Emissions Nuclear Policy Renewables

CO2 abatement cost with electricity generation options in Australia

Guest Post by Peter LangPeter is a retired geologist and engineer with 40 years experience on a wide range of energy projects throughout the world, including managing energy R&D and providing policy advice for government and opposition. His experience includes: coal, oil, gas, hydro, geothermal, nuclear power plants, nuclear waste disposal, and a wide range of energy end use management projects.

A 10-page printable PDF version of this post can be downloaded here.

An Excel worksheet showing the calculations (allowing you to change inputs/assumptions) is also available.


What is the cost of carbon dioxide (CO2) emissions abatement with the various electricity generation technologies being considered for Australia?

The abatement cost of a technology depends on many factors such as the engineering characteristics of the electricity grid to which the new technology will be connected, the geographic location and many others.  One important factor often not mentioned is the reference case against which the abatement cost is calculated.  The abatement cost for a new technology is only meaningful when compared with another new technology or with an existing generator it would ‘displace’; e.g. nuclear compared with a new coal power station or nuclear compared with an existing power station.

The Electric Power Research Institute (EPRI, 2010) report for the Australian Department of Resources, Energy and Tourism provides data that allows CO2 abatement costs to be estimated for a range of new technologies. Unfortunately, the report is complex and opaque in parts.

The purpose of this paper is twofold:

  1. to summarise in tabular form the relevant information from the EPRI report so others can access it easily and produce levelised cost of electricity (LCOE) figures under differing assumptions, particularly using the NREL LCOE calculator .
  1. to calculate and compare the CO2 abatement costs for a range of new technologies for each of three ‘displaced’ technologies.

This paper does not attempt to calculate the effects of carbon price on the LCOE or CO2 abatement costs, because:

1)     the EPRI report does not include the effects of carbon price — nor feed in tariffs, renewable energy certificates and other subsidies — so incorporating the effect of CO2 pricing, and other incentives and disincentives in the analysis would require many additional assumptions, and

2)     the purpose of this paper is to show the abatement costs for the various technologies so options can be compared and so the cost of incentives and disincentives (including carbon pricing), which would be needed to make each technology viable, can be made visible.


The CO2 abatement cost is calculated for seven new electricity generation technologies, selected from the EPRI report.  The seven new technologies are:

  1. Coal (black, without CCS).
  1. Coal (black, with CCS)
  1. Nuclear
  1. CCGT (Combined Cycle Gas Turbine)
  1. OCGT (Open Cycle Gas Turbine)
  1. Wind (wind class 5, 100 x 2 MW)
  1. Solar thermal (Central Receiver, 6h storage, DNI = 6)

The abatement cost for each is calculated by comparison with each of three ‘displaced’ technologies:

  1. Hazelwood, brown coal power station, Victoria (1,600 MW, commissioned 1964 to 1971)
  1. Liddell (see photo above), black coal power station, NSW (2,000 MW, commissioned 1971 to 1973)
  1. A new black coal plant, withoutCCS; (this is same as #1 in the list of new technologies).

Most input data are taken from EPRI (2010) ; these are summarised in Appendix 1.  To bring the figures up to date and to aid in international comparisons, costs presented in Table 1 have been converted from 2009 A$ to 2011 US$; these are in Appendix 2.  Details of the costings, including the exchange rates and inflation rates used, are included. The calculation steps and results are presented.

CO2 Abatement Cost is the difference in LCOE divided by the difference in CO2 emission intensity (EI):

CO2 abatement Cost = (LCOEnew – LCOEdisplaced) / (EIdisplaced – EInew)

The data needed for calculating LCOE for each technology, using the NREL simplified LCOE calculator, are provided in the Appendices.

The capital cost is one of the inputs needed for the LCOE calculation.  The capital cost figure needed is the Total Capital Required (TCR). But theTCRfigure is not given in the EPRI report.  As such, the method of estimating it, including the inputs and intermediate calculation results, are presented in Appendix 1.

The CO2 emissions intensity (EI) presented in the EPRI report includes only the emissions from burning the fuel in the generator. Fugitive emissions are not included.  Nor do the emissions intensities include the higher emissions intensities produced when load-following; e.g. when cycling power up and down to back-up for intermittent renewable energy generators.

