Earlier this year author Ramez Naam took me to task for what he calls a lack of context in a piece I wrote comparing the land footprint of solar thermal with that of small modular nuclear.

It’s an interesting discussion. He’s not arguing with my numbers. I’m not arguing with his. Both of us support deployment of nuclear and renewable technologies.

The disagreement appears to be based on whether the land use issue for solar is even an issue.

The context knife cuts both ways on that one.

Naam puts the land use required for a solar United States in the context of the area of the whole United States, coming to a figure of 0.6 %.

This figure is low because Naam uses (as did I) average electricity output per unit land for a system with no storage. It was suitable for my comparison of one facility with another. It is not suitable in a comparison of powering the entire US. In that case, average output and no storage is irrelevant. The worst possible period of output will govern the size and economics of the solar thermal requirements. As Naam says he doesn’t believe 100 % solar is going to happen. Nonetheless, it pays to understand this as many commentators expect a big role from solar with storage.

In the same article, Naam reminds the reader that agriculture roughly uses 30 % of the land of the United States, the built environment is using 166 % of the area that would be required for solar and coal mines are using about the same area as would be required for solar (an interesting quantification, to be sure, which I won’t dispute). National defence areas are raised as another example, and one could go on and on.

There is a serious flaw in this reasoning.

The point was never that the world is literally too small for solar. The point was that land is scarce in the economic definition of the word: it is subject to many competing uses and demands and it must be allocated efficiently. The use that most often gets shafted in our human civilisation is biodiversity. Put another way, we get amazing biodiversity outcomes when we make land near-valueless to humans for anything else. For example Naam highlights disused farm land in the US to assert that the size required for solar is relatively small. The interesting question, surely, is what should we do with this disused land? Give it over to energy production of some form like energy cropping? I would hope not. I would hope it might be returned to habitat as has been the case for New England forests.

Lurching from one land-intensive energy supply to another does not further the land-sparing outcome. The way coal consumes country is horrible as I pointed out in this video. Naam asserts that solar uses the same amount of space, with lesser disturbance. I regard that as faint praise.

Solar thermal won’t work on just any old land. Naam acknowledges that the efficiencies of the system matter. A first-order consideration for economic output from solar is the right area with the best solar resource.

That’s why Ivanpah is in the Mojave Desert, where it displaced an endangered species, not on disused farmland in the eastern United States. Naam’s quantification that a solar USA would require half the Mojave is getting closer to the point. That’s also the reasons why it is on a flat area, not the mountainous Mojave terrain which is much of the terrain. Again, the suitable area is constrained and the relative pressure rises for scarce space.

Pointing out the other (often destructive) ways humans have used land, and the amazing scale of this use, is an argument for constraining our footprint in everything we do from here: agriculture, human habitat and energy to name the big three. Leveraging it to say “therefore this impact doesn’t matter”, well… that’s the sort of corporate, environmental impact assessment logic which time and time again drives the death of a thousand cuts of one area after the next. When an option for massively smaller disturbance is available, as there is in the comparison of nuclear with solar thermal, we should take it. To assert the difference doesn’t matter is a blind spot.


  1. Smil’s “Power Density” book does the numbers on this for the US. He uses a mix of PV/CSP/Wind to provide 320 GW will require ~2 million hectares, biofuels for 1,100 GW will use 380 million hectares and another 78 million hectares to grow biomass to replace solid and gaseous fossil fuels to the tune of 700 GW.

  2. I’m of the opinion that retrofitting solar PV to all suitable existing built environments would be immensely constructive and would not displace or alter living environments- virgin ones or those being remediated/revegetated after prior human use. How much power could be generated by this approach? PV has the advantage of generating some energy during overcast periods, where my understanding is, CST has a less forgiving threshold below which it simply won’t generate electricity.

    1. Yes.

      The proposition for PV on existing structures, right on top of loads, is an entirely different matter to large solar thermal plants. It’s a real plus.

      In my opinion the economic value/utility will be strongly boosted with small amounts of storage to ensure output matches summer evening peak demand. The quantity may remain somewhat modest but the system utility would be very much greater

  3. I came across a story on Pt Augusta’s Sundrop Farms
    which pointed out some differences of opinion by the founders on the use of a winter gas boost. Pt Augusta is often cited as a good location for solar thermal. As with Ivanpah in the US we have to ask if it’s really solar boosted gas not vice versa.

    I recall some early solar thermal thermal plants with molten salt storage were only expected to achieve a capacity factor around 40%. In contrast the new Coonooer Bridge wind farm in Victoria is said to have a c.f. of 48% with no storage such as batteries. Since most cities were once farmland they get plenty of rain and cloud. An s.t. heavy electricity mix will require thousands of km of outback transmission. Solar thermal is like the high maintenance girlfriend and you have to ask if it’s worth the bother.

  4. The solar alternative to thermal with storage is PV + batteries. The most serious trial so far appears to be by Ergon in Townsville Qld
    Townsville has close to 20% PV annual capacity factor despite the odd cyclone. The batteries are up to 12 kwh capacity the size of a family fridge (not wall mounted) and Ergon will replace them if the leases go long enough. Ergon will send phone signals to each module to feed back into the grid when required. That is to displace peaking plant not baseload. The required rooftop PV is about 5 kw. Leases cost about $1100 a year. However I think if you bought your own system a few years ago and got the generous 44c feed-in-tariff you might be better off without batteries until the f.i.t. expires.

