Clean power from deserts

Overview of CSP
History of CSP
Activities and events
Press releases
Letters and comments
Mailing list
Site map

QUESTIONS ABOUT the desertec proposals

Back to Concentrating Solar Power

Questions are often asked about possible problems with CSP and the Desertec proposals. Several of the main questions are discussed on other pages. Some others are listed here, with some answers.

How can this be of benefit to the UK?

Although the UK does not have enough direct sunlight for CSP, and the deserts of the Middle East and North Africa may seem far away (see next), the Desertec proposals are highly relevant to UK needs:

  • Indirect benefits:
    • Given their worldwide potential, the Desertec proposals can have a major impact on cutting worldwide emissions of CO2, thus helping to safeguard us all from changes in climate that may prove dangerous.
    • Since CSP electricity is expected to become one of the cheapest sources of power in Europe (including the cost of transmission), imports of solar power from the Middle East and North Africa (MENA) can help to hold down the cost of electricity throughout Europe.
    • Flooding the world market with relatively cheap supplies of clean energy means fewer worries about the security of energy supplies or about the risks of conflict over "a global grab for energy".
    • By alleviating worldwide shortages of energy, water, food and land, the Desertec concept may help to reduce tensions arising from poverty and immigration.
    • The Desertec project will create many business opportunities for “UK plc”, with corresponding potential for the creation of new jobs in the UK.
  • Direct benefits:

These and related points are made in Desertec and UK policies (PDF, 135 KB, 6 sides).

Isn't the Sahara too far away from the UK?

There are three points here:

  • The UK and other countries in the north of Europe can benefit from 'desert' electricity without it being necessary for electrons to travel all the way from the Middle East or North Africa. This is explained on our page about the cascading principle (see also the figure below and Kick-start and upgrade).
  • It is possible to transmit electricity efficiently and economically over long distances using low-loss 'HVDC' transmission lines (see Electricity transmission grids).
  • Given a single market for electricity throughout Europe or EUMENA, low-cost solar electricity that is fed into that market can help to hold down the cost of electricity throughout Europe or EUMENA.
Cascading principle
In some respects, a transmission grid is like a lake. Water (electricity) may be added at one place and the same amount may be taken out at another place without it being necessary to move that quantity of water (electricity) all the way from one place to the other. Solar electricity fed into southern Europe can have an immediate benefit for countries throughout Europe.

Aren’t CSP developments something for the future?

No, things are happening now:

  • The PS10 and PS20 CSP plants are already in operation near Seville in Spain and the Andasol 1 CSP plant near Granada has also recently come on stream. Other CSP plants are operating in California, Australia and elsewhere around the world. More CSP plants are under construction in Spain, Morocco, Algeria, Egypt, Israel, and elsewhere—and many more are being planned around the world.
  • A consortium of blue-chip companies, the Desertec Industrial Initiative, has been formed with the declared intention of implementing the Desertec concept.
  • The Union for the Mediterranean has initiated a "Mediterranean Solar Plan" that has been inspired by the Desertec concept.
  • The World Bank is already providing funding for CSP plants that are under construction in Morocco, Egypt and Mexico, and it is likely that more projects will be funded by the bank in the future.

CSP plants can be built in 2 or 3 years. Build times are likely to decrease as designs are rationalized to facilitate speedy construction.

It is true that priority should be given to the needs of people in the host countries but, with the right political and financial impetus, it should be possible to develop the local and export markets in parallel.

Is there a list of CSP projects to show take-up of the technology?

CSP projects—planned, under construction and up-and-running—may be seen on Google Earth via a link from our Resources page.

News about CSP projects may be seen on our News pages.

Isn't Desertec just a part of the discredited idea that we can have continual economic growth?

Absolutely not! Desertec is not intended to be a comprehensive set of policies for solving the world's problems. And it is entirely compatible with a "limits to growth" perspective.

In The Population Bomb (1968), Paul Ehrlich gives the formula:

I = P × A × T,

where I = Environmental Impact, P = Population, A = Affluence, and T = Technology. In those terms, Desertec is largely about T and does not attempt to say anything about P or A.

The Desertec concept can and should have an important role to play in the development and implementation of worldwide policies for truly sustainable living, including sensible policies to stabilise and then bring down the size of the human population, and bringing our demands on the planet into line with what it can provide without compromising the future.

There is a small selection of ideas and discussions of other policies that will be needed on New models for economics and the environment.

Technological fixes will not get us out of the mess we are in—we should simply learn to use less

Some people say that we should simply learn to use less of everything and others believe that, in addition, we should return to the simpler ways of living of the past.

Here are a few points that seem to be relevant:

  • Yes, we should certainly learn to use less of the things that are causing damage to the environment (such as fossil fuels and nuclear power).
  • But demographers say that the population bomb has already been detonated. Although fertility rates are falling around the world, the fact that there are many people of breeding age or younger means that there is little or nothing that can be done to prevent the world's population growing to 8, 9 or 10 billion by about 2070 (depending on the provision of birth-control facilities and other factors—see below). Hopefully, it will then stabilise and start to decline.
  • This unprecedented situation means that, even if the world were to use the same amount of energy as it is using now, per capita use of energy would shrink substantially. Unless we decide do without electricity altogether there will be a need for large amounts of electricity and it will have to be clean. Desertec means the use of relatively simple technologies that can meet that need.
  • Jefferey Sachs argues quite persuasively (in Common Wealth: Economics for a Crowded Planet) that raising standards of living of poorer people is not only an absolute good but that it is, amongst other things, necessary to persuade people that they will benefit from having smaller families. Since the provision of clean electricity is one of the things that will raise living standards of poorer people, we have another reason for promoting the Desertec developments.
  • In summary, the population explosion will force most of us to use less of things whether we like it or not. And, for the same reason, returning to the ways of living of the past is probably not an option.

See also New models for economics and the environment.

Environmental and social impacts of CSP plants

From at least as far back as Walt Disney’s The Living Desert, wildlife films have made us aware that deserts have their own vibrant ecology. If the world’s deserts were all to be covered with CSP plants, there would indeed be cause for concern about the animals and plants that live there. But less than 1% of the world’s deserts would meet current world demands for electricity and even in pessimistic scenarios, it seems unlikely that more than 5% would be needed in the future. It should be possible for CSP plants and wildlife to co-exist.

In a similar way, it seems likely that CSP plants would have a minimal impact on the lives of people that live in deserts. Indeed, by providing employment, electricity, water, shading from strong sunlight, and opportunities for horticulture (see CSP bonuses), any impacts may be more positive than otherwise.

Having said all that, it is, of course, very important that both the environmental and social impacts of solar developments should be carefully assessed. And solar developments should not be permitted where the environmental or social impacts are unacceptable.

Environmental impacts of the Desertec proposals are discussed in Section 5 of the TRANS-CSP report from the German Aerospace Centre and socio-economic impacts are discussed in Section 4.


Environmental and social impacts of HVDC transmission lines

Environmental impacts of HVDC power lines (overhead, underground and submarine) are discussed in Section 5.2 and 5.3 of the TRANS-CSP report from the German Aerospace Centre and also in a thesis by Nadine May (PDF, 5.8 MB).

Overhead power lines can be unpopular because of their visual impact. However, there is a variety of ways in which transmission grids may be upgraded so that visual impacts are minimised, as described on a separate page.

Electrical and magnetic effects from power cables follow the inverse square law which means they fall off rapidly with distance from the cables. With a single cable, it appears that electrical and magnetic effects fall to background levels within 5 or 6 metres.

However, with submarine cables and underground cables, it is now widely accepted that there should be a 'return' cable for each 'outwards' cable (instead of using the sea or the ground as a return cable) and in that case it is normal practice to lay the return cable next to the outwards cable so that electrical and magnetic effects from either cable is largely cancelled by electrical and magnetic effects from the other.

Do we feel comfortable with large industrial organisations being involved in the Desertec Industrial Initiative?

It is inconceivable that the Desertec vision would be realised without the involvement of large industrial companies.

Naturally, Desertec developments should be done in a responsible way (as discussed here and here) but we believe that, with Desertec, the temptation to do things badly is less than with the extraction of oil from tar sands, for example.

Most large engineering companies now have policies for environmental protection and social responsibility.

Isn't this just another neo-colonialist raid by rich countries on the poor?

A possible objection to the Desertec proposals is that they are just another case where rich countries take what they need from poorer countries leaving little for local people, except pollution.

There are good reasons to think otherwise:

  • Several of the benefits of CSP are purely local and cannot easily be exported or expropriated. These include local jobs and earnings (including earnings from exports), local availability of inexpensive, pollution-free electricity, desalination of sea water, and the creation of shaded areas under the solar collectors with several potential uses including horticulture. These various benefits and bonuses from CSP are described on a separate page.
  • Unlike oil, coal or mineral extraction, where there can be substantial negative impacts on local communities, there is no pollution from the operation of CSP plants. It is true that the early plants do tend to use quite a lot of water but this is not really necessary and CSP plants can be designed to operate with little or no net consumption of water.

The TREC initiative is a collaboration amongst several different countries, including countries in the Middle East and North Africa. There are large potential benefits both for countries in the sunbelt and other countries that may buy solar electricity from there. Desertec can and should be a win-win collaboration that will be of benefit to all.

Here (from the Genesis Morocco blog) is an amusing comment on the idea that Desertec is a neo-colonialist project where Europeans steal African sun:

A young man named Mustapha sums it up with a bit of light-hearted irony. He's selling phone cards on a Rabat street corner, on a baking hot day. "The sun belongs to everybody," he jokes [in Arabic]. "I don't mind if foreigners come for it, just as long as they leave us the shade."


How do CSP plants cope with sandstorms?

It seems that sandstorms are not all that frequent but that may vary from one desert to another. On the Sahara Wind website ( it says "According to our measurement recordings, sandstorms are fairly rare in occurrence. Winds blowing eastward from the desert account for less than 5 % of the normal wind regime. Sandstorms make only one part of this 5%."

With the parabolic trough type of system, it seems that the operators simply turn the mirrors upside down during a sandstorm. Other CSP systems may also protect the mirrors in a similar way. If the CSP plant is a hybrid (with gas firing as a backup) or if it has heat storage, then electricity generation may continue during the sandstorm.

CSP plants have been generating electricity successfully in California since the mid 1980s without significant damage from sandstorms.

Since it can be cloudy near the coast, doesn't this undermine the idea that CSP may be used for the desalination of sea water?

The AQUA-CSP report from the DLR confirms that, for example, "on the Western South-American coast and the Western South African coast, there is the phenomenon of dense fog banks from the ocean covering several kilometres inland for several months per year." (p 44).

The report suggests that, if there are cloudy conditions near the coast, the desalination plant itself may be located at the coast but the CSP plant may be located at a distance, where there is direct sunshine. In that case, the CSP plant simply provides electricity to power the desalination plant and it is not possible to use waste heat for desalination. Of course, in coastal regions where there is plenty of direct sunshine, then it will be possible to use waste heat from solar generation for the desalination of sea water.

What is the energy return on energy invested (EROEI)?

How long does it take for a CSP plant to recover the energy that was needed to create the equipment and install the plant?

Evert du Marchie (p 10) gives an energy payback time for the Fresnel mirror type of CSP plant as 6.7 months.

SCHOTT AG (p 13) suggests that the energy payback time for a parabolic trough type of CSP plant is about 5 months.

Could there be shortages of materials to construct CSP plants or HVDC transmission lines?

Generally speaking, CSP plants are constructed from materials that are plentiful and relatively cheap: steel, glass, concrete, aluminium and the like. At present, most of these are made using 'dirty' technologies that release CO2 into the atmosphere but since they are so widely used in the world, it will be necessary to find new ways of making them, or good substitutes for them, that do not pollute the atmosphere.

Assuming the world continues to use at least as much electricity as now, the world's entire fleet of power plants needs to be replaced every 30 or 40 years anyway. To a large extent, materials like steel can be recycled from the old obsolete power plants to make the new clean power plants that we need now.

Do CSP plants need more or less materials for their construction than old-style types of power plants that they would replace? Taking account of the complete power-generation cycle, it seems likely that CSP plants would require less materials than traditional alternatives. Although CSP plants need solar collectors, they do not need any of the relatively large infrastructure that is needed for the mining, processing and transportation of fuels such as coal or uranium.

Some figures given by Evert du Marchie (p 10) suggest that, if all of the world demand for electricity were to be met using CSP plants, then it would take about three years at current rates of production to make all the necessary steel. Although this scenario is conceivable, it is likely that CSP would always be one amongst several renewable sources of electricity, that construction of the necessary plants would be spread over a much longer period, and that ways will be found for constructing CSP plants without so much steel.

Although CSP may use less resources overall than are needed for traditional sources of power, there are concerns about worldwide demands for materials for all purposes and the often unscrupulous way in which they are being exploited (see links below).

HVDC transmission lines can be made using either copper or aluminium for conducting electricity. Aluminium is more plentiful than copper but it appears that there is enough of both metals available to allow for the expansion of HVDC transmission. If copper were to become scarce and expensive, it is likely that new supplies could be obtained by replacing copper piping in houses and other buildings with plastic piping.


Since wind power is relatively cheap, isn't this the thing we should all go for?

Owing to imaginative and forward-looking policies by the Danish and German governments, amongst others, wind power has received the financial support that has been needed to enable the industry to grow (to about 120 GW worldwide in 2009) and to bring down costs. In the USA, where commercial-scale CSP was first developed, the government has not had the same enlightened policies. As a result of this and, until recently, a general lack of awareness of the technology and its potential, there is still less than 1 GW of CSP worldwide, and costs are still relatively high. However, as CSP expands, costs will fall with economies of scale and refinements in the technologies, as with wind power. The TRANS-CSP report from the DLR has calculated that CSP is likely to become one of the cheapest sources of power in Europe, including the cost of transmission. Google Inc has identified CSP as one of the key technologies in its quest to develop "Renewable Energy Cheaper than Coal" ("RE<C"). See also our page about the costs of Desertec-related technologies. The CEO of CSP company eSolar claims that they can already generate electricity as cheaply as gas-fired power stations.

There are several reasons for developing a whole range of different renewable technologies:

  • Given the massive task of decarbonising the world's economies, it is likely that we will need clean energy from a range of renewable sources.
  • Different sources of clean electricity have different characteristics. By connecting them up over a wide area, we can balance their strengths and weaknesses and thus create a robust and resilient system for the supply of clean energy. In that connection, CSP has the particular advantage that, with heat storage and backup sources of heat, it can deliver "power on demand". This is not true of wind power.
  • There are far too many uncertainties in how technologies will develop for us to start picking winners now.

Combining a wide range of renewable technologies is a key part of the Desertec concept, as described in the TRANS-CSP report and shown on the Desertec map.

In sunny deserts, wind power has clear potential and should certainly be developed. But, taking a medium-to-long-term view, CSP is likely to have much more potential, and, with heat storage and the use of backup sources of heat, it has the advantage that it can deliver power on demand. PV is likely to have a role too but, as things stand now, it is relatively expensive and it doesn't have the same potential as solar thermal power to provide power on demand.

Isn't Desertec a case of "picking winners"?

A possible objection to Desertec is that it involves governments making decisions about particular technologies rather than leaving it to the market to decide which technologies are best. There are two answers to this objection, one to do with the proposed large-scale supergrid and the other to do with CSP.


A supergrid could be constructed as a series of privately-funded projects, one for each transmission link. But there are problems with this approach, as described by Michael Goggin in an article in (2008-07-30):

To use terminology from the field of economics, our inability to build new transmission is fundamentally a public goods problem. In most regions, policies require wind plant developers that want to connect to the electric grid to pay for the full cost of an upgrade to the grid network, even though the majority of the benefits of this upgrade would accrue to millions of electricity consumers and other power plants that could piggyback on this investment. Across the country, hundreds of wind projects comprising tens of thousands of wind turbines are on hold because no one wants to step forward and pay for upgrades that will primarily benefit others.

Probably the simplest way round this dilemma is for governments to pay for transmission grids in much the same way that they pay for the roads network. This is a case where government backing for a particular kind of technology is probably justified.


At present, CSP is having to compete against electricity from coal-fired and nuclear sources that is subsidised in various ways and is not yet paying the proper price for emissions of CO2 (see our page about CSP costs and the quote below). Governments need to take action to remove these distortions in the market place so that CSP can compete on a level playing field (see Clean power from deserts: what governments can do, PDF, 68 KB).

It is widely recognised that renewable energy technologies need public support until costs can be brought down via economies of scale and refinements in the technologies. Until the worldwide installed capacity in CSP has increased from its current low level, there will be a case for support via feed-in tariffs or a "renewables obligation" or some such scheme, together with support for other renewable technologies that are finding their feet.

Thus government actions are needed on two fronts to protect and support CSP until it is properly established but neither case can be classed as "picking winners".

"More than half of the subsidies (in real terms) ever lavished on energy by OECD governments have gone to the nuclear industry."

From "Nuclear power out of Chernobyl's shadow", The Economist, print edition, May 6th 2004.

Wouldn't it be better to generate our energy locally in a decentralised way?

There is quite a lot that can be said on this subject. In brief:

  • Local generation of electricity (with PV, combine heat and power, micro wind turbines etc) certainly has a role to play. But so do large-scale but remote sources of renewable energy such as offshore wind farms, wave farms, tidal stream generators—and CSP. Only time will show what is the best balance between the two.
  • There is so much solar energy available in the Sahara that it makes good sense to generate electricity there. Researchers at the German Aerospace Centre have calculated that solar electricity from the Middle East and North Africa is likely to become one of the cheapest sources of electricity in Europe, and that includes the cost of transmitting it. However, it also makes good sense to generate electricity locally, and increasingly so as PV becomes cheaper.
  • Quite apart from the transmission of "clean power from deserts", there are several other benefits from large-scale transmission grids:
    • They provide access to other large-scale but remote sources of renewable energy such as offshore wind farms, wave farms, tidal stream generators etc.
    • Without the provision of a transmission grid, local decentralised generation of electricity would be wasteful—because surplus power in any area cannot be moved to where it may be needed.
    • Large-scale transmission grids help to smooth our variations in supply and demand.
    • And they improve energy security because any shortfall in one area can usually be met from one or more other areas.
  • Some people suggest that large-scale generation of electricity and large-scale transmission grids simply concentrate power into the hands of a few large energy companies. This has been true in the past and is still true in some places but there is no need for it to be true in the future:
    • The provision of a single market for electricity throughout the EU (which is favoured by the European Commission and the British government and is being developed now) should promote competition between suppliers and between different sources of electricity.
    • There are good reasons to extend the single market for electricity to EUMENA and beyond.
    • There must be appropriate provision to ensure that local, decentralised generation of electricity can develop properly alongside other sources of electricity.
  • A well-designed transmission grid can serve the needs both of local, 'decentralised' generation, and large-scale generation in remote locations.

Hermann Scheer

There is, to our knowledge, only one prominent person or organisation that is actively hostile to the Desertec concept (apart from some representatives of the fossil-fuel and nuclear lobbies). Despite his several achievements in promoting renewable forms of energy, Hermann Scheer has been quite outspoken in his criticisms of the Desertec concept.

Herman Scheer's objections to the Desertec concept seem to be mainly these:

  1. PV has delivered in Germany (thanks to the feed-in tariffs that Hermann Scheer has been so successful in promoting) but CSP has not yet delivered much electricity anywhere in the world ("The approach Khosla is supporting led only to 400 MW of installed concentrated solar power in California during the beginning of the 1980s. The figures did not rise since. It amazes me that he does not ask himself why the concept that I advocate has developed dynamically, although it is supposed to be more expensive than the technologies he is supporting. Surely he must ask himself why his approach did not experience a breakthrough yet. Is that realism?").
  2. CSP is a "big business" approach to electricity supply and, as such, it is liable to deliver electricity at inflated, monopolistic prices ("The level of the German electricity rates is determined by the supply monopolies of the few big suppliers of conventional power--they are the major profiteers. ... The same goes for huge concentrated solar power (also known as CSP or solar thermal) plants in the North African desert, proposed to supply Western Europe with energy. The ambition behind such strategies is to keep the existing supply monopolies in the hands of a few big international power players and out of the reach of the democratic control of our societies and the market.").
  3. CSP requires a large-scale transmission grid but this would be too expensive in many parts of the world ("Most of [the developing countries] are unable (and will most likely be so in the future) to shoulder the expenses for a grid connecting the 2 billion people worldwide still living without access to electricity.").
  4. CSP requires a transmission grid and this means losses of power during transmission.
  5. Apart from climate change, there are several other energy-related problems (such as inequalities and injustices, atomic waste; reactor accidents) and these cannot be solved with "conventional energy structures".

There has been quite a ding-dong debate about these things between Hermann Scheer and US venture capitalist Vinod Khosla. An example of what Vinod Khosla has to say may be seen in "Scheer nonsense"—the damage idealistic environmentalists can do.

Here are some answers to the points that Hermann Scheer makes, and some other related points:

  1. It is true that feed-in tariffs have delivered big increases in renewable sources of electricity in Germany and this is very much to be welcomed. But it is misleading to suggest that CSP is dead in the water. After those CSP plants were put up in California in the mid-1980s, there has been a 15-year gap where no more plants have been built. This has been partly because fossil fuels were too cheap and partly because there were no policies in place (like feed-in tariffs) to make CSP attractive to investors. But now things are changing fast. Our News page has frequent announcements of new CSP projects around the world. These are marked 'CSP-PROJECT'. All the CSP projects that we know about—plants that are planned, under construction or up and running—may be seen on Google Earth. Most entries have links back to relevant reports on our News page. There is a large number of new CSP plants now planned in many parts of the world, some are being built and several are on stream. In 2009, the World Bank has estimated that, worldwide, about 9 GW are in the pipeline. Emerging Energy Research puts the figure at about 14 GW.
  2. We don't share Hermann Scheer's views about big business. It is true that companies like ExxonMobil have disgraced themselves in the way they have responded to the challenge of climate change but we are never going to solve these problems without having big (and small) businesses on board. The PV panels that can be seen on roofs all over Germany as a result of Hermann Scheer's initiative in developing feed-in tariffs were not the product of some cottage industry. They are all produced in factories run by big business. Could "big business" create some kind of monopolistic control over solar power from deserts? We doubt it. Less than 1% of the world's hot deserts could produce as much electricity as the world now uses. There is plenty of desert to go round.
  3. It is true that CSP only works properly where there is lots of direct sunlight (as in sunny deserts) and this means that large-scale HVDC transmission grids are needed to bring the electricity to where people are living. But it is wrong to suggest that such grids are not affordable by developing countries. What matters is the overall cost of the delivered electricity, including the cost of transmission. The TRANS-CSP report from the German Aerospace Centre calculates that CSP electricity is likely to become one of the cheapest sources of electricity in Europe, including the cost of transmission. The same is likely to be true in many other parts of the world, including countries like India and China. The TRANS-CSP report estimates that transmission costs would be about 20% of the delivered cost but that cost is likely to be more than offset by the anticipated low price of CSP electricity from desert regions. It is also relevant to mention that worldwide arrangements for cutting emissions of greenhouse gases are likely include increases in the transfers of funds from richer countries to poorer ones. There should be money available to pay for transmission grids.
  4. The idea that transmission grids are bad because there are losses in transmission is seriously misleading. A system without transmission grids would lead to greater wastage of electrical energy than a system that has transmission grids. This is because a grid makes it possible to move electricity from areas that have a surplus to areas where there are shortages. Without that facility, and without facilities for economical bulk storage of electricity, excess electricity that may be generated in any one area is simply wasted. That wastage is likely to be far in excess of any losses in transmission. In any case, with HVDC transmission, losses are very low (about 3% per 1000 km, plus small AC/DC conversion losses at each end). This issue has arisen in connection with Greenpeace's promotion of decentralised energy. They don't actually say that transmission grids are bad but they imply it very strongly. They are right to promote the advantages of CHP but they are quite wrong in what they have led people to believe about transmission grids. Apart from reducing wastage of renewable electricity, large-scale transmission grids have several other benefits.
  5. We believe it is quite wrong to suggest that the kinds of energy-related problems identified by Hermann Scheer (item 5 above) would not find good solutions in the Desertec proposals. Several of the problems he mentions are to do with nuclear power and it is clear that, under the Desertec scenario, the world could phase out nuclear power altogether.
  6. Hermann Scheer has nailed his flag to PV and is attacking CSP but CSP has major advantages compared with PV:
    • With facilities for storing solar heat and with the use of gas or biofuels as backup sources of heat when there is not enough sun, CSP plants can deliver any combination of base-load power, intermediate load and peaking power. In short, they can deliver "dispatchable" power according to consumer's needs. This is a major advantage compared with PV which can only deliver power during daylight hours. It is true that there are systems for storing electricity (eg flow batteries or pumped storage) but these are a great deal more expensive than systems for storing solar heat that can be integrated with CSP plants.
    • CSP is currently a lot cheaper than PV. It is true that the concentrating technique can be used in conjunction with PV and that can dramatically reduce the amount of PV that is needed (and the cost) but systems like that have no means of storing solar heat and cannot be hybridised with gas-firing so they lose the advantage just mentioned.
  7. Having said all that, everyone seems to be agreed that we need a diversity of different kinds of renewable energy, including both PV and CSP. The TRANS-CSP report describes a scenario up to 2050 in which Europe meets all of its needs for electricity using a wide variety of low-carbon sources (but phasing out nuclear power) and with CSP providing just 15% of the total.

Is it possible that geo-engineering solutions to climate change might have an impact on CSP?

The short answer to this question seems to be "yes", the injection of particles into the atmosphere to reduce the amount of sunlight reaching the ground could have an impact on CSTP and CPV plants (see Atmospheric ‘sunshade’ could reduce solar power generation, Earth System Research Laboratory, US National Oceanic & Atmospheric Administration, 2009-03-11): "On average, for every watt of sunlight the particles reflect away from the Earth, another three watts of direct sunlight are converted to diffuse sunlight. Large power-generating solar plants that concentrate sunlight for maximum efficiency depend solely on direct sunlight and cannot use diffuse light." This could reduce output from CSP plants by as much as 20%.

Possible answers to this problem include avoiding the need for shading by bringing down concentrations of greenhouse gases in the atmosphere, using sunshades in space rather than particles in the atmosphere, and the use of non-concentrating solar technologies such as PV which can make use of diffuse light as well as direct sunlight.


Could the proposed supergrid make us vulnerable to the effects of a coronal mass ejection?

In the article Space storm alert: 90 seconds from catastrophe (New Scientist, 2009-03-23), Michael Brooks describes how transmission grids may be vulnerable to catastrophic failure caused by plasma from the sun (a 'coronal mass ejection'). Since the development of large-scale supergrids is an important part of the Desertec concept, it is reasonable to ask whether the development of the Desertec concept might be storing up trouble for the future.

With or without the Desertec concept, the world already depends heavily on transmission grids, so this problem clearly deserves attention. It is reasonable to ask what may be done to guard against the disruptive effects of a worst-case scenario in which one or more national or international transmission grids were knocked out completely. Here are some tentative answers:

  • Identify the critical things that need to keep working (eg power for hospitals, power for water pumping) and then provide alternative sources of power that don't depend on the grid. Hospitals already have emergency generators to cope with 'normal' power cuts and the same may be true for water companies. These backup sources of power may need beefing up so that they can keep the critical things working until transmission grids can be repaired.
  • Make sure that the mix of power sources (as described in the TRANS-CSP report) includes 'decentralised' sources of power (such as PV) that will keep on working even if the whole grid is knocked out. This may happen naturally because PV is likely to become a lot cheaper in the future and is likely to be much more widely used than it is now.
  • Since the destruction of transformers appears to be one of the main problems to be overcome, there may be a case for keeping 'strategic' stores of transformers to cope with this kind of situation, in the same way that there are strategic reserves of food and fuels. 

Clearly, these things need careful study and appropriate policies need to be developed.

A possible answer?

In Michael Brooks's article, he says "The power companies need about 15 minutes to prepare their systems for a critical event ..." and suggests that existing warning systems may not work fast enough. However, a more recent report in the New Scientist (Solar 'double vision' aids space weather warnings, 2009-04-13) says that a new system, using a pair of satellites to create a stereo image, could provide a warning as much as 24 hours before a coronal mass ejection. Let's hope systems like this can be developed in good time.

By trapping solar heat, could CSP plants be adding to global warming?

CSP plants are designed to capture solar energy, first concentrating sunlight to create heat and then converting a proportion of that heat into electricity. The heat that is not converted into electricity will be dissipated, and the energy in the electricity will ultimately be converted into heat. Those sources of heat, which are much the same as would be produced by a coal-fired, gas-fired, or nuclear generating plant, may conceivably be a significant factor in global warming, although they are widely-regarded as not significant.

But it has been suggested that the heat-trapping effect of CSP plants could be much worse than that. In a blog (in French—which may be translated with Google Translate), it is argued that, because CSP plants are only about 15% efficient, the remaining 85% of the sunlight falling on the solar collectors of a CSP plant is dissipated as heat.


The argument just described implies that the albedo of CSP solar collectors (the proportion of incident sunlight which is reflected from them) is close to 0 (on a scale from 0 to 1). This means that, viewed from above, the mirrors of a CSP plant should appear to be black, or nearly so. But this is clearly not the case:

In a FAQ about CSP (PDF, 437 KB), researchers at the German Aerospace Centre (DLR) say "A CSP collector absorbs about 60-70% of the solar radiation and converts it to useful heat. This is its typical collector efficiency. That means the rest (30-40%) is rejected. In fact it has practically the same albedo as desert sand, but a preferred direction of reflectance."

The blog has overlooked the fact that the mirrors of a CSP plant receives solar radiation from the sky (in addition to the radiation that is received directly from the sun), and much of the light from the sky will be reflected past the heat collectors and back up into the sky. Also, some of the light that is received directly from the sun will be scattered by dust or dirt on the mirrors or scratches and other defects in their surfaces. Overall, the albedo of CSP solar collectors will certainly be higher than 0. It appears to be about the same as a typical desert.