The emissions intensities (EI) for Liddell and Hazelwood power stations are 1.08 t/MWh and 1.53 t/MWh (sent out) respectively (ACIL-Tasman (2009), Table 18 ).  These EIs include fugitive emissions (whereas the EPRI EIs do not). This causes an error in the calculated abatement costs. In the ACIL Tasman report, fugitive emissions comprise 10% to 27% of EI for gas, 2% to 9% for black coal and 0.3% for brown coal.

The LCOE for Liddell and Hazelwood are ‘Commercial in Confidence’, so I’ve used $30 and $28 respectively, which are figures I’ve seen stated for the ‘equivalent LCOE’ for the remaining plant life.


The CO2 abatement costs are summarised in Figure 1.

Figure 1: CO2 abatement cost for seven selected new technologies (named on the horizontal axis) compared with each of three ‘displaced’ technologies (named in the legend).   Abatement costs are in US$/tonne CO2 (constant, 2011US$).

The inputs and intermediate calculation results are in Appendix 1 (in 2009 A$) and Appendix 2 (in 2011 US$). The data in Figure 1 is from Table A2-5.

Table A1-2 and A2-2 show the proportion of “Capital” (i.e. TCR) that EPRI apparently assumed for ‘Owners Costs’, including ‘Allowance for Funds Used During Construction’ (AFUDC).

The ratio TCR/TPC is given in Tables A1-3 and A2-3.  This ratio shows how much higher the TCR is than TPC for each technology.  For example, for nuclear the TCR is 1.93, or 93% greater than TPC.


This report uses the EPRI (2010) figures for LCOE and emissions intensity.  These are the figures being used in Australian government reports such as ABARE (2010) and for the Treasury modelling of the carbon tax and ETS.  Some discussion of the figures and assumptions is warranted.

The Total Plant Cost figure in the EPRI report is confusing because it is not the full capital cost used to calculate LCOE.  The capital cost figure needed for calculating LCOE is the Total Capital Required, which includes Owner’s Costs.  Back-calculating from the figures provided reveals the amount of Owner’s Costs EPRI used in their LCOE analyses. This cost is significant. It is 93% higher than the Total Plant Cost for nuclear, 88% higher for CCGT, 45% higher for coal, and 41% higher for solar thermal. The EPRI report does not make clear the basis of the Owner’s Costs or the assumptions. For example, the construction period is not stated?

EPRI uses 85% for the average lifetime capacity factor for mature technologies such as coal, gas and nuclear. However it also uses 85% for immature technologies such as carbon capture and storage, and assumes capacity factors for Wind (36.6%) and Solar Thermal (31.6% with 6 hours storage) that appear to be based on the best possible figures, rather than the average achievable over a plant’s life. It is difficult to understand how these capacity factors could be realized in practice over the plant life.

The emissions intensities do not include fugitive emissions and appear to be for the technology running at optimum efficiency, rather than average efficiency. The abatement costs for Wind and Solar are probably understated, because the capacity factors assumed seem to be unreasonably high.

The reason the OCGT abatement costs are high is because EPRI used a capacity factor of 10% for the calculation of LCOE.  This is because OCGT is economic at capacity factors up to about 14% due to its high fuel costs (IPART, 2004, Exhibits 1-2 and 1-3 )

If we assume wind or solar are backed up with OCGT, it is clear, without needing to do detailed calculations, that wind and solar with back-up are a high-cost way to avoid emissions.

Of the options considered, CCGT is clearly the least cost way to abate CO2 emissions.  For example, if we are making a decision about new baseload capacity we might compare between a new baseload coal plant withoutCCSand other options.  From Figure 1, the CO2 abatement cost, compared with new black coal, is $44/t CO2 for CCGT and $107/t CO2 for nuclear.

Based on the EPRI figures, nuclear cannot be justified inAustraliaat this time because it is too expensive.   For nuclear to be an economically viable option, the impediments that are causing the EPRI estimates for the cost of nuclear in Australia to be several times higher than in Korea need to be removed.


Of the options considered, CCGT is clearly the least cost way to abate CO2 emissions, given the EPRI assumptions.

The abatement cost with CCGT is about 40% of the abatement cost with nuclear.

Based on EPRI’s estimates, nuclear is not economically viable in Australia because it is too expensive.  This situation will remain while the impediments to low-cost nuclear remain in place.


