Santos chief’s gassy vision Part 2 – is gas almost as good as nuclear?

This is part 2 of an analysis of some recent public statements by Santos CE David Knox. If you haven’t already done so, read this first.

In Part 2, I address the following statement:

One thing I would note about advocates of nuclear is that they often ignore natural gas and its role in power generation. Often, they gloss over the existence of gas and simplistically abbreviate the debate to one of ‘if renewables fail, then we have to go nuclear’ as was reported in last week’s Advertiser. Gas already delivers close to 70% of the carbon intensity reduction that a shift from coal to nuclear would achieve in eastern Australia, but at far less expense and with none of the sociopolitical challenges. In short, the real competitor to nuclear power in Australia will be natural gas.

Is this claim credible? Let’s crunch the numbers, working from first principles.

For simplicity, I’ll compare the most efficient gas technology (combined cycle gas turbines; CCGT) and Generation III nuclear power (thermal reactors — the type of reactor that will be most commonly deployed over the next few decades).

First, consider the greenhouse gas intensities. I’ll derive numbers here from the recent meta-analysis in Energy by D. Weisser (2007). These include upstream and downstream emissions, so are more conservative than the optimistic figures Knox gives for gas in his talk, which only considers operating stage emissions (aside: these are zero for nuclear):

Coal = 950 to 1,250 kg CO2eq / MWh
CCGT = 440 to 780 kg CO2eq / MWh
Nuclear = 3 to 24 kg CO2eq / MWh

Next, consider Australia’s current electricity demand. As noted here, this is 250,000 GWh per year. Of this, 78% comes from black/brown coal, 15% from gas, and 7% from hydro (see Garnaut Review). So, switching from coal to gas, or coal to nuclear, involves replacement of ~200,000 GWh of electricity per year (23 GWe average).

First, based on Weisser’s review, 200 TWh of coal-fired electricity should generate 190 – 250 million tonnes (Mt) of CO2eq. If this coal-fired electricity was to be entirely replaced by CCGT, it would result in 88 – 156 Mt of annual emissions from power generation (an 18 to 65 % reduction [incorporating the full range of uncertainty given by Weisser's figures]). If the coal was instead replaced by nuclear, it would be 1 – 5 Mt (a 97.5 to 99.8 % reduction).

Alright, let’s take the best-case scenario for gas, and the worst-case for nuclear and coal. In this case, nuclear results in a reduction of 245 Mt of CO2eq. Gas results in a reduction of 162Mt of CO2eq. On this basis, gas would result in a 65 % of the reduction that a shift from coal-to-nuclear would achieve. That’s the absolute best-case scenario for CCGTs (the worst case is just 18 % of what nuclear could achieve).

However, looking at it another way, in this best-case scenario for CCGTs, Australia’s cumulative yearly GHG emissions from the electricity sector would be 17 times greater than if we went for nuclear (88 Mt vs 5 Mt). Food for thought.

But look again at what Knox actually says:

Gas already delivers close to 70% of the carbon intensity reduction that a shift from coal to nuclear would achieve in eastern Australia

Already? I don’t think so. From the above figures, gas generates roughly 37 TWh of electricity per year. Since most of this comes from steam and open cycle turbines, the upper bound on emissions intensity for CCGT of 780 kg CO2eq / MWh  is more appropriate here. On this basis, gas currently saves ~20 Mt per year. A shift from coal-to-nuclear would save ~245 Mt.

Okay then, let’s look at it from another angle. Australia’s total GHG emissions inventory from the energy sector (fuel combustion + fugitive emissions from electricity, transport etc.) in 2007 was 408 Mt (see table above). What if we use this as the reference point? We can easily work out that if our current gas-fired electricity had instead been supplied by coal at the high emissions intensity level, this figure would have been closer to 428 Mt. On this basis, gas already results in about a 5 % reduction in our energy-sector carbon intensity. A full nuclear replacement of electricity only) would drop this to closer to 160 Mt — a 60 % reduction (prior to replacement of transport fuels etc.).

Sorry Santos, but I just can’t make your strange gassy claim work. (A prize to any BNC reader who can!).  It simply doesn’t stack up and I suggest you don’t repeat it in public in the future. Indeed, I conclude that natural gas is already doing very little to mitigate our carbon intensity, and even a massive wholesale switch to CCGT for electricity generation would be nowhere near as effective in cutting emissions as a dash to nuclear. Nor is gas a medium- or long-term solution for energy security, as I explained here.

Update: The (probable) answer, from Martin Nicholson:

Gas already delivers close to 70% of the carbon intensity reduction that a shift from coal to nuclear would achieve in eastern Australia

Let me give you another (marketing) spin on what I think David Knox meant here. First he is talking about operating emissions not full life-cycle. This is what is important to him because that’s what he expects to be paying a carbon cost on.

His “gas already delivers” statement refers to a new CCGT plant installed today. The best currently available CCGT operating emission intensity is around 400 kg/MWh. He believes this will come down to 350kg/MWh (based on his slides). On the same slide he has the average coal operating emission intensity at 1030 kg/MWh. So new gas plants will reduce coal emissions by 680 kg or 66%. 70% is the closest ‘round number’.

So, I’m left to reiterate the following statement — now hardened with the above analysis:

The UK is now paying dearly for their dash for gas, following the coal mine closures of the 1980s. Their once-abundant North Sea fields are rapidly depleting. Again, Australia should take note of this warning. We must not go down the natural gas-for-coal substitution route. It would be long-term economic suicide. Also, gas is a carbon-based fossil fuel, releasing 600kg of carbon dioxide per megawatt hour. Unlike the situation for uranium power, the electricity price is strongly tied to the fuel price for gas. A spike in the gas price means big jumps in power prices. Cheap uranium energy is a much more secure proposition. Gas is best reserved to meet occasional peak power demands, not baseload needs.

Add to FacebookAdd to NewsvineAdd to DiggAdd to Del.icio.usAdd to StumbleuponAdd to RedditAdd to BlinklistAdd to TwitterAdd to TechnoratiAdd to Furl

About these ads

90 Comments

  1. The ultimate issue for a full conversion to natural gas fired electric generation will be sustaining supply. In the 80’s this became an issue in North America when the rush to gas caused an overload on the transmission networks (in particular from Canada to the US) and created a huge backlog in drilling production wells in the known gas fields. Basically they could not punch holes fast enough to meet demand.

    This was caused by shortages in rigs and skills and is a concern now with the move to develop shale gas, as it requires very specialized equipment, and knowledge to drill production wells in these fields, and have them yield economically.

    These questions often remain unasked in this debate, and this is too bad, because this will be a driving force for price rises.

  2. David Knox stated:
    Gas already delivers close to 70% of the carbon intensity reduction that a shift from coal to nuclear would achieve in eastern Australia, but at far less expense and with none of the sociopolitical challenges. In short, the real competitor to nuclear power in Australia will be natural gas.

    I suspect that “already delivers” refers to gas-fired generation already existing and in use somewhere in the world, if not Australia. But using the same logic, one could say that nuclear “already delivers” CO2 reductions of about 99%.

    The “far less expense” phrase certainly holds true for construction of the power plant, but is a lot less certain when talking about providing base-load power, especially over longer periods of time as fuel costs rise.

    As far as “none of the sociopolitical challenges”, this is something we will have to get over. Statements like this are certainly not helpful in overcoming such challenges.

    I think there is a role for natural gas powered generation – providing power for peak demand periods. The low cost of plant and relatively high cost of fuel makes this the ideal place.

    If there is to be any dash for gas, it should be in the transportation energy sector. The turn-over in automobiles is faster than the turn-over in power plants, so this sector of energy consumption can follow more closely the abundance of various energy sources. That said, I still strongly favor the use of electricity (from nuclear sources) for transportation.

  3. I think this sort of question gets clearer if we think about cumulative emissions.

    How much climate change we will experience basically depends on the proportion of all the world’s fossil fuels we burn between now and when humanity reaches carbon neutrality.

    That gets especially worrisome when you consider the potentially huge stocks of unconventional gas that exist. Gas may be cleaner than some other options, per megawatt-hour, but it has a potentially huge stock of fuel to draw upon. That means it can potentially generate a huge amount of climate change.

