“Nuclear power would not be worse for our water security than coal. It would be approximately neutral, because the consumption is basically the same. But might it actually be better?”

This post is inspired by a message that got back to me from my Mum, that a relative of mine (who shall remain nameless. Hi Auntie Jackie!) will “never agree with me because nuclear power plants use so much water”.

It’s an important point. In Australia, we can never take water for granted. The most recent and very severe drought pushed water supplies to the brink around the country. Investing heavily in a new power source that would stretch those supplies even more would seem a risky thing to do.

So this question is worth examining. In the mix of Australian energy generation, would nuclear power be:

  • Better for our water security?
  • Worse for our water security?
  • Neutral for our water security?

I’m going to answer that question by comparing it with coal fired power generation, which is the most prevalent form of generation in Australia, and the form I want to displace quickly with nuclear power.

To start the answer, it’s worth understanding a little about nuclear power generation. Basically, it’s a thermal electricity generation method. This means you have a heat source, you add water to create steam, draw the steam through a turbine to generate electricity, then condense the steam back to water and repeat the process. This is the way it works for coal, gas, nuclear, and concentrated solar thermal electricity generation. The principal difference among these methods is the heat source. Fossil fuel plants burn large amounts of fossil fuel to create the heat (producing large amounts of greenhouse gas and other pollution in the process). Concentrated solar thermal uses concentrated sunlight, (producing no direct pollution to speak of, but taking up a lot of space and unfortunately costing a lot of money). A nuclear plant uses a nuclear reaction for heat (producing a small amount of waste which is captured and stored).  Aside from that, the basic mechanism doesn’t change.

So there is water being used in two ways. Firstly to create the steam. This water is pretty much in a closed system. It is turned to steam, run through the turbine, condensed back to water, and then recirculated. So that water consumption is not really of concern.

The second way is to provide the cooling for the condenser. The condenser is, simplistically speaking, a metal barrier on the far side of the turbine from where the steam is coming. By keeping it cool with water, a temperature gradient is created between the two ends of the turbine. This acts like a vacuum to suck the steam through the turbine at speed and create the electricity. This is where the water consumption occurs. The water that is used for cooling the condenser evaporates, which is what we see leaving the power plants chimneys as steam. Allowing for better and worse, older and newer designs, because the mechanism is the same whether the heat source is fossil, nuclear, or solar thermal, there is no substantial difference between the quantity of water consumed per unit of energy created; the same job needs to be done.

To that end, we have a partial answer to the question. Nuclear power would not be worse for our water security than coal. It would be approximately neutral, because the consumption is basically the same. But might it actually be better?

Water itself is not scarce. There is more water on this planet than we know what to do with. Fresh water on the other hand, is very, very scarce indeed. It’s less than 3% of the total water on the planet, and a good deal of that 3% is captured in ice. Come to Australia, especially the south, south west and south east, and that water scarcity is regularly on display. Fresh water is way more useful to us than sea water, for drinking, industry and agriculture. Our water security is determined by our supplies of fresh water. Provided we can tolerate some localised impacts (like brine discharge from desalination processes), we can use sea water to our heart’s content without worrying about water security.

So that’s another potential difference between the power sources that we need to understand; what type of water can be used for cooling: fresh water or sea water? The answer is less technical than you might think. It depends on where the plant is.

The 2.2 GW operations of the Loy Yang coal fired power plant consumes 60,000 tons of coal per day. There is no way of moving that much fuel economically. So where do you build the power plant? Next door to the mine. The only water that is available for cooling in the middle of Gippsland is fresh water. Lovely fresh water that could be used for human consumption, agriculture, or river health is instead evaporated in a coal fired power station. For Loy Yang, the quantity of fresh water it consumed in 2009 was 37 GL (Sustainability Report 2009); 14 GL from aquifers, 25 GL from the La Trobe River, and the remainder from the Maroondah Reservoir. Is that a lot? Well, in 2006 Melbourne as a whole consumed 369 GL of water. So the coal power station consumes approximately as much water as 360,000 people  in a modern Australian city(1/10th the 2006 population of Melbourne). So yes, it’s a lot. You can repeat this exercise for the Hunter Valley coal region in NSW, or Queensland or South Australia’s coal generation centres. [Author’s note: I have since learned and shown in another post that South Australia does in fact transport a very large amount of coal a very long way and use sea water for cooling. It merely requires the longest coal train in the world…]

Uranium on the other hand, is a super dense energy source. The quantity of fuel required  for a nuclear plant is miniscule. Some modern designs are fuelled up, started, and refuelled 4.5 years later. This is a big advantage.  It means you can put the plant more or less wherever you want, such as closer to the centres of demand, which reduces the wastage of electricity that is lost in the transmission lines. In Australia, it also means we can put them in reasonable proximity to the ocean, with some sensible buffer from storm surge and tsunami, and cool the condensers using pumped sea water.