OCGT – Open Cycle Gas Turbine

CCGT – Combined Cycle Gas Turbine

CCS – Carbon Capture and Sequestration

CST – Concentrating Solar Thermal

EPRI – Electric Power Research Institute

NREL – National Renewable Energy Laboratory

LCOE – Levelised Cost of Electricity

TCR – Total Capital Required

TPC – Total Plant Cost

AFUDC – Accumulated [or Allowance for] Funds Used During Construction (Capitalised Interest)


ABARE (2010), Australian Energy Projections to 2029-30:

ACIL-Tasman (2009), Fuel resource, new entry and generation costs in the NEM:

EPRI (2010), Australian Electricity Generation Technology Costs – Reference Case 2010:

Independent Pricing and Regulatory Tribunal (2004) The long run marginal cost of electricity generation in NSW:

NREL (2011), Levelized Cost of Energy Calculator:

South CarolinaElectric & Gas Company (2011), VC Summers Nuclear Station Units 2 and 3 (June 30, 2011):

Appendix 1 – Input data and intermediate calculation results with costs in ‘constant 2009 A$’

Appendix 1 summarises the significant data from the EPRI (2010) report for the seven technologies selected for this study.  Costs are in ‘constant, mid-2009 A$’.

Table A1-1 lists the values needed for input to the NREL LCOE Calculator, .

The Capital Cost figure listed in Table A1-1, needed for calculating LCOE, is ‘Total Capital Required’ (TCR). But the TCR figure is not given in the EPRI report.  So it must be back-calculated from the other data available in the report. The EPRI report provides the breakdown of LCOE by Capital, O&M and Fuel (Tables A1-2 and A2-2). This data was used to calculate the value EPRI used for TCR. The results are in Tables A1-3 and A2-3.  These tables also give the ratio TCR/TPC. This shows how much higher the TCR is than TPC for each technology.  For example, for nuclear the TCR is 1.93, or 93% greater than TPC, whereas for coal it is 48%.

Appendix 2 – Input data and intermediate calculation results with costs in ‘constant 2011 US$’

The cost figures in Appendix 1 are in ‘constant, mid-2009 A$’.  In Appendix 2 they have been converted to ‘constant, mid-2011 U$’.  The conversion factors are in Table A2-6.

Table A2-1 lists the values needed for input to the NREL LCOE Calculator.

By Barry Brook

Barry Brook is an ARC Laureate Fellow and Chair of Environmental Sustainability at the University of Tasmania. He researches global change, ecology and energy.

262 replies on “CO2 abatement cost with electricity generation options in Australia”

The UK Dept of Energy has a new version of its ‘2050 calculator’ out, that for the first time attempts some costing of the user’s choices for reducing CO2 emissions.

The official least cost plan, which they are calling Markal 3.26, is marginally cheaper than business as usual, but that is because they are assuming $130/bbl imported oil and $110/Te imported coal, so reducing fuel use pays for the infrastructure changes.

Not much wind or solar in the least-cost mix….


Climate Spectator, yesterday, included an article under the heading “Clean energy’s Affordable Makeover”. It described a perception that public enthsiasm for renewables is giving way to a desire for affordable energy.

The 2050 Calculator Tool you brought to our attention appears to be heading further in the same direction if it indeed does reduce the emphasis on wind and solar.

My comment on the other site included:

” “So Britain’s energy minister Chris Huhne used his annual statement to parliament on energy strategy on November 23 to shift the focus onto clean energy as affordable energy.”

That could be the first step towards the exit. Once “affordable energy” becomes the primary goal, solar and wind have much to fear and everything to lose.

Will the Minister’s next annual statement to parliament retain the same “energy strategy” but a different (non wind, non solar) means of attainment? Bear in mind that the CCS project in Scotland sank without a trace a couple of months back.

Without exception, each UK energy program that relies on government financial support, is challenged by this new strategy.”


UK Levellised Cost Study.

Digging into Luke_UK’s reference, above, I found a cost study which is very much relevant to this thread.

Click to access 2153-electricity-generation-cost-model-2011.pdf

Peter Lang will find PB’s methodology and results for FOAK and NOAK across a wider range of technologies than most others.

It may provide perspective for the discussion upthread about Australian FOAK nuclear. Having only just found it, I have not digested its contents.

Thanks, Luke_UK. The site you linked to is a treasure trove. It even provides a link to the Minister’s statement which I referred to in the preceding comment.


State of Connecticut Nuclear Analysis – 2011
“The situation might change significantly if the nuclear industry establishes good performance on the early new plant construction projects being conducted in other states and/or if significant carbon emissions penalties are implemented on generating facilities. *****Elimination of the financing premium for nuclear power plants provides a levelized cost of electricity (LCOE) that is very competitive to a CCGT power plant.*****”

In the United States we have decided that CO2 reduction is not a good enough reason to build new carbon free power generation. What we have decided is to “keep the nuclear option open”. This means that the construction of two pairs of AP1000 reactors in the SouthEast will be allowed to continue with completion dates between 2016 and 2019. If all goes well with these projects then other projects might be funded at a lower interest rates during the 2020s.

By 2020 the Chinese should have mastered AP1000 modular construction techniques by FINISHING somewhere between 14 and 24 AP1000 construction projects. I’m not sure that the Chinese think about levelized costs at all so they may not try to reduce the construction time for a single plant to 36 months. Instead they may try to optimize the construction of their fleet of reactors. It will be a fascinating/frustrating process to watch unfold in extreme slow motion.


@Peter and others: Thank you for this post and discussion. It got me interested in learning more. I wanted to get some estimates on how the system wide costs vary as we add carbon free sources into the mix. Adding wind or nuclear displaces other sources and influences capacity factors so costs elsewhere change. You can read my attempt to understand the scale of the changes:

Basically, I find that while the nuclear path ends up with only modest changes in the system wide LCOE all the way to the decarbonized electricity supply, the renewables path leads to escalating costs. I didn’t even postulate any particular costs reductions and used costs which appear somewhat reasonable today where I live. Of course since climbing is hard, it becomes politically harder and harder to follow to costly route as soon as the costs start to bite.


I like your approach to the issue, Jani.

EL referenced a study by Heide in the latest thread (see the Guardian questions thread comments) that looked at a 100% wind and solar Europe electricity supply. It was theoretically feasible but needed a 50% overbuild of wind and solar capacity, plus half a week of average power storage. Either of which is cost prohibitive, but especially the half a week of energy storage. The energy storage alone, even making optimistic assumptions on cost reduction of batteries, at euro 100/kWh sodium sulphur batteries, would cost 20 cents per kWh by my overly simplistic calculation. At today’s (real) cost it is 2-3x as much.

One thing that bites even harder is that if Europe needs to get more than half its electricity from wind, there won’t be enough good wind speed sites so expensive offshore and less windy (costly) onshore will have to be used. And that then needs to be overbuilt 50% (again without considering storage losses). This is already cost prohibitive even before the 20 cents per kWh half a week electricity storage system cost.

Half a week of energy storage would also add over 1 ton of batteries per capita. What is the CO2 impact of that? Lots of energy intensive chemicals such as sodium produced by electrolysis and metals produced with coal.


These cost discussions are confusing. In Jani’s link, with a number of simplifying assumptions, LCOE for hi penetration wind is about twice that of hi penetration nuclear. cyril, your numbers seem much higher.

excuse any stupidity on my part, but is that 20 cents/kwh an addition to the LCOE? That’s a huge addition, and likely underestimates according to you.

Still, someone mildly scared of nuclear might be attracted to hi penetration renewables if electricity costs “merely” doubled. I was under the impression that hi penetration renewables would be in fact cost prohibitive, but twice as much for electricity is not cost prohibitive.

I recall Peter Lang’s demonstrations involving costs far greater than double.

also: fossil fuel use seems to decline much more with hi penetration in Jani’s model than in others I have seen here. I’d really like to get clear on this.

Intuitively, I have a hard time understanding how massive overbuild would but double electricity costs. I realize there are other big issues here, like footprint, etc.


Gregory, looks like Jani actually gets very similar results. The Heide study looks at wind and solar combined in an optimal way to meet the power demand as much as the time as is possible. This should get better results than Jani, which looks at only wind.

What is interesting in the Heide study, is that it is possible to reduce the storage requirement drastically by overbuilding wind and solar. Still even with that optimization step, half a week of energy storage would be required for a completely renewable powered system for Europe.

I should note that most of Europe has very poor wind resources. Basically the vicinity of the North Sea, is excellent, much of the rest is very poor:

This is a serious problem for powering most of Europe with wind. There is also NIMBY that is already very strong at today’s low European wind energy penetratioin. If one is forced towards offshore then the costs rise.

For energy storage we have to be realtic. There are not enough hydro and pumped hydro resources available so we’ll need innovative solutions such as drastically cost reduced NaS batteries and underground pumped hydro. You’ll find that there are no plausible paths for $300/kWh. Riverbank Power had a $2 billion project for 6000 MWh of underground pumped hydro storage, that’s coming in at $333/kWh. It got cancelled due to lack of suitable geology. The local residents weren’t happy about it and an activist group formed. That’s for just one project. Lots of resistance.


is that 20 cents/kwh an addition to the LCOE?

Yes, its in addition to the normal cost of wind and solar. It does not include storage losses and almost neglects maintance costs. And it use favorable 5% average interest rate for the combined loan from investors and the bank. Realistically you’d be pushing 25 cents per kWh at least. If the average cost of wind and solar is 12 cents per kWh and you add 50% to that, then that makes for a busbar cost of 43 cents per kWh – no transmission and delivery costs included. Even if the average cost of wind and solar is half that (6 cents per kWh, very optimistic), the cost is still 34 cents per kWh, without transmission and delivery.

Clearly the energy storage cost is the limiting factor. So a realistic system would use less (if any) energy storage and just burn fossil fuels. And 50% overbuild isn’t interesting, so we won’t do that, either. So just burn lots more fossil fuels. And that’s the end of the transition. A mixed wind and solar and fossil grid that hasn’t gotten us the required deep (90%) reduction in fossil fuel use. We’ll have failed.


so with a 43 cents/kwh busbar, what would the real cost be, with profit added? (hope this question makes sense?)

thanks, cyril. so storage is main “limiting factor”? more than drastic overbuilding to limit storage? as you point out, of course, drastic overbuild, costs aside, runs into site/space/resistance from population issues.

thanks, cyril.


Cyril R. wrote:

It was theoretically feasible but needed a 50% overbuild of wind and solar capacity, plus half a week of average power storage.

You are misreading this study. It does not call for a half a week of average power storage. 1% is an aggregate measure (when all of the charge and discharge cycles are added up over a 365 day period). There is no other way to read this in the study, or in others of the same looking at storage as an enabling technology for renewables or any other grid or generation support application: DOE, EPRI, Sandia, NREL, and various professional workshops. Let’s get this right, finally. If you think this is the contrary, please point out where this is the case.


@ EL. Storage equal to 1% of average demand means you need half a week of average energy demand flow worth of storage. This is what the Heide study says. Literally. I cannot see for the life of me how you can warp a statement as clear as this.

If, as EL suggests, the charge and discharge cycles are all added up, adding up to 1% of demand, this implies that you need only an hour or two of energy storage – when multi day low wind periods are very common. In a 100% wind and solar grid.

By deduction we can conclude this cannot be the case.

You need some more common sense to see the forest. Think about what you are saying and look at the real wind data before you claim that not much energy storage is needed in a wind and solar grid.


For refererence, here is the Heide study:

At 50% excess generation the required long-term storage energy capacity and annual balancing energy amount to 1% of the annual consumption. The required balancing power turns out to be 25% of the average hourly load.

Note the word “CAPACITY”. Note also, 25% of the average annual load comes from storage (or fossil fuel, which is not acceptable of course). This is a major contributor to keep the grid together, a far cry from 1% that EL suggests.


student mode again:

there are three terms here and their relation is not clear (to moi).

long term storage energy capacity; annual balancing energy; required balancing power.

How do these terms relate to one another? what is the difference between annual balancing energy and balancing power? cyril, with your last statement about balancing power, are you suggesting that the annual balancing energy number is dubious?


Cyril R. wrote:

Note the word “CAPACITY”. Note also, 25% of the average annual load comes from storage (or fossil fuel, which is not acceptable of course). This is a major contributor to keep the grid together, a far cry from 1% that EL suggests.

Yes, energy capacity + annual balancing energy amount to 1% of the annual consumption (when all charge and discharge cycles are aggregated). I don’t know how you can have energy storage output without storage capacity? In the paper, they refer to this as “the average balancing energy.” From a direct quote: “The storage energy capacity and annual balancing energy are normalized to the average annual consumption” (p. 2517). No matter how many times you try and misread this study, it does not mean what you claim it does. 25% of the average hourly load does not mean 3.5 days of continuous replacement of energy output for an entire grid. Can we close this debate now, or risk cluttering up the thread with a lot of back and forth (when the evidence on this matter is straightforward, clear, and incontrovertible).

In Heide study, none of the annual load comes from fossil fuels … it looks at an operational model for a 100% renewables test case on a European scale (and tweaks the variables to better understand the relationship between excess generation and energy storage requirements, and generation mixes among renewables optimized for roundtrip storage, hydrogen storage, or annual balancing energy).
Agreed – you should both end it here. You are never going to agree and this is becoming a circular argument.


Upon a quick review, the study I looked at, , concludes, for storage, 16TWh of storage and 240GW of capacity. That’s highly suspect to start with, as average demand would be about 370GW, and peak I assume would be twice that.
Regardless, it then lists 105GW of salt and hydro capacity and says that’s about right(240 being right), and adds, “For one-way storage reservoirs and [renewable power at 1.5 times demand] the required annual balancing energy and power based on daily, instead of hourly power mismatches turn out to be 30TWh and 90GW, respectively.”
Well, that’s nice. Not having peak demand is, I suppose, a great way to deal with supply meeting peak demand. 90GW of storage capacity still wouldn’t be able to meet an an average demand of 370GW much of the time – but it’s less than the 105 that they found could exist …so problem solved
By the way, they do end up stating 25TWh of annual storage based on annual demand of 3240TWh – and that is about 0.8% which does equate to around 3 days. Obviously.
The problem isn’t that, in this paper, the problem they gloss over is meeting demand when demand exists to be met – which they choose not to do.


Just in case I am wrong (and some of these terms are a bit confusing in the paper), here are some relevant quotes from the conclusion. Annual European consumption in 2007 is reported as 3240 TWh in paper.


“At zero excess generation Y = 1 the required European annual balancing energy and balancing power result to be 510 TWh and 265 GW, respectively. These numbers are reduced down to 160 TWh and 200 GW once the excess generation is increased to Y = 1.5” (p. 2523).

“For balancing reservoirs and Y = 1.5 the required annual balancing energy and power based on daily, instead of hourly power mismatches turn out to be 30 TWh and 90 GW, respectively” (p. 2523).

On meeting these energy storage targets:

“The storage lakes in Norway, Sweden, Austria and Switzerland currently have an annual balancing energy of about 150 TWh with a balancing power of about 55 GW” (p. 2523).

[For hypothetical hydrogen storage in salt caverns] “A typical large cavern field has a volume of 8 x 10e6m3, which, given the volumetric energy storage density of 170 kWh/m3 for hydrogen, would provide a storage energy capacity of 1.3 TWh with a discharge power of about 2.6 GW. Mainly North Germany, but also Denmark, the Netherlands and Great Britain certainly have the potential for some more of these salt cavern fields” (p. 2523).

For me, the main utility of this paper (and likely the purpose intended by the author, although this is pure conjecture on my part) is to quantify the effect of excess generation and resource mix on energy storage variables. I do not think it’s a useful guide or a practical model for a fully developed technology pathway to a 100% renewable energy grid. It’s simply looking at resource mix and excess generation, and making some “hypothetical” statements on how we should look at these variables in more concrete, cost-effective, and practically oriented studies (likely looking at much smaller renewable energy targets in future cost effective and balanced energy portfolio options and mixes). In short, what types of questions should we be asking. If you think I am recommending a 100% renewables option, you would be wrong. I do not think this would be a good solution for a cost-effective or practical strategy for deep GHG reductions, or make the best use of capacity reserves or storage (which by most measures, and future expectations, will continue to have a high cost).


EL, I thought you said a few comments ago you were going to close this rather than going back-and-forth?

I don’t see much practical utility in the paper (and I see you don’t either). When discussing balancing energy, it is talking about existing hydro, not hypothetical pumped hydro which would be required if the overbuilt renewable infrastructure was to dump its excess generation into — and as Peter Lang and others have shown, turning the former into the latter is a mammoth undertaking.


Scott Luft wrote:

By the way, they do end up stating 25TWh of annual storage based on annual demand of 3240TWh – and that is about 0.8% which does equate to around 3 days.

Thanks for direct link to article, Scott! This is what they seem to be saying, but this does not directly follow from the quote in the paper. The direct quote: “Twenty of them would provide a storage energy capacity of 25 TWh with a discharge power of 50 GW” (p. 10 above reference).

25 TWh of energy would amount to some 500 hours of continuous output at 50 GW storage capacity. How do you understand this odd discrepancy?

They appear to have very different meanings for the term “capacity” and “discharge power” in the paper.


EL, I thought that the first figure refers to how much power needs to be stored (ie. wind generated in the spring would allow reservoirs to stay full until the summer).
I think the second number was calculated without regard for the first, which was how much energy might have to come from storage at any one time.
I don’t think the paper closes that gap – which is problematic. With the little time I have to look at it, it seems it put a little drain on a big bathtub.


Leave a Reply (Markdown is enabled)

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

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

Facebook photo

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

Connecting to %s