Pumped hydro

In my Conversation article on the Turnbull government’s plan to keep coal-fired electricity alive, I said that most of the opportunities for hydro-electric power had already been exploited. I was thinking of primary power generation, and in this respect, I maintain my view. However, I neglected the option of pumped storage, where water is pumped uphill when excess electricity is available, then run downhill through turbines to (re)generate the electricity when it is most needed.

My old university friend, Andrew Blakers, now with the Research School of Engineering at ANU emailed me to point out this study, looking at the large number of sites potentially available in Australia, more than enough to backup all the renewable energy we will be generating in the foreseeable future.

This isn’t just a theoretical proposition. The Kidston hydro storage project in the advanced stages of planning, will offer 2000MwH of storage combined with a co-located 270MW solar PV project. The same report mentions some big wind + storage projects.

Still, if Labor is silly enough to endorse Turnbull’s NEG idea, it’s hard to see any more progress being made.

34 thoughts on “Pumped hydro

  1. Didn’t JQ leave CSP with hot salt out of the storage portfolio? It’s conceptually messy: the storage only makes sense tied to an intermittent CSP thermal generator, which constrains the size and location, but within that the developer can pick the optimum amount of storage. The typical scale falls between grid batteries and pumped hydro. So, importantly, does theconstruction time.

  2. Capital Costs dominate the cost equation in all renewables technologies including pumped storage. For long-term assets, the discount rate dominates the cost equation. Instead of using a discount rate in the calculations we can take the capital cost and divide by the life of the asset to get the cost of capital. As pumped storage capital costs are dominated by the cost of assets that last at least 100 years the cost of pumped storage will remain cost competitive for at least 100 years because the longer we keep using it the cheaper it gets.

    The same logic applies to all Capital Expenditure. Using it removes the cost of interest. We can still give investors a return on their money by paying them back more money. So for a 100-year asset give investors an inflation-adjusted annuity for 15 years of twice as much invested and the capital cost for the next 85 years is zero. If you pay back in half cost electricity the business case becomes even more compelling.

  3. Hydro in most regions of Australia, whether pumped or river flow fed, will be constrained during persistent drought conditions. Pumped systems may go longer in big dry’s but they would not be immune.

    Sea water based systems near to coast would probably be the more reliable alternative, although having it’s own limitations, including different and probably more expensive requirements for materials used in pipes, pumps and generation turbines as well as more limited geographic locations. Wind and solar are the current success stories but some others are not entirely out of contention, including geothermal and tidal – improved cost effectiveness of those may impact the cost effectiveness of batteries and pumped hydro. Even for transport there may be improvements in electrically produced chemical – ie liquid or gas – fuels, like H2, methane or ethanol – that can compete with batteries.

    I think storage of all kinds will see innovations and improvements but the end game for the last 20% of electricity requirements is still too distant for any kind of certainty. Up to around 50% isn’t that problematic; it is 50-80% that is the more immediate and looming challenge. Whilst using methods that are scalable past that is desirable I don’t think it can be a major consideration for our nearer term decision making.

  4. JQ refers to Andrew Blakers report identifying some 22,000 potential sites in Australia for pumped hydro. To bring some of those to operation would be something of a project, and those whose first reaction is cost might get the vapours.
    However, in 1950 Australia, population ~8 million, embarked on the Snowy scheme at a projected cost of $820 million. In 2017 figure that is around $8.4 billion. If we extend that number by three times to reflect a population of 24 million, then the per capita total would be ~$25.2 billion. Not such a huge amount compared to the NBN at $40+ billion. If we add in Kevin Cox’s (@#4)treatment of capital cost pumped hydro gets even better.
    I can’t say what we might get for that kind of money in terms of reduced emissions. Hopefully though those possibilities will be explored.

  5. @Kevin Cox

    Yes, in the report linked above by Fred the tables of payback times for various scenarios suggest that in only a decade or so the costs of building a pumped hydro facility would be paid off. After that the costs over the next century or so would merely be a spot of periodic maintenance, no? This, it seems, ought to bring the costs of pumped energy storage during that lengthy period to a point near to being too cheap to meter.

  6. Australia’s peak grid demand is about 35 gigawatts late in the afternoon and early in the evening during summer heatwaves.

