A Sustainable Future? The limits to Renewables

by David Elliott, Professor of Technology Policy at the Open University

Although vast quantities of energy arrive daily from the sun, capturing it is always going to be limited by technical, ecological and land-use constraints. The most serious barrier to capturing it at present, however, is that other energy sources are artificially cheap.

Can renewable energy save the world from climate change by replacing fossil fuels? It is relatively easy to outline a series of 'technical fixes' for the climate change problem which would allow most of us to continue to live much as at present, at least for a while. Shell's 1995 scenario1 suggested that, in theory, renewables could be supplying possibly 50% of world energy by 2050 and, in 1993, the Stockholm Institute's scenario for Greenpeace suggested that, if we wanted to, we could have a world system based almost entirely on renewables by 2100, even assuming continued growth at 2% a year in energy use.2

Since these studies emerged, renewables have developed rapidly - for example, there is now 24,000MW of wind power in use around the world - and it has been argued by Amory Lovins that demand for energy can be dramatically reduced by clever 'Factor 4' and even 'Factor 10' energy efficiency measures3. So the prospects for a shift to a sustainable future are looking promising.

Indeed, there is something of an emerging consensus that, as the UN/World Energy Council 'World Energy Assessment' report, published in 2000, put it "there are no fundamental technological, economic or resource limits constraining the world from enjoying the benefits of both high levels of energy services and a better environment". A little more cautiously, the report adds "A prosperous, equitable and environmentally sustainable world is within our reach, but only if governments adopt new policies to encourage the delivery of energy services in cleaner and more efficient ways"4

However, the consensus is not complete. Although renewables are seen as playing a rapidly increasing role in this optimistic future, the strategy that is seen as being required also relies on continued use of fossil fuels, albeit more efficiently, and possibly also on the expanded use of nuclear power. Most environmentalists cannot countenance the latter option: they argue that, quite apart from the uncertain economics, why try to solve one problem (climate change) by creating another (radioactive pollution)? In addition, there is the possibility, argued forcefully by Colin Campbell and others, that the economically extractable reserves of oil and gas, may not be sufficient for their continued use on a large scale for very long. If that is so, we will have to move even faster to renewables.

Certainly, the WEA's fairly leisurely approach to replacing fossil fuels with renewables may not be adequate in the face of the climate change threat. We may not simply be able to wait for fossil fuels to run out (or rather to become prohibitively expensive). Sheikh Yamani is alleged to be the original source of the now familiar view that 'just as the Stone Age didn't end because people ran out of stones, the Oil Age won't end because we run out of oil.'

So there are plenty of reasons why we should consider moving rapidly to a more sustainable approach, based on the use of renewable resources and the adoption of more efficient ways of using energy. That will not be easy. The development of the new green energy systems involves many technical challenges, and many believe that it will in practice be difficult to actually achieve major energy efficiency gains. In addition, there are many strategic and political battles to be won - for example, that to obtain the necessary funding. However, in this paper I will try to explore the basic resource problems that face this approach. For not everyone believes that there will be sufficient renewables energy resources to meet growing demand, especially as the developing countries industrialise.


My first question is - how much renewable energy will be available? Looking a long way ahead is obviously difficult. But some broad patterns are clear. The chart below produced by energy analyst Gustav Grob shows the relatively short period during which industrialisation occured, based on fossil fuel5. It is followed, after the projected demise of fossil fuels, by continued and accelerated expansion of energy use, based on renewables, up to about twice the current level of energy use.
From an historical perspective, the use of non-renewable energy sources appears as a brief spike (the dark area above).
Will renewable energy sources be developed to take over as fully as the chart shows?If true, that is good news. That period of expansion could allow the developing world an opportunity to catch up with the industrial countries, although of course, alternatively, it could allow the industrial world to continue to expand ahead of the rest.

But, either way, subsequently, according to this chart, growth can continue but not at such as rapid rate. Technical, ecological and land-use limits impose what Grob calls a 'natural limit' on the amount of additional energy we can obtain from renewable sources, although we can raise this limit as we develop better renewable energy technologies and learn how to use natural energy flows more efficiently. Estimates vary as to what the ultimate limit actually is. Some, like the Australian ecologist Ted Trainer, put it much lower than Grob6; others, mainly the technophiles enthused by the potential of renewables, put it much higher - maybe ten times or even more.

