Bradfield in reverse

Australians have long been captivated by the idea of turning coastal rivers back, to irrigate the dry inland. The most famous advocate of such a scheme was John Bradfield, designer of the Sydney Harbour Bridge.

Such schemes have always foundered on the ugly physical fact that water is heavy. Any significant amount of uphill pumping is prohibitively costly, so schemes of this kind require lots of expensive tunnelling, and can only get access to water from the upper slopes of the Great Dividing Range.

As coastal populations have grown, we’ve seen increasing interest in the reverse option, of taking inland water to the coast. Adelaide is already buying water back from irrigators on the Murray. Current Victorian policy prohibits Melbourne from doing the same, but it’s hard to see this ban being sustained as water restrictions are tightened, and the technical difficulties are not great.

Today’s Australian has a proposal from Dennis O’Neill of the Australian Council for Infrastructure Development to take water from the Tantangara Dam on the Snowy to Sydney, via Googong dam near Canberra. This sounds a lot more expensive than the options for Melbourne and Adelaide, though not in the same league as Colin’s canal. And at least the scheme works with gravity rather than against it. I’ll look forward to seeing a more detailed proposal.

All of this will certainly make for interesting times for irrigators on the Murray-Darling, the main focus of my research and modelling. On the one hand, those who are able to sell water rights stand to do very well. On the other hand, those wanting to buy will be competing with the swimming pools and gardens of the big cities (the uses currently subject to the tightest restrictions).

20 thoughts on “Bradfield in reverse

  1. Details of the original ACTEW proposal can be found here.

    ACTEW are talking about transferring purchasing about 20 GL per annum, which, after losses, would augment their supply by 15 GL pa.

    If that 15 GL pa were transferred to the Wollondilly River at Goulburn, and thence to Warragamba, additional losses would reduce that amount to something less than 10 GL pa. Sydney’s current usage is about 630 GL pa, with current supplies of around 600 GL pa. The extra 10 GL pa from the Murrumbidgee might help a bit, but it’s not really an answer to the problem.

  2. Thanks for this, SJ. It sounds as if the 20GL is determined by the presumed needs of the ACT rather than being an inherent constraint. Still, it’s hard to see the volume being increased greatly, and I don’t imagine it would be economic to take water out of any storage further south than Tantangara.

    On a very quick scan, I can’t see the cost of water to Sydney coming in under $1.50/kl which is about the point at which desalination becomes competitive. More on this later, I hope.

  3. “On the other hand, those wanting to buy will be competing with the swimming pools and gardens of the big cities (the uses currently subject to the tightest restrictions).”
    True and the first State Opposition to advocate the price mechanism for distributing water, rather than the policy stance of the current quantity control freaks, will romp it in. Assuming of course they advocate the cheapest source, rather than costly canals, tunnels and pipelines.

  4. John:

    The proposal probably makes more sense for Canberra and/or Goulburn than it does for Sydney. Those cities don’t have the option of desalination. However, Canberra and Goulburn have viable options other than inter-basin transfers using costly pumps and pipelines.

    Bottom line: I don’t think the proposal will get a guernsey in the near future.

  5. Mind you, I’m not saying that inter-basin transfers are always unviable.

    There are, I think, four of them in current operation in NSW.

    – The Snowy scheme, which transfers water from the Snowy river to the Murray and Murrumbidgee Rivers (about 1,100 GL pa)

    – The Shoalhaven scheme, which transfers water from the Shoalhaven River to the Wollondilly and/or Nepean Rivers (up to 280 GL pa)

    – The Barnard River scheme, from the Barnard/Manning Rivers to the Hunter River (up to 20 GL pa)

    – The Fish River scheme, from the Fish River to the Blue Mountains (Grose/Hawkesbury Rivers, up to 14 GL pa)

    The second and third of these schemes were built in the late seventies/early eighties, and sat largely unused until quite recently. Their value was rediscovered in the current drought.

    Some out-of-date info on inter-basin transfers can be found here. (535 KB PDF file)

  6. Dennis O’Neill assumes we want to continue to centralise our population in the Sydney Basin.
    Surely some thought must be given to the development of population centres elsewhere.
    Burrendong dam can hold about 60% of Warragamba’s capacity.
    With an upgrade of rail and road connections to the central west of NSW, Bathurst/Orange become as close to Sydney as Newcastle/Woolongong.
    Enhanched development would extend to all regions west of the Divide where lifestyle would be at least as good as in the alternate western suburb sprawl[ my opinion].
    The citycentric Governments we are now blessed with must, at some time, see the practical advantages of decentralisation.

  7. “Burrendong dam can hold about 60% of Warragamba’s capacity. ”

    But it’s currently only about 14% full, indicating that it can barely cope with the demand it already has.

  8. SJ
    It is only 14% full as it makes economic sense to use your allocation and grow an annual crop in the current year as the dam may fill for the next year. Annual cropping is a different beast to the perennial needs of a city population.

    I am assuming that water allocated to cotton etc would be reallocated, via the market, to higher value domestic/industrial use. The changed status water would entail a greater buffer being kept in the dam for the perennial needs of population centres.