    Australia currently has approximately:

    9.4 gigawatts of gas capacity.
    7.9 gigawatts of hydroelectric capacity.
    2 gigwatts of pumped hydro storage.
    .85 gigawatts of diesel and kerosene capacity.
    .4 gigawatts of firm wind capacity.

    This comes to about 20.6 gigawatts and leaves the grid around 14 gigawatts short of meeting peak demand if coal power is eliminated. Increasing solar capacity can help with this a little, but sunset is a pretty firm barrier as far as direct solar generation is concerned. Demand management, which is nothing new in Australia, can also help. But we’d probably still want at least 12 gigwatts of dispatchable generating capacity to meet evening heatwave demand.

    If we built another 10+ gigwatts of wind capacity we could knock it down to say 11 gigawatts.

    11 gigawatts of gas turbines would cost around $4 billion if we don’t get anything fancy. Money would have to be spent maintaining them and gas would have to be paid for, but while Australia’s gas capacity would more than double, we wouldn’t necessarily increase the amount of natural gas we burn, as the the gas turbines would mostly stand idle.

    Pumped storage is an option, but isn’t cheap with Snowy 2 costing around $1,000 per kilowatt of power output if its estimated cost is correct and Kidstone $800 or more.

    When it comes to stored energy rather than power output, pumped storage performs far better, but the first 2 hours of stored capacity has the most economic value. Being able to continuously supply energy for potentially weeks is a nice feature, but not worth a great deal.

    One hydroelectric option is to increase the power output of existing dams. The cost of this will vary on a case by case basis and I don’t know what it is likely to cost, but it may be reasonably low, at least in some cases.

    The large battery storage facility that will soon be completed in South Australia will provide power at $640 per kilowatt. This is cheaper than pumped storage, but the amount of energy it can provide is much less as it will go flat after about an hour at full output. But we can expect the cost of battery storage to fall a long way. Once the cost falls low enough, people may start installing large amounts of them in homes and businesses.

    A solar thermal power station that is to be built in Port Augusta will have molten salt storage. The cost of this type of energy storage doesn’t have far to fall until it starts competing with gas.

    Thermal storage, whether molten salt or other, can be used without being connected to a solar power station. Converting electricity into heat and later back into electricity tends to be very inefficient, but if the cost of electricity is low enough then it can pay for itself. With increased solar and wind generation capacity there may often be periods where the cost of electricity is very low or free.

    So there are plenty of options for meeting grid demand and it looks like pumped storage may have a difficult time competing with them on price. Gas would be less attractive if it had to pay for environmental harm it caused, but with a reasonable carbon price we’d be eliminating coal so hard and fast we wouldn’t be looking to cut back on less dangerous gas capacity until it was gone.

  7. I’ll concede frustration with views that start with nixing stuff out because they are two costly. I have two problems with that sort of assessment.
    The first is that we still limit our views to the economics, ignoring all else, especially environment. If we are looking at stuffing the climate up, surely the first concern should be the environmental worthiness of a project. Is anyone game to put a dollar value on the worth of our climate?

    Second, it is a great tactic to dismiss a concept by throwing a massive “estimate” at it. Not meaning to be rude but Ronald (@#9 prices pumped hydro at $1,000 per KW for the Snowy at least. That may be informed for the Snowy, but would not necessarily be any sort of yardstick for other sites. And as others have pointed out, the lifetime is virtually endless.

  8. There are plenty of options for renewable energy and energy storage. Australia could do it quickly and easily if we had the right policies. Considering that we face a planetary emergency very soon we should be doing it very quickly.

    It is clear that a form of runaway climate change has already started and will radically change the climate in as little as the next 20 years. In this context I mean runaway to a chaotic climate state which will persist for hundreds if not thousands of years. I do not mean a runaway to a Venus state.

    The current consensus among climate scientists is that the officially sanctioned consensus in the IPCC reports is far too optimistic… or at least all developments to date indicate we are fully on track for the most dire of the IPCC scenarios. Our climate is already destablising and there is clear evidence of extreme events getting worse. Droughts, floods, hurricanes, heatwaves and different weather patterns from what we need to grow our food are all manifesting now. Climate change is now. The damaging effects are already here and will get rapidly worse from now on. The most concerning development of all probably the melting of the permafrost underway and the massive methane releases which will be associated with that.

  9. @Ikonoclast
    Ikonoclast maybe it’s time we introduced the social scientists into the mix. They seem to be very quiet about climate change. After all, they will be the ones to render social help as required.

  10. Kevin Cox is quite correct that capital is the major cost with pumped storage. This is another reason why John Quggin’s idea of renationalising the power industry should be pursued, as the cost of capital is lower for government as there is no need for the risk premium that the private sector demands.
    The cost of pumped storage can be minimised by installing the pumped storage near transmission lines. When the sums for Snowy Mountain 2 are done, it may be the cost of new transmission lines that rule it out.
    By the way, pumped storage fits in well with the NEG, as pumped storage contributes to fulfilling the reliability standard

  11. @gmhendo

    Yes. In time of war nothing is too costly in money terms. It all comes down to real persons, real resources and marshaling of the same. We are entering a period of global emergency and the world economy will have to put on the equivalent of a war footing, except to fight climate change, not other people. In times of real war (example USA in WW2) the government moves to a command economy and it is better for the existential survival task.

    “Few economists remember that after the Japanese bombed Pearl Harbor in 1941, the US government, facing the need to rapidly restructure the peace-time economy to a war economy, suspended the market for the duration and set up a system of central planning. The results were highly successful, soon producing the ships, planes, tanks and other weapons — and food and clothing — needed to win the war, while incidentally finally bringing the Great Depression to an end.” – David M. Kotz

    Similarly, we will need to move to a much more statist command economy phase to survive and win the war against climate change. The challenge to our civilization will be severe.

  12. @John Goss
    “By the way, pumped storage fits in well with the NEG, as pumped storage contributes to fulfilling the reliability standard”

    I’m not at all clear that “the reliability standard” has been specified in any detail. Has it?
    It seems to me that simply talking about dispatchable power “solves the reliability issue in the NEG.

    I’m not even sure that “dispatchable” power has been clearly specified. It’s well known that the “spinning reserve” of any turbogenerator is “dispatchable” so a fraction of any coal-fired capacity is dispatchable, if pumped storage hydro is dispatchable. So specifying “dispatchable” suitably will never rule out coal, if that is the hope.

    There are still issues of fast-frequency response, rotational inertia, and reactive power management, which have been glossed over it seems.

  13. @totaram
    totaram I don’t know if “glossed over” covers the discussion. It more likely means “not yet publicly raised” for example.
    As far as dispatchable power goes, pumped hydro takes about 60 seconds to ramp up, a bit more than say, a Tesla style battery but pretty fast. Even gas needs more ( 8 minutes?). And surely pumped is reliable even in a specialised definition.
    Issues of power quality such as frequency and voltage consistency are important, especially when power is in short supply, but less difficult to maintain when there is sufficient power available.
    Sorry, I’m not sure what you intended “rotational inertia” to mean.

  14. @gmhendo

    I only pointed out that no matter how fast you can ramp up pumped hydro, a steam turbine running at 60% of its rated power can be ramped up equally fast to 100%. Hence (a percentage of) coal-fired is as dispatchable as pumped hydro. That’s a fact which we should not try to gloss over.

    Rotational inertia of the TG set allows it to overcome any sudden surge in demand, as opposed to a PV system which has no such property. However, a wind turbine does have some rotational inertia. Rotational inertia contributes to total system stability. A large battery performs the same function. If a “cold” gas turbine needs about 8- 10 minutes to come on line. Would it be considered dispatchable?

  15. OK thanks totaram for that. I guess I discount coal fired and gas fired options as being the thing we want to be rid of. I expect that will take some time yet but my god the progress is slow, dangerously slow.

  16. Link WA to the eastern states. Eastern states evening summer peak can then be handled by solar in WA. Likewise WA mornings can be handled by the eastern states. HVDC

  17. If only we had some Germans running the show. Humorless, but they get their sh*t together.

  18. @David Allen
    Aah! So how does this “market” incentivise anyone to put up a HVDC link anywhere? Especially if it was to replace or upgrade an existing AC interconnect (e.g. between SA and VIC)? Would we need another State-owned entity (from another foreign state e.g. Singapore) to find it profitable to put this in and then milk it for profits, while all Australian govt.s sat on their hands and cheered from the sideline? I suspect that would be it.