It is relevant to point out that, with the clear skies of desert regions, much of the light from the sky will normally be at the blue (short-wavelength) end of the spectrum (including UV). This means that, after it has been reflected, the light will have a relatively good chance of escaping out into space (because it is the long wavelengths—infra red—that are trapped by greenhouse gases in the atmosphere).

Getting things in proportion

Using CSP, about 0.6% of the world's deserts would be needed to generate as much electricity as the world is using now. If CSP plants, including the spaces between the mirrors and other areas, were completely black, they would reduce the albedo of the desert that they cover from about 0.4 to about 0.0. In other words, in the unlikely scenario that all the world's electricity were to be generated by CSP plants in deserts, totally black CSP plants would reduce the overall albedo of the world's deserts by 0.6 × 0.4 = 0.24%.

Of course, as we have seen, the solar collectors of CSP plants are far from being black. Also, the other areas in a CSP plant are not black. And CSP will always be just one part of the range of renewable sources of electricity. Overall, any reduction in albedo would be much less than 0.24%.

Any positive or negative effect of CSP plants on the albedo of deserts is likely to be very small, especially in comparison with the potential benefits of CSP in helping to phase out the use of fossil fuels in the generation of electricity. Any negative effects can certainly be offset by installing upwards-facing mirrors in neighbouring areas of desert.

A study

A comprehensive study of these issues would be useful, covering at least the following points:

  • Taking account of light from both the sun and the sky, what proportion of that light is reflected by the solar collectors of a CSP plant, out past the heat collectors and into the sky? Likewise for a typical desert.
  • What is the distribution of wavelengths of the light that is reflected from CSP solar collectors, and from deserts?
  • Taking account of the way in which greenhouse gases in the atmosphere trap different wavelengths, how much of the light that is reflected from CSP collectors is likely to escape into space? Likewise for deserts.
  • If there is any excess heating effect, how much of that excess heat would be radiated into space during the night?
  • Overall, what is the warming or cooling effect of CSP plants?
  • How does that warming or cooling effect compare with the benefits of generating electricity without emissions of CO2 (assuming that CSP plants are replacements for CO2-emitting source of electricity and not simply adding to the overall generating capacity)?
  • Does the heat from thermal power plants (coal-fired, gas-fired, nuclear, or CSP) add significantly to global warming?


From the evidence available now, it appears that the albedo of CSP solar collectors is about the same as that of a typical desert. It is possible that, compared with a desert, the wavelengths reflected by solar collectors may be biased in favour of those with a relatively good chance of escaping into space. Any net albedo effects of CSP plants (compared with desert land) are likely to be very small compared with their potential benefits in helping to decarbonise the world's energy supplies. And any negative effects may be offset by the installation of upwards-facing mirrors.

Wouldn't it be better to wait until CSP and PV are cheaper before starting the Desertec project?

Bjorn Lomborg and others have argued that, since CSP and PV are getting cheaper, it might make sense to wait until they are more affordable before starting on something as large and ambitious as the Desertec project. In answer to that suggestion:

  • The thing that drives cost down is economies of scale. Unless steps are taken to expand the PV and CSP industries (by providing subsidies), the industries will not expand and costs will not fall. If everyone were follow the 'wait' policy, there would be a kind of stalemate. Few people would be buying PV or CSP (because it would be considered too expensive) and so costs would not fall. We should be grateful to countries like Germany for providing subsidies for PV—which have had the effect of driving down prices. Now the UK government has introduced similar policies, showing that they are willing to share the responsibility for driving down prices even further. The same kinds of policies should apply to PV and CSP in desert regions.
  • We really don't have the luxury of waiting 10 or 20 years in case costs fall. If we are to minimise the risk of dangerous climate change, we should be introducing new clean sources of power urgently. We need to factor in those risks and corresponding costs, and Nicholas Stern's argument that delays in taking action on climate change are likely to mean (much) higher costs later.
  • A related point is that power stations of all kinds wear out after 30 or 40 years. There is a constant rolling programme of replacing old power stations with new ones. In that context, Desertec is not an enormous project at all. It is similar in magnitude to the normal process of renewing power stations as they wear out. If we are replacing power stations, it is better to build clean ones than dirty ones.

Back to Concentrating Solar Power

Last updated: 2011-01-10 (ISO 8601)