For example, since some solar PV cells can convert sunlight to electricity at 15% efficiency, compared to the 1% efficiency of photosynthesis, then, given the huge solar input to earth, there are potentially very large amounts of extra energy available. In reality, for good or ill, the amount of energy that can in practice be obtained from natural renewable sources like solar may not be as large as these figures imply. While the amount of solar energy falling on the earth is very large (around 90,000TW equivalent), given the limitations of geographical access, only about 1000 TW is in any way actually available to us to use7. That is around 70 - 80 times current global energy generation (13TW). However, in practice there are technical limits on how much of this can actually be converted into useful energy. This is due to constraints on the efficiency of conversion and the diffuse and intermittent nature of much of this resource, as well as land access limits.


Well before the use of renewables begins to expand to the energy limit, there will be landuse conflicts and, in particular, conflicts between the natural ecosystem and the emergent human-managed ecosystem. It has always been that way, ever since we started farming. As we have spread our influence across the planet this issue has become crucial to the survival of the planet - indeed many 'deep greens' say it may already be too late.

At the same time, if Grob's estimate is anywhere near right, there is not an indefinite amount of room for economic expansion available ahead, so surely, within this more limited arena, it must be possible to have some sort of co-evolutionary balance between the human system and the natural system. Clearly they must not conflict: they must be part of the same overall system.

We are already fighting out some of these issues in terms of the debate over the location of wind farms and similar issues could emerge over other renewable energy options such as the growing of energy crops. The key point is that renewable energy sources are mostly diffuse and the energy collection system must therefore cover large areas.

Some are however worse than others. For example wind turbines cover relatively small areas and wind farms can produce up to twenty times more energy per hectare than energy crop plantations of short rotation coppice. Growing oil seed rape and the like for liquid biofuels is even worse in terms of energy per hectare - by a factor of perhaps ten. However energy crops have the advantage that they can be stored, although so far they look like generating electricity at much higher prices than wind8.

The opposition to wind projects in the UK is quite serious. It has meant that around 70% of project proposals have been blocked in recent years, so that the UK is falling behind in its attempt to obtain 10% of its electricity from renewables by 20109. Opposition is also mounting to large hydro around the world not just on the grounds of the social dislocation resulting when large areas are flooded for reservoirs, but also since it now seems that some hydro projects in warm climates can generate methane gas which makes a major contribution to climate change10.

Most of the other renewables are seen as more benign and as having less land-use implications. For example most PV solar modules would be located of rooftops and on walls and so have no land take implications. Offshore wind, wave and tidal stream system obviously have no land use implications.


It is sometimes argued that the main limitation to renewables will be that more energy will be needed to construct the equipment than it will produce over its lifetime. Fortunately this is a fallacy - a misreading of arguments about the 'embedded energy' debt. As it happens, the embedded energy costs associated with renewables are mostly low and usually less than for other energy technologies.

Thus, a review of energy payback times by Hydro Quebec has indicated that, over their full lifetime, typically, wind turbines generate around 39 times more power than is used in their construction and operation. For comparison, nuclear power plants are estimated to only generate around 16 times the energy needed for construction and operation, including the provision of fuel (which of course wind turbines get free). Combined Cycle Gas Turbines are even worse, only generating fourteen times the energy needed for their construction and operation. 11

Source: Luc Gagnon, Hydro Quebec, April 2000

It is true that some renewable options are less attractive in this sense but even PV solar, the most energy intensive renewable energy technology, still manages to generate 9 times more energy than is needed for cell fabrication, and that is using current types of cells. The newer PV technology now emerging is far less energy intensive.

Large hydro, whatever other problems it may have, is about the best deal, generating, according to the same study, around 200 times more energy than is consumed in construction - presumably because of the large capacity of the plants and their very long lifetimes (perhaps 100 years or more before major equipment replacement is needed).

Interestingly, however, energy crops do not come out very well on this analysis, presumably due to the high requirement for mechanised energy for planting, harvesting and in particular transportation of the bulky fuel to power plants. Biomass plantations are estimated to only return five times the energy needed to grow and collect them. As noted above, liquid biofuels have even lower energy output to input ratios than solid biofuels.

However, the use of forestry residues seen as much better, yielding 27 times the energy needed to collect them (growing is presumably seen as free).

Of course these energy calculations need to be set in the context of the value of the energy produced. Electricity from energy crops would replace electricity produced from fossil sources that produce carbon dioxide gas. As long as the rate of energy crop planting balances the rate of harvesting, the overall process can be roughly neutral in carbon dioxide terms since plants absorb carbon dioxide while growing. If land is not scarce then electricity generation from energy crops could therefore be seen as valuable in environmental terms. Of course, in a market system, there are also other types of value. For good or ill, fuel for vehicles commands a high price at present, so it may be that liquid biofuels will be the preferred energy crop despite the high energy input to output ratios. That certainly seem to have been the case in continental Europe.