  4. Another problem with gas will be matching supply to regions of highest demand. What I’d call south eastern Australia consisting of SA, Vic and Tas all rely on mature basins that will run out early. The remaining reserves will be coal seam gas in central Qld and the offshore WA natgas basins. Long pipelines will need to be built to the southeast. Rex Connor predicted this 30 years ago.

    Thus to switch or replace Vic brown coal stations to gas will eventually need backup gas lines to WA or Qld. Loss of energy independence and cost is why it won’t happen. Integrated gasification combined cycle IGCC of coal must seem like the gas-poor man’s alternative. At one time an IGCC coal developer could promise the govt they would add on carbon capture and storage, but now the govt doesn’t care.

    Apparently Rudd will announce a beefed up MRET pIan (ghost written by Santos) which will set the scene for more gas fired generation. I expect the public will just accept it and pay through the nose because they don’t know any better.

  5. “Gas already delivers close to 70% of the carbon intensity reduction that a shift from coal to nuclear would achieve in eastern Australia”

    Let me give you another (marketing) spin on what I think David Knox meant here. First he is talking about operating emissions not full life-cycle. This is what is important to him because that’s what he expects to be paying a carbon cost on.

    His “gas already delivers” statement refers to a new CCGT plant installed today. The best currently available CCGT operating emission intensity is around 400 kg/MWh. He believes this will come down to 350kg/MWh (based on his slides). On the same slide he has the average coal operating emission intensity at 1030 kg/MWh. So new gas plants will reduce coal emissions by 680 kg or 66%. 70% is the closest ‘round number’.

  6. Martin, I think you’re correct that this is what he or his spin doctors meant. Indeed, I’d calculated the 66% figure also — on an economy-wide basis — in the post above (with very generous assumptions for gas). Statements like this are disingenuous and highly misleading.

    So you get the prize — which is a new set of bootstrapped figures (Martin knows what I mean!).

  7. Barry, Excellent article.
    Very interesting and constructive comments by all (DV82XL, donb, Milan, John Newlands and Martin Nicholson).

    Martin Nicholson. Thank you for explaining David Knox’s spin in the statement

    Gas already delivers close to 70% of the carbon intensity reduction that a shift from coal to nuclear would achieve in eastern Australia.

    How dishonest is that!! This is the sort of intentionally misleading spin we’ve come to expect from the renewables advocates.

    David Knox has compared a new CCGT with the existing coal fleet. The existing coal fleet includes power stations that are over 40 years old. Hazelwood, for example, emits 1.53 t CO2/MWh and is a large power station at 1600MW. There are many other such power stations included in the average that David Knox has used in his comparison with a new CCGT generator.

    He has also chosen to ignore all the gas emissions enroute from gas well to the generator. This is a ploy often used by the gas advocates.

    What David Knox should do is to compare new technologies on a fair and comparable basis. To do this, all technologies need to be able to produce power on demand and at the required quality. So, for example, wind power cannot be compared as a technology on its own; instead a mix of technologies comprising wind and its back up generators must be considered when comparing emissions and costs for wind generators.

    David Knox should have compared: new CCGT, new coal, new nuclear and new wind with gas back up. The gas backup must be mix of efficient CCGT and inefficient but more responsive OCGT generators. The emissions factors for these new technologies, for each Australian Electricity Market region, are listed in Table 41, p63 here: http://www.aemo.com.au/planning/419-0035.pdf. In summary the emissions intensities are: In t CO2-e/MWh”
    New CCGT = 0.4 – 0.5
    New OCGT = 0.66 – 0.76
    New Coal Black = 0.85
    New Brown Coal = 1.0
    New nuclear = 0.02*
    *refers to Gen II

    Funnily enough the comparison David Know should have done has already been done, here: http://bravenewclimate.com/2010/01/09/emission-cuts-realities/

    Then David Knox would conclude:

    1. Coal emits the most CO2 but is the least cost option (if the cost imposts on nuclear remain in place);
    2. Nuclear cuts emissions the most and is the least cost way to reduce emissions
    3. Gas is an in-between option, but with the fastest cost growth. Gas would cut emissions from coal by about 50% (not 70%), but the electricity prices is very susceptible to the price of gas. Gas price is expected to grow rapidly as we run out of oil, because gas is seen as a replacement for oil and the gas and oil prices are related.
    4. Wind is a total dud. The emissions from gas generators shadowing and backing-up for the intermittent wind generators is about the same as gas generators alone (some say more). So wind with gas backup is a far more costly option than gas alone. There is no real benefit of wind power.

    David Knox’s statement that:

    Gas already delivers close to 70% of the carbon intensity reduction that a shift from coal to nuclear would achieve in eastern Australia

    is misleading disinformation.

  8. Barry,

    Slightly off topic, but I have a suggestion.

    The suggestion is for your group of gurus to extract electricity data from the AEMO web site and chart the data in GapMinder or some other freely available tool.

    If adopted, the suggestion should deliver benefits to society (both world and Australian) and to all who want to analyse Australia’s electricity and emissions data.

    Background

    For some time I have wanted to extract the electricity power output (MW) for each generator, by half hour or shorter period. I want to compare the fuel use (coal and gas) and emission in periods when the wind power output is fluctuating rapidly with the periods when there is little wind power being generated. I am hoping to be able to estimate how much more gas and coal is used when backing up for wind power. (By the way, this has already been done for Texas).

    Yesterday, I wanted to do something much simpler. I wanted to extract from the AEMO web site the total electricity generated by Hazelwood power station in 2009. To do this, I’d have to download, in txt format, 365 files for one year, each 35MB in size. Then I would have to extract the output from the eight Hazelwood units from the 365 files, and sum them for each time period. The files look like this: http://www.nemweb.com.au/REPORTS/CURRENT/Daily_Reports/

    That is an example of ‘open access’ to our ‘publicly available electricity data’ (sarcasm alert)

    What I would like:

    I’d like to be able to extract and chart our electricity information like GapMinder does (example of what I mean showing for electricity consumption versus life expectancy): http://www.gapminder.org/world/#$majorMode=chart$is;shi=t;ly=2003;lb=f;il=t;fs=11;al=30;stl=t;st=t;nsl=t;se=t$wst;tts=C$ts;sp=5.59290322580644;ti=2007$zpv;v=0$inc_x;mmid=XCOORDS;iid=pyj6tScZqmEcKxvG4lnIreQ;by=ind$inc_y;mmid=YCOORDS;iid=phAwcNAVuyj2tPLxKvvnNPA;by=ind$inc_s;uniValue=8.21;iid=phAwcNAVuyj0XOoBL%5Fn5tAQ;by=ind$inc_c;uniValue=255;gid=CATID0;by=grp$map_x;scale=log;dataMin=5.71;dataMax=28213$map_y;scale=lin;dataMin=23;dataMax=86$map_s;sma=49;smi=2.65$cd;bd=0$inds=
    (Click on the play button and try changing the axes).

    The GapMinder software is a free download. What we need is to extract the AEMO data.

    Here is an example of what has been done by Andrew Miskelly, in his own free time, to extract data from the AEMO web sites. In this case it is the wind power data only. http://www.landscapeguardians.org.au/data/aemo/ This site produced Figure 1 here: http://bravenewclimate.com/2009/08/13/wind-and-carbon-emissions-peter-lang-responds/

    This demonstrates it is possible to extract the data.

    What a great benefit this would be for educating and assisting Australians to become knowledgeable about electricity and our largest single source of greenhouse gas emissions.

  9. Thanks for the prize Barry :)

    Another thing that I don’t think has been discussed about gas is that it can only be a short-term fix to reducing the emission intensity (EI) of electricity generation. According to Treasury figures we need to get the average EI down below 50 kg/MWh by 2050. Gas will not be able to do this without CCS and it seems a stretch to think it could even with CCS despite the rather optimistic 40 kg/MWh on the Santos slide (which I assume is based on capturing 90% of the emissions but someone forgot about the 40% extra energy needed in the capture and storage process).

  10. Barry …

    There’s an argument going round the blogs about the contribution of wind to South Australia.

