Is it possible to deliver low carbon energy to 9 billion people?


For some time I’ve been thinking about the question – how will we deliver low carbon energy to the 9 billion people that will be living on this planet in 2050? I mean, that’s the crux of addressing global climate change isn’t it?

I’ve been astonished at how difficult it is to navigate through all the hyperbole; the claims and counterclaims; the miracle cures and magic bullets. When someone says that nuclear is the answer – is it? When someone says that biofuels will fuel our transport systems – can they? And I was also surprised to find that when I considered these questions myself, I didn’t really have a clear and unambiguous response that I could back up with numbers.

So I set myself a challenge: to explore the possible contribution that different low carbon energy technologies could make to delivering 150kWh of energy per person per day – which is what we need to do if we want everyone to enjoy the same per capita energy consumption as I do in Europe.

It’s a thought exercise; I’ve deliberately focused on what is possible from first principles. For the time being I’m putting aside the technical, economic and resource questions. That is not to say that these aren’t important – clearly they are – but I just want to look at what might be possible, before considering what would be best. Once I’ve excluded what is impossible, I can then consider whether the possible makes sense technically and economically.

First some statistics.

Global population in 2050 will be 9 billion persons. The land surface area of the planet is 149 million km2, of which 17 million km2 is arable land. The sea surface area of the planet is 361 million km2. The global coastline is 356,000km.

Per capita energy consumption varies form country to country. In France our per capita energy consumption is 176kWh/per/day; in the UK it is 145kWh/per/day; in Australia it is 188kWh/per/day and in India it is 17kWh/per/day. For the time being, I have assumed that we want to maintain our energy consumption at roughly European levels (which would require our American cousins to make some substantial reductions) and that developing countries will aspire to that level. That being the case, we will need a global energy supply of 150kWh per person per day – or 492,750,000 GWh per year. (How we might reduce this consumption is another discussion).

To compare the relative contribution from different renewable technologies, I’ll make reference to an area the size of 1% of the earth’s land surface. For comparison, this is an area about twice the size of France, about half the size of India, about the same size as the State of Queensland in Australia, or the equivalent of about 8 million football stadiums.

The results are interesting.


The IEA estimate that there is 847Gt of known coal resources in the world. Assuming that we wish to make this last 100 years and that the average energy content is 26.7GJ/t and that a reasonable efficiency of a coal-fired power plant is 40% – then coal has the capacity to deliver just 8kWh/per/day if we want the coal reserve to last 100 years. However, this will release about 5kgCO2 per person per year or about 30GtCO2. If we want the solution to be low carbon we will need to combine Coal with carbon capture and storage. Adding the collection and storage of carbon will reduce the efficiency of the plant by about 25%. So Coal + CCS has the capacity to contribute 6kWh/per/day, which would be 4% of our objective. If the actual coal reserve was double existing identified reserves (as some suggest) then the contribution of Coal + CCS would be double, but we would still be faced with the need to find a replacement for coal in 100 years.


Natural gas has the potential to make a similar contribution to that of coal. Global gas reserves are approximately 175 trillion m3. If we wish to extend these reserves over 100 years, and if we use this resource in a combined cycle turbine with an efficiency of 60%, then gas could deliver 7kWh/per/day, or 4% of our objective. If we add carbon capture and storage to this technology, and therefore reduce the overall efficiency from 60% to 52%, then the contribution from Gas + CCS would be about 6kWh/per/day. Therefore together, Gas and Coal +CCS could make a combined contribution of 12kWh/per/day, or about 8% of our objective.


Wind is currently the fastest growing renewable energy technology. Current global wind capacity is 121GW. Data and studies from the NOAA/ESRL, the US National Climatic Data Center and Risø National Laboratory show that large areas of the planet enjoy average onshore wind speeds of 6m/s and average offshore wind speeds of 8m/s. If we first consider onshore wind. Average wind speeds of 6m/s would give us a power density of 2.2W/m2 for a typical wind farm. If we were to cover 1% of the earth’s land surface with onshore wind turbines this enormous wind farm would have the potential to deliver 9kWh/per/day, or 6% of our objective. If we added an equivalent area of offshore wind turbines, with average wind speeds of 8m/s giving us a power density of 5.2W/m2, this offshore wind farm would deliver 21kWh/per/day, or 14% of our objective. So combined, a giant onshore and offshore wind farm covering 2% of the earth’s land surface area would contribute about one fifth of our objective.