    Nothing to do with Germans, and don’t sweat their humour, it’s just a little different. More to do with ideological stupidity that trumps engineering and science. The Germans do that as well, but only if it screws other countries like Greece.

  19. @Ronald
    Helpful numbers.

    I don’t have any for hot salt,but it stands to reason that the cost must be much lower when connected to a thermal solar plant than simply to the electric grid. CSP initially produces heat, which has to be fed through a steam turbine at 40% efficiency to produce electricity. Heating the salt electrically (with a CSP generator) implies two trips through the turbine, roughly twice the losses, and a net efficiency of about 20% assuming no heat losses from the storage. Not good. Might work though with nearly free curtailed wind or solar electricity

    The Holy Grail for CSP is getting it hot enough to run a gas turbine directly, at up to 60% efficiency, and more compact to boot. This is a large technical challenge. The receivers are just about fixed: porous ceramics can run at over 1000 degrees C. SFIK they are still looking for ways to store heat at that temperature. Salts won’t do, so they are trying liquefied sand.

  20. @totaram
    There is a well-known study by the Texas (socialist!) grid regulator ERCOT, summarised on the AWEA blog, finding that wind power actually reduces the (instantly available) spinning gas reserve because wind is fully predictable a few hours in advance, unlike equipment failures in large fossil plants. (Equipment failures can be ignored for wind, because the low risk is spread over a large number of individually quire reliable turbines.) Wind needs a larger gas reserve overall to cope with weather fluctuations, but the predictability means that this reserve can be cold, with your ten-minute warmup time. The price difference between spinning and cold reserve fully offsets the greater need for the latter, so overall wind has a lower reserve cost than fossil generation.

  21. @Ikonoclast
    True that a WWII socialist war economy would do the job in the energy transition. But it does not look at all necessary. Capitalism is generating 40% per annum growth rates on solar and electric vehicles already. Growth in wind is lower, but there is no reason to think the existing suppliers could not easily ramp up. This is not a broken system. The war economies had to deal with severe constraints on critical raw materials – oil, rubber, and so on, so civilian demand had to be cut, and rationing was fairer than taxes.

    In WWI, the belligerents (possibly apart from Russia) managed a comparably massive expansion in armaments with much less direction of the economy. The system did allow extremely unpopular profiteering. Our current political climate is much closer to 1914 than 1939. The limited appetite for state intervention needs to be focussed on areas where it is essential: regulations against air pollution and deforestation, for example, and above all carbon sequestration.

  22. Gmhendo, when it comes to energy I think reducing emissions is of paramount importance. This is why we should not build energy storage yet because at this time energy storage will increase emissions.

    In the future, when we have so much renewable capacity that some clean electricity is frequently wasted, then energy storage can reduce emissions. But we’re not at that point yet.

    Right now the world is far better off if we spend our money on increasing solar and wind capacity or improving efficiency, which definitely reduce emissions, rather than on energy storage that increases them.

  23. James, thank you. I hope the numbers I used are correct. I don’t think the figure for hydrocapacity includes pumped storage, but I could be wrong.

    Thermal storage of heat generated from electrical resistance would only be used when the economics are right. For example, when the price of electricity is very low, zero, or negative. Or alternatively, when the owner is confident they can make a profit no matter what they pay for electricity. For example, during a heatwave, thermal storage could heat up using electricity at night for 10 cents a kilowatt-hour and sell it in the day for $10 a kilowatt-hour. Even if the thermal storage is only 25% efficient, they are still $9.60 ahead.

    We have a considerable amount of peak capacity with low efficiency that sits idle most of the year. Low efficiency thermal storage could fulfill a similar role.

  24. “most of the opportunities for hydro-electric power had already been exploited”

    But not all. We could always build a dam on the Gordon River, just below the Franklin. šŸ™‚

  25. Ronald I see what you mean. But one of the characteristics of solar and wind is their indeterminacy. Of course the sun is gone at night and the wind blows when it does. The supply side needs to deal with those gaps formed when sun or wind are not available. Storage provides a method of providing power at those ties and also maintaining power quality (especially frequency and voltage). Storage provides greater reliability to the grid.
    But a variant on your approach might be to develop interim storage to manage our grid now, and develop larger structural storage like pumped hydro a little later.

  26. Gmhendo, because at this moment we can reduce emissions more per dollar by subsidizing solar and wind generating capacity than spending money on energy storage, I think energy storage should be left up to the market until it becomes clear that subsidizing it would reduce emissions by more than subsidizing wind and solar would. (It shouldn’t take long. Under 100 days for the world’s largest battery storage.)

    As for reliably meeting demand, our electricity market is supposed to result in that occurring anyway. If it can’t manage it, then Government should bite the bullet and admit it was a bad idea in the first place. At the current time, gas generation is the cheapest way to provide the dispatchable power required to meet demand. This may change.

    Even though I don’t think we should subsidize energy storage now, we can still consider what we might need in the future and prepare, but eliminating coal use by building out solar and wind capacity should be where virtually all our current available resources go.

    Actually, I would prefer to see coal and gas pay the full cost of their externalities than see solar and wind power subsidized, but apparently we can’t do that because it makes Tony Abbott cry.

  27. Ronald perhaps I am more existential than you. I see market failure in your context and government fail across all dimensions. Whatever value neoliberalism had or has, that value is now seriously depleted.

    To rely upon government intelligence and action to act effectively is not a good bet – look how they are managing energy futures right now.
    To trust the “market” to act in our best interests is a dream. They have a different agenda.

    Expecting fossil fuels to pay for their externalities? I don’t think they have that sort of money. And in any event, they would have to keep burning coal to fund their reparations.

  28. The shortcomings of our present government as it flails around trying to find a pathway forward in electricity (and NBN, and climate change, and tertiary education, and NDIS, and just about everything else) are not necessarily inherent shortcomings of government. There was a time when governments were able to think through the consequences of optional policy initiatives, reach a prudent choice and establish a functioning apparatus to support that strategy.

    We have seen the hollowing out of the administrative and analytical capacity of the public service, meaning that issues are now gravitating upwards to the political level where a political bunfight ensues without the benefit of in-depth previous research and policy legwork.

    This analytical incapacity was predictable – and predicted, by scholars of public administration like Profs Quiggin and Michael Pusey who didn’t sign up to new managerialist/neoliberal orthodoxy- as much as three decades ago. It will take as long to rebuild administrative capacity, if we can ever do it. I doubt that we can, because the forces driving the rundown of governance capacity are still operating.

  29. I have to say that I’m very doubtful of those potential storage sites.

    I had a look at the claimed sites near where I grew up, and I don’t like the odds of getting environmental approval for building dams on any of them.

  30. @Robert Merkel
    Rob I suppose that you having seen one or two of the sites you have it over me, I’m still in my armchair. But really, out of 22,000 potential sites, there is some probability that many are actually OK as pumped hydro sites. I’m thinking that the sites were identified using some GIS handiwork and that the selection criteria was intended to look for basins with an altitude difference of 250 metres or so. Closer inspection will quickly sort the less likely sites.

    It may be difficult, as you say, to get environmental approval. But is not the goal to solve a huge environmental issue, one that acts against all mankind way into the future. Environmental approval? Queensland is clearing land hand over fist as we speak. Clearing some space for pumped hydro plumbing should not be an issue at all, given the net gains possible. Anyway, I think State planning ministers can call in a project and wave it through at their discretion.

  31. @Robert Merkel

    Yes, there’s a systematic error in the site identification. Broadly, pumped hydro sites are undeveloped hills. Typically they are currently used as nature reserves because they can’t be developed. Coming up with a way to develop them doesn’t negate their conservation value, or produce other non-developed land to replace any loss of conservation value. It’s thus likely that almost all the X-thousand sites suffer from the problem that they are residual wild spaces. Whether we value those more than air conditioning is something I suspect I disagree with most Australians on.

    The flip side is that in many cases pumped hydro is fairly low impact, and the buffer zone between the hydro and any other development can provide a better defended conservation area. Although looking at the high-priced suburb of Lucas Heights, I wouldn’t count on “it’s scary living downstream of a dam” having any value at all – the buffer zone will be a few metres around the actual infrastructure.

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 )

Connecting to %s