  9. The new principle seems to be those nearest the source have the strongest claim. I guess that’s why insurgents blow up oil pipelines to stop the wrong people getting the oil. Since I presume Tantagara is networked to the Snowy scheme that would be another diversion after the Snowy River giveback. Remember Adelaide with one million people is 60% dependent on the Murray. I also think the Murrumbidgee downstream of Canberra vitally needs the treated effluent in dry times so we already have major water re-use. Perhaps a chunk of Sydney could relocate to the Ord River basin.

  10. Long term water security plans for coastal cities, inland towns and agriculture has to make assumptions about future energy prices. John says water is heavy which means it’s energy intensive and vulnerable to energy price rises. I don’t think anyone seriously expects to ever again see cheap prices for oil, gas or coal, regardless of whether oil production peaked last year or will, as the French Government estimate, in 2013.

    Neither will renewables of adequate scale be cheap. The real costs of nuclear power are unaffordable even making the unreasonable assumption of no security issues or accidents in generation or waste disposal.

    Long distance movement of water other than by gravity can only be a temporary solution. Perhaps real water security depends on local capture, use and re-use and levels of efficiency orders of magnitude higher than at present. Which gets downs to better price signals and public policies.

  11. You need to put some numbers to it to really get a feel for whether the energy costs are significant or not.

    A kilolitre of water weighs one thousand kilograms, and it takes 0.27 kWh of energy to raise a thousand kilograms to a height of 100 m.

    In the Shoalhaven scheme, mentioned above, the water has to be lifted about 850 m. That takes 2.3 kWh for each kL. Assuming that the pumping and other losses make the process only 50% efficient, we’re looking at 4.6 kWh. If we can get off-peak electricity for 6 c/kWh, each kilolitre costs about 28 c in energy.

    For the ACTEW proposal, the difference in elevation is only about 120 m, costing around 4 c/kL in energy.

    Compare that to the $1.50/kL price John gave for desalination.

  12. For Melbourne (and would guess for Sydney and Canberra) there are so many free kicks in the system in terms of water efficiency that require pretty modest asks of the population, that much of the problem is a chimera.

    Compulsory low-flow shower heads and front-loadign washing machines wouldn’t lose any votes or cost any jobs, but would deliver ~30% savings with a snap of the fingers. And just install third pipe recycled water and/or rainwater tanks for garden water and you’re already beyond the savings we are likely to need.

  13. SJ is correct: desalination is currently more expensive than the pipe and pump options, especially if inshore brine dumping is allowed. Relative costs will depend on which costs and impacts are internalised and the scope for accesing cogenerated or baseload energy.

    My general point is that energy intensive options are not good considering where we’re getting our energy from, and worse if its nuclear. And the real costs of energy seem likely to rise, even without inernalisation of externalities.

    Perhaps comparisons of long term strategy should be based on resource intensity rather than current dollar costs which are less predictable especially long term, than resource availability.

  14. funding is given as main reason for inactivity, study the present socio-econom. fraud and see that under S83, govs can appropriate any requirement, and under S51.13, money can be costless.
    signed, secretary.

  15. Homer, I understand the point you’re getting at.

    I don’t know at the moment what proportion of the cost of water from a desalination plant is attributable to energy cost. I’ll see what I can find out.

  16. OK, according to this paper (580 KB PDF file), the typical process uses 5.36 kWh/kL.

    Assuming the same 6 c/kWh I used before, that means an energy cost of 32 c/kL, which isn’t all that much more than the 28 c/kL for water pumped from the Shoalhaven River.

    The result is a bit surprising. I expected that the energy cost of desalination would be higher.

  17. Thanks SJ, that’s interesting. The 5.36 kWh/kl probably assumes inshore brine dumping. If the politics go against this I understand you can assume much higher energy (and capital costs) to pipe it to deep water.

    I’m inclined to think desal makes more sense as a peak demand or emergency response, not base load eg. Wivenhoe Dam system provides 80% of the water in SE Qld and is vulnerable to blue green algae which could take it off line at short notice especially as it nears depletion.

  18. Gentlemen you all make a lot of sense of the subject you are discussing, however please don’t lose sight of the effect of daming our northern rivers.
    Flood mitigation alone of our northern east coast brings benefits to all forms of agriculture. This fact alone improves the lot of employers / employees and subsequently the country’s G.D.P.

  19. Gentlemen please don’t lose sight of the forrest because of the trees.
    By daming our far north rivers we have the immediate benefit of flood mitigation for all east coast agriculture. This provides full time employment rather than stop / go as currently exists due to flooding with the obvious reduction of unemployment benefits, increase in taxes paid to government and better G.D.P.
    Having said that, if these dams could be gravity fed to west of the divide as John Bradfield surveyed that they could, all the above benefits would be made available to the beef industry and others west of the divide.
    Benifit three would be adding water to our inland river system not wasting water to the ocean as at present. Each stage would benefit. It does not rely on the total projec before benefits are seen.
    If only governments would bite the bullet!

Comments are closed.