Rather than embedded energy being a constraint, the main limit to the rapid expansion of the renewables could be financial. Most energy technologies are capital intensive to some degree. Quite apart from the reticence of individual investors and companies to back new technologies like renewables, there may not be sufficient financial resources available overall to permit the expansion, much less the renewal, of even the present conventional types of energy system.

In part, of course, this is due to the high embedded energy content of energy technologies, but there are also other key elements - especially in the more advanced conventional energy systems which use expensive high technology, large amounts of rare materials, and highlyskilled and expensive construction personnel.

In addition to being complex, most conventional projects are also physically large and take many years to plan and build. This adds to the cost of borrowing money. The result is that, to put it simply, energy technology is expensive. Worse still, it seems to have become more expensive over the years.

In his celebrated 1976 book The Poverty of Power, Barry Commoner argued that in its drive to increase the rate of profit, capitalism relied on ever more capital-intensive forms of production and energy production was no exception12. Indeed it was one of the most capital intensive. However the gains made in each successive wave of investment were falling - or, rather, the cost of each productivity gain was growing faster than fresh capital (that is, resources) could be created. Basically, he argued, the capitalist system, which has to keep improving productivity and expanding to survive, was running out of the resources it needed to do so. Commoner's quasi-marxist model of endemic economic crisis may be less fashionable these days, in part because the capitalism has learnt how to increase productivity with technologies that permit lower levels of resource (that is, capital) use.

In the energy sector, the end point of the old resource-intensive model was nuclear power, with reactors costing up to $3000/kW - three times as much as coal plants. By contrast, modern combined cycle gas turbines (CCGTs) can be installed at around $500/kW. It interesting in this context that, in its current attempt to get back in the game, the nuclear industry is trying to develop new plants with the target of getting capital costs down to $1000/kw. That seems some way off, with, for example, the much hyped South African pebble bed modular reactor being perhaps ten years away. Its costs are still very speculative, too. By contrast, wind projects are now being installed at $750/kW and $500/kW is seen as likely soon13.

Even so, given that most of the world's power plants will have to be replaced over the next few decades because they are reaching the end of their lives, there could be shortages of the resources needed to do so. This problem is clearly worsened by the huge expansion in energy demand both in the industrial countries and in the developing ones. There may simply not be enough financial resources to permit this expansion, whatever type of technology is used.

Some renewable energy technologies are less resources-intensive and thus cheaper than conventional technology. Wind power for example is now marginally competitive with CCGT's in some contexts. But most renewables are more expensive, at least for the moment. So their widespread adoption may be difficult - unless companies are willing to choose green options for their longer-term environmental (and commercial) benefits and consumers are willing to pay more for green power (for their, or their 'descendents', longer-term welfare).

Clearly it is unfair that clean green energy technologies have to compete with dirty fossil fuel based systems, but for the moment, in the absence of a system reflecting the environmental costs in the price we pay, the playing field is far from level. This limitation is however not like the ones I have discussed before - it's a construct of our society and its economic basis, and as such, it can be changed.


One of the reasons why renewables sometimes look expensive is because of the way we value energy. The conventional system is based on large centralised power plants that are usually some distance away from the power user. They are seen as providing instant power reliably at a flick of a switch. Their environmental costs are assumed to fall somewhere else.

Most renewables by contrast have very low environmental costs but are intermittent, offering only variable sources of energy. Some renewables, like hydro and biomass, are reliable, although weather dependent to some extent, but energy is only available erratically from the sun, winds and waves, and that from the tides depends on the lunar cycle. Let us take these issues one at a time. First, the environmental costs issue. There are of course various ways of reflecting the environmental costs and benefits of energy technologies in prices - by adding a surcharge, by some sort of energy and carbon tax, or more generally by subsidising options seen as desirable. This is not the place to explore all the ramifications of green pricing. But, as a striking example, on the basis of the figures produced by the EU EXTERNE study on environmental impacts, renewables could be condoned on environmental grounds even if they cost twice the price of conventional power.

Extra cost resulting from environmental damage (to be added to conventional electricity cost - assumed as 0.04 euro/kWh average across the EU) in Euro cents/kWh:
Coal 5.7
Gas 1.6
Biomass 1.6
PV solar 0.6
Hydro 0.4
Nuclear 0.4
Wind 0.1
Source: The ExternE 'Externalities of Energy' report, 2001, European Commission ExternE Programme, DG12, L-2920 Luxembourg.

As can be seen, the extra environmental cost associated with the use of wind is miniscule compared with coal or even gas, and four times less than nuclear. Given that there are actually wind projects going ahead at below 2.5p/kWh (less than 4 euro cents/kWh), then clearly there is something wrong with our current way of valuing these options.

Part of the problem is the second issue mentioned above, the belief that renewables are unreliable, due to the intermittency of the energy sources. In fact this intermittency is not too much of a technical problem. If renewables only supply up to around 20% of the total electricity generated in a country and their power is fed into the national grid, then the local variations in renewable availability are balanced out. However, for larger proportions of renewables we would need some way to store the energy. But by the time we have reached that point it should be possible to use hydrogen as a storage medium- generated from renewable sources by the electrolysis of water and then transmitted along gas pipelines, perhaps initially mixed in with natural gas, to the point of use. Hydrogen can be burnt as a heating fuel or in a power station to generate electricity, with no emissions except water, and can also be used to power fuel cells to generate electricity.

These, and other generation systems like small gas-fired combined heat and power (CHP) units, are small enough to be used to supply power to individual homes. The same is true for photovoltiac solar cells although they are still very expensive.

However, we can look forward to an energy system which has a range of sizes of generation plant, some quite large (e.g wave, offshore wind) some small enough to be in individual homes (PV solar, fuel cells, micro-CHP units). These micro power systems would all be linked via the electricity grid which would help to balance out local variations in energy production.

They would be backed up by power from nonvariable renewable sources like energy crops and by gas supplies generated increasingly from renewables too14.

The end result would be a robust decentralised energy system with the advantage that, on average, much of the power would be generated locally, from local sources, with only excess being exported via the grid and imports only being required to meet occasional shortfalls. Surprisingly, such a system could also supply cities as well as rural areas since most cities have sufficient roof space to provide for most of their power needs, averaged out, via PV solar, with this energy being backed up by waste digestors and pyrolysis units, converting the cities' wastes into energy. (waste is one thing in which cities are self-sufficient!).

Such a system would avoid the large energy losses incurred by shunting large amounts of power over very long distances via the grid, as happens with the present system. Currently, the advantage offered by generators embedded in local power systems is not recognised in the way we value power. Indeed, small local generators are often penalised as offering only small amounts of unreliable power. If renewables are to expand rapidly and replace fossil fuels, we need a new approach to the economic evaluation of distributed and dispersed renewable energy sources.

The precise mix and size of technologies will depend on the context. In some areas of the world off-grid generation from renewables makes sense. Indeed, for most of the 2 billion or so people who currently do not have access to electricity, it is likely to be their only option. PV solar is the obvious option, along with direct solar heating and cooling, and modern biomass technologies. Micro hydro also has a lot of potential. But there are also locations where larger grid-linked options make sense. For example India and China both have ambitious wind programmes. However, plans for large, environmentally invasive hydro projects could be replaced by large tidal current projects, like the 2.2 GW tidal fence being considered as part of a causeway between a series of islands in the Philippines.
Tidal Power in the Philippines: Rapid tidal currents between islands could be used to generate large amounts of electricity with tidal turbines installed along causeways. The UK has developed some pioneering designs for free-standing tidal current turbines like the ones shown.

In most of the industrialised countries, wind, on and offshore, looks like being the largest single option, with offshore wave and tidal power being the next largest for those with access to this resource (for example they could each supply 20% of UK electricity). Wave and tidal are still relatively expensive but as the technology develops, prices should drop as is already happening for solar PV. Energy crops remain uncertain economically as I have described, and, although the resource is vast in many parts of the world, it requires land and a good water supply.

Direct solar heating has many attractions even in the cloudy north and heat supplying options could begin to make headway as fossil fuel prices increase. The main problem so far has been the difficulty in competing in the heating market with cheap gas.

Basically, it is the same story in every field. The main current limits are economic - as reflected by the current market valuation of conventional fuel sources. Only when that problem is resolved will we have to face up to the other limits I have identified.


As we have seen, while for the moment the constraints are mainly economic, in the longer term, there could be relatively tight environmental and technical limits to renewable energy- based human economic activity, at least of the current sort, ultimately imposed by the constraints of natural energy availability but reinforced by social and economic factors. We may be able to deal with some of the social and economic factors - for example, we may be able to convince people that it makes sense to pay more for green energy and to accept some visual intrusion from wind farms for the greater good of the planet. Then come the wider environmental limits - the need to maintain biodiversity and not impinge unduly on the natural processes that maintain the Earth's ecosystem. That could get increasingly hard if populations and affluence grow.

