008) shows that even in Britain where it rains a lot development of all possible sites couldn’t plug gaps in wind supply. Hydro electricity provides only about 15 per cent of world electricity, and 6 to 10 per cent of Australian electricity (i.e., only 2 per cent of all our energy), so it couldn’t meet anything like total demand when there is no wind or sun (even if all dams could be adapted to it and few can be because you need a low and a high storage space). Using electricity to compress air is viable, but you have to burn gas to heat the compressed air or efficiency is quite low and the availability of caverns is a problem. New batteries are being used to store wind energy, but at present only on a minute scale (30 MW compared with what would be needed, e.g., 96,000 MWh to get a solar power station through a four day cloudy period. Exetec is aiming for batteries costing $500/kWh, but that means storing for night time supply from a 1000 MWPV power station would cost you $8b, about four times as much as a coal-fired power station.

The options?

Lenzen’s review of renewable concludes that it is not possible for wind to contribute more than 20 to 25 per cent of electricity demand because problems caused by variability increase steeply after that point, setting integration difficulties. He suggests that a slightly higher figure for PV. But this is debateable. This means that wind and PV can at best supply 55 per cent of the 20 per cent of energy that takes the form of electricity. Where are we going to get the other 89 per cent?

Lets briefly consider the options.

Biomass In an era when land is being lost and a food crisis is developing, the world is very unlikely to find as much as a 1 billion hectare on which to plant biomass energy crops. The loss of habitat is the cause of the holocaust of extinctions we are now causing so we should be returning vast areas nature, not thinking about taking more. If that area was put into producing ethanol we would probably get 50 EJ which is around 5 per cent of the world energy demand figure we are heading for by 2050.

Geothermal Even the renewables-optimistic WWF Energy Report (2010) and the claims of Jacobson and Delucci (2011) assume geothermal can contribute about 4 per cent of world energy. Australia has much better hot dry rock heat resources than the rest of the world but it is anything but clear how effectively they can be tapped, if at all. How much energy will it take to bore holes 5 km deep through rock, fracture rock down there, pump water down and force it 500 metres across to the nearest rising hole? What will be the temperature and rate of flow of the water that comes up, and what generation efficiency will that enable? And what will be the dollar and energy costs of constructing very long transmission lines from the deserts where the hot rock is? The answers are not known yet. The only operating plant in Australia (not at the most promising location) achieves 6 per cent efficiency, one-sixth the value for a coal-fired power station. Early in 2010 the much-publicised Geodynamics venture abandoned its efforts, writing off $350 million. 

Solar thermal Here’s the back-of-envelope calculation. The world is heading towards needing 700 EJ/y of final (not primary) energy by 2050 (Moriarty and Honnery, What energy levels can the earth sustain, 2009). Let us assume a 33 per cent reduction in demand due to energy efficiency effort. My review of solar thermal systems found that in mid-winter both central receivers and Big Dishes could probably deliver at distance a continual flow of about 25W/m2 of collection area. Probably the best strategy. Big Dishes using ammonia for heat storage, might cost $600 per square metre in future. This means we’d need 1,980 million of them, the total cost would be $475 trillion, i.e., $19 trillion p.a. assuming a 25-year lifetime. If we assume world GDP will treble by 2050 this sum would be 13 times the present ratio of energy investment to GDP in developed countries. Note that other costs such as the transmission lines, thousands of kilometres from the deserts, have not been included. And we would still have a problem of intermittency; i.e., what to do when there is little or no sun on the solar thermal fields for days at a time...pay for huge excess heat tank storage capacity

Hydrogen How about using huge numbers of windmills, the cheapest renewable source, to produce and store hydrogen. The energy efficiencies of a) producing hydrogen from electricity, b) compressing, pumping and distributing it, and c) re-generating electricity via (very expensive) fuel cells are optimistic, meaning that for each kWh your windmills generate you end up with .22 kWh to use via this path. So a crude estimate is that to supply 89 per cent of that 700 EJ/y this way we would have to produce 2,8232 EJ/y, and we would need 179 million windmills each of 1.5 MW peak capacity (each producing on average .5 MW or 15.8 TJ/y and costing $3 million), at a total cost of $534 trillion, i.e., about the same cost as the solar thermal option. We would have to add the cost of the hydrogen production compression/liquifaction, distribution and huge storage capacity.

Whatever option you choose, you might have to multiply the total by 1.75 to pay the interest on the capital borrowed to build all that renewable plant. Finally, the cost of energy and materials are now rising fast and will be much higher than is assumed in the above exercise. All this has been a long-winded way of saying that we couldn’t possibly afford it.

By the way, if your goal was to provide to all people, probably 9+ billion by 2050 the energy per capita Australians are heading for, your target would be 5 times as great as the 700 EJ/y assumed in the above exercises. Do you still think the world can all live affluently on renewables?

What is the answer then?

The point is, there isn’t one. If the question is how can we provide the energy to keep going the energy-intensive, growth and market driven societies we have in rich countries today, let alone to enable the continuous and limitless pursuit of ever-increasingaffluent living standards, then the answer is that it cannot be done. For decades many of us have been trying to get the mainstream to grasp that this quest is suicidal. 

We Australians now have a productive land footprint that is ten times as big as would be possible for all people in 2050. It is precisely the mania for affluence and ever-greater levels of production, consumption and GDP that is causing all the big global problems, most obviously resource depletion, Third World deprivation, the greenhouse problem, the destruction of the environment, and international conflict.  Such a society cannot be fixed. For instance you cannot reform a growth-based society so that it can have a zero growth economy, let alone one producing at a small fraction of present levels. Sustainability is not achievable without scrapping and replacing several of the fundamental structures of this society.

For fifty years mainstream society has refused to face up to this case, and their delusion has been strongly reinforced by the unexamined faith that renewable energy can be substituted for carbon fuels and enable us all to go on pursuing affluence and growth. 

This has not been an argument against transition to renewable energy sources. It is an argument that they can’t run energy intensive societies. We must move to full dependence on renewable energy sources as soon as possible. We can all live well on them...but not in consumer-capitalist societies. For detail on the radically different path that must be taken, see The Simpler Way website http://ssis.arts.unsw.edu.au/tsw/ .