Jeff Gibbs and Michael Moore have been finding out what it's like to throw a sacred cow onto a BBQ with their documentary Planet of the Humans. I've already written a previous short piece on this.
Long time renewable advocate and anti-nuclear activist Ian Lowe has added to the chirruping critiques with a piece on The Conversation with the seemingly even handed title "3 times Michael Moore’s film Planet of the Humans gets the facts wrong (and 3 times it gets them right)".
How does Lowe do?
(1) Solar panels take more energy to produce than they generate (False according to Lowe)
Lowe is correct to cite a simple payback argument to refute the claim. In brief, the cost of panels is largely a function of the embodied energy it took to build them. Once you save that amount on your bills, then further savings all demonstrate the falsity of the claim. Interestingly, the same claim is often made about mining and enrichment of uranium ... that it takes more energy than it yields ... and again the claim is obviously false when you consider the price of uranium.
At least one person on Twitter tried to defend Ozzie Zehner (who made the claim in the film) by asserting that he didn't actually say this at all ... except that he did. Zehner asserted "You use more fossil fuels to do this than you are getting benefit from it." Which is about as clear as it gets ... unless by "benefit" he didn't mean energy. What other benefit would he mean from an energy source? I don't know. If he didn't mean to say it, then it should have been edited out.
So Lowe is right, but it hardly matters; the low EROEI for solar still hangs on its neck like a bag of steel reinforced rocks.
The twittersphere has suffered some significant discussion about the problems of EROEI. The concept is intuitively obvious but turns out to be surprisingly tricky in practice, but that's true of many metrics. Grid parity is another related and slippery term. If the so called levelised cost of the electricity (LCOE) from a technology matches that of electricity from the grid, this is defined to be grid parity. Unfortunately, LCOE isn't levelised with respect to many significant costs, so "grid-parity" is tainted by association. Here's my alternative definition. Take two houses, connect one to the grid and look at the price on your bills of your electricity. Then supply the other one (without connecting it to the grid) with electricity (at the same reliability) as the grid, using the technology in question, and then compare the price. That definition isn't perfect either, but it's significantly less brain-dead than the normal definition.
(2) Renewables can’t replace fossil fuels (False according to Lowe)
What's interesting here isn't the claim, which is obviously false, but the things Lowe's regards as evidence against it. Of course you can replace fossil fuels with renewables, but the down sides are far from trivial and go deeper than Gibb's documentary.
We know that the sun doesn't shine at night, but what about the wind? It typically has a minimum output, even when the weather is relatively still. That minimum is roughly 10 percent of the windy maximum. So you can always guarantee enough electricity by "simply" building 10-15 times more wind capacity than you need. You may well end up throwing almost all of that electricity away, but if you don't mind doing that, then it's easy. I'm hoping nobody is interested in that kind of solution ... I'm hoping people want to know if there is a system which is affordable and where you aren't throwing away most of your electricity.
Here's some data from the first of South Australia's two 4-day heat waves in December 2019. The first of these had 4 days with maximums over 42 degrees C. Here's a chart of SA's demand and the output of our wind farms. You can see how the minimum output corresponded almost perfectly with the maximum demand; not good.
I first saw a graph like this in an AEMO report shortly after a long heatwave in 2009. The scales on the axis are very different, and it would have been tricky to spot it when we had much lower levels of wind power, but somebody at AEMO spotted in and produced the graph.
So what does Lowe give as evidence that renewables can replace fossil fuels?
Does he give a counter example? Does he name one or more non-trivial grids now operating exclusively on renewables? No. Why not? Because there aren't any. Here's his evidence:
[Lowe] To disprove this claim we need look no further than Australia, where wind turbines and solar panels have significantly reduced our dependence on coal.
So he thinks that evidence that we can replace some fossil fuels is evidence that we can replace them all. It isn't.
Here's Australia's electricity by fuel type ... the black is black coal, and the brown strip beneath is brown coal. Is this graph evidence that renewables can totally replace fossil fuels? Of course not. For Lowe to imagine otherwise is bizarre.
Lowe's next sentence doesn't nail it either.
In South Australia, for example, the expansion of solar and wind has led to the closure of all coal-fired power stations.
The state now gets most of its power from solar and wind, exporting its surplus to Victoria when its old coal-fired power stations prove unreliable on hot summer days.
SA has certainly closed her coal plants, in favour of another fossil fuel, gas. Not only that, but SA added a new gas generator in 2019.
Now lets look at a random week ... the week before I started writing this piece.
So here's an image of that week's electricity production by source in South Australia.
As you can see, South Australia did indeed get plenty of wind and solar power in that week; we even had a few days where it supplied our entire load. And yes, we even exported quite a bit (2nd, 6th, 7th and 8th of May). But we also imported rather a lot on the still days and nights (3rd, 4th, 5th, 7th and 8th of May). Are you persuaded that renewables can deal with our entire load ... on every day? I didn't bother looking for a week with even less wind, but they are quite common.
But what about batteries? Lowe mentions them in relation to claim (3) about whether renewables need backup, which is really just another form of claim (2).
South Australia has what is often touted as a "huge" battery; touted as the world's biggest when installed. You can see it working on the graph if you look really really closely ... it's the little strips of blue on some mornings.
Lowe praises the growing role of batteries by citing an AEMO news article predicting 5.6 gigawatts of battery capacity in Australia by 2036/7.
Perhaps Lowe doesn't realise how small this number is? Or why the AEMO article hasn't provided the second number? ... what second number?
Batteries are characterised by two numbers; the first is their maximum power output (5.6 gW in this case) and second is the amount of energy they store; which tells you how long they can supply that maximum power for. 5.6 gigawatts for 1 second isn't a lot of use.
The battery in South Australia has a maximum power output of 100 MW, and it can provide a total (as of March 2020) of 185MWh. A MWh is one megawatt of power supplied for one hour, so, theoretically, a 185MWh battery could supply 60MW for 3 hours, or 92.5MW for 2 hours ... but in practice, they are configured for a specific maximum power output. So you need to know both that maximum power and the time period it can supply it for in order to understand what a battery can do. You might have a 180MWh battery which is limited in its maximum output to just 20MW; which it could supply for 9 hours.
AEMO didn't provide the second number in the news article because they understand that it's largely irrelevant ... because big batteries aren't about energy storage, but frequency control ... as mentioned in the news report. When people simultaneously start cooking of an evening, there's a massive increase in the requirement for electricity. The result, if you do nothing, is a rapid drop in the the alternating current frequency of the supply. Too big a drop can break some very expensive machinery. A big battery can kick in very quickly to prevent that and buy the time you need to add additional power to the grid ... a gas turbine generator might manage to add 100MW per minute of power, so you only need a few minutes. That's what batteries are for ... AEMO know this, but not, it seems Ian Lowe.
A little data makes it even clearer. The US Department of Energy hosts a global energy storage database on which it keeps details of all large battery installations around the world. As of February 2020, (see table below) it lists 300 operational large Li-ion batteries with a median power output of just 500 kilowatts and with median energy storage of 430 kilowatt-hours. If these batteries were used for storage, the energy number would be much larger than the power number.
The following table of the (non-hydro) battery technologies with 20 or more operational projects shows how tiny the global "big" battery industry is.
Not content with two poor pieces of evidence, Lowe continues.
What’s more, a report released this week by the Australian Energy Market Operator (AEMO) said with the right regulations, renewables could at times supply 75% of electricity in the national electricity market by 2025.
The two key words here are "at times". What does this mean? 5 minutes? 5 hours? 5 days?
What 100% renewable advocates need to do is not to demonstrate the practicality of generating most of the electricity some of the time, but all of the electricity all of the time.
This isn't at all simple.
Here's what the South Australian electricity demand looks like on a tough week; meaning during a heatwave. Here's a graph showing data from the second of those two heatwaves in in December last year. Let me walk you through it.
The red line is the demand. You can see that it peaks at about 3000 MW in the evening of the 27th of December.
The orange line is the total output of wind plus solar (the green line is solar). The mismatch between wind output and demand is considerably better than during the first heatwave, but not on the day we needed it most.
The gap between the orange line and the red line is the shortfall ... and it was supplied by fossil fuels.
The vertical shaded boxes cover 3 hour slabs around the peak demand of each day and the number on the top of each tells you how many "big" batteries we'd need, fully charged, to cover the shortfall ... meaning just that peak shortfall, not the entire shortfall. I give a figure for the total shortfall at the top ... 1269 batteries.
I'm preparing another post which looks at what happens to the above graph if you add a really big battery to the system. Look for it in a few days!
In summary, Lowe's evidence is lazy and irrelevant. It's remarkable that anybody thinks they can get away with trotting out such trivia.
(3) Solar and wind need fossil fuel back-up (False according to Lowe)
This is really just a rewording of (2), so many of the same comments apply.
Solar power doesn't work at night and wind power is highly variable, so obviously something has to plug the holes on windless nights. But what?
Big dams, assuming you have the water, can obviously store the kind of quantities of energy that are required to get big cities through still nights; especially when those cities have industries that do things like make aluminium. Of course, if you don't make anything and just ship or fly your manufactured goods in from somewhere else, then you need considerably less electricity.
Lowe refers to batteries and dams (pumped hydro) as his evidence that non-fossil fuel sources can backup wind and solar.
The batteries claim is simply wrong as discussed above. It's not that you couldn't build big enough batteries, but the costs will always make it improbable; not just the financial costs, but the ecological and social costs. Lowe seems oblivious to these. Do batteries grow on trees? No. Are they made with materials found lying around on beaches and in local parks? No. To build a Lithium-ion battery which will output say 2 gW for 3 hours (which would have covered the shortfall during the 3 hour peak on 27th of December last year, you'd need about 23,000 tonnes of battery (at a current energy density of 265 Wh/kg) at a dollar cost of about AU$1.4 billion plus whatever value you place on the lives of children mining cobalt in the DRC.
Batteries don't make energy, but they can certainly make money and headlines (think Elon Musk). But if you want to build a clean electricity infrastructure, then they are not a good choice. Dams will certainly do the job; assuming that you regard dams as both safe and environmentally acceptable. And also assuming that you can afford them.
Lowe also claims:
And the New South Wales government’s pumped hydro plan shows how by 2040, the state could get 89% of its power from solar and wind, backed by pumped hydro storage.
Except that it doesn't.
Go and visit the page Lowe hyperlinks.
It isn't a plan in any real sense of the word. If I say Australia has plenty of desert available for solar farms. Is that a plan? The page Lowe links to doesn't call itself a plan because it isn't. It calls itself a "road map", which itself is quite a stretch. The page Lowe links is just broad aspirations. It links in turn to a so-called RoadMap and a Handbook. The Handbook contains guidelines for people who want to invest in pumped hydro projects. The RoadMap has a little more detail, mostly based on an ANU study which claims to have found tens of thousands of potential sites for pumped hydro schemes.
The problem for would-be dam builders isn't sites, it's renewable energy penetration.
Think about it. As renewable energy penetration goes up, the marginal cost of that electricity goes down. But a big dam needs a good return on investment. Plugging shortfalls won't provide that. This is the mechanism which has driven coal out of business in South Australia.
The economics is just like that of batteries. If you pay a billion dollars for a battery and it has a finite lifetime, then you'd better find customers willing to pay plenty to use it and you'll want them to use it frequently; not just to keep as very expensive insurance for a rainy day. We've seen this kind of problem in other areas of our lives; railway systems. Who wants to build a railway system in a city replete with cars? The marginal cost per passenger kilometer may be wonderful, but the capital cost will dominate decision making.
Consider the Middleback Ranges pumped hydro project. This is a $452 million dollar project for just a maximum power output of 110 MW with a total storage capacity of 1500 MWh (13 hours). Will it ever get built? Or will it go the way of the Kanmantoo project currently paused after AGL pulled out. Keep in mind that pumped hydro schemes don't generate electricity, they have to buy it when it's plentiful and sell it when it isn't.
Provided you are okay with flooding valleys, or old mines, then pumped hydro is technically very good mature technology. But, at low levels of renewable penetration, it doesn't have cheap energy to fill it's dams, while at high levels of renewable penetration, it may have trouble selling enough to cover the capital costs.
Lowe's view on what Planet of the Humans got right?
1. We need to deal with population growth (Lowe agrees)
The problem both with the film and Lowe is that both seem to think that population growth hasn't been dealt with. Campaigns to reduce population growth have been among the most successful on the planet over the past 60 years. They have halved the global average number of children per woman! Have any campaigns halved the number of cars per capita globally? Or in even a single country? Or the number of kilometers travelled by car? Or the energy use per capita? No ... not even close.
I can't imagine how insulted the many people must be who've driven these campaigns. The number of (usually) rich people calling population "The elephant in the room" is as astonishing as it is ignorant.
Does the number of people on the planet make for a bigger impact on the biosphere and make the problems tougher? Of course. But pretending that somehow there is some kind of magic solution that could reduce the problem if only people would pay attention to the elephant in the room is just wrong and rude.
2. Biomass energy does more harm than good (Lowe agrees)
Lowe's response to Gibb's attack on biomass is to firstly agree that it isn't green at all:
While the film unfairly criticises the environmental benefits of solar energy, it’s true that some so-called clean technologies are not green at all.
But then to step back from any real criticism and seek refuge in a conciliatory point that biomass merely has "limitations".
Most advocates of cleaner energy systems recognise the limitations of biomass as an energy source.
Is there any evidence for this claim?
Germany in 2019 got 45.5 terawatt-hours of electricity from biomass and 46.5 from solar power. This is roughly a 50 percent increase in biomass electricity during the past decade. The German Green's have substantial power over energy policy in Germany, and biomass is clearly getting plenty of support. That 50 percent German increase was also matched globally ... see the BP Statistical Review in World Energy 2019. Clearly the energy from biomass industry has plenty of friends, including in environmental circles. In Australia, biomass was an essential component of the highly regarded Zero Carbon Australia 2020 report, praised frequently in environmental circles. Similarly, Mark Diesendorf has been a strong advocate for 100 percent renewable energy and his work isn't shy about including biomass among the sources.
3. Infinite growth on a finite planet is suicide (Lowe agrees)
Not much to say here. Does anybody on the planet advocate infinite growth? Not as far as I know. Most politicians advocate growth in the context of the next electoral cycle, because they don't have another model for economic prosperity. Many countries have been actively decoupling substantial sectors of their economy from the environment. For example, there has been a net growth of forests in the US and many countries of Europe over the past 200 years (Germany, France, Switzerland ... to name a few). There is plenty of work to be done to complete the decoupling and to find models of stationary prosperity, but it would be more useful to do some of that work than simply bang on about such an obvious problem.
Conclusion
The Conversation isn't a peer reviewed journal, but its charter says it will:
Inform public debate with knowledge-based journalism that is responsible, ethical and supported by evidence.
Lowe's article doesn't in my view, conform to the charter at all. It's incredibly thin on evidence and positively rude to the generations of people who've done so much incredibly successful work on reducing population growth in so many countries.