The 2013 update to the Australian Energy Technology Assessment has the potential to generate breathless headlines in favour of renewable technologies and downplaying the potential of nuclear. The truth is a good deal more complex. This valuable resource is imperfect and evolving. This update provides important improvements and highlights areas of ongoing deficiency in comparisons. It needs to be applied in an informed way to promote good decision making on energy and to deflate the techno-triabalism that is hampering our move away from the most polluting fuel sources.
The Australian Energy Technology Assessment (AETA), published by the Bureau of Resources and Energy Economics (BREE) provides near-to-medium term forward cost estimates of the levelised cost of electricity (LCOE) of a wide range of electricity generation technologies in Australian conditions. This is a valuable service, so the 2013 update is therefore of considerable interest. Here is a link to the document.
Aside: LCOE is, simplistically speaking, the price at which electricity can be sold by a generator into the market. The lower the better. It’s a very useful source of comparison of the economic competitiveness of different electricity generating technologies and a decidedly imperfect one also. It only accounts for the marginal cost of new generation. Admirably, the AETA is upfront about this limitation, with ample discussion on page 13. See further discussion here for how over-reliance on LCOE can generate conclusions that are precisely wrong. End aside.
The update provides, superficially speaking, a poorer cost outlook for nuclear power and a better outlook for renewable technologies. It has the potential to generate fairly breathless headlines along these lines. Indeed, it already has.
A cooler reading of the update suggests a more complex picture. The report provides direction on where efforts are best directed in building a case for a fuller, more rapid decarbonisation using all technologies including nuclear power.
This post addresses four important discussion points for nuclear power:
- First year of available construction
- Revised FOAK and NOAK costs for nuclear
- Forward cost trajectories for nuclear
- Overall message from outputs
1. First year of available construction
Where previously nuclear power was shown as an available technology now, in the update it is not shown as available until 2020. The justification is as follows (page 16):
Additionally, it was submitted that the first year available for construction of nuclear technologies be extended. In support of this the submission argued that nuclear technologies are examples of generation technology where legislation, regulation and public policy planning is highly region-dependent and must therefore precede deployment by several years. This was considered an appropriate submission and therefore the first year available for construction in the model has been changed from 2012 to 2020.
This is a disappointing and poorly justified change. AETA 2012 was a breath of fresh air among documents from Government and Government-owned corporations in Australia for its impartial and disinterested inclusion of nuclear power in energy considerations. Most often, nuclear power is simply left out as though it were never invented, perpetuating ignorance and dearth of discussion.
As a technological assessment, the original judgement of the readiness of nuclear power for construction now was completely correct. Nuclear power is mature technology, providing around 10% of global electricity, in over 30 nations, with about 50 reactors under construction and a competitive market of suppliers. There is no technological barrier and in preparing AETA that is what BREE should be principally concerned with. Otherwise the document is on the slippery slope of politicisation of energy. Were knee-jerk planning regulations to come down that effectively shut the door to further developments of wind or utility solar, removal of their consideration from AETA would be entirely inappropriate.
The fact that nuclear is technologically mature and we refuse to use it for political reasons is important knowledge. The simple inclusion of nuclear as ready to go now makes this statement in simple and powerful terms. It should be retained.
2. Revised FOAK and NOAK costs for nuclear
There are been a substantial increase in both FOAK and NOAK costs for nuclear in this update, shown in the table below:
The justification is as follows:
Although there are cost adjusting factors used for other technologies studied in the AETA 2012 report, there is no adjustment of the US cost base for Australian labour costs and for adjustment of Australian equipment costs. Submissions indicated that an adjustment to costs should also be used for nuclear to ensure there is commonality in the comparison.
This suggestion was considered appropriate and adjustment to the costs for nuclear technologies was made. Based on the factors to convert US rates to Australian rates used in the fossil fuel technologies, i.e. the factors from the Thermoflow software, the new nuclear capital costs were calculated. The multiplier from Thermoflow for labour costs is 2.05 and the factor for equipment costs is 1.3, both US to Australia.
I find little to disagree with in the above and, to be honest, I welcome the update. The costs in the 2013 update represent a more realistic cost starting point for Australia and, if nothing else, a much better base from which to manage expectations. Those costs may be bettered in the eventual process; I believe they could be. There was nothing to gain by published estimates that were not a true comparison.
3. Costs trajectories for nuclear and renewable technologies
While the starting point for nuclear cost is now more realistic, the trajectory reveals something curious. It’s perhaps best expressed in a graph showing the low end LCOE estimates for a few technologies over the modelled period.
As you see, the low end LCOE for nuclear is virtually flat, from now to 2050, similar to the 2012 modelled outputs. That’s a heck of an outcome, especially for SMR nuclear. It would appear these talented designers have exhausted every opportunity for learning before commercial manufacturing has even begun!
The high end of the estimates provides a more mixed picture. While it’s all good news for fixed solar PV, otherwise it seems the BREE model regard all as susceptible to upward price pressures.Note particularly the 2050 high estimate for solar thermal with storage of $260, giving a range of $210 from the low estimate! I suspect the output for this technology is highly sensitive to some key assumptions.
These forward LCOE findings for nuclear need to be treated cautiously. The raison d’etre of SMR nuclear is to take advantage of the well-known cost lowering potential of standardised, factory built manufacturing. Even for GW scale nuclear, we are only relatively recently into the period where novel design is no longer the norm, and great learning is being seen where nuclear is being built in committed programs like China and South Korea (which has exported these excellent prices as a turn-key solution to the UAE).
So what is going on? I suspect it’s pretty simple based on this passage of the report:
As part of the Model Update, special emphasis [highlight added] was placed in this report on the Operational and Maintenance (O&M) costs and O&M improvement rates for all wind, solar thermal and solar PV renewable technologies covered in AETA 2012. As mentioned earlier, this took into account relevant Australian and international reports, public domain information, private sources as well BREE and WorleyParsons’ in-house experience.
This emphasis is evident in the extent of the evidence and literature review cited above, the bulk of the update report being committed to this purpose, and the list of references and stakeholders consulted in this review. It seems pretty clear that wind, solar thermal and solar PV had the vast bulk of the focus in this review, with little attention paid to nuclear beyond the recalibration of pricing described in point 1.
It’s awfully difficult to find what is not being looked for. An effort to seriously review the forward LCOE of nuclear power would require a similarly exhaustive literature review. An effort at SMR costing would, at the minimum, take in consultation with the US Department of Energy, the International Atomic Energy Agency, GE-Hitachi, Babcock and Wilcox and NuScale. It is certainly regrettable that in Australia we lack an official domestic presence to advise on matters nuclear.
On the basis of this report and model being reviewed on an annual basis, which appears to be the case for now, this is all forgivable provided it is addressed adequately in future.
4. The overall message of the costs
The further downward revision of costs across the renewables technologies has pushed modelled LCOE outputs for both on-shore wind and fixed solar PV into some seriously low territory. That passes the sniff test for me. These technologies are increasingly mature, globally very popular, growing very quickly and factory made.
From a planning and decarbonisation point of view, it would seem to me appropriate that from here Australian Governments plan, as dispassionately as possible, for the successful integration of larger amounts of these two renewable technologies. Unless direct barriers are erected to suppress their deployment, we should expect more of them and while install rates for solar may have peaked, there will be more installations. Whether it is a success or not depends how smart we are about it. Variable output is nothing but downside in my opinion. However, we are learning to manage it. We should be planning our network to accommodate more, and our market designs to allocate costs sensibly and minimise disruption for other generators whom we need for reliability of supply. There is a great deal more uptake of these technologies that can be intelligently integrated before system costs become overwhelming.
However that all reaches its own limit in time. Wind and PV are subject to variability and short-term fluxuation of supply and the tendency of it all to come at once, wanted or not. This makes the penetration of these technologies into networks economically self-limiting, provided we intelligently measure and publicise total system costs of including variable renewable energy at rising penetrations. This is no secret. You cannot build and run a reliable network on these sources of supply alone. If it were so, no renewable advocate would care much about solar thermal with storage, offshore wind, batteries, demand shifting, biomass, geothermal, ocean power or expanded transmission networks. All of the above is in the name of achieving the stability and reliability provided a priori by fossil fuel or nuclear generation.
The message here for the nuclear advocates in Australia is to maintain reasoned criticism but essentially let these markets run their course, as they are not competing with nuclear in any serious way, and one may argue a complimentary relationship. If policy is relatively technologically neutral we can expect at least one large generation of this technology in Australia. Whether it will be replaced like-with-like at end of life is, for me, the next interesting question.
Nuclear is reasonably compared with other large, dispatchable generation sources. That being the case, it’s important to see how nuclear fared alongside other large dispatchable zero-carbon generation in this update. Even out to 2050, and even with the conspicuous lack of research attention, AETA ranks gigawatt scale nuclear as the cheapest dispatchable zero-carbon generation on offer in Australia (with the exception of landfill gas and sugarcan waste, with obvious scaling limitations). SMR Nuclear is faring less well, but remaining in the second approximate bracket of generators along with the solar thermal, off-shore wind and hot sedimentary geothermal aquifers.
It is important to note at this time that the default energy storage assumption of the AETA model is currently 6 hours. This is decidedly inadequate for making a genuine comparison of reliability between solar and nuclear. The model remains a little dumb in this regard. Storage is a major element of cost in these systems however at this time BREE considers the addition of variable storage levels “is likely to add significant complexity to the report and to the model, the benefit of which at this stage does not warrant the changes” (pg 23). It is also important to note that all costs in this study have been amortised over construction+30 years. After 30 years, a nuclear plant will be paid for and cheerfully ready to give another 30 years of rated service. It is likely that solar thermal will have met its expected lifespan of 25-30 years, and must be built again. The other inarguable advantages of land utilisation, and flexibility in siting close to networks and load remain exclusively that of nuclear generation.
In summary, the AETA report and model remains an important enabler of informed discussion for upcoming energy choices in Australia. It is limited by the limitations of LCOE itself as a metric and the report is admirably transparent in that regard.
BREE need to be cautious to maintain the purity of the AETA lest its usefulness devolve. Removal of nuclear as a “now” construction option is a good example how this may occur. The regular reviews must also be sure to spread the research effort priorities and broaden the stakeholder consultation if readers are to maintain confidence in the outputs for all technologies. The lack of focus on SMR nuclear in this review, evident in the static forward LCOE, gave an outcome that defies credibility. This can and should be redressed in future. It is inevitable that these outputs will be leveraged to argue the case for certain technologies. As such more transparency and visibility is required regarding the default energy storage assumptions for solar technology. In the long term the model should include this flexibility. In the short term indicative cost multipliers should be provided for additional hours of storage.
Like what you see here? Please subscribe to the blog, Like Decarbonise SA on Facebook and follow @BenThinkClimate on Twitter. Read more about the potential for nuclear power in Australia at Zero Carbon Options.