    If this trial goes well I think it will be the kibosh for solar thermal as it requires no new areas. If not it could mean solar is just our fair weather friend and not at night or in overcast weather.

    1. That’s pretty sensible in QLD. But still leaves around 5 GW of baseload which is, of course, quite comfortably met by black coal.

      Brisbane hosted a nuclear aircraft carrier last month, as it does regularly, without worry.


      1. It’s remarkable how many people have now drunk the battery Kool Aid. It includes the CEOs of ANZ, NAB and Morgan Stanley banks as well as ex pollie John Hewson. However AEMO disagrees with all of them
        They also look at EV charging and switching gas appliances for electric. Bottom line seems to be the payback period is too long for batteries absent subsidies so will get limited takeup. Also for those who can remember the troubles with pink batts (heat stress and electrocution of installers) I wonder if kids will stick screwdrivers into these big battery cabinets.

          1. OK maybe that’s not the problem. What about big fan cooled boxes sitting on the porch with tropical heat, torrential rain and humidity? If the homeowner is a heavy aircon user I presume the utility Ergon puts a limit on power draw. That is they can only use so many kwh in order to leave plenty in reserve if needed for the grid. Product info http://www.sunverge.com/product/

            1. From my understanding the grid in the battery trial is a single wire and a transformer. The batteries are meant to be used to reduce peak draw. (Also I am not sure the connection would even be able to handle a heavy air conditioner draw anyway)

          2. Of course. Which is why most homes now have high tech safety switches. Bloody good thing to.

            There’s an (apparently) old joke somebody told me the other day: “George died yesterday. ” … “Oh, dear, what did he die of?” … “Oh, nothing serious”. Herein lies the problem of roof top solar. Some who froth at the mouth over the dangers of radiation and nuclear power, don’t consider
            falling off a roof and ending up dead, or worse, as a serious matter. It’s dying of nothing serious.

            Try it next time you are talking
            to somebody … just tell them that rooftop solar is far more dangerous than nuclear power because people fall off roofs. Watch them laugh. But anything involving being on a ladder is dangerous and needs to be treated seriously. Similarly, big batteries have all kinds of safety issues that need dealing with. And I’m sure they will be. But its the little batteries that cause the most harm precisely because people don’t take the risks seriously of button batteries … and kids do swallow them.

    2. Where did the FIT money come from? How would you propose to design the program where a large portion of the population expected that money to arrive in order to pay for their solar power systems?

      Don’t get me wrong, I like batteries. I like them because they offer the potential to make better use of reliable, low marginal cost power plants by providing peaking capability. They would have a much greater utilization rate in a situation where they can be recharged anytime power demand drops below output. Batteries supplied by unreliables can only be recharged if the sun is shining or the wind is blowing, if they empty before the “natural” energy flow comes back, the owners are out of power or forced to rely on fossil fuel generators.

  5. Succinct article and interesting discussion thread. Well done, again.

    The link to AEMO was worth following.

      1. 10km^2 (5 times your report to allow excess)exclusive wind compared with 2km^2 exclusive for nuclear (And if we go offshore wind-farms may actually create marine sanctuaries)

  6. Ben – another aspect of the comparison between the footprint required for solar and that required for nuclear is the margin available for improvement.

    Solar collector space for a given power output will only be reduced by marginal improvements in efficiency. If they were ever able to achieve the ability to convert 100% of the incoming solar energy into useful electricity, they would be at their smallest possible size per unit output. They would still take up a lot of relatively flat land.

    I know many people, including at least one of your commenters, have pointed out that solar panels can be added to the roofs of existing structures. That is true to a certain extent, but like the reality of mountains in the Mohave, most roofs have a substantial portion that is either tilted the wrong way or in use for building mechanical systems.

    When people suggest covering vertical walls with solar cells, I try to help them understand a little bit about geometry and the relationship between the arrival angle and the energy that can be collected.

    In contrast, if lack of space is a real consideration, nuclear plant designers have a lot of room to maneuver. They don’t have to be spread out to collect their energy from a diffuse source. Take a hard look at the visible portions of a US aircraft carrier, for example. Then consider that ship contains two powerful nuclear plants that provide all of the propulsion power the ship needs plus all of the electrical power, aircraft launching power, and has enough left over to produce fresh water for 5,000 or more residents.

    The nuclear power plants are in the ship’s “basement.”

  7. Maralinga A-bomb tests were shown on two SBS TV shows last night ‘Uranium; Twisting The Dragon’s Tail’ and ‘Fallout’. It seems spooky that the huge Olympic Dam deposit was discovered on the edge of the exclusion zone 20 years later. For SA to get into an expanded nuclear fuel cycle will be a natural progression.

Leave a Reply

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

WordPress.com Logo

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

Google photo

You are commenting using your Google 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 )

Connecting to %s

%d bloggers like this: