Roger Pielke, Jr. has looked closely at Australia’s ETS targets and helpfully put some numbers into the hypotheticals.
With all their subsidies, goodwill and fervent wishes, solar, wind, and geothermal produce just 3% of our energy needs. Fossil fuels produce a whopper 94%. And “energy” on these grand continental scales is measured in quadrillion BTUs which is known as “one quad”. Australians use about 5 quads / year, and to make that we pump out about 400 Mt of carbon dioxide per year. (These kind of big-picture numbers are often hard to find, so I wanted to capture that to keep things in perspective.)
Population growth is a big factor in Australia
Carbon Dioxide Emissions= Population * Per Capita GDP * Energy Intensity *
…carbon accumulating in the atmosphere can be reduced only by reducing (a) population, (b) per capita GDP, or (c) carbon intensity of the economy.
One of our biggest hurdles is that the population is growing so fast. Our numbers are increasing, on a per capita basis, about four times as quickly as the UK and the US. (Hence, after nearly two decades of rapid growth with no population debate, it’s finally become an election issue — because everyone has noticed how the roads are clogged, hospital queues are longer, airport parking is a joke, and house prices are astronomical.) The rapid growth of the population means that “cutting” by 5% looks deceptively easy, but even if we held carbon emissions per capita exactly constant (at 19.4 tons per person) by 2050 we’d be pumping out 55% more CO2 than we were in 2000. Ouch.
It seems to compensate for population growth and GDP growth we’d need to cut emissions by about 5% per year each and every year.
Using a bottom up analysis, the combined effects of Australian population and per capita economic growth of 3.0% per year imply that to meet a 2020 emissions reduction target of 5%, 15% or 25% below 2000 levels would require an that the combined effects of increasing energy efficiency and reduced carbon intensity of energy occur at an average annual rate of 4.3%, 5.0% and 5.9% respectively to 2020…
Pielke calculates that we’d pretty much have to stop all coal consumption, and replace it with nukes or renewables. That means 35 nuclear power plants (up from zero) by 2020.
At the moment our total nuclear industry adds up to one small plant — which makes radioisotopes for medical work, and which the Greens would like to shut down. There are exactly no proposals to build a nuclear plant anywhere.
Pielke works the numbers:
It is straightforward to convert the energy mix into greenhouse gas emissions by multiplying the amount of energy consumed (measured as quadrillion BTUs or one quad) by the amount of
carbon emitted per quad for each fuel.7 According to the US Energy Information administration(2008), in 2004 Australia emitted about 391 mT of carbon dioxide from 5.3 quads of consumption, with the mix shown in Figure 4. Multiplying the carbon dioxide generated per quad (from footnote 17) by the proportion of energy from each fuel source results in 390 mT of carbon, essentially the same as that reported by the US Energy Information Administration. With this information it is then possible perform a simple sensitivity analysis describing what it would take to decarbonize the Australian economy to a level consistent with a particular emissions reduction target. In 2004 Australia produced 0.83 tonnes of carbon dioxide emissions per $1,000 (U.S.) (essentially the same as in 2006). For this to be cut in half over the next decade or less – as implied by the 5%, 15% and 25% 2020 targets – would require that nearly all Australian coal consumption be replaced by a zero-carbon alternative such as nuclear or renewable. If an average nuclear plant provides 750 megawatts of electricity (World Nuclear Association, 2007) and one quad is equivalent to 11,000 megawatts of electricity (American Physical Society, 2010) then about 15 nuclear power plants would provide one quad. Coal provided 2.4 quads for Australia in 2004, meaning that this could be replaced by about 35 nuclear power plants.
Bonus: If Australia’s energy demand increases (as it has been) we’d need another 21 plants.
Of course, Australia’s energy consumption has increased since 2004 and is expected to increase in the future. If Australia’s demand for energy increases by 1.5% per year to 2020 then an additional 1.4 quads of energy will be needed, implying the equivalent of 21 additional nuclear power plants, or a total of 46. These assumptions can be adjusted to explore the implications of aggressive energy efficiency programs or expansion of renewable energy technologies (or other assumptions, such as the expansion of natural gas). For instance, if demand is held constant at 2004 levels and renewable energy comprises 20% of the total mix, then only 13 equivalent nuclear power plants would be needed by 2020.
Hey lets go solar:
The Cloncurry Solar Thermal Power Plant in Queensland provides 10 megawatts
of electricity (Renewable Energy Development, 2008). One quad (at 33% efficiency) of energy implies 3,333 Cloncurry plants. Providing 3.8 quads implies 12,667 Cloncurry equivalent plants, or about 24 such plants coming online every week from 2010 to 2020.
Instead of getting 2.4 quads from 35 nuke stations, we could get it from solar. I make that 8,000 Cloncurry equivalent solar plants (just 16 new ones each week for the next ten years!)
Building just ONE Cloncurry solar plant is proving to be difficult
The Cloncurry Solar Plant was announced in 2007, supposed to be working early in 2010 and meant to have 8000 mirrors heating a graphite block so water could be heated in it day and night regardless of the weather. Last month, the Mayor of Cloncurry said the town had been “kept in the dark” and “fed lies” about the shortcomings of the solar plan. [Courier Mail]
Months after it was meant to be finished the Solar Plant seems to be missing 7996 or so of it’s 8000 mirrors, and has hit technological issues so bad, it may have to be moved, or dare I say… abandoned.
…three years after its launch… the project consists only of four test panels and a fake tower behind a locked gate.Boffins are now looking into concerns that residents could be exposed to blinding light from the plant.“There was a glare issue exceeding what they consider to be appropriate levels,” he said. “If the glare issue cannot be addressed the project will be moved somewhere else in Cloncurry or it will not proceed.” [from the Courier Mail August 12, 2010]
Commenters think that just 18 Nuclear Plants would do the trick (not 35). Perhaps that’s true. It isn’t any more likely to happen…
Bernd # 9 says:
Based on the four 1.4GW nuclear plants ordered by the UAE from South Korea’s KEPCO at the start of this year, with the last to be completed by 2020, about half the number of plants would suffice; being almost double Pielke Jr’s assumptions. i.e. about 18.
DandyTroll #10 agrees:
In France the new EPR design of 1650 MWe per reactor is “todays” standard for the coming decade, the old “newest” that went into full operation 2000 was the four 1450 MWe.
So 1650 MWe per reactor, four reactors per plant, you need two plants to make sure you meet the 11 000 MWe or 1 quad. 35 reactors at 1650 MWe would cover 5 quads with some room to spare at maximum output, so 18 reactors to cover 2.4 quads.
co2isnotevil explains why we would need 35 plants
The number of plants required must meet peak demand, not average demand, which is where the larger number comes from. Solar and wind will never satisfy our energy requirements until there’s the ability to economically store and release electrical power at the Gigawatt scale. You will always need to keep to coal and oil plants hot so they can instantly be brought online to keep the grid from collapsing as a cloud passes over. You can’t just shut them down when they are still required transiently, as they will often take hours or days to come up to operating temperature. Keeping them hot burns energy, so you might as well turn it into electricity, rather than dissipate the heat in a cooling tower and it would be better yet to run the plant at a capacity level which optimizes it’s fuel efficiency. I really see no point in solar/wind as a solution to CO2 emissions (not withstanding the fact that CO2 emissions are benign). Owing to spinning reserve requirements, which increases as more transient, i.e. ‘green’ sources are added to the grid, little to no emitted CO2 is actually eliminated.