Solar insolation on the equator, at midday, in full sunshine is 960W/m2. Actual solar insolation varies according to the incident angle of the sun at different latitudes, cloud cover, night, etc. Solar insolation in Brisbane is 207W/m2; in Toulouse it is 156W/m2; in Edinburgh it is 94Wm2; in Bangalore it is 216W/m2; and in the Saharan desert it is 260W/m2. If we take a value of 150W/m2, build a solar photovoltaic park covering an area 1% of the earth’s surface and assume an efficiency of 15% for the solar panels (which is higher than the current average of about 10% and lower than emerging technologies that can achieve 20%) then this gigantic solar park would deliver 89kWh/per/day, which represents 60% of our objective (and more than ten times the potential contribution from Coal + CCS). If we extend the above analysis to solar concentrating technologies with a lower efficiency of 10%, but deployed in regions with higher insolation of about 200W/m2 (which we find in dry tropical or desert areas for example) the same size solar farm would have the capacity to deliver 79kWh/per/day, or about one-half of our objective.


Tidal power options have a wide range of power densities depending upon the nature of the site. First let’s look at in stream tidal technologies (effectively large underwater turbines that work on similar principles to wind turbines). With such technology an average tidal flow of 1m/s would deliver a power density of about 60W/m2. If we could identify 100 x 500Km2 areas around the world that had average tidal flows greater than 1m/s, these projects would contribute just over 1kWh/per/day to our objective. Tidal power may also be generated from tidal barrages, such as that in place at Rance in France. However, the potential global resource is highly unlikely to exceed the contribution from tidal flows discussed above, so we may double the total contribution from tidal technologies to 2kWh/per/day – but it remains a negligible contribution.


The power density of waves in the Atlantic is between 40 and 60kW/m; in the US the EPRI has identified sites in the Pacific Ocean off California with power densities ranging from 20 to 35kW/m. If we assumed we could find sufficient wave resources with a power density of 30kW/m (optimistic) and a mechanical to electrical conversion efficiency of 25% for the wave energy collector (also optimistic, and much larger than current performance of 5%) and if we installed wave power plant off 10% of the world’s coastline, such a giant wave power project would deliver less than 1kWh/per/day, a negligible contribution to our objective. Even if wave power technologies were to reach 50% efficiencies and we were to install them along every meter of coastline, they wouldn’t contribute more than 14kWh/per/day.


Of the 960W/m2 of solar energy that falls on the earth, only 45% of that is photosynthetically active. Of that 45% some of the energy is reflected, some is used by the plant for respiration and some simply fails to perform photosynthesis. As a result, the maximum theoretical efficiency of converting solar energy to biomass is between 3 and 6%. The actual efficiency is further reduced because the sun does not shine all the time, because nutrient levels differ from one region to another, cultivation techniques vary, etc. etc. Therefore, actual biomass power density varies considerably: for switch grass grown in the southern United States it is about 0.92W/m2; for eucalyptus with an annual yield of 20m3/ha it is about 0.7W/m2; and for unirrigated jatropha with a 40% oil content it is about 0.07W/m2. If we use an optimistic power density of 1W/m2 for our biomass and plant our energy crop over an area 1% of the earth’s land surface, taking into account the energy conversion efficiency of the electricity/thermal plant which we can assume to be 50% efficient, our biomass crop would deliver about 1.5kWh/per/day, or less than 1% of our objective. Indeed, if we wanted biomass to make the same contribution as the solar farm discussed above we would need to plant 50% of the earth’s total land surface area. And this analysis ignores the energy that is needed to grow, harvest and transport the crop, as well as the fugitive emissions from land clearing and from the soil.


According to the Nuclear Energy Agency’s Red Book, there is 5.5 million tones of conventional identified uranium resource (which they classify as reasonably assured resources and estimated additional resources costing less than US$130/tonne uranium to extract) and a further 10.5 million tones of as yet unidentified uranium resource (which is estimated additional resources and speculative resources that are as yet undiscovered). There is also an estimated 20 million tones of low grade uranium to be found in phosphate deposits. In current generation light water reactors, 1 tonne of uranium will generate about 50GWh of electricity. If we use such reactors to exploit the identified uranium supply over 100 years they would deliver less than 1kWh/per/day. The new generation of fast breeder reactors (which are yet to be deployed commercially) are about 60 times more efficient than the current light water reactors. Using such technology 1 tonne of uranium can deliver more than 3,000GWh of electricity. Again, if the total identified uranium resource is exploited over 100 years in a new generation of fast breeder reactors, they would have the capacity to deliver 55kWh/per/day, or one-third of our objective. If uranium supplies are indeed much larger than current reserves would indicate (which is probable – though we do not know how much larger they might be) and we were able to exploit both the identified and unidentified uranium and the uranium in the phosphate reserves, then these reactors types would contribute 6kWh/per/day and 380kWh/per/day respectively. Current generation light water reactors therefore have very little to contribute. Whereas, next generation fast breeder reactors, if they can be commercialized, have the potential to contribute a very significant portion of our objectives.


So what can we conclude?

1./ That the challenge to deliver 150kWh/per/day of electricity to 9 billion people can be met – at least in theory. For me this was a surprise. I didn’t expect to get anywhere near these numbers.

2./ Nevertheless, it is still extremely difficult, and in reality we will have to reduce consumption as much as possible. It just makes sense – we should start doing this right away.

3./ The best big picture solutions would appear to be solar and next generation nuclear fast breeder reactors. We need to get to work on these sources right away. In particular we need to work on large scale solar farms and associated transmission lines, developing fourth generation nuclear technologies and on expanding uranium reserves.

4./ The problem is so large that there will always be a role for any low carbon technology. But we have to be honest, some technologies are more important than others. We can forget about wave, tidal, biomass and nuclear light water reactors as big picture solutions – these technologies are simply fiddling at the edges. They will have a role to play in certain circumstances, but they aren’t going to deliver substantial low carbon growth to developing countries. Light water reactors may have a role to play as a stepping stone to fast breeder reactors – but alone their contribution would be negligible.

5./ I’m tempted to say we should forget about Coal + CCS. Its contribution is relatively small and the technology is still on the drawing board. However, in certain countries where coal reserves are abundant (Australia and the US for example) Coal + CCS would be an attractive solution and could make a significant contribution in those circumstances. The same is true of Gas + CCS, which has the potential to make a small, but important contribution to our objective. We should also bear in mind that with a large portion of renewable (high variability) and nuclear fast breeder reactors (low flexibility) in our electricity portfolio we will always have need of the flexibility that coal and gas can provide in the generation mix.

6./ Onshore and offshore wind will have a role to play, but it will be secondary to solar and nuclear fast breeder reactors.

7./ There are some other longer term technology options that I haven’t included here. They also need to be considered: what about lingo-cellulose ethanol? Geothermal? Algae? Fuel cells? We need to look at these as well and I’ll try to add them into this analysis if I can come up with a method for doing so.

8./ The above analysis is focused on the supply side – so there is obviously another piece of work looking at the demand side and here the question is: how low could we get our per capita energy consumption without compromising basic requirements, such as health, security, mobility, etc? I hope to look at this shortly.

9./ The above analysis s also global in scope – I have made no attempt to relate it to a particular country. However, it is clear that each country will approach the supply side problem differently and I think it could be interesting to consider what the analysis would look like if we applied it to Australia, France, the UK and India. This is something for another weekend.

10./ We will be dependant on large scale solar installations and nuclear fast breeder reactors coming on line. If they don’t – if we can’t overcome the technical challenges or aren’t willing to pay the price, or if we simply lack the imagination, energy and political will to make this happen – then we will be in serious trouble.

Finally, what about the technical, economic and resource questions that I’ve put aside for the time being. Obviously, I need to look at these issues more closely, but my initial impressions are:

Technology – I don’t think that I’ve used any pie in the sky technologies in the above analysis. Some technologies such as wind and solar are there now. Others technologies such as wave or tidal are emerging and are at the early stage of deployment. Perhaps the biggest question marks linger over carbon capture and storage technologies and fourth generation nuclear reactors, which are still on the drawing board, but these technologies do not appear impossible.

Economics – In part this is also a function of the maturity of the technology and economies of scale. Having avoided the pie in the sky technologies and noting that these technologies will have to be rolled-out on a massive scale, I don’t believe that the economics will be impossible. But that is not the same as saying that these solutions will be cheaper than current solutions – they will not. And I’m not ignoring the investment that will be required – it will be massive. But in truth these investments and costs are going to be borne one way or another.

Resources – For me, this is the main problem. We may theoretically be able to get sufficient energy from these technologies, but can we get that energy to where it is needed, when it is needed? Not every country enjoys access to the high quality solar, wind, wave or tidal resources that would be needed to meet the above scenarios. Such technologies also place additional demands on the electricity system to balance variable renewable sources and electrical energy may not be the best form of energy in certain circumstances.


One Response to “Is it possible to deliver low carbon energy to 9 billion people?”

  1. 1 Nuclear Power - A Quick State of Play « Think Carbon

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: