The S-Curve: Why I’m Optimistic about the Clean Transition

May 28, 2021

The S-Curve of technology advancement is explained in my new video, by Tony Seba and Jonathan Koomey, both energy experts. But the idea is not a new one. See below video.

Harvard Business Review:

Many people suggest that rates of new product introduction and adoption are speeding up, but is it really, across the board? The answer seems to be yes. An automobile industry trade consultant, for instance, observes that “Today, a typical automotive design cycle is approximately 24 to 36 months, which is much faster than the 60-month life cycle from five years ago.”  The chart below, created by Nicholas Felton of the New York Times, shows how long it took various categories of product, from electricity to the Internet, to achieve different penetration levels in US households.  It took decades for the telephone to reach 50% of households, beginning before 1900.  It took five years or less for cellphones to accomplish the same penetration in 1990.  As you can see from the chart, innovations introduced more recently are being adopted more quickly.  By analogy, firms with competitive advantages in those areas will need to move faster to capture those opportunities that present themselves.

53 Responses to “The S-Curve: Why I’m Optimistic about the Clean Transition”

  1. ecoquant Says:

    These are empirical adoption curves, but there is a conceptual theory behind them. The curve is actually a logistic curve and it is applied to the diffusion of innovation (Rogers). But mathematically, the same form is seen population ecology.

    Simplifying, the curve can be seen as the solution of a particular first order differential equation

    \frac{da(t)}{dt} = k_{1} + k_{2} a(t)

    The k_{1} represents adoptions per unit time due to people who, say, adopt solar PV on their own (“early adopters”) and k_{2} is the additional adoptions per unit time per number of adoptions in place.

    This grows without bound, and that’s why the notion of a carrying capacity K is useful, even for a new technology, giving

    \frac{da(t)}{dt} = k_{1} + k_{2} a(t) (1 - \frac{a(t)}{K})

    Because there is a saturation point, the curve eventually flattens.

    People aren’t familiar with this kind of behavior, because adoption starts out slowly. But in the middle of the curve, not only is adoption superlinear, it is superexpornential, although barely slow, and then begins its plateau.

    There are other analogies with physical diffusions and percolations.

  2. tildeb Says:

    Expornential makes math doubly exciting; superexpornential makes it climactic… especially if it’s climatic!


    • And I shall exponentialize your use of the words “climactic” and “climatic” with the word “climacteric“, which is used in my post entitled “Misquotation Pandemic and Disinformation Polemic: Mind Pollution by Viral Falsity“.

  3. pendantry Says:

    This is very encouraging. But I am becoming increasingly concerned about what happens when the current generation of solar panels (for example) reaches the end of its life (in about 25 years).

    This ‘S-curve’ is just a depiction of exponential growth, and this presentation omits several important considerations, including the exponential growth in extraction of (finite) resources, the super-exponential growth of waste — and the lack of growth in our ability to recycle that waste.

    ‘Away’ is a place on Earth (and it stinks).

    • ecoquant Says:

      I don’t know the disposition of used renewables across of the industry, but I do know that wind turbine blades now have a viable and tried recycling pathway, and owners of SunPower PV panels or Teslas don’t get to do anything they want with the panels or the batteries when they reach end of life.

      In both the case of SunPower panels and Teslas, owners are licensed to use the panels and batteries, but both companies very much want them back at end of life, and, in fact, it’s a breach of contract not to return them. (I don’t know about Tesla’s policy on their solar panels.) This an example of extended producer responsibility more of which we need to see.

      However, I’m a little disturbed at the intimation that somehow the possibility recovering or “wasting” these panels is comparable or equivalent to the prodigious pollution that comes from fossil fuel extraction, refining, transport, and use, or even from something as mundane as production of cement. Panels and turbine blades, even if they were put into landfills, are not as toxic, as as Mr Musk has observed, why would anyone do something so stupid. Metals are the easiest materials to recycle at lowest additional energy inputs (glass is horrible), and Carbon fiber in turbine blades is easier still.

      For more information, see:

      Pehl, Michaja, Anders Arvesen, Florian Humpenöder, Alexander Popp, Edgar G. Hertwich, and Gunnar Luderer. “Understanding future emissions from low-carbon power systems by integration of life-cycle assessment and integrated energy modelling.” Nature Energy 2, no. 12 (2017): 939-945.

      Mahmud, M. A., Nazmul Huda, Shahjadi Hisan Farjana, and Candace Lang. “Environmental impacts of solar-photovoltaic and solar-thermal systems with life-cycle assessment.” Energies 11, no. 9 (2018): 2346.

      • pendantry Says:

        I totally agree that the concept of ‘extended producer responsibility’ is one that should be encouraged.

        However, I see nothing in your offering that suggests that recyclability of solar panels will not be an issue in the future. Please do correct me if my understanding is wrong (which, I fully accept, it may well be).

        I currently see us jumping from the frying pan into the fire.

        • ecoquant Says:

          Hi @pendantry,

          It’s just that solar PV panels and their manufacture is no where near as dirty as extraction or mining tailings from fossil fuels, or byproducts of refining.

          See Mahmud, M. A., Nazmul Huda, Shahjadi Hisan Farjana, and Candace Lang if you want details, or the other reference. I have more. Zero Carbon energies even perform attractively compared with fossil fuels, even on land use consumption:

          Fthenakis, Vasilis, and Hyung Chul Kim. “Land use and electricity generation: A life-cycle analysis.” Renewable and sustainable energy reviews 13, no. 6-7 (2009): 1465-1474.

          Of course, fossil fuel land use is often in remote areas and in communities which are too poor and challenged to put up much of a fight, unlike the white privileged suburban communities which typically fight utility scale solar installations, but, nonetheless, apples are apples.

          • pendantry Says:

            I want to believe what you’re saying, but, given the misinformation that’s the norm these days, say in respect of nuclear power for example, I seriously wonder whether considerations of things like recyclability and longevity have been factored in — or are just conveniently ignored, as is so often the case.

    • rhymeswithgoalie Says:

      I note that the ownership of clothes washers took a hit during WW2, in part due to the high demand for metal (both in scrapping existing washers and in availability to make new ones).

      Likewise, few people remember the scrapped cars lining many roads in the 1960s because demand from Japan had junkyards collecting, smooshing and selling them on.

  4. jimbills Says:

    Oh boy.

    Okay, I’ll go over this again and try to be clear. The s-surve scenarios listed in the chart above are apples, and renewables are oranges. In every single example in that chart, they were technologies that were either far better or completely new to those they replaced. This is not the case with renewables. They are better in terms of environmental footprint, but as a power source they are largely a 1:1 replacement.

    Secondly, none of those examples required the same level of industry and materials that will be needed for an 80-90% renewables takeover from FF. The scale for that is so massive it’s largely impossible to conceive.

    Thirdly, none of those examples in the chart were ideologically opposed by a combined and concerted effort from long-established corporations, think tanks, politicians, and larges swathes of the public.

    Fourthly, for renewables, we need to hurdle several major infrastructural issues on a global basis. Some of those examples like cell phones and autos required that, but we’re talking peanuts in those compared to what is required for a global renewables replacement. Adding electricity in the 1920s is the closest example, but it was taking place in a much smaller world. We need to get a real handle on storage and transmission, and these issues will hamper the steep upslope part of the s-curve.

    Fifthly, it’s assumed by using the chart above that renewables will automatically 80-90% replace FF. But, look at China – it’s not the case. They have been growing renewables exponentially for over a decade now, but they are stuck at renewables being a 20% total part of their electricity mix. They keep adding FF at the same time. Growth doesn’t exist in a vacuum for one thing (like renewables) only – it grows everything at once.

    Germany, a better example, is at almost 50% renewables energy mix, but they are on more of a linear curve. They never hit the steep part of the s-curve:
    https://www.cleanenergywire.org/news/germanys-renewable-power-share-climbs-46-percent-2020-preliminary-data

    And they are having major issues at 50%:
    https://www.greentechmedia.com/articles/read/germanys-stressed-grid-is-causing-trouble-across-europe

    All of this is pointing to the nonsense of Seba’s prediction of 80-90% renewables in the very near future (by 2030-2035). And I strongly think it’s dangerous nonsense, because it creates an illusion that we are close to a solution when we are not, and because of that it dulls impetus for strong governmental policies (why have government stand in the way of inevitable technology-spurred exponential growth?) that are far more realistically REQUIRED for fast adoption.

    I don’t doubt that there will be some date when we’ve fully to almost fully replaced FF. But it’s far more likely to be more linear (with multiple bumps) in the part of that theoretical s-curve that shoots steeply upwards. This pushes that replacement date further out, and policy can bring that date closer.

    I know you like to reach for positive news. There has been positive news in terms of technological progress (the opposite is true for environmental progress, though). But this Seba stuff is too far. I’d strongly suggest backing off it. I know I’m a nobody commenting on a blog, and my impact is akin to an ant farting in a crowded subway station, but it’s my two cents on the matter.

    • rhymeswithgoalie Says:

      Consumer goods fit a prettier S-curve than both electrical connections and the Internet, which are dependent upon both geography and public budgets. I can see EVs being adopted fairly quickly in the US, but don’t know how the market for old ICE vehicles would play out, since new cars are very expensive for a lot of people (and the rest re-sold to Mexico?)

      I look forward to fancy after-market devices for, say, roll-out solar mats to recharge vehicles out in the wild (where gasoline isn’t available, either).
      Let the games begin.

    • John Oneill Says:

      ‘…as a power source they are largely a 1:1 replacement.’
      Not even that – more like a 0.2:1 replacement ( solar ) or a 0.4:1 replacement ( wind) – unless you include massive overbuild, unprecedented storage, or vastly increased transmission – none of which coal or gas need. Hydro, nuclear or geothermal either.

      • ecoquant Says:

        @John Oneill,

        It’s not “massive overbuild” if it’s what is needed to make the technology viable. And otherwise your criticism is apples to oranges comparison: What crucially matters is the dollars for the overbuild versus the dollars for new construction of what it’s replacing.