    I suspect as many do here that wind is a stalking horse for gas and the roughly 900 MW of installed wind capacity in SA has contributed a lot less than the 270MW that a CF of 30% would imply but I’d like to have some basis of quantifying how much extra gas is used to allow 900MW of name plate in SA.

    Is there any easy way of finding out?

  11. Ewen I’m not really into conspiracy theories about the gas industry but in answer to your question, Mark Diesendorf argues in his book that you need a little gas backup for wind (as you would expect him to say) but he then goes on and says it is 25% of the installed wind capacity. So 900MW of wind needs 225 MW of gas backup. As this number came from Diesendorf you can only assume that is the minimum figure.

  12. Mark Diesendorf argues in his book that you need a little gas backup for wind (as you would expect him to say) but he then goes on and says it is 25% of the installed wind capacity. So 900MW of wind needs 225 MW of gas backup. As this number came from Diesendorf you can only assume that is the minimum figure.

    That’s ridiculous. How could anyone not see through that assertion instantly?

  13. Martin Nicholson,

    We have been discussing the amount of back up required for wind for a long time. At one extreme we have Mark Diesendorf saying we need only 25% of the wind capacity to be backed up.

    At the other extreme we have groups within the industry saying we need 100% back up and perhaps even more. These groups argue it can be more because of the ratio of CCGT to OCGT and how this ratio changes from the high wind to low wind period throughout the year. (new OCGT emits only 15% to 25% less CO2 than new black coal, most of Australia’s gas generation is OCGT)

    Explaining this a bit more, if we want to maximise the CO2 emissions cuts we need to maximise the amount of CCGT. However, we also need sufficient OCGT to be able to shadow and back up for the wind power when it is fluctuating wildly. And we need to keep sufficient OCGT available to come on line when we need it when the wind power drops suddenly. So we would need to build as much CCGT as possible to maximise the CO2 reductions in the low wind periods and we must also overbuild the OCGT to allow us to follow the wind’s volatility in the high wind periods. The result is that the amount of gas capacity needed to back up for wind power can exceed the amount of wind capacity installed.

    I lean closer to believing this view than to the wind power industry’s spin.

    This provides more detail: http://www.masterresource.org/2010/02/wind-integration-incremental-emissions-from-back-up-generation-cycling-part-v-calculator-update/#more-7271

    Recently produced studies of the Texas and Colorado electricity systems show they are emitting more CO2, as a result of the cycling of the coal and gas plants to back up for wind, than would be the case with no wind power.
    http://www.windaction.org/documents/26827

  14. Barry,

    Finrod, I think Martin means 900 MW of peak wind must be shadowed by 225 MW of gas backup. At a 30% CF, that’s 300 MW of delivered wind + 225 MW of gas. Not such a great deal.

    Yes, that is what the wind advocates are saying, but many do not agree. I suspect, for simplicity we should assume that 1GW of wind needs to be backed up by 1GW of gas if these are being promoted as an alternative to a baseload plant. To argue otherwise should be taken as spin by the renewables and gas industry.

  15. Finrod, I think Martin means 900 MW of peak wind must be shadowed by 225 MW of gas backup. At a 30% CF, that’s 300 MW of delivered wind + 225 MW of gas. Not such a great deal.

    In that case there will surely be times when the only power available will be the gas station… unless they’ve gone back to the “The wind is always blowing somewhere” pitch and propose to interconnect every wind farm on the continent, which as I understand it is too small a region for the pitch to be true.

  16. Finrod, Barry, Martin,

    The argument being pushed by the wind advocates is a nonsense. They are usiing average power output from the wind power station instead of the total capacity. This is a fudge. The wind plant occassionally produces up to its full poewer, so we do have to be able to shadow for the wind power all the way from zero to 100% output. We do not have the transmission systems and the enetgy storage systems to avoid having to back up for the full range of wind power output within each region. This is being shown all over the world.

    Its worth studying this to understand all this:
    http://www.masterresource.org/2010/02/wind-integration-incremental-emissions-from-back-up-generation-cycling-part-v-calculator-update/#more-7271

  17. The wind plant occassionally produces up to its full poewer, so we do have to be able to shadow for the wind power all the way from zero to 100% output.

    I figured from the size of the backup gas plant that they were sanguine about just throwing away the extra power from the times of higher production. As far as I’m concerned, my original impression stands.

  18. Perhaps there is some misunderstanding about capacity credit allocated to wind. Capacity credit (or capacity value) is the amount of ‘guaranteed’ capacity that a generator can contribute to system reliability. This is not to be confused with capacity (or load) factor which is the ratio of the actual energy output from the generator over a year to the output it would produce if it operated non-stop at full capacity. System reliability is top of mind for electricity network system operators. They need to ensure they have sufficient total capacity to reliably supply the anticipated load.
    In SA, AEMO allocated 8% capacity credit for wind. So the 900 MW of wind is allocated only 72 MW of capacity credit. New gas plants could have a capacity credit of 99% (1% risk of failure). To consider wind as contributing 300 MW toward system reliability the system operator needs to have (300-72)/0.99 = 230 MW of gas system reserve capacity. This is the way system reserves are calculated. There is no attempt to backup all the wind capacity. Some networks might chose to allocate zero capacity credit to wind. In this case wind would purely be seen as negative load and there is no need for any backup.

  19. It’s the height of stupidity to have to back up dilute, discontinuous wind from 0-100% [100% is the most likely scenario] with any emissions producing generating form.[gas/coal] Overall cost for the base load power will be greater and emissions reductions will probably be zero to precious little. Check out what has happened in Denmark.

  20. Martin,

    You are talking about something different, which I think is used to confuse and befuddle the unwary.

    The renewable advocates are suggesting we can have 20%+ of our energy provided by wind power.
    When the wind power drops from 100% to 0% of installed capactiy in an area, something has to back up for that. Either we cycle coal or gas.

    I am talking about the amount of investment needed in capacity to back up for wind power when it is a significant proportion of our grid. When that is the case, my belief is that the capacity credit is close to zero. I know what the wind industry is saying, but I don’t believe it.

    Regarding AEMO allowing 8% capacity credit for wind in SA it is worth noting that they allow only 3% in Victoria. However, importantly, we are dealing with wind at about 1% of total installed capacity, so at the moment, wind is bludging off the other generators and they are not paying their fair share.

  21. Peter what I am describing is how I understand the SO calculates the needed gas backup capacity and I am certainly not trying to befuddle anyone. The more the wind capacity penetration increase, the greater the potential impact on the network and the lower the capacity credit given to wind. Beyond 20% penetration, wind will probably get no additional capacity credit. So it may prove to be a maximum point of penetration. Remember also we are talking about ADDITIONAL backup not total system reserve. The networks already has 20% system margin to handle unforseen events (including a complete dead calm across all wind farms). The SOs are not going to risk such event happening without having sufficient system reserve to handle it.

  22. Really, assuming you could find a site for industrial scale wind near a major source of demand and you could reliably get a CF of 30% with only infrequent dips below that figure, then the best option would probably be mostly OCGT with CCGT for any anomalously long lull.

    Sure it’s more Co2-intensive but providing you aren’t doing it all that often and only on short notice, you’re probably ahead. Cycling those CCGT plants isn’t all that good for their life, AIUI.

    Still, one has to ask — why would one bother tying oneself in knots trying to wring at least the 30% out of the wind resource when one could simply have bog standard nuclear plants with 90% availability that require no net back up at all.

    Given the low marginal cost of running the plants why wouldn’t you run them at near optimal efficiency and where needed, dump any surplus power cheaply into the market. In the end, that makes a lot of other usages a lot more economically viable. Some of these would be taxable.

    In particular it would underpin the conversion to PEVs since a lot of recharging will take place off peak. One would expect a narrowing of the difference then between peak and base load, making the offpeak business more viable.

    That’s something that gas simply can’t do at comparable cost or CO2 efficiency.

  23. Further to my previous comment, let’s consider that we want to replace Hazelwood, a 1600MW coal power station. Without getting into the details of the system reliability, and the effect of an unscheduled shutdown of one or more units, letss just consider what investment is required to replace Hazelwood with equivalent power quality, and what are the emissions.

    One option is to replace Haselwood with 1600MW of Combined Cycle Gas Turbines (CCGT). This would provide the equivalent base-load power that Hazelwood provides now. This option would reduce the emissions from 1.53 t/MWh to about 0.41 t/MWh.

    Alternatively, we could replace Hazelwood with say 1600 MW of wind power with gas back up. I’d argue that we’d probably install 1600MW or more of gas to back up for the 1600MW of wind power. This would be needed to optimise the emissions while providing the same power quality.

    If we replaced Hazelwood with gas only, we’d replace with CCGT. However, if we replace with wind and gas, then the gas must be a mix of Open Cycle Gas Turbines (OCGT) and Combined Cycle Gas Turbines (CCGT).

    In the low wind periods we’d like as much CCGT as possible to minimise emissions – say 75% of the 1600MW. Then the average emissions from the 25% OCGT and 75% CCGT would be would be 0.52 t/MWh. Factor this up to say 0.60 t/MWh to allow for the cycling to follow the wind power intermittency. That is the situation in the low wind period.

    What about the high wind period? We’ve committed to 75% x 1600MW of CCGT for the low wind period, or 1200MW. For the high wind period we are going to need sufficient OCGT to follow the rapid wind power fluctuation from near 0% to 100% of wind capacity. The CCGT can follow some of the changes, and better than coal. We will need more than 25% OCGT to work with the CCGT to follow the wind power fluctuations throughout the cycles. So the total amount of CCGT+OCGT installed will be more than 1600MW.

    During the high wind period let’s assume that OCGT generates half of the energy generated by gas. The emissions from CCGT+OCGT are 0.54 t/MWh, and factored up to say 0.7 t/MWh to allow for the cycling. (all numbers very rough to get the concept across).

    Now let’s assume wind power generates 30% and gas 70% of the electricity. In the low wind period emissions are 0.42 t/MWh (70% x 0.6 t/MWh) and in the high wind period they are 0.49 t/MWh. If half the year is high wind period and half the year is low wind period then the average emissions for the year are 0.46 t/MWh.

    In summary:
    Emissions:
    Hazelwood as it is now = 1.53 t/MWh
    CCGT replacement = 0.41 t/MWh
    Wind and gas (OCGT &CCGT) = 0.45 t/MWh

    Capital investment:
    $1314/kW for CCGT only
    $3895/kW for wind plus CCGT and OCGT back up.

    On the basis of these rough figures, wind with gas back up has higher emissions than gas alone and costs three times as much. Wind power makes no sense.

  24. Ewen,

    Given the low marginal cost of running the plants why wouldn’t you run them at near optimal efficiency and where needed, dump any surplus power cheaply into the market.

    In particular it would underpin the conversion to PEVs since a lot of recharging will take place off peak.

    Not just PEV. Dont’ forget storing energy in pumped-hydro plants in the early hours of the morning when demand is low

  25. The reality is Peter no rational person once they really understood about wind power and electricity networks would ever suggest trying to replace Hazelwood with wind power. It’s not going to happen. It seems unlikely that wind will directly replace any baseload generators. It might replace a small fraction of baseload capacity from time to time but actually I doubt it – and certainly not before we have a price on carbon high enough so that the SO decides to part-load a few of the baseload coal plants on a really windy night to reduce costs.

  26. Quite right Peter …

    and we might add

    a) desal
    b) sewage treatment
    c) pumping water
    d) heavy industrial usage

    Doubtless if the power were cheap enough in the summer, people would run their aircon at night and get a good night’s sleep.

    Quality of life counts.

  27. Martin Nicholson

    The reality is Peter no rational person once they really understood about wind power and electricity networks would ever suggest trying to replace Hazelwood with wind power.

    Who is rational? The rational guys are in the industry, and they are not being listened to. The irrational people – NGO’s and green activists – have the ear of governments (throughout the developed world).

    You reckon no one will say wind power could and should replace base-load power stations. I expect about 80% of the population believes wind power should replace our base-load power stations – thanks to all the hype by the green activists and propaganda agencies like our national broadcaster.

    Diesendorf and Jacobson, and the organisations like Greenpeace and ACF are continually preaching that wind farms can and must replace base-load power stations.

    Then we have the senior politicians saying, for example, Capital Wind Farm will provide the power for the Sydney Desalination plant. (I haven’t seen the Capital WF turbines turning in numerous trips past.)

    And, of course the Victorian desalination plant is going to be powered by wind turbines, so we are told.

    It might replace a small fraction of base-load capacity from time to time but actually I doubt it

    If wind power only ‘replaces base-load capacity from time to time’, then it is not replacing base-load capacity at all. To replace base-load it must be reliable, and available all the time. If we are requiring the distributors to buy power from wind farms instead of from the base-load generators, then there is an extra cost for the baseload power stations that must be added into the cost of electricity to keep the power stations viable.

    The distortions such as RET, feed in tariffs, subsidies, etc are what we need to remove, completely, if we want to get rational.

    and certainly not before we have a price on carbon

    I expect you may have missed the discussion on previous BNC threads this subject. Raising the cost of electricity through another distortion like ETS or CPRS is the last thing we need until all the distortions have been removed. Raising the cost of electricity is bad for Australia and bad for humanity throughout the world. Bad! Bad! Wash your mouth out. :)

    And, yes, I understand what the politicians, NGO’s, gas industry, and renewables industry are saying. But we are talking about ‘rational’ policy, not the wants of special interest groups.

    so that the SO decides to part-load a few of the base-load coal plants on a really windy night to reduce costs.

    Oh my Gawd! That wont reduce costs. It will raise costs. Base-load at $30/MWh versus Wind at $125/MWh (including the extra transmission cost externalities) plus an inflated price for base-load because the generator knows that from time to time it cannot sell its power – thanks to the lovely government which has made laws that say the distributors must buy wind power whenever it is available. That’s rational?

    You said “no rational person …”. The problem is there are very few of them in the positions where they are needed.

  28. Peter Lang:

    I have a challenge for you. You and others have done an excellent job of demolishing the reputation of wind’s utility as a potential source of baseload power. I have asked you in the past about wind’s potential as a stranded asset and you have tended to be dismissive of it’s role here too. I have had a thought but can’t take it much further without your analytical capabilities. Here it is:
    1) Let’s start by thinking about wind installations already extant – we’re hardly likely to chop them all down.
    2) Wind’s ERoEI is goodish, CO2 emissions excellent and economic costs (excluding backup) quite good.
    3) Wind makes electricity directly. Fossil fuels and nuclear make it through steam, the low grade heat from which is generally wasted at centralised plants.
    4) CHP schemes are therefore more efficient but depend upon distributed generation which has its own inefficiencies and extra costs.
    5) Centralised storage of electricity is an economic no no. (barring pumped storage for expensive peaking power).
    6) Low grade heat has few uses but one is to heat and/or absorption chill domestic accommodation and this represents a very large energy demand.
    7) Low grade heat can be stored quite cheaply and efficiently (I think) but is not cost effective to move around over anything other than very short distances (hence must be widely distributed – each contributing household having an insulated, accumulator hot water tank)
    8) Heat pumps are efficient gadgets that can get 3kW/hth from 1kW/he.
    9) Electricity is priced according to demand. Although one can predict that baseload plants might be producing cheaper power at night, predictability from wind can’t be assessed.
    10) Could central power providers develop a sophisticated and affordable enough system to be able to provision householders with unpredictable but cheaper power, only at the times of its availabily,to smooth out their other power demands? This could (should?) be dedicated to powering heat pumps to heat water to be stored in accumulator tanks.

    There, Peter, is my idea – not backed by numbers. I suspect that you might be able to blow it out of the water straight away. If not, number crunching might be required and it’s over to you.

  29. Douglas, Wise,

    Thank you for the comment and challenge, and for the supportive comments in your introductory remarks.

    I’ve gone through your suggestion. In my opinion if we are rational we would not build any more wind farms, but we wouldn’t ‘chop down’ the ones we have already built. We’d honour the contracts that allowed them to be built. I expect, just like the American ones (see photo in link below), we wouldn’t maintain them when they fail. We’d just let them fall into disuse. If we focus on the big picture, wouldn’t focus on the 1% or so of wind power we’ve already built, we’d focus on the other 99% of our electricity generation assets we need to deal with.

    Regarding your main point which I understand is:

    Could central power providers develop a sophisticated and affordable enough system to be able to provision householders with unpredictable but cheaper power, only at the times of its availabily,to smooth out their other power demands? This could (should?) be dedicated to powering heat pumps to heat water to be stored in accumulator tanks.

    That is happening, has been developing for at least two decades, and will continue to progress. But it takes time. Yes, we will continue to develop smarter control systems and there will be lots of changes. So, yes to your main point. The distributors are already moving to do what you suggest. The main reason, however, is to reduce their risks and costs, not ours! I believe, for a long time to come, central generators will be the most economic way to provide our power needs, except in remote locations. I believe Franc has the near ideal system with about 76% nuclear and the other 24% comprising hydro and gas.

    I’ve provided responses below to some of your assumptions. Some of my responses are a bit sarcastic; no offence meant.

    1) Let’s start by thinking about wind installations already extant – we’re hardly likely to chop them all down.

    (sarcasim alert)When they die we’ll move them to old-wind-farm-homes like the pictures about half way down here: http://webecoist.com/2009/05/04/10-abandoned-renewable-energy-plants/

    2) Wind’s ERoEI is goodish, CO2 emissions excellent and economic costs (excluding backup) quite good.

    Definitely not. CO2 emissions from wind are not excellent; they are about the same perhaps worse than gas alone when the emissions from the back up are included, and they must be included for a sensible comparison.

    Wind is ridiculously uneconomic. At $110/MW plus transmission costs of about $15/MWh. Furthermore, the energy they provide is of low value energy, some would say it is more trouble than it is wortyh valueless. It is subsidised by about 150%. If it wasn’t mandated no one would buy it. The alternative is basload power at about $20 to $50/MWh. Why anyone rational person would believe wind power is economic beats me (that’s not a dig at you, but it is at the polies and media and the NGO’s that have driven us to build wind farmsa, solar panels, bio fuel and to ban nuclear).

    3) Wind makes electricity directly. Fossil fuels and nuclear make it through steam, the low grade heat from which is generally wasted at centralised plants.

    This is not relevant

    5) Centralised storage of electricity is an economic no no. (barring pumped storage for expensive peaking power).

    Pumped hydro is not suitable for storing energy from intermittent generators such as wind – see http://bravenewclimate.com/2010/04/05/pumped-hydro-system-cost/#comment-58646 and other discussion on the pumped hydro thread.

  30. Peter,

    Thanks for your reply. I take no offence whatsoever and, as you appreciate, was considering what to do with the power from windfarms already built.

    You may, however, have misunderstood some of what I attempted to say. ERoEI, CO2 sparing and economics of conveniently situated onshore wind (which, I think, describes most of current capacity) compare favourably with corresponding figures for Western built new nuclear UNTIL on factors in backup.

    I was trying to think up something that wind power might be used for without recourse to backup. My solution was to feed it into the central grid and get rid of it again into distributed storage in the form of low grade hot water, simultaneously enhancing efficiency with heat pump technology. In fact, I believe you acknowledged that this could be technically feasible. I’m sure you’ll accuse me of rootling round in the weeds again and thinking more about the 1% than the 99%. However, before the momentum to stop it gathers pace and achieves its effect, we might have 5-10% of it to deal with. Anyway, I think my idea is a better fit for the UK than Australia for reasons that will be obvious to all. It also seems a better fit than storing it in the battery of one’s electric car which, if practicable, is better matched to reliable surplus nuclear not needed for baseload at night.

  31. Ewen,

    Energy storage can be centralised or localised at each generator. For intermittent generators, I believe the energy storage has to be at the generator (such as at the wind farm), not centralised storage.

    The options for storage at the generator are chemical, hydrogen, fly wheel etc. Chemical is the nearest to being economic. The chart included in the Pumped hydro lead article compares the cost of the various storage options. If you want to dig deeper, you might want to look at this site (and perhaps book mark it): http://www.electricitystorage.org/tech/technologies_comparisons.htm

    The Solar Power Realities article compares the cost of pumped-hydro with NaS batteriies.

  32. Douglas Wise,

    … CO2 sparing and economics of conveniently situated onshore wind … compare favourably with corresponding figures for Western built new nuclear UNTIL on factors in backup.

    I am having trouble with this statement. First, I don’t agree that you can compare nuclear and wind, unless the wind is isolated from the grid. Perhaps you are asking for a use for wind energy if isolated from the grid. I can’t think of any economic use for it. If there was, I reckon we would have thought of it long ago. The fact wind has to be subsidised by such a large margin shows that it cannot provide power as cheaply as can be obtained from centralised powers stations via the grid. So the only use would be in locations with no grid supply. And then it would be very expensive electricity. Some of the European wind farms were decommissioned, towers and turbines removed to be replaced with larger one. They tried to give the old ones to developing countries. No one wanted them. The cost of building, maintaining, building transmission and then getting only irregular power supply meant they were of no value.

    Second, I don’t see wind power as economic compared with nuclear. To me wind generates electricity at about $125/MWh (transmission costs included) for power which is virtually worthless, compared with nuclear at $100/MWh (for a FAOK plant in Australia, with all imposts remaining in place and based on prices before the UAE contract was awarded) for reliable, dispatchable baseload power. The cost of nuclear could and should be less than coal if we had a genuine level playing field. I can’t imagine anyone buying energy from a wind farm voluntarily.

  33. I think it is interesting that Germany has just also reneged on its Kyoto commitments
    http://www.spiegel.de/international/germany/0,1518,691194,00.html
    Perhaps to their credit they put their money where their mouth is and massively invested in renewables with something like 5 GW of solar and 24 GW of nameplate wind power if I recall. That puts Merkel in a different league to Rudd who talked big but did nothing.

    The point being that a non-nuclear country can invest in renewables until it hurts yet they still need to burn more carbon.

  34. John Newlands,

    Good points. Now with the PIGS (Portugal, Ireland, Greece and Spain) in financial strife (and Iceland and UK and Italy also in trouble), I am wondering how much or this is due to irrational government policies. I’d equate high expenditure on renewable energy (including all the buried subsidies, tax breaks, feed in tariffs, mandatory targets, etc) as an indication if irrational policies.

    I was thinking, last night, of charting debt/GDP versus renewables expenditure/GDP for all the EU countries, or something like that. It might be easier to chart debt/GDP versus % or electricity generated by renewables.

  35. Martin Nicholson, Barry, Finrod, Douglas Wise,

    Regarding the discussion about capacity credit and the amount of conventional generation needed to back up for wind capacity, several of the questions and comments below this article explain this well. They are also interestinga s they mention some of what is being done to attempt to quantify the effects.

    http://www.masterresource.org/2010/02/wind-integration-incremental-emissions-from-back-up-generation-cycling-part-v-calculator-update/comment-page-1/#comment-9878

  36. The Santos coal seam gas wells in the Qld Darling Downs were shown in the ABC Landline show, no link yet. Maybe by year 2020 or so WA and Qld will be the remaining gas rich States with NG and CSG respectively. I wonder how interstate rivalry and the absence of a national ETS could affect the actions of the future gas-poor States. Victoria for example could dig its heels in over brown coal which is still 10% the cost of gas if upthread figures are correct.

    On the other hand some kind of crisis could trigger a sudden NP decision. Outer Melbourne got to 48C last summer perhaps it will break 50C next summer. In that event maybe political pressure could force an abandonment of brown coal. Probably not.

  37. Peter Lang:

    I am not sure why you are referring me to the above link which I have previously read. It implies that you take me to be a supporter of wind energy. I have made it abundantly clear that I am not and, also, that you have, in part, been instrumental in my conclusion.

    My challenge to you related to how best to use the electricity from wind energy that is currently fed into the grid and the extra that will be created before politicians wake up to the realities that we both accept.

    My suggestion was to store it in the form of distributed low grade heat (hot water) for domestic use which I thought might handle its intermittency problems better than anything else I could think of.

    Your response didn’t match my challenge. Instead, you responded with another rant against wind with which I agreed but which didn’t further the debate.

  38. My suggestion was to store it in the form of distributed low grade heat (hot water) for domestic use which I thought might handle its intermittency problems better than anything else I could think of.

    Rather than reconfiguring the whole domestic grid to ensure that wind power goes to domestic water heating, how about channelling whatever wind power is close enough to be delivered without excessive transmission loss to a preheating unit for the water to be cycled through the dteam plants for coal power stations? This would reduce the amount of coal burned for a given power production level.

  39. Douglas,

    I thought I answered your last comment on 29 April 2010 at 23.04,.

    My post today at 2 May 2010 at 12.03 adressed to several people was in relation to earlier posts and especially to the discussion with Martin Nicholson and Barry about capacity credit and how much back up capacity is required (i.e. investment in back up capacity as distinct from system reliability). I addressed it to you as well for your interest.

  40. Finrod and Douglas Wise,

    I’d agree with with this suggestion as long as it is not proposing that the wind farms feed into the national grid. The wind farm investors would provide their own grid to transmit the power to the user. And all this would be done on a fully commercial basis with no subsidies. Sounds fine to me on that basis.

    I guess there would be quite a few externalities that should be included – such as the loss of land value due to to the transmission towers and the wind turbines. All this should be included.

    The point that Douglas Wise seems to miss is that wind is more expensive that nuclear power, even without back up, and supplies low value power, some would say almost worthless power. So why would we want to spend any more on building wind turbines?

    The ones we’ve built already provide only about 1% of our energy, so dealing with the stranded assets is not worth us wasting our time on. There are far more important issues to address – such as how are we going to supply the other 99% of our electricity, cheaply, securely, and cleaner and safer than now?

  41. The point that Douglas Wise seems to miss is that wind is more expensive that nuclear power, even without back up, and supplies low value power, some would say almost worthless power. So why would we want to spend any more on building wind turbines?

    Hmm. Peter, I don’t think you’ve quite grasped what Douglas is proposing here. We’re not talking about going out and developing wind farms for the sake of doing these things, just trying to find something, anything useful to do with the otherwise useless already-existing wind infrastructure, if it can at all be done. It would not suprise me to learn that the proposals are impractical.

  42. Without knowing the retirement schedule of existing fossil fuelled generation I’d hazard a guess how different Australian States could source increased energy needs. This assumes that emissions reductions are off the political radar for at least the next 3-4 years.

    WA – new natgas fired plant >500 MW
    Qld – CSG fired plant >500MW
    NSW – CSG plants < 500 MW and supercritical coal
    Vic – no change to cheap abundant brown coal
    Tas – import more Vic coal power via Basslink
    SA – increase imports from Vic and NSW

    Absent a recession it is unlikely emissions will decrease. Renewables will get just token support from either a future Abbott or Rudd government.

    Incidentally the ABC Landline program reinforced that the ETS was probably always going to be a political football. Before the ink was dry the farm lobby was hellbent on getting exaggerated and unverifiable carbon credits.

  43. Thank you Finrod and my apologies to Douglas Wise,

    Peter, I don’t think you’ve quite grasped what Douglas is proposing here. We’re not talking about going out and developing wind farms for the sake of doing these things, just trying to find something, anything useful to do with the otherwise useless already-existing wind infrastructure,

    I suggest a rational approach would be to let the existing wind farms and those already approved to remain until they are no longer economic. All existing contracts should be honoured.

    However, we should cease, from now on, the mandatory renewable energy schemes, feed in tariffs and all other subsidies and tax breaks.

    The amount of electricity wind farms contribute to the grid is small now, and can be handled in the grid. The proportion will become less over time, if stop building them.

    Since the wind farms are only a very small part of the existing electricity infrastructure, the issue of them being a stranded asset is a relatively trivial issue. I suggest we need to focus our attention of the far more important issue of what should we do going forward from here. IMO we need to focus on the rational approach to providing our increasing electricity needs, at least cost, with the highest level of energy security, and with improving health and environmental outcomes.

  44. John Newlands,

    I think this is a pretty good summary. We could discuss some points (like new coal in NSW), but this would be down in the weeds.

    I’d like to focus on what we could do, going forward, that would be both rational and achievable politically.

    I know I’ve said this before, but this would what I’d like to do:

    1. We (the government) will provide policy for a 20 year time frame. The policy will provide energy security at least cost. It will meet our international commitments, as they evolve over time. We will provide continually improving health and environmental outcomes.

    2. To achieve these outcomes it is essential our outcomes must be rational. They must be based on rational analyses not on beliefs, ideology and emotive issues

    3. To achieve energy security at least cost together with improving health and environmental outcomes, we will have to adopt nuclear energy, and it will be come our main source of electricity.

    4. To implement nuclear power at least cost we will have to remove many of the imposts that are causing nuclear power to more costly than it needs to be. Many of the imposts on nuclear power are irrational.

    5. To find the optimum way to implement low-cost, low-emissions electricity in Australia, we will establish a single organisation to manage the transition. You could think of it like a modern equivalent of the Snowy Mountains Scheme. It will be an engineering organisation.

  45. I accept that carbon pricing is out of contention thus removing a force for change. It fortunately takes away some dead weight as well. However I think political shocks could be on the way

    1) a high Green vote in the 2010 Federal election
    An unstable Liberal or Labor government might have to cut deals with the Greens. I suspect that could kill new coal but the public would have to pay for an expensive renewables fronted gas build. A backlash would follow.

    2) peak oil and a shift to gas for transport
    Credible analysts are expecting world liquid fuel prices to escalate 2012-2014. If there is a massive uptake of CNG for trucks and cars and GTL for aviation fuel that will make gas for generation markedly more expensive. Gas could quickly double in price.

    3) China woes
    The Chinese economic balloon could burst either due to monetary instability or running short of domestic coal with imports unable to plug the gap. They may drastically cut other imports like iron ore. Perhaps this will make Australians rethink energy issues.

    My feeling is that something major will force the issue of fossil fuel dependence in the next 5 years.

  46. John,
    I think you are probably correct that we will have a high green vote at the next federal election, with the liberals rolling their pro emission reduction leader and labor delaying ETS.
    The likely outcome will possibly be the continued expansion of new wind capacity by about 0.5GW per year, to 3-4GW by 2014. This would require at most 1GW of new OCGT back-up or an upgrade of the Bass-link to access more of the 2.6GW of TAS hydro back-up that is presently not being used or building a lot of the new wind capacity in TAS.
    Unfortunately starting on nuclear will probably be delayed beyond 2014 so we are unlikely to have any nuclear completed until 2022 or 2024. If new wind capacity continues to be build at 0.5GW/year we would be looking at 9GW wind(3GW average) and perhaps an additional 2-3GW of OCGT (20% capacity) by 2024. this would seem to be a fairly low CO2 option, not as good as having 4GW nuclear but I cant see how 4GW of nuclear is going to be built before 2025-2030.

  47. Neil
    the State ALP member pushing a 2nd Basslink cable got rolled at the recent Tas election. The ALP-Green coalition appears very unstable and might be a foretaste of the Federal scene by year’s end. Tas is at least 25% coal powered since Basslink, up from 0% in 2006. Over 300 MW of new wind (Musselroe, Lake Echo) is on the cards. The gas supply comes from underwater Bass basin and should be good for at least a decade.

    Note I’m pushing your idea of an HVDC cable across the Nullarbor. For different reasons it’s also been taken up by Desertec and Siemens.

  48. Hi Neil Howes,

    Welcome back. I missed your posts.

    Do you have any figures on how much CO2 is avoided by wind power (per MWh) when the emissions from the back up generators are included?

    What do you think of this? http://www.masterresource.org/2010/02/wind-integration-incremental-emissions-from-back-up-generation-cycling-part-v-calculator-update/comment-page-1/#comment-9899

    Do you have any figures on the CO2 avoidance cost for wind power, with back-up and allowing for the higher cost of transmissions and grid stabilisation?

  49. John,
    If it makes economic and technical sense to expand the Bass-Link ( to say 1000 or 1500MW) it should not depend upon one state member. Adding a link from Norseman to Pt Augusta would also enable the very considerable OCGT capacity in WA to be used more efficiently( ie lower CO2 output) and most importantly allow WA wind to be somewhat balanced with the east coast wind.
    Both projects may make sense when local wind capacity exceeds off-peak demand.
    Peter,
    I have had some other distractions but am working on a scenario of all NEM coal-fired power being replaced by 16GW av wind, 6GW av CSP(12GW peak capacity, with 24h storage), 1.5GWav hydro and approx 1.5GW av OCGT(present 4GW capacity) , using your figures for present NEM demand. I am using the data from the 18 wind farms in operation at present( 1600MW capacity, scaled x29) to calculate pumped storage requirements and load shedding losses (wind output above 80% capacity during low demand periods). Adding more OCGT capacity doesnt really help, as the critical requirement is to be able to store excess wind during low demand periods. OCGT capacity would be used mainly to assist hydro in balancing daily peak and seasonal variations in solar and wind. This would require about 16GW of additional pumped storage (>1200GWh over a 5 day period), a little more than double the capacity of the Tantangara/Blowering project you detailed. A 1.5GW HVDC link to WA would reduce pumped hydro peak pumping capacity by 1.5GW but require a larger total storage capacity( which is available).

  50. Neil
    these analyses are getting too complicated to understand without flow charts and time cycle graphs. My reasons for an east-west HVDC cable are

    1) in the long run WA will have the cheapest gas
    Rather than transcontinental pipelines as envisioned by Rex Connor it seems more prudent to send peaking power to the east from the west.

    2) it could fit in with a SA west coast NPP/desal
    I’ve suggested that Olympic Dam could get its 690 MW and 187 ML/d desal from a plant near Ceduna for example. Surplus power from a large NPP could be fed in to the HVDC cable and the new ‘national’ grid. Others have suggested a more distributed approach.

    3) it would help the economics of Desertec type proposals. CSP and so on may make more sense if there is adjoining transmission. It might buy peace from the Greens who could end up holding the cards politically.

  51. John, I agree that eventually we will get an E-W HVDC link.

    The cost however will be substantial at around $1 million/km minimum. HVDC also has a disadvantage in that you cannot easily connect a CSP plant in the desert (or a remote town) to an existing HVDC E-W link in the way you can connect to an AC link. AEMO may need to break the link at several places to include AC busbars requiring DC-AC-DC conversion at each busbar. This all costs in terms of capital and power losses and may make AC more attractive if we expect to have multiple desert based systems connecting along the way. Multi-terminal DC links may be available in the future but are not available today.

  52. Neil Howes,

    The scenarion you are analysisng looks really interesting. Will you be posting an article on BNC. I hope you will.

    I am looking forward to it. Will you be putting costs on the system?

    Did you see this article I wrote, mainly as a result of your prompting? http://bravenewclimate.com/2010/01/09/emission-cuts-realities/

    You and Alexei were both arguing, correctly, that my earlier papers were all very well but they a) considered only the end state after replacement of the existing system with the new system, and b) applied to a single technology rather than a mix of technologies.

    You pointed out that we need to consider the transition period and consider total accumulated emissions during the transition period. You also pointed out that a mix of RE technologies might be more effective and lower cost than the nuclear option. So I did this exercise, to the limits of my capability, to attempt to address your questions – but in a very simple way.

    I presume you have seen this site for the NEM wind farm outputs or you have a better source:
    http://windfarmperformance.info/

    Your scenario analysis looks like it will start a lively and fascinating discussion here.

  53. Martin I note Broken Hill NSW has had a short low powered ‘HVDC Light’ cable for some time, with I guess most of the input attributable to Hunter Valley coal stations. That must work out cheaper than the previous diesel generator.

    The Siemens paper doesn’t go in to the fact that the transcontinental rail track is the logical transmission corridor but it is mostly some distance from the coastline. That won’t help wave power and cliff-top wind (should they be economic) and seaside towns like Ceduna. However a water pipeline to Roxby Downs from a desal somewhere on the coast (currently Whyalla but unlikely to be approved) would create an easement that would intersect the rail line and hence any E-W electrical cable. The number of converter stations or junctions could be kept to a minimum.

  54. I don’t know about Broken Hill but the Terranora to Mullumbimby HVDC light link (I actually live near Mullumbimby) is underground or run in conduit along the old rail line. It is a low power link (105MW/245MW). Most of the energy actually comes down from QLD rather than up from the Hunter. It was very expensive apparently – I suspect a ‘pork barrel’ implementation to solve power shortages in Byron Bay. One of the engineers involved says it runs close to 90 dec C so I wonder what the losses must be on that link.

  55. Early evidence for my belief that natural gas and not electricity will replace most oil based transport
    http://money.ninemsn.com.au/article.aspx?id=1049261
    Major semi trailer routes along the east coast will offer LNG filling stations. I believe the cryogenic tanks need to be kept at -160C and under 22 bar pressure. I presume the government will offer rebates for engine conversions. The LNG network might then expand to CNG filling (ambient temperature, 220 bar) for smaller trucks and private cars. If as some predict petrol and diesel hit $2/L by 2012 then the shift to gas powered transport will be on.

    Thus truckers and motorists will be prepared to pay around $50 per gigajoule of fuel. I derive this on a rough petrol or diesel energy equivalence as (1000 MJ/40 MJ) X $2. I understand long term domestic gas supply contracts are now priced around $6 per GJ. This suggests as oil supplies to transport tighten the non-transport gas market could be thrown into turmoil. Gas fired generators and industrial gas users will have to compete with customers prepared to pay several times as much for gas.

    Note the LNG truck stops will be built by gas company BOC. Santos doesn’t seem to have anticipated transport demand.

  56. John Newlands,

    Are you aware of the LNG trials in the early 1990s. We had LNG powered semi trailers running between Adelaide and Melbourne, refueling at a half way point. We also had the coal trucks running between Appin and Wollongong (they still are, I think). We extracted the Methane from the coal before it was mined, liquefied it and fueled the trucks. We also had LNG powered trucks running between Alice Springs and a gas field (I’ve forgotten where).

    It is very easy to write articles like this. Getting the whole infrastructure in place is not as easy.

    I am sure it will happen eventually, but that does no mean that LNG or CNG will be the main substitute for oil for most of the road transport. I am open on this, but whatever happens, electricity is going to grow its market share of total energy consumptions. It will replace a lot of gas for heating and replace a lot of oil for road transport. I expect nat gas will help out as an interim solution.

  57. @Nicholson, Newlands, Howes: you talk eg of 1.5 GW HVDC lines within AU.

    There have been coronal mass ejections, CME, destroying transformers (as in eg Carrington Event of 1859; or 1921; or the more recent Quebec event) via induction.

    The workshop report on severe space weather events is at:

    http://www.nap.edu/catalog/12507.html

    Antenna ie HVDC line length is said currently on BNC by blogger DV28XL to be a problem of renewables only. Because PV or wind or hydro generation (he cites Quebec hydro) is often a long way from where it is consumed.

    But it seems to me that only if there were small-capacity NPPs dotted around a country and not a few big ones with long transmission lines is the HVDC-antenna-induction problem lessened. So should you not factor warehoused replacement transformer costs into any calculations?

  58. !0% nameplate as baseload, wow I’m impressed. This is grasping at straws, 10% dispatchable power makes the cost of wind per KWh incredibly high, and that’s the way to look at this, if indeed it is so. Frankly I still don’t see how any network of wind turbine generators could guarantee even that dependably.

    There was a CNG conversion program in Canada during the 70’s too, and I knew one person who went ahead with it. He kept it for about two years or so as I recall, until the one filling station in the area that had set up for this fuel, gave it up from lack of customers.

    I drove the car once and I wasn’t impressed by its performance, and the engine didn’t sound happy.

  59. Peter Lalor – Long transmission lines are an issue regardless it is true, however I would suggest that nuclear generating stations (NGS) can be placed at reasonable distances from their markets as installations like Ontario’s plants.

    Thus it would be pointless in my opinion to build NGSs so far away that this would be an issue.

  60. On gas for cars I speculate that if Australia uses 50 Mt a year of oil (now mostly imported) we could use 50 Mt a year of local gas. The graph in Part 1 of the article shows 2020 gas output at 4500 PJ or say 90 Mt. Can Australia go from 90 to 140 Mt (=90+50) without cutting back on LNG exports? We could be selling ourselves short by flogging too much gas overseas without a policy for oil replacement.

    On long transmission lines my view is they should cut the $43bn broadband rollout in half and spent it on key projects. One would be the east-west connector 1400 km X $2m/km = $2.8 bn. It would
    – create a national grid
    – link eastern Australia to WA gas peaking power
    – give the Desertec proposals a fighting chance
    – enable electricity export from a Nullarbor coast NPP/desal that primarily serves Olympic Dam.

  61. We could be selling ourselves short by flogging too much gas overseas without a policy for oil replacement.

    With Rudd as Provincial Governor of this little Satrapy of the New Middle Kingdom, it is doubtful that this issue will be seriously considered.

  62. On lack of zip of gas powered cars some new models have turbo charging. Dual fuel petrol/CNG cars can run on future expensive petrol if there are few NG filling stations around for the early years. It would be good to get say 300km range on $20 of gas. Contrast that to the plug-in Chevrolet Volt that will cost $US45k and get just 65 km on an overnight battery charge. I don’t think battery swap cars will be popular or practical.

  63. I’m not saying that CNG was a great idea – I think Muldoon did it in NZ because he thought it was necessary. In that way, he was something like Pickens, only 30 years ahead of Pickens.

  64. Lawrence I’m saying cut the broadband budget in half and spend it on national projects of which the E-W connector would be just one. I don’t think every remote corner of Australia needs fibre optic. I have satellite and it mostly works OK.

    Re NP on the Nullarbor coast. In just a few weeks time an announcement will be made on the expansion of Olympic Dam mine which is currently limited by the use of local groundwater and diesel generators. BHP Billiton has a site prepared for a desalination plant at Whyalla which will pump fresh water some 300 km to the mine. Due to objections over elevated salinity in a sensitive fish breeding area that desal is unlikely to be approved. Secondly either the company or the State government must provide up to 700 MW of power to run crushers, a larger township (Roxby Downs) and so on.

    If the desal and new generation don’t go ahead it it is possible that uranium bearing ore from OD will be sent to China for processing, thereby costing Australia jobs and profits. If neither happens long term I would expect it would affect the world uranium price (~ $100/kg U3O8) since the expanded OD would be the world’s largest producer. Therefore there should be both a large desal and a large new generator somewhere in that region.

  65. Speaking not just as the liberal I am, but even as the shareholder wealth increaser that I’m not, building out broadband access seems like a very good idea to me.
    Let’s get our priorities straight.
    Fucking with the capability of people to participate in the conveyance of information is as anti-democracy as you can get.
    I don’t support it, and I can’t believe you do.

  66. Lawrence I’m saying the govt should pay for more 4 Mbps satellite or wireless broadband instead of connecting cable to more areas. It’s a huge improvement over copper wire and enables streaming video albeit with bandwidth penalties. The cable broadband rollout also seems to be preoccupied with 100 Mbps speed which could be where a lot of the money is going. If a cutback frees up even $10bn that money would be handy elsewhere.

  67. I have spent some time in areas where there was supposed to be decent broadband, and it wasn’t.
    What if you want to work from home, and you need decent connectivity. In a future deprived of oil, there’s going to be a need to work from home, with a reliable connection, not a satellite dish that loses connectivity everytime something waves in front of it. Upload at dial up speed is crap too.
    I shouldn’t go beserk, I guess, but I’m really REALLY sick of listening to people be polite about the bastards in this world.
    There are the idiots that clog up our lives, like Eclipsenow, and then there are the corrupters that inflict massive damage like the monsters who get elected and then prevent any progress towards a sustainable future.
    Maybe we’ll all be lucky and such Cassandras as myself will be wrong, but I think Richard Branson is right – we are facing a crisis of global scale within 5 years, just in energy alone, let alone the degradation of the biome.
    As far as I’m concerned, when I read in 1983 that CO2 warmed any atmosphere it was in that was the end of the time to talk softly. Now we need to be blunt.

  68. I also think Branson is right. Things could get crook within the next few years with gas unable to save us. Watching the ABC Landline segment on Queensland CSG
    http://www.abc.net.au/landline/content/2010/s2894275.htm
    the segment noted the growing problem with saline discharges associated with CSG. It is unknown whether it will affect the fresh water layers needed for crop irrigation. It showed Federal Minister Ferguson and Premier Bligh apparently at a banquet in Beijing celebrating a liquefied CSG export contract. My question for the honorable MPs is what plan does Australia have to replace 50 million tonnes a year of oil?

    There was footage of two new combined cycle plants in Queensland. I expect they paid less than $10 a GJ for gas. However truckers and motorists may be more than happy to pay 5c per MJ or $50 per GJ. Will gas generators be prepared to match that, a five fold increase? This why I think Branson is right.

  69. http://www.cbc.ca/canada/story/2010/05/09/rural-canada-broadband-internet.html
    http://jutiagroup.com/2010/05/02/peak-oil-are-the-lights-about-to-go-out-on-western-civilization/
    “Warning: 80 Mb audio file. It’s a two-hour show. The Matt Simmons interview starts at 38 minutes in. Among other things, he says peak oil was five years ago.”
    Matt Simmons is only on for about 25 minutes. The interviewer talks over him a lot because the interviewer is too dumb to shup up once Simmons starts answering the question.
    What Simmons has to say is scary. He won’t give timelines, but he says it could cost 100 trillion to put enough money into oil and gas to give it any future at all. In other words, we’re very close to the beginning of the big irreversible decline of oil and gas. You can see the figure on p25 of this presentation
    http://www.simmonsco-intl.com/files/Dallas%20Committee%20On%20Foreign%20Relations.pdf
    Simmons specifically says the decline rate of fracked gas is early and vertical, so the market sentiment that gas can be valued cheap is dead wrong. It will not save us from the decline in oil.
    I look around at all the cars on the road, so many of them look new, and it’s very difficult to believe this is going to end abruptly, but I am convinced by numbers not what my eyes see now.
    One small thing, an acquaintance of mine’s daughter is going to University soon, I have suggested that she can pursue chemistry (what he says she likes) by going to Carleton and asking David LeBlanc about the chemistry of the LFTR. Just a suggestion.

  70. Peter Lang, on 11 May 2010 at 9.49 — I thought the question was what to do with existing wind turbines — sunk cost.

    In this region (PNW) from less to more expensive now goes
    (0) geothermal (some small potential)
    (1) CCGT
    (2) wind
    (3) NPP
    (4) coal burners
    assuming adequate wind backup (we have it) or else none required.

  71. One thing that counts against Simmons is his apparent belief that off-shore wind is the only viable future energy source. I hope he is wrong about everything, although I think that’s a faint hope.
    Perhaps he is wrong about offshore wind because he is not expert enough – he should be expert enough about oil.
    One thing he said about nuclear was that he’d heard others say that enough energy is embodied in the building of nuclear plants that they were energy negative for as much as the first 15 years of their operation. I find that hard to believe, but he apparently thinks otherwise.

  72. One thing he said about nuclear was that he’d heard others say that enough energy is embodied in the building of nuclear plants that they were energy negative for as much as the first 15 years of their operation. I find that hard to believe, but he apparently thinks otherwise.

    Sounds like whoever he’s listening to is using Stormsmith as their source. If he can’t do more thorough fact checking than evidenced by that, I wouldn’t take anything else he says too seriously.

  73. David B Benson,
    http://bravenewclimate.com/2010/04/28/a-gassy-vision-2/#comment-64397

    You seem to have missed the discussion. Your list is back to front and Wind is incorrectly placed. Wind cannot be compared in you list of baseload generators without back up or energy storage. If you include the cost of back up or energy storage so that wind power is comparable with the others, then Wind power is by far the most costly – perhaps 3 to 5 times the cost of the next highest (CCGT).

  74. Pingback: Switching from coal to natural gas would do little for global climate « BraveNewClimate

Leave a Reply (Markdown is enabled)

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