The implication for water security is suddenly clear. Displacing all coal fired power stations for nuclear power stations would lead to a dramatic improvement in Australia’s water security by freeing up huge quantities of scarce fresh water for more beneficial uses. It is yet one more criteria in which nuclear trounces coal for sustainability. So thanks for the question Jackie. Australians need to know the answer.

24 comments

  1. A nuclear power plant uses heat to boil water to drive a steam turbine to make power.
    A coal power plant uses heat to boil water to drive a steam turbine to make power.
    A solar thermal power plant uses heat to boil water to drive a steam turbine to make power.
    A combined cycle natural gas plant uses waste heat from a gas turbine to boil water to drive a steam turbine to make power.

    They all make power from a steam turbine, and they all use roughly the same quantity of water per unit energy produced by the steam turbine.

    There’s nothing special about the nuclear plant in this regard, other than the siting and other advantages Ben described.

  2. All good points made in this article.
    However,when siting any structure near salt water it is well to remember that we have an increasing rate of sea level rise due to global warming.This rise in now built into the system and will only respond slowly to reductions in greenhouse gas emissions.And we are not doing too well in that regard are we?

    That said,a nuclear power station near the ocean or estuary will be able to run a desalination plant with the excess power generated off peak.So,in that sense nuclear will enhance the water supply,not diminish it.

  3. “It means you can put the (nuclear) plant more or less wherever you want, such as closer to the centres of demand” i.e. near population centres. And that’s the problem – nobody wants to live and work in close proximity to a nuclear power station, and when things go wrong, such as at Fukishima, tens or hundreds of thousands of people need to be evacuated for long periods of time, at enormous financial and human cost.

    “Concentrated solar thermal uses concentrated sunlight, (producing no direct pollution to speak of, but taking up a lot of space and unfortunately costing a lot of money).” Nuclear is also very very expensive. Very keen to see some accurate, verified costs for nuclear so we can compare the technologies better.

  4. Andrew,presumably people don’t mind living in close proximity to a coal fired generator.You surely know that pollution from coal of many different types is much worse than nuclear could ever be.

    Your Fukishima argument and the cost of nuclear are all standard,off the cheat sheet arguments from the renewable lobby.They are no more valid now than they were 10 years ago.

    Let’s see you produce some accurate,verified costs for solar thermal and wind,including the massive grid expansion and provision of a huge amount of storage.
    And,at the end of the day you will still not have a reliable,despatchable power supply.Unless you have backup from fossil fuels.
    That sort of defeats the object of the exercise,doesn’t it?

    1. I do appreciate that the proximity issue of nuclear, though I agree with the rebuttal from thirra, remains a serious barrier and there is no point being naive about that. I choose to address it though, not accept it. Realistically, in SA , some compromise position will probably be reached as we have the luxury of space. I just hope we don’t incur too much unnecessary cost and waste by dramatically over reacting to peoples fear and ignorance instead if building strategies to tackle it.

    2. “Andrew,presumably people don’t mind living in close proximity to a coal fired generator.” That’s not what I said at all (although I do have experience with residents around a large coal fired power station – they seem to be more ingratiated with the plant’s employment opportunities than worried about its emissions, which are horrible and are uncontained).

      “Your Fukishima argument and the cost of nuclear are all standard,off the cheat sheet arguments from the renewable lobby.They are no more valid now than they were 10 years ago.” Oh please. Cheat sheet … renewable lobby? My point about people being evacuated from Fukushima isn’t valid? How many people have been (and still are) evacuated from the area around Fukishima, for how long and at what cost? And what cost will the Japanese society bear to limit TEPCO’s financial liability? Billions.

      My point is that when things go wrong with nuclear plants, which they don’t do often, but they do occasionally, the risks and the costs can be enormous. If you seek to gloss over this fact, you’ll struggle to convince a dubious public about the upsides of nuclear.

      “Let’s see you produce some accurate,verified costs for solar thermal and wind” No problem – that’s easy. And you do the same for nuclear. Make sure you include all the hidden costs which the public ultimately bears, such as limits to financial liability offered to the nuclear industry by governments (eg http://en.wikipedia.org/wiki/Price%E2%80%93Anderson_Nuclear_Industries_Indemnity_Act).

    3. Three more points:

      “Let’s see you produce some accurate,verified costs for solar thermal and wind,including the massive grid expansion and provision of a huge amount of storage.” Several people have alluded to ‘massive costs’ associated with grid expansion and backup generation required for renewable generators. Up to a certain level of renewable generation, this argument does not hold.

      Viable sites for renewable projects are partly chosen based on the availability of grid capacity (and sufficient resource), and don’t require grid expansion. Once that capacity has been used, the argument stated above becomes valid, as grid needs to be expanded to allow further generators (renewable or otherwise) to connect to the grid.

      On the issue of backup generation, the generally accepted rule of thumb is that up to a wind energy penetration of approximately 20% wind, the intermittency issue is manageable without additional backup generation. A greater penetration is possible where electricity grids are meshed with other grids (eg through interstate or inter-country interconnectors), so excess power can be exported, or power shortfalls can be imported. This is the case in Denmark, which is highly connected with the rest of Europe.

      “And,at the end of the day you will still not have a reliable,despatchable power supply.” A point of semantics: renewable generators are not unreliable. They are intermittent. There’s a big difference.

      1. Some very good points Andrew. I am valuing you participation, and I understand those penetration levels to be about right; I think the Essential Services Commission of SA had quite a bit to say about incorporating higher levels of wind in a report a couple of years ago. As I understand it though, the transmission infrastructure in the areas where these resources can be found is pretty Spartan, and reaches capacity quite fast, bestowing a major first mover advantage for developments. Coordinated investment in bigger transmission for areas of high resource, with cost that can be spread evenly over, say, 10 new developments, seems a smarter, fairer way to go that will attract more sustained growth in renewables. But then we are incurring some major additional costs. As for the interconnection… well, I think Denmark needs to shift some pretty massive quantities of unwanted energy when the wind is really howling at little or even no cost. That’s not necessarily a financially wise solution. Frankly, I think it still remains very clear that a renewable-only strategy is enormously high risk verging on impossible given what is at stake, and really quite unnecessary given that nuclear is actually a great technology in practically every respect. I would rather see governments invest a tiny fraction of the money required to built transmission on bringing Australian’s around to these facts about nuclear, and get the decarbonising job done faster and simpler. But I think your input is going to help in understanding the degree to which renewable will contribute further to the outcome Decarbonise SA is seeking.

        Ben Heard Director

        ben.heard@thinkclimateconsulting.com.au M- 0411 808 202 W- http://www.thinkclimateconsulting.com.au

      2. Ben, in reply to a couple of your points:

        “As I understand it though, the transmission infrastructure in the areas where these resources can be found is pretty Spartan, and reaches capacity quite fast, bestowing a major first mover advantage for developments.” That really depends. In South Australia’s case, there is significant wind resource on the Eyre Peninsula but there is no significant grid capacity left there. Further north, there is significant solar thermal, solar PV and geothermal potential, but limited or no grid. To address this, there is a proposal called “Green Grid” to expand the grid on the Eyre Peninsula to allow several thousand MW of new wind farms to potentially be built there. Ultimately Green Grid may extend northwards and include a new interconnector to the east coast, which would allow solar thermal, solar PV and geothermal projects (and nuclear too, if that ever happens) to export their energy from South Australia to east coast consumers.

        “Frankly, I think it still remains very clear that a renewable-only strategy is enormously high risk verging on impossible given what is at stake, and really quite unnecessary given that nuclear is actually a great technology in practically every respect.” I don’t regard renewables as high risk, but I don’t believe that renewables will become the sole source of generation in South Australia, unless a lot more changes than just generation.

        The existing grid has been designed around the “centralised generation / consume as much as you want whenever you want” approach, and you’re essentially proposing to continue that approach, but to replace fossil fuel generation with nuclear generation. An alternative proposal is to distribute / decentralise generation, and to become much smarter in how and when we consume energy. Both approaches have benefits and downsides.

        Ultimately it may not be an ‘either/or’ outcome: it may be both.

        1. Yes, that discussion of the Green Grid is pretty much what I was talking about as the solution. I have no issue with it, except for significant cost that doesn’t need to be borne by technologies that do not have critical dependency on location. It may indeed be both though, that would be fine, especially since I want to see electric vehicles too which can have a nicely synchronous relationship with wind power. We just can’t afford to focus on Green Grid at the expense of just solving 90% of the problem by replacing the fossil fuel baseload with nuclear.

          Using renewables as the centrepiece of efforts to decarbonise is a high risk strategy if one understands, as I do, that we have no time to waste and cannot afford to fail. Yet at present, that’s all Australia is considering, and failing is exactly what we are doing. 1,000 MW of installed wind in SA has coincided with about a 25% increase in emissions from electricity generation in the same period. That’s a failed strategy. It is not up to the task. It needs support from nuclear to get the job done.

          “The existing grid has been designed around the “centralised generation / consume as much as you want whenever you want” approach, and you’re essentially proposing to continue that approach, but to replace fossil fuel generation with nuclear generation.” Absolutely and I make no bones about it. We don’t have time to waste and I want South Australians to be happy to just give the plan their support, and get on with their business. “An alternative proposal is to distribute / decentralise generation, and to become much smarter in how and when we consume energy. Both approaches have benefits and downsides”. Do that too, but let’s be honest. The first approach, alone, would work if the goal was rapid decarbonisation while remaining essentially neutral about lifestyles and the underlying system itself. The second approach, alone, would not work and is decidedly not neutral about lifestyles and the underlying system itself, and that creates loads of friction. It would fail over a long, slow period of time while emissions kept climbing; a repeat of the last 20 years in SA. We have 9.1 million ton of CO2-e from electricity every year that we need to quickly cut. 8 million of it is from baseload. You cut it quickly by replacing the baseload with zero carbon sources.

          Ben Heard Director

          ben.heard@thinkclimateconsulting.com.au M- 0411 808 202 W- http://www.thinkclimateconsulting.com.au

  5. “Let’s see you produce some accurate,verified costs for solar thermal and wind” No problem – that’s easy.

    err… where are they? Not trying to being facetious, I’m genuinely interested.

    1. Large scale wind power costs between $1.8m and $2.2m per MW (installed). The Snowtown Wind Farm, 2 hours’ drive north of Adelaide, is a 98.7MW project and cost $220m (http://wotnews.com.au/like/trustpower_opens_snowtown_wind_farm_in_australia/2689105/). It was commissioned in 2008 and would be quite a lot cheaper today.

      Large scale solar thermal is somewhat in its infancy, but costs between $3.5m and $5.5m per MW (installed). We’ll have some definite figures for a large scale solar thermal project in Australia in the next month or two, when Solar Flagships program projects are announced.

      1. Hmm, That’s pretty expensive. I’m simplifying here but with a 30% capacity factor you’d need to build 35 of these wind farms at an eventual cost of $7.7bn just to equal the output of your average 1000MW FF plant, and that doesn’t even take into account the necessary storage costs (pumped hydro?) to smooth intermittency or extra transmission lines to each farm, both of which could add billions. That is not very encouraging.
        What kind of capacity factor are they getting for the solar thermal (with salt storage?)?

  6. I plan to address costs in a more structured post soon. Where possible, can we try to provide costs in terms of Levelised Cost of Electricity? I really think that is the most suitable metric for comparing energy sources. Talking in terms of MW installed leaves a lot of work to do accounting for capacity factors/ fuel costs/ other issues before you know whether something is actually going to provide electricity at a good price. In the absence of that, I accept MW installed is good information to start with.

    My solar thermal costs are pretty close to Andrew’s ($3m-$6m MW) based mainly on what is proposed for Blythe in California, but I’ve not had much luck in getting a better estimate than that. Only a 24% capacity factor expected though. You can imagine the difference that makes.

    The recent paper in the peer reviewed journal Energy by Nicholson, Biegler and Brook (yes, that one), is a must read for this topic, and very useful having been based on a meta review of costs across 25 different studies.

    1. Thanks Ben, I posted mine before I saw yours. Will try to stay OT. That said, 24% for ST is very disappointing!

      1. Hi Marion. Yes it is low isn’t it? The figure is taken from Commission Decision report of the California Energy Commission, September 2010 for the Blythe Solar Power Project. It’s 1,000 MW, 6 acres/MW (which is apparently pretty good), for 2,100,000 MWh per year. I get a 24% capacity factor out of that. For between $3bn-$6bn, I’m not really liking the value proposition there. I will be looking at costs specifically soon, but to date it seems that of zero carbon power sources that can provide reliable baseload power at some sort of scale (being hydro, nuclear, and solar thermal), hydro is a non-consideration for South Australia, and in cost terms there is quite a bit of daylight between nuclear and solar thermal.

        Ben Heard Director

        ben.heard@thinkclimateconsulting.com.au M- 0411 808 202 W- http://www.thinkclimateconsulting.com.au

    2. This article is very interesting: http://climateprogress.org/2011/04/06/does-nuclear-power-have-a-negative-learning-curve/

      To quote a section:

      “Why did (nuclear) costs escalate? Why was there “negative learning.” He offers this theory:

      … with increasing application (”doing”), the complexity of the technology inevitably increases leading to inherent cost escalation trends that limit or reverse “learning” (cost reduction) possibilities. In other words, technology scale-up can lead to an inevitable increase in systems complexity (in the case of nuclear, full fuel cycle management, load-following operation mode, and increasing safety standards as operation experience [and unanticipated problems] are accumulating) that translates into real-cost escalation, or “negative learning” in the terminology of learning/ experience curve models. The result may be a much wider cost variation across different technologies than so far anticipated.”

  7. Should mention, I don’t have a link for the energy paper, but if you contact Barry Brook through Brave New Climate he will send you a PDF.

  8. The other thing about water is, which is part of this discussion that Ben and Andrew Dickson have both alluded to, is where it is is pretty important. It’s expensive stuff to pump around in any great volumes.

    While I would have thought that the northern SA desert would be relatively cloudless (is it?) and good for solar thermal in those areas, hopefully reaching better than 0.24 capacity, that’s also an area lacking water*. Try to move solar thermal closer to adelaide and you run into space and land cost constraints long before you get close to decent water supplies. For a nuclear power plant, that’s not such an issue, they have been successfully sited in major cities around the world.

    While Andrew Dickson rightly says “but nobody would ever want to live near one of them!”, that’s actually the whole raison d’etre of this website, to try to puncture some of the myths and fear surrounding this relatively benign technology.

    I wouldn’t want to live near a nuclear power plant either, but that’s strictly an aesthetic decision, I don’t want to live near a Bunnings or any large ugly industrial-looking thing. The almost fanciful risk of radiation does not faze me one iota.

    * Oh another thing I just remembered – CST and PV both in fact require more water than standard steam turbine operation. That’s because if they’re to provide anything close to their capacity, they have to be washed every week or two. Using water of course. The dust does bad things to their power.

  9. “While I would have thought that the northern SA desert would be relatively cloudless (is it?) and good for solar thermal in those areas, hopefully reaching better than 0.24 capacity, that’s also an area lacking water.” That’s true – generally anywhere north of Port Augusta has good enough solar insolation for solar thermal plants, and there are a number of sites with sufficient grid and water supply.

    Capacity factors for good solar thermal plants are around 30% (and for a good wind farm in South Australia, high 30s).

    “Oh another thing I just remembered – CST and PV both in fact require more water than standard steam turbine operation. That’s because if they’re to provide anything close to their capacity, they have to be washed every week or two. Using water of course.” It is true that solar mirrors/arrays need to be cleaned, but it’s easy to overstate the washing and water requirements. Good solar sites aren’t necessarily particularly dusty, and water requirements for mirror/array washing are pretty minor.

    1. Just strictly confining ourselves to the water matter, do you think it’s true that on a per megawatt delivered basis, CST isn’t terribly water efficient, i.e. this is an issue for this technology?

      1. Wilful, I’m not sure that CST is much different from other thermal generation sources. Sure, it uses some water for mirror cleaning, but that’s a relatively small volume. Large volumes of water are required for condensing steam.

        An alternative approach to condensing steam is using “dry cooling”, which essentially uses large fans to blow air over the steam tubes to achieve condensation. This requires a lot of baseload power (eg 30MW for a 700MW thermal plant).

        So cooling is a choice: it’s either a lot of energy for dry cooling or a lot of water for wet cooling. CST (and presumably nuclear) can use either. If there’s water available, generally wet cooling is preferable.

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 )

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