The overall renewable energy resource limit is even harder to deal with. Of course you might say it provides a welcome ultimate limit on our ability to damage the ecosystem by continued economic expansion. Technical fixes, like devices that use energy more efficiently, can obviously help stretch these energy resources. However, most of the easy and cheap energy saving opportunities will be rapidly exhausted early on and it is hard to see how efficiency gains, through clever new Factor 4 type innovations, can continually keep pace with the seemingly inexorable rise in energy demand of around 2% yearly. If we want to expand human energy use beyond these limits then we would have to find other sources of energy. Some people look to nuclear fusion, hot or cold, others to as yet even more unproven options like the socalled 'free energy' techniques. There is even talk of their being large amounts of hydrogen gas produced by bacteria deep underground.

Options like this are very speculative and might have their own environmental, economic and social limits. For the moment then, we are stuck with trying to operate within the technical, environmental and social limits of the climate and weather system related renewable energy sources. The simple point is that, on a finite planet with a finite energy flux coming in from the sun, there are inevitably resource limits, and renewables cannot help us escape these.

Some people fear that, sooner or later, we will have to face up to radical social, economic and cultural changes. Not everyone sees change in lifestyles as a problem - some say that we would all benefit from a shift in emphasis from the quantity of consumption to the quality of consumption. Some say we should do this sooner rather than later since the environmental and social problems associated with our current way of life are becoming urgent. Indeed, some say we have already gone beyond the ecological carrying capacity of the planet, and are living on borrowed time - borrowed from future generations. But rising material expectations are locked into, reinforced by, and reinforcing, the global sys- tem of economic expansion. We all seem to want more! Even some of the altruisticallyminded argue that global economic growth is the only hope for the developed world - if only in terms of allowing for some 'trickle down' to the less well off!

With billions of new consumers potentially joining the race as the developing countries industrialise, it is easier to think in terms of just changing the technology and then just hoping for a more enlightened approach to consumption to emerge. That surely is not good enough. We cannot just keep trying to rush blindly forward believing that we can fix any problems that crop up. Most technical fixes have a downside - they create unexpected problems themselves. And they clearly cannot allow us to continue with materialistic growth for ever. To put it simply, it certainly looks as if environmentally sustainable technology can be developed and provide a technical fix for a while but what we also need is to create a sustainable society - and that's a larger project.


1. Shell, The Evolution of the World's Energy System 1860-2060, Shell International, London, 1995
2. Greenpeace, 'Towards a Fossil Free Energy Future', Stockholm Institute report for Greenpeace International, London, April 1993.
3. von Weizsacker, E. Lovins A, Lovins, H., Factor Four, Earthscan, London, 1994
4. UN /WEC, World Energy Assessment: Energy and the Challenge of Sustainability, Development Programme, UN
Department of Economic and Social Affairs and the World Energy Council, 2000.
5. Grob, G. 'Transition to the Sustainable Energy Age', European Directory of Renewable Energy Suppliers and Services, James and James, London, 1994.
6. Trainer, T., The Conserver Society, Zed Books, London, 1995.
7. Jackson, T. 'Renewable Energy: Summary Paper for the Renewable Series', Energy Policy, Vol.20 No.9, pp 861-883, 1992.
8. Elliott, D. 'Land use and Environmental Productivity' Renew 133, Sept-Oct 2001, pp 22/24
9. Elliott, D. Windpower in the UK, NATTA Compilation report Vol.IV, Network for Alternative Technology and Technology Assessment, Milton Keynes, 2002
10. World Commission on Dams, Dams and Development: A new framework for decision making, Earthscan, London 2001.
11. Contact: gagnon.luc@hydro.qc.ca
12. Commoner, B. The Poverty of Power. Jonathan Cape, London, 1976.
13. Milborrow, D. evidence to the Performance and Innovation Unit's Energy Review, see Renew 136, March-April 2002, p.29
14. Hewett, C., Power to the People, Institute for Public Policy Research, London, 2001.

This is one of almost 50 chapters and articles in the 336-page large format book, Before the Wells Run Dry. Copies of the book are available for £9.95 from Green Books.

Green Books banner 1

Continue to Section B of Part Two: Using the net energy concept to model the future

Sitemap for Before the Wells Run Dry

Copyright © Feasta. All rights reserved.