        Coal and gas and nuclear and especially hydro all need gobs of transmission.

  5. ecoquant Says:

    @jimbills and all,

    I am not going to try to rebut the comment about S-curves and Professor Tony Seba being too optimistic. I’ll note, however, that the examples offered regarding Germany and China operate in an environment where the prices per kWh are still expensive. Professor Seba has never said 80%-90% renewables would happen by 2030-2035. What RethinkX has observed is that it is likely marketplace nonlinearities will be in evidence if as projected the capital cost of construction of solar+storage or wind+storage per kWh becomes a sixth of just the transmission cost of electricity from the grid. It’s possible to engage in informed speculation on what that will look like.

    On the other hand, it is possible to achieve 80% renewables by 2035, but only, as @jimbills suggests, with federal government intervention in a big way. The actions and statements from President Biden, and the twin climate czars, Kerry and McCarthy, are not at all encouraging. Greta Thunberg is correct: Magic Beans.

    Professor Seba, on the other hand, is much more aggressive about EV adoption.

    I have a couple of comments, however.

    First, people should know that I have pretty much given up on the U.S. Constitution and any derivative (state) government being able to satisfactorily address the problem of climate disruption. I think these apparatuses are just not up to the problem. They are too slow, and they are designed to be slow. I’m not a pessimist, however, because I think the problems have a good chance at being solved, but the means left to us are those which the Green New Deal/Exinction Rebellion/Senator Warren/Senator Sanders folks are exactly the wrong ones: Large multinational corporations. Unfortunately, of course, because the political mechanisms, the public, and to a large degree, environmental and climate activists, have completely missed the ball, the solution where we end is likely to be far from the universal buy-in egalatarian solution that might have been. Sure, delay was encouraged by fossil fuel lobbying and PR, but it was the environmentalists who bought into the personal Carbon budget ideas and still think growing forests are going to fix everything. They have not done their arithmetic.

    I challenge several claims, however, which look like dual layers of red herrings and comparing incomparables. First, that

    This is not the case with renewables. They are better in terms of environmental footprint, but as a power source they are largely a 1:1 replacement.

    There was no way which the Internal Combustion Vehicle (“ICE”) was a 1:1 replacement for horse drawn transportation. ICE depends upon a long and extensive supply change to extract, refine, transport and sell fuels. It took building paved roads. That took a good deal of time to develop, but not as long as people imagine. Also, the infrastructure to power everything with electricity is not that difficult, once premiums are properly recognized by those who want to invest in these things. These premiums might not be explicitly monetary, but they made be good bets. Hence, electrifying the production of steel is something which is feasible, but the case needs to be accepted that it’s something in which to invest. The big thing is that capital costs for solar+wind+storage are low per kWh, and will continue to decrease. Even transmission, should that want to be invested in, is less expensive than thought, but only so if the accounting is done correctly.

    From an investment perspective, the opposition

    by a combined and concerted effort from long-established corporations, think tanks, politicians, and larges swathes of the public

    is less ideological opposition but a natural opposition by interests who have a good deal of money in existing assets and don’t want to see it go to zero. When disruptions happen, however, and I had a ring side seat seeing it happen at IBM, the wishful thinking doesn’t matter. In the end, the assets do go to zero, and the delay in realizing that is entirely an accounting accident. The accounting is valuing things as they always were, coasting on the success which the company, whether IBM or Kodak, or whoever, already had. The floor drops out but even then it takes big corporations a long time to fail, and their wracking is painful to watch, as they take silly gambles, and convince themselves it cannot be happening.

    The comparison of a nascent all electric culture and economy with a fully established ICE and fossil fuel economy, built up over years with government supports is illegitimate. The all electric economy, when it comes, will look nothing like the existing one. I don’t think we can even be sure transmission of the kind we have now will survive, as I hinted above. There will be use of it at the outsiet because, well, people haven’t imagined any other way, and they have not yet figured out how to make decentralized generation and usage scalable. If that happens, it will require a major architectural change to what we now consider “the grid.” At some point, people who have access to electricity and electrical means of doing business and living will simply stop buying the stuff that funds fossil fuels and conventional utilities. Any efforts to force them to pay extra when they do access the grid will simply incentivize them farther to leave it completely. Jurisdictions which prohibit leaving it will find businesses and residents leaving simply because the cost of hooking up to the grid will be too high, and I know how crazing that sounds.

    There will be a need for some fossil fuels for a while in order to build the zero Carbon energy systems and distribution mechanism, and mine the materials for them, etc, but it needn’t be as large as the present fossil fuel industry. In fact, as their market begins to abandon them, fossil fuels will get cheaper. For consumer and small business, however, this never match the cost advantages of things like EVs because properly built EVs are simply better more reliable products, and are primarily software systems, not conventional vehicles. (And there are lessons there for people who think Ford and GM will soon mount a serious challenge to the primary markets for EVs.) This is predictable. Remember Andreessen’s 2011 assertion “Software is eating the world.” That’s a characteristic old school utility people don’t understand … It’s even a characteristic long before renewables which Norbert Wiener wrote about in his Cybernetics, constrasting the two EE schools, one of “strong currents”, one of “weak currents”. Wiener thought “weak currents” would ultimately win. Again, this is counterintuitive, but note that the biggest coal and mining extraction engines on the planet are powered by electricity, not ICEs.

    It isn’t Professor Seba people should worry about. It’s Joseph Schumpeter’s creative destruction. The world to come will be pretty ugly, with rotting hulks of abandoned pipelines power stations, transmission lines, that no one wants to pay for ripping down. It will look like a large scale version of the empty and abandoned factories IBM had all over the Hudson Valley, in Boca Raton, FL, and Burlington, VT, after its manufacturing collapsed when it failed to appreciate the threat of microcomputing and powerful small servers.

    “Small currents” are control currents. A proper transmission system matched to zero Carbon generation is a digital system, largely self-healing, and self-regulating, and one which uses smart controls and computers to balance across synoptic scales. The formula for a self-sustaining system like that is pretty well known: To have reliable zero Carbon energy, take the median of the maximum draw for the region to be supplied, call it M. Build out a smart (meaning sensitive to availability) deployment of wind and solar generation on a synoptic scale which is 4M. (Capital costs for wind and solar are low, so why not?) Ensure transmission is in place or generation is close to consumption. And provide 4M-7M  (depending) of storage. Architect to keep the storage at 100% except for the (mostly rare) demands. 

    The design allows for generation at spatial scales typically bigger than weather systems. The overbuild, as considered from a conventional energy perspective, permits generation from a fraction of the deployment which suffices to keep the storage ahead of demand.  It’s cost competitive because wind+solar+storage are cheap in their capital costs compared with fossil fuels and all their infrastructure (e.g., pipelines, train lines, trucking, etc).  And most of the time, the system has more energy than anyone can use, which means it can be put into manufacturing uses, or stored as Hydrogen, developed via electrolysis of water, for heavy trucking, shipping, etc. 

    That people insist there remains a need for natural gas means they did not do that. 

    What can go wrong? Plenty, and @jimbill’s skepticism could well be warranted. Zero Carbon energy generation demands land and if you want to have efficiency, it needs to be generated close to consumption. (That’s not really clear, but, for now, let’s say it is, as the alternatives aren’t going to be available for a decade or two.) With their faux environmentalism, objectors in the Northeast U.S., whether to build-outs of solar on basically anything you can put it on, or anywhere it can go, or objectors slowing construction of onshore wind turbines, whether for aesthetic reasons (or the “they gave me a heart attack”), or objectors to impediments to fishing (which is not environmentally sound done as it’s done), are effectively climate change denialists. As people have commented, it doesn’t matter if they agree climate disruption is a threat or not, if they act like they don’t they are climate change denialists. “I believe in solar PV power, but just not putting it here” is a non-starter.

    Still, they can slow it down, can extend the need for fossil fuels. And, if the delay is long enough, and as the climate deteriorates, eventually it may be necessary to indulge in the horrific but possibly necessary project of albedo hacking” and, then, not only will there need to be a continuing flow of kerosene to power those planes spraying sulfuric acid droplets all year long in the Stratosphere, for every year forever, this may disincentivize further action. (It won’t impede solar PV that much … Some, yes. It may increase winds. We won’t know until the U.S. National Academy of Sciences conducts its research on whether or not “solar geoengineering” will work, and what the lateral consequences might be.) And that world could be uglier, still, than anything anyone commonly imagines: The oceans will eventually die (so those fishermen won’t have a profession any longer), and resorts may collapse, because no one will be able to go sunbathe or hike due to excessive UV (H_{2}SO_{4} is believed to destroy ozone), but the world won’t get as harm. Hoorah. Big win.

    So @jimbills could be correct, and worse.

    • ecoquant Says:

      (Apologies for the failure to close em tag at albedo hacking.)

    • jimbills Says:

      I didn’t say ICE cars and horses were 1:1. I said renewables vs. FF when strictly viewed as the power service they provide are largely 1:1. On concerted pushback by established corporations, we’re saying the same thing using different words.

      I do think there will be massive creative destruction – unplugged NG well heads, etc.

      On: “Professor Seba has never said 80%-90% renewables would happen by 2030-2035.”

      He’s more than implying 100% renewables by 2030 here:

      And, he says that happens without subsidies. Additionally, super fast adoption is all but outright stated when mentioning the s-curve. The math makes it a 10-15 year window for most to all of the transition. We’re at the end of the bottom tail, therefore, the steep slope is supposed to happen now.

      The 6x or more price plummet in renewables construction in the near future is economic theory. Fine, I can accept that. One can engage in fantasies of the implications of that. I don’t think it accounts at all for realities on the ground, the very real material and political impediments to that – even if that happened on the purely economic level, which is a big ‘if’ to me.

      But doesn’t Seba go a bit past purely speculating? Does anyone else see the dangers in putting any faith in technological and free market cure-alls?

      • ecoquant Says:

        @jimbills,

        I agree that Professor Seba’s projections seem to be data-fitting-only, and as both a statistician and a scientist I can see the dangers of doing that. Problems with doing that have been recognized. However, he is grounding these on a deep study of technological disruptions of the past, far deeper than we can gloss here. And, the troubling thing about his projections is that in contrast with the authoritative projectionists, such as the IEA and U.S. EIA, he has been correct and they have been wrong, not about 2030, but year-over-year about solar and wind and storage. To assess him means you need to have an explanation for that.

        My explanation is that he does have something in his Schumpterian understanding of technological disruptions. And, unlike many economic historians or scholars, Seba understands nonlinearities. All the standard projections are linear, as are government projections. Frankly, most don’t know what to do with nonlinearities. These are common in science and in engineering, and simply require mastering there.

        But, right now, I’m betting Seba is correct and everyone else is wrong, and that’s based upon everything I know about physics, statistics, maths, and engineering. And, if he is, there’s a world of economic and other pains awaiting people when his projections come true and have been ignored.

        And he’s not the only expert: Professor Mark Z Jacobson.

        Place your bets, people! It’s going to be an interesting ride.

        • jimbills Says:

          My money, for the record, is not on renewables won’t grow. They will, and faster than many think. But my bet is that they won’t grow at nearly the rate promoted by Seba and others (or, conservatively, 80-90% replacement by 2030-2035) – and that we’re engaging in dangerous hopium to plan on that.

          On his predictions, I’ll just say it’s a lot easier to be right about exponential growth at the bottom tail of the s-curve. The steep slope engages levels of industry which severely test physical and human bounds. Energy is scales of magnitude more complex and massive than cell phones replacing land line phones (which haven’t disappeared, either).

          A recent J.P. Morgan paper goes into some of the difficulties:

          Click to access future-shock-amv.pdf

          “The overarching message of this paper is not climate nihilism; it’s that the behavioral, political and structural changes required for deep decarbonization are still grossly underestimated. ”

          (The paper is co-authored by Vaclav Smil, who is more than a little on the pessimistic side, but it also raised points that poke more than few wholes in what is essentially mathematical theory from Seba’s optimistic side.)

          • ecoquant Says:

            One can dispute, but Seba’s assertions are, from my study, based a great deal more than upon “mathematical theory.” In fact, that claim is laughable. If there is any criticism of Seba’s work is that it is based (too) much upon empirical Bayesian estimates, not that it is too a priori theoretical.

            In my opinion, from what I’ve studied, Smil is a shill. Period. I have no respect for him. But, then, what do I know. I’m not a celebrity.

            As I noted, place your bets. We’ll see who wins.


        • This looks like the safest bet to me:

          • Brent Jensen-Schmidt Says:

            With all the impressive and enviable science around here, note the practical successful solution above. Free of ‘ifs’ and ‘could’s’and other wishful desires.

            If ifs and could’s were peas and puds, nobody would go hungry.

          • rhymeswithgoalie Says:

            Through 2019:

            Cooling-water systems need to be retrofitted onto existing nuclear power plants to avoid the power shutdowns they had during the heat waves.


      • Seba is a snake oil salesman! He’s throwing out a lot of whizbang technical sounding words like U-curve and superpower and promising near zero marginal cost. Can you say Supershell with platformate?

        • ecoquant Says:

          Your maths are wanting.


          • “Your maths are wanting.”

            Perhaps, but I think my summaries and assessments are fairly accurate, … along with Michael Shellenberger’s:

      • pendantry Says:

        Does anyone else see the dangers in putting any faith in technological and free market cure-alls?

        /nod

        We cannot solve our problems with the same thinking we used when we created them. — Albert Einstein

    • rhymeswithgoalie Says:

      The world to come will be pretty ugly, with rotting hulks of abandoned pipelines power stations, transmission lines, that no one wants to pay for ripping down.

      Defunct power stations (take the copper!) have some advantages which make them attractive for grid storage sites:
      – they’re physically patched into the grid
      – they’re pre-permitted for power generation on the site

      There will be plenty of pressure to use existing transmission lines, in part because they typically connect things that will stay connected.

      Pipelines are typically made from high-quality steel, and those which won’t be re-purposed for, say, bio-fuel will be attractive for scrap, if not direct re-use of the formed pipe. Some buried pipeline might be forgotten by the system.

      The big problem is really the abandoned wells, which pretty much vent methane from fossil-fuel deposits directly into the atmosphere, with no benefit to humanity or the planet.

    • John Oneill Says:

      Your math uses dissimilar terms. ‘..And provide 4M-7M (depending) of storage.’
      M must be measured in watts. Storage will be in watt/hours. Effectively, for modern cities, it would be gigawatt/days, at least. New Zealand is investigating a pumped hydro scheme that would store nearly a Gigawatt/year – the world’s largest – and that for a nation of only five million, that already has 85% renewable electricity.

      • ecoquant Says:

        No, M is a unit of energy, not power. That’s a kWh. Storage is only needed as buffering and when generation of wind+solar drops so low demand is higher. The 4X design anticipates two things. First, generation will never go completely to zero. If so, the placement of the generation was not done properly. Second, on most days, because there is 4X overbuild, there will be so much electricity generated it will essentially be free, drive electricity prices including from all other sources to zero (or even negative!), and having such will incentivize doing things like electrolyzing water into Hydrogen and Oxygen. The Hydrogen might be used for storage, might be used for fuels, could be used for a lot of things, because it’s fungible.

        This all happens because the cost of generating an extra kWh of electricity once wind+solar are constructed is essentially zero. (You do need to recover initial capital costs, but those get paid quickly.) That’s the other reason why proponents of fossil fuel generation and other sources like nuclear do not want solar+wind+storage to go big. If it does, they are economically dead. Not just non-competitive, dead. They would need huge, prohibitive subsidies from government to continue to operate.


  6. The top of the S-curve for renewables is where capacity meets peak demand. After that, it hits a solid wall of storage expense and underutilization. This traps the top of the S-curve somewhere in the midrange of generation. Batteries and the smart grid are not going to save the day. They are utterly dependent on an increase in mineral use and are headed for a mining bottleneck. You have an increase in demand headed for limited supply which increases the price of a myriad of essential minerals and there’s a lot of them. This is already starting to happen, especially with copper.

    • ecoquant Says:

      Not with the synoptic scale 4X formula. There, most of the time, there’s generation because weather systems are smaller than the synoptic scale. At 4X deployment even if but a fourth of the wind+solar is generating (seldom in the same place from every minute to every minute) all’s good. The storage, whatever form it takes (and it may not be LiON batteries, it could be compressed air), acts more as a buffer than the store which supports when “wind doesn’t blow and sun doesn’t shine”. In a properly architected zero Carbon generation system that almost never happens. See Jacobson’s 100% Wind Water Solar and Storage. That’s not “snake oil” at all.

      Yeah, if this is implemented the shareholder value lost by fossil fuel businesses, their supporting industries, and utilities will be huge, not to mention their jobs. Frankly they deserve it.

      • John Oneill Says:

        ‘At 4X deployment even if but a fourth of the wind+solar is generating (seldom in the same place from every minute to every minute) all’s good.’
        Nonsense. Solar goes to zero over a hemisphere every night, and doesn’t always come back either – the rainclouds we have here at the mo. probably knock it back to 10% of nameplate, for a short day. Winter, you know. Wind can also drop to ten percent or less of what it says on the packet, over large areas, for days. West Denmark, European wind capital, is getting 3% of its nominal capacity right now, Germany was getting 8% a few hours ago, and Ireland is at 10%. All are at 0 solar. Spain is nearly 20%, but an eight-times overbuild there sure wouldn’t power most of Europe, even with a hundred-fold increase in transmission lines. https://www.electricitymap.org/zone/DK-DK1?wind=true&solar=false

        • ecoquant Says:

          @John Oneill,

          In most places wind and solar are anti-correlated, and you neglected to respect the notion of building out over a synoptic scale. If that’s done, the generation is independent of most weather systems.

          • John Oneill Says:

            If ‘synoptic scale’ means ‘ bigger than a large high pressure system’, you’re talking about an immense amount of hardware, to catch that wind at wherever it chooses to manifest itself, and then to transmit it to where the users actually live. I’m talking wind because if you mean solar, you’d need world government and trans-oceanic superconducting cables – if Russia shuts off gas to Europe, at least all the lights don’t go out straight away.
            Germany already has about a 2X overbuild of wind and solar – peak demand today was a bit over 70 GW, versus 117 GW of wind and solar. Their emissions per kw/h were still about seven times higher than France, next door. You’re telling them that if they just keep the faith and build twice as much again – and if everyone in ‘synoptic range’ does the same, and sends them the juice when their turbines stop, they’ll be able to shut their lignite mines ? How cheap did you say all this stuff was going to be ? And remember it all has to be replaced in thirty years – you might recycle the metals, but you still need the embedded energy to make all those towers and panels.
            By the way, I mentioned the near Gigawatt/year pumped hydro scheme that New Zealand is considering, to back our hydro and wind up. A freshly fuelled nuke has one and a half Gigawatt/years stored right inside that pressure vessel, and sometimes another one on hand, just in case. That’s a bit more bankable than ‘ The wind’s always blowing somewhere.’

          • ecoquant Says:

            The turbines and solar don’t have to be right next to each other. In fact, they oughtn’t be. The formula is the same: 4X median maximum daily demand for the synoptic region. Spread that out spatially over the region using a spatially balanced randomized scheme like GRTS

            Stevens Jr, Don L., and Anthony R. Olsen. “Spatially balanced sampling of natural resources.” Journal of the American statistical Association 99, no. 465 (2004): 262-278.

            and you’re set. Just need the 4X. And it’s not a lot because of the low capital costs.

          • Brent Jensen-Schmidt Says:

            The recent Texas debacle consisted of ten bad days in a row from what appeared to be a half continent wide weather system. Really makes a mockery of ‘if the suns not shining the wind is blowing’ when the most densely populated half of USA is socked in. Neither unique or particularly unusual.

    • rhymeswithgoalie Says:

      Batteries and the smart grid are not going to save the day. They are utterly dependent on an increase in mineral use and are headed for a mining bottleneck.

      (1) Fossil fuels represented ongoing and increased extraction since the 19th century for something that was continuously consumed to produce the energy. Materials for renewable energy are the functional equivalents of reusable storage tanks.
      (2) Many energy storage technologies are not based on rare chemical substances.

      As for copper, I don’t see how RE represents exceptional consumption over any other type of power production and distribution. If anything, local-community cluster power generation will reduce the need for the long power-lines that get stolen for their copper in developing nations.

      Of course, I’ve long been a fan of PV solar for its easy installation independent of water sources. You can pack a remote-site PV installation onto the backs of a few donkeys heading into the mountains.

      • ecoquant Says:

        Excellent thoughts.

      • John Oneill Says:

        ‘As for copper, I don’t see how RE represents exceptional consumption over any other type of power production and distribution.’
        ‘Copper usage averages up to five times more in renewable energy systems than in traditional power generation, such as fossil fuel and nuclear power plants.’ according to Wikipedia. Power lines usually use aluminium, not copper. Wind farms in India have already fallen victim to organised gangs of copper thieves armed with power tools. ‘The copper content per installed wind turbine is 2.5–6.4 tonnes per megawatt, as follows: Generator: 0.7–4.0 tonnes of copper. Cabling: 0.7–1.0 tonnes of copper. Transformers: 0.7–1.4 tonnes of copper.’
        https://copperalliance.org.uk/knowledge-base/education/education-resources/copper-wind-power-2/#:~:text=The%20copper%20content%20per%20installed,0.7%E2%80%931.4%20tonnes%20of%20copper.

        • rhymeswithgoalie Says:

          Do wind turbines use more copper for generation than the equivalent megawattage of thermal?

          Of course the the giant high voltage lines use aluminum; the volume is too great. I’m talking about smaller scale power lines that transfer powers along roads and between villages out in the bush. My point was that self-sufficiency is a better idea for villages in developing countries because they typically can’t rely on stable national power and distribution.

          Tangent from Zimbabwe: In an ironic twist of fate, a Bulawayo man, who sustained horrific burns while allegedly attempting to steal a Zesa high voltage transformer at Mpilo Central Hospital, could not be fully attended to for three days as the hospital had no electricity following theft of cables by unknown suspects.

          • ecoquant Says:

            Update from research done for a recent post: In the BP Statistical Review of Energy, the fraction of Lithium, Cobalt, and Copper mined in out of the way places has dropped between 2018 and 2020Q1, and increased in places like the United States, some parts of Europe, Australia, and China.

            Also, if my other response to your post is ever Moderator-approved, you’ll see that the larger requirement for metal for solar PV is Aluminum, not Copper. And transmission demand is expected, per the papers I read to compile responses here, is no bigger than it is now. After all, it’s not like a future transmission grid will be a crossbar network.

            That claim someplace above that

            Copper usage averages up to five times more in renewable energy systems than in traditional power generation, such as fossil fuel and nuclear power plants.

            is puffery from The Copper Alliance as to why their companies are good investments. It is true per share prices of Copper ETFs have bid up appreciably. I thought that might be why some of the solar energy stocks have depreciated. I investigated and I found that, among other things, solar installers and builders are experiencing a pretty severe shortage of electricians. The big projects pay more than residential ones do, and electricians are in demand for new housing starts as well. So they have a worker shortage, as many other industries are experiencing.


  7. Synoptic — now that’s a high falutin sounding word. Using Google, I see it’s often associated with Gospels. The only way anyone could take Jacobson’s numbers seriously would have to be in some sort of religious sense. My favorite example of his ridiculous numbers is his 387 100MW CSPs for his famous New York State plan (no longer in his subsequent 50 state road map).

    There is absolutely no way Jacobson and his ilk have taken a proper detailed look at the mining requirements. I’d love to see him respond to Mark P. Mills:

    The IEA assembled a large body of data about a central, and until now largely ignored, aspect of the energy transition: It requires mining industries and infrastructure that don’t exist. Wind, solar and battery technologies are built from an array of “energy transition minerals,” or ETMs, that must be mined and processed. The IEA finds that with a global energy transition like the one President Biden envisions, demand for key minerals such as lithium, graphite, nickel and rare-earth metals would explode, rising by 4,200%, 2,500%, 1,900% and 700%, respectively, by 2040.

    The world doesn’t have the capacity to meet such demand. As the IEA observes, albeit in cautious bureaucratese, there are no plans to fund and build the necessary mines and refineries. The supply of ETMs is entirely aspirational. And if it were pursued at the quantities dictated by the goals of the energy transition, the world would face daunting environmental, economic and social challenges, along with geopolitical risks.

    https://www.wsj.com/articles/bidens-not-so-clean-energy-transition-11620752282

    • ecoquant Says:

      Synoptic scale“:

      Used with respect to weather systems ranging in size from several hundred kilometers to several thousand kilometers, the scale of migratory high and low pressure systems (frontal cyclones) of the lower troposphere.

      Mike Dombroski’s knowledge of meteorology is about as good as his knowledge of engineering. Here’s another source of atmospheric knowledge for you:

      Jacobson, Mark Z. Fundamentals of atmospheric modeling. Cambridge University press, 2005.

      As far as Mills assertions go, Jacobson and Delucchi synthesized answers to that question in 2011:

      Jacobson, M. Z., & Delucchi, M. A. (2011). Providing all global energy with wind, water, and solar power, Part I: Technologies, energy resources, quantities and areas of infrastructure, and materials. Energy Policy, 39(3), 1154–1169,

      in their section 5, and their work was based upon that of several others. The matter was examined in detail in 2019 by

      Giurco D., Dominish E., Florin N., Watari T., McLellan B. (2019) Requirements for Minerals and Metals for 100% Renewable Scenarios. In: Teske S. (eds) Achieving the Paris Climate Agreement Goals. Springer, Cham.

      and by

      Moreau, Vincent, Piero Carlo Dos Reis, and François Vuille. “Enough metals? Resource constraints to supply a fully renewable energy system.” Resources 8, no. 1 (2019): 29.

      So if Mills and Dombrowski have a problem with the analysis it isn’t with Jacobson. In fact, it’s a lot closer to them.


      • I’m not that versed in the terminology of meteorology so I missed that you were referring to some commonly used scale. I mistakenly interpreted the word synoptic as abject adjective sophistry. I’m sure most of the time there will be generation (by renewables) from weather systems smaller than the synoptic scale, … but not always. And besides that, getting constant generation from renewables to meet demand is still going to be expensive and require lots of land and resources and probably a lot of factors that Jacobson or anyone else has never considered because getting near synoptic scale areas to run exclusively on renewables has never been done before.

        Also, there’s no way for Jacobson to accurately access mining effects, because a lot of them are hidden away in toxic lakes in places like Mongolia. When he gathers data for his studies, he’s going to cherry pick sources he likes such as Lazard and ignore or dis sources he doesn’t like IEA. Here’s Mark Mills’ take on a new IEA report:

        The International Energy Agency, the world’s pre-eminent source of energy information for governments, has entered the political debate over whether the U.S. should spend trillions of dollars to accelerate the energy transition favored by the Biden administration. You know, the plan to use far more “clean energy” and far less hydrocarbons—the oil, natural gas and coal that today supply 84% of global energy needs. The IEA’s 287-page report released this month, “The Role of Critical Minerals in Clean Energy Transitions,” is devastating to those ambitions. A better title would have been: “Clean Energy Transitions: Not Soon, Not Easy and Not Clean.”

        https://www.wsj.com/articles/bidens-not-so-clean-energy-transition-11620752282

        • ecoquant Says:

          Oh, and Lazard is considered an unreliable source? What do they have to gain from biasing their results?

          If anything, while IEA and U.S. EIA may not be biased, they project based upon rearward looking analyses. This is why both badly missed the plummeting levelized costs in solar and wind during the last decade, because they were not taking advantage of good projections.

          People can take Lazard’s projections to the bank. For example, it’s expected EV sales will explode around 2025 because a bunch of battery manufacturing plants will come online globally, driving battery costs down deeply. And TSLA, for instance, is sourcing their Lithium from Nevada, not Chile.

          We’re expecting that, so will be leasing an EV for 3 years to replace a rickety 2005 Corolla. We already have a Tesla 3 (and solar PV on the roof, and getting more of that, and air source heat pumps, etc). The PV system is a profit center, considering both avoided costs and sales of SRECs.


          • What would Lazard have to gain by being biased towards renewables? There’s probably a lot of green-minded investors that they are marketing to. They’re also probably run by green-minded people themselves. There’s undoubtedly a lot of confirmation bias in favor of renewables.

            Could there be an electric car boom by 2025? Possibly, but I don’t think it’s clear. If there is, there’s going to have to be a lot of stuff that gets mined. Lithium is not the only thing EV batteries use. When demand for a mineral goes up, so does its price.

        • ecoquant Says:

          Another thought: Some mastery of meteorology is necessary to understand wind and solar generation. That’s untrue of fossil fuels. There it’s necessary to have some mastery of geology.

  8. ecoquant Says:

    The simplest thing to do is invest your money where and how you think will win.

    In Lazard’s defense, they invest as they say, and they make a lot of money doing it. I guess the securities markets must consist primarily of self-deluded investors as well. Of course, everyone else is wrong but Messrs Dombroski, Smil, Mills, and Shellenberger.

    We’ll see who comes out ahead.

  9. rhymeswithgoalie Says:

    “There’s undoubtedly a lot of confirmation bias in favor of renewables.”

    Do you even know how this blog got its name?

  10. Brent Jensen-Schmidt Says:

    Shortage of minerals is a boogie man excuse for inaction. This will be overcome due to the profit motive. Failing that, alternatives will be found, or we fry,
    As for recycling and environmental damage, they are trivialities compared to CAGW and the standard shitty situation of the world now. Problems to be addressed, not excuses for inaction.


Leave a Reply

Please log in using one of these methods to post your comment:

WordPress.com Logo

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

Google photo

You are commenting using your Google 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 )

Connecting to %s

%d bloggers like this: