The thing I always considered obvious, that so many people I talk to seem to miss, is that every additional marginal reduction in CO2 emissions *buys more time* to solve the remaining problems. Keep investing, keep building, as fast as you can justify, and then maybe you have until 2060 instead of 2050. And that's a *huge* advantage. After all, those 2050-2060 builders have an extra decade of technological development and economic growth (10-30% higher world rGDP?) behind them to take the next step, and the remaining problems they're directing them towards are smaller!
This piece is almost like an echo of my own mind. Greta Thunberg may be an admirable climate activist. But she'd be a lot more useful if she stayed in school and got a related STEM degree and went to work on any of the thousands of technological problems we have before us. No dumber words were ever uttered by a human than stating that addressing carbon reduction in the atmosphere is just a matter of political will. "Keep it in the ground" it in the ground runs a close second.
I could, on the basis of knowledge of how electrical grids operate, criticize the idea that natural gas fired plants can be viewed as backup power supplies to wind and solar. But ultimately it does not matter. If you build excess generating capacity it must be paid for. It must be built and maintained, ready to go into action, at a moments notice. Which means those costs have to be reflected in your power bill whether they produce power or not. The same rules apply to the distribution grid and all those power lines.
There is a reason the Texas power grid was essentially isolated from adjoining grids. A very sound financial reason as it happens. They want to maintain the lowest costs of power by closely aligning potential supply with potential demand. They do not want to carry the costs of excessive redundancy in supply. They don't want other grids dumping cheap power into their grid regardless of source. That weakens their internal suppliers cash flow. And makes grid management more difficult. Cheap power from outside sources drives their overall costs higher. Those domestic, if you will, natural gas plants still have to be staffed and maintained while they aren't producing power.
The real elephant in the room is we need the cost of batteries to go way down from where it is currently. If the grid could store gigawatt-hours of excess solar/wind energy in batteries, we could easily run on renewable power 24/7. We also need cheaper batteries to convert all of those cars, trucks, and other machines that run directly on burning fossil fuels over to electricity.
But, of course, the world we actually live in today is not like that. Instead, we have big bottlenecks in supplies of battery materials, rendering the battery capacity necessary to really turn wind and solar into 24/7 power for entire cities, prohibitively expensive. Hence the need for wind and solar to be complemented by nuclear and gas, which is a fine solution for going from 0% renewable electricity to 20%, but won't work for going from 20% renewable electricity to 100%.
Until somebody comes up with a new battery technology that is far cheaper than lithium ion at massive scale, reducing CO2 emissions without drastic cuts in standard of living is going to be nearly impossible. The good news is that there is a lot of active research in this area, and some ideas have showed early promise. The bad news is that everything is just early prototypes right now, and what works at small scale in a lab isn't always economical at large scale in the real world. But, the solution to this is going to be key to getting world carbon emissions down.
Two points: 1) please don't confuse kilowatts (kw = power) with kilowatt-hours (kwh = energy). Cost issues are usually a function of kwh, not kw. (Your conclusions are still valid). 2) CCS will be in play; DAC is almost certainly never going to be cost effective, energy efficient, and land efficient. Trying to remove CO2 at 400 ppm out of air requires too much air movement for too little return, which will just waste voluminous amounts of energy (and probably land), even if it's green energy.
I like this post, but gritted my teeth each time Mr. Yglesias mistakenly used "kilowatt" instead of "kilowatt-hour".
Most units of flow specify a unit of time, e.g., "gallons per minute" or "dollars per year". The metric unit for energy flow is a joule per second. Confusingly, a joule per second is also called a watt.
Saying that a solar panel produces 1 watt per month is nonsensical:. It means 1 joule per second per month.
Saying that a solar panel produces 1watt-hour per month does make sense:. It means energy production of 3,600 joules per month (1 hour = 3,600 seconds.)
All this seems pedantic, but it underscores an important distinction between a generator's maximum possible output (flow) of electric energy per second, measured in joules per second or watts, and the amount of electric energy actually generated during some time period, measured in joules or watt-seconds or watt-hours.
The best path forward for new low cost nuclear is: Mini CANDU reactor licensed by Canada to India, India produces reactors at scale with cheap labor and ships them around the world, fuel cycle is shifted to a Thorium/HALEU mixture from a company like Clean Core which will prevent proliferation risk, and domestic construction crews are reused across sites to prevent another Vogtle boondoggle
There are only two proven methods of getting to a >90% decarbonized grid: pure hydro and/or pure nuclear. France, Ontario, Quebec, Washington state have all achieved this. I can't think of any decarbonized grid that has more than 30% of electricity coming from PV/wind.
At some point I hope PV/wind proponents will honestly assess the report card of Germany, which has by far the largest renewables capacity per capita in the world for an industrial economy. Their grid is polluting as much as it did before they purchased $500bn of wind and solar.
You seem to be missing a very important factor related to carbon sequestration. We don’t need new technology or fancy methods to do carbon sequestration. We already have a way to do so that is easily scalable and cost effective.
We need to replace conventional farming methods with regenerative farming methods. With regenerative farming the excess C02 is pulled from the air via plants and then used to build topsoil. This is doubly important because conventional farming methods have been destroying topsoil at an alarming rate, and without topsoil we can’t grow food.
Regenerative farming practices do this by restoring the natural liquid carbon pathway that plants and soil microbes have. Basically plants take sunlight and carbon and then trade it to soil microbes and fungi who in turn help the plants by providing nutrients and other services (like pest and disease fighting).
Regenerative farming does this through a combination of methods. Primarily through combining no till, cover crops, rotational grazing along with reducing or eliminating synthetic fertilizer, herbicies, pesticides etc.
For example, we’ve caused lots of environmental damage by removing animals from farms and off to feed lots. Instead bring those animals back to the farm. Use rotational grazing methods were the animals are densely packed but only on the section of land for a brief time (the same way herds of bison would have moved on the plains.
Likewise you should never leave soil bare. You need to put cover crops in to protect the soil and feed the microbes and store carbon by building the topsoil.
Tillage breaks up the soil structure and the fungi networks that are so crucial for building soil. Likewise synthetic fertilizers disrupt the natural liquid carbon cycle because plants get their nutrients from the synthetic fertilizer instead of working together with natural soil microbes to build the soil. And of course pesticides and herbicides also destroy soil microbes.
It’s been estimated that if the entire world switched to regenerative farming practices we could absorb all the excess carbon created since the start of the industrial revolution. Plus you would be creating healthier nutrient rich foods (nutrient density drops greatly under conventional agriculture practices).
It looks like there are several companies looking to build wind-powered cargo ships, but I think there will definitely be an uphill battle. What's cool is that we've done it before (unlike planes)! But the idea of shifting global supply chains into a lower gear? No wonder it's a tough sell to investors.
Sorry, I just re-read your reply, and I think I misunderstood it. I thought you meant pirates would ruin the wind-powered cargo ships idea, but now I think you meant the same thing as my point about hijacking a nuclear-powered cargo ship.
If Democrats are so smart about tradeoffs, why was Diabo Canyon even considered being taken down? Why has the international left been against building nuclear? I think it's actually still up in the air how captured Democrats are by anti growth environmentalists, and it seems clear to me that Diablo Canyon is only open because of a last minute energy crisis the Democrats did not forsee because that's just not something on the top of their minds.
IMO, in the end Diablo Canyon was saved by the nuclear subsidies in the democrats' Inflation Reduction Act.
Environmentalists have some sway at the margins, but these decisions seem to be driven first and foremost by plant economics and local grid conditions. We see this play out over and over again all across the country. If the economics are marginal for the plant, the environmentalists win and it gets shut down. If the plant economics are solid, the plant stays open.
Originally, Diablo Canyon needed several billion in subsidies for the owners to be interested in keeping the plant open. I believe that was on top of the cost of the new cooling towers. Higher nat gas prices came along and made the economics better, but still not good enough for the math to work. Finally, IRA came along and its nuclear power subsidies were enough of a boost to save the plant. Once the rest of the math worked, Newsom was fine giving the plant a waiver on the cooling towers.
I guess I wonder then, without IRA, does CA shut down that plant and 10% of its electricity in this energy environment, or does it find the money somewhere else? Because it seems pretty bad for that plant to be closing right now.
And say that the timing didn't line up and we closed it a year ago, and CA just burned a bunch of coal to meet its energy demand. Doesn't that look horrible for Democrats who are supposed to care about the environment? Shouldn't their finger be on the opposite side of the "margin" here?
Great article Matt! One more thing to comment about renewables to expound on your comment about the tradeoffs in building renewables and the incredible amount of space they take up. I recently listened to an excellent episode of the Ezra Klein Show where his guest was climate expert Jesse Jenkins of Princeton. When talking about the area needed provide all the electricity our country needs from wind and solar alone, Jenkins said, "The most cost effective of our net-zero scenarios spans an area that is equal to Illinois, Indiana, Ohio, Kentucky and Tennessee put together. And the solar farms are an area the size of Connecticut, Rhode Island and Massachusetts."
>>California, probably the leading-edge state on renewables, actually built a small number of emergency gas generators that they turned on for the first time on September 4 of this year to meet rare peak demand and avoid blackouts.<<
The problem during California's big heat wave was actually a lot more daunting than this suggests. On Sept. 4, that 120M of emergency gas generation accounted for 0.5% (sic) of natural gas-generated electricity at the day's peak, and about 0.3% of total electricity demand at that time. And that wasn't the highest day, which came a couple days later.
The big heat wave proved unfortunately just how dependent California is still at this time on natural gas, not only because of its total size but because of its reliability. The heat wave put this big stagnant dome of air over the state, causing the winds to die down and wind-generated electricity to tumble. Smoke, ash and clouds from various fires caused solar-generated electricity to vary erratically. The sustained drought has emptied reservoirs making large hydro-generated electricity a much less important contributor. Battery power is increasing but only so far hitting a max of around 3% of demand. Nuclear power is a stable but small source (around 5% of demand) and can't ramp up for peak demand. Thus, natural gas.
More wind, solar and battery power has to be the way forward, but the package will always be subject to lower sustained reliability. We need to do everything obviously, including expanding nuclear, but the key factor will be linking into an integrated national network, so we can draw on solar, wind and battery power from surplus capacity around the nation. And that means passing the permitting legislation which environmentalists are partnering with Republicans to defeat. As Yakov Smirnoff would say, "What a country!"
This is only adjacent to the crux of the piece, but the "wasted electricity" part really seems like something we just kind of tolerate for no real good reason. Why does it only depend on energy storage? It seems like it should be somewhat easier to simply find uses for this energy - desalination seems like the most obvious opportunity but it's of course not plausible everywhere for various reasons (matching power generation to areas with water to desalinate or actual water problems, time to build out, etc).
"Because massively overbuilding renewables would not only cost a lot of money but wastefully consume vast tracts of land, it seems like a better idea would be to use long-term batteries. If you had really big batteries that stored electricity for a long time, you could simply store surplus power in the high season and unleash it in the low season."
In the spirit of focusing the energy debate on numbers, not adjectives (https://bit.ly/3LJLFUZ), Iet's look at some of the literature we have on this trade-off.
Perez et. al, 2020 find that "while unconstrained, intermittent renewable generation will achieve very low [cost] targets-they are already below grid parity-transforming PV into the firm, effectively dispatchable resource needed by world economies will be very costly if done with storage alone, even when considering the most aggressive future cost projections for storage. Overbuilding renewables can reduce the storage requirements to the point where true below parity firm generation will be attainable." (https://bit.ly/3SE0YRq)
Concretely: in Minnesota, 0% overbuilding of panel requires so much storage to deal with seasonal intermittency that cost of power shoots up to 28 cents/KWh, compared to 5 cents/KWh today. By contrast, 50% overbuilding (implying you'd throw away a third of the summer surplus) drives the cost down below grid parity, 5 cents/KWh. (The mix assumed here is 55% solar / 40% wind / 5% natural gas). (https://bit.ly/3SE0YRq) They've observed the relationship between overbuilding and cost across other geographies (https://bit.ly/3ULxov9).
The reason this happens is that 1.5x overbuilding of panels allows you to use 10x less storage (https://bit.ly/3SDZ3fG). That's the tradeoff.
This is just one study. I'm it has limitations (it's not clear whether the growing marginal cost for summer-surplus panels is factored into the cost figures above), and I'd love for folks to chime in with other evidence. The key claim here that solving seasonal intermittency via storage requires such profound amounts of storage that it will never be cost-effective.
The thing I always considered obvious, that so many people I talk to seem to miss, is that every additional marginal reduction in CO2 emissions *buys more time* to solve the remaining problems. Keep investing, keep building, as fast as you can justify, and then maybe you have until 2060 instead of 2050. And that's a *huge* advantage. After all, those 2050-2060 builders have an extra decade of technological development and economic growth (10-30% higher world rGDP?) behind them to take the next step, and the remaining problems they're directing them towards are smaller!
This piece is almost like an echo of my own mind. Greta Thunberg may be an admirable climate activist. But she'd be a lot more useful if she stayed in school and got a related STEM degree and went to work on any of the thousands of technological problems we have before us. No dumber words were ever uttered by a human than stating that addressing carbon reduction in the atmosphere is just a matter of political will. "Keep it in the ground" it in the ground runs a close second.
I could, on the basis of knowledge of how electrical grids operate, criticize the idea that natural gas fired plants can be viewed as backup power supplies to wind and solar. But ultimately it does not matter. If you build excess generating capacity it must be paid for. It must be built and maintained, ready to go into action, at a moments notice. Which means those costs have to be reflected in your power bill whether they produce power or not. The same rules apply to the distribution grid and all those power lines.
There is a reason the Texas power grid was essentially isolated from adjoining grids. A very sound financial reason as it happens. They want to maintain the lowest costs of power by closely aligning potential supply with potential demand. They do not want to carry the costs of excessive redundancy in supply. They don't want other grids dumping cheap power into their grid regardless of source. That weakens their internal suppliers cash flow. And makes grid management more difficult. Cheap power from outside sources drives their overall costs higher. Those domestic, if you will, natural gas plants still have to be staffed and maintained while they aren't producing power.
The real elephant in the room is we need the cost of batteries to go way down from where it is currently. If the grid could store gigawatt-hours of excess solar/wind energy in batteries, we could easily run on renewable power 24/7. We also need cheaper batteries to convert all of those cars, trucks, and other machines that run directly on burning fossil fuels over to electricity.
But, of course, the world we actually live in today is not like that. Instead, we have big bottlenecks in supplies of battery materials, rendering the battery capacity necessary to really turn wind and solar into 24/7 power for entire cities, prohibitively expensive. Hence the need for wind and solar to be complemented by nuclear and gas, which is a fine solution for going from 0% renewable electricity to 20%, but won't work for going from 20% renewable electricity to 100%.
Until somebody comes up with a new battery technology that is far cheaper than lithium ion at massive scale, reducing CO2 emissions without drastic cuts in standard of living is going to be nearly impossible. The good news is that there is a lot of active research in this area, and some ideas have showed early promise. The bad news is that everything is just early prototypes right now, and what works at small scale in a lab isn't always economical at large scale in the real world. But, the solution to this is going to be key to getting world carbon emissions down.
Two points: 1) please don't confuse kilowatts (kw = power) with kilowatt-hours (kwh = energy). Cost issues are usually a function of kwh, not kw. (Your conclusions are still valid). 2) CCS will be in play; DAC is almost certainly never going to be cost effective, energy efficient, and land efficient. Trying to remove CO2 at 400 ppm out of air requires too much air movement for too little return, which will just waste voluminous amounts of energy (and probably land), even if it's green energy.
I like this post, but gritted my teeth each time Mr. Yglesias mistakenly used "kilowatt" instead of "kilowatt-hour".
Most units of flow specify a unit of time, e.g., "gallons per minute" or "dollars per year". The metric unit for energy flow is a joule per second. Confusingly, a joule per second is also called a watt.
Saying that a solar panel produces 1 watt per month is nonsensical:. It means 1 joule per second per month.
Saying that a solar panel produces 1watt-hour per month does make sense:. It means energy production of 3,600 joules per month (1 hour = 3,600 seconds.)
All this seems pedantic, but it underscores an important distinction between a generator's maximum possible output (flow) of electric energy per second, measured in joules per second or watts, and the amount of electric energy actually generated during some time period, measured in joules or watt-seconds or watt-hours.
The best path forward for new low cost nuclear is: Mini CANDU reactor licensed by Canada to India, India produces reactors at scale with cheap labor and ships them around the world, fuel cycle is shifted to a Thorium/HALEU mixture from a company like Clean Core which will prevent proliferation risk, and domestic construction crews are reused across sites to prevent another Vogtle boondoggle
There are only two proven methods of getting to a >90% decarbonized grid: pure hydro and/or pure nuclear. France, Ontario, Quebec, Washington state have all achieved this. I can't think of any decarbonized grid that has more than 30% of electricity coming from PV/wind.
At some point I hope PV/wind proponents will honestly assess the report card of Germany, which has by far the largest renewables capacity per capita in the world for an industrial economy. Their grid is polluting as much as it did before they purchased $500bn of wind and solar.
You seem to be missing a very important factor related to carbon sequestration. We don’t need new technology or fancy methods to do carbon sequestration. We already have a way to do so that is easily scalable and cost effective.
We need to replace conventional farming methods with regenerative farming methods. With regenerative farming the excess C02 is pulled from the air via plants and then used to build topsoil. This is doubly important because conventional farming methods have been destroying topsoil at an alarming rate, and without topsoil we can’t grow food.
Regenerative farming practices do this by restoring the natural liquid carbon pathway that plants and soil microbes have. Basically plants take sunlight and carbon and then trade it to soil microbes and fungi who in turn help the plants by providing nutrients and other services (like pest and disease fighting).
Regenerative farming does this through a combination of methods. Primarily through combining no till, cover crops, rotational grazing along with reducing or eliminating synthetic fertilizer, herbicies, pesticides etc.
For example, we’ve caused lots of environmental damage by removing animals from farms and off to feed lots. Instead bring those animals back to the farm. Use rotational grazing methods were the animals are densely packed but only on the section of land for a brief time (the same way herds of bison would have moved on the plains.
Likewise you should never leave soil bare. You need to put cover crops in to protect the soil and feed the microbes and store carbon by building the topsoil.
Tillage breaks up the soil structure and the fungi networks that are so crucial for building soil. Likewise synthetic fertilizers disrupt the natural liquid carbon cycle because plants get their nutrients from the synthetic fertilizer instead of working together with natural soil microbes to build the soil. And of course pesticides and herbicides also destroy soil microbes.
It’s been estimated that if the entire world switched to regenerative farming practices we could absorb all the excess carbon created since the start of the industrial revolution. Plus you would be creating healthier nutrient rich foods (nutrient density drops greatly under conventional agriculture practices).
Great post -- count me as being on agnosticism bandwagon -- need "all of the above"
Is it just stupid to talk about sail for shipping? This is a genuine question.
It looks like there are several companies looking to build wind-powered cargo ships, but I think there will definitely be an uphill battle. What's cool is that we've done it before (unlike planes)! But the idea of shifting global supply chains into a lower gear? No wonder it's a tough sell to investors.
Pirates would likely ruin it for everyone, but if we can have nuclear warships then we should be able to have nuclear cargo ships.
Sorry, I just re-read your reply, and I think I misunderstood it. I thought you meant pirates would ruin the wind-powered cargo ships idea, but now I think you meant the same thing as my point about hijacking a nuclear-powered cargo ship.
Yep, that's right. Although wind powered cargo ships would be utter sitting ducks for today's pirates!
The flipside is the blackmail potential for pirates if they seize a large cargo ship with a nuclear reactor powering it.
The flipside to that flipside are all the Hollywood blockbusters that could be made with that premise!
^^ this person gets it
If Democrats are so smart about tradeoffs, why was Diabo Canyon even considered being taken down? Why has the international left been against building nuclear? I think it's actually still up in the air how captured Democrats are by anti growth environmentalists, and it seems clear to me that Diablo Canyon is only open because of a last minute energy crisis the Democrats did not forsee because that's just not something on the top of their minds.
IMO, in the end Diablo Canyon was saved by the nuclear subsidies in the democrats' Inflation Reduction Act.
Environmentalists have some sway at the margins, but these decisions seem to be driven first and foremost by plant economics and local grid conditions. We see this play out over and over again all across the country. If the economics are marginal for the plant, the environmentalists win and it gets shut down. If the plant economics are solid, the plant stays open.
Originally, Diablo Canyon needed several billion in subsidies for the owners to be interested in keeping the plant open. I believe that was on top of the cost of the new cooling towers. Higher nat gas prices came along and made the economics better, but still not good enough for the math to work. Finally, IRA came along and its nuclear power subsidies were enough of a boost to save the plant. Once the rest of the math worked, Newsom was fine giving the plant a waiver on the cooling towers.
I guess I wonder then, without IRA, does CA shut down that plant and 10% of its electricity in this energy environment, or does it find the money somewhere else? Because it seems pretty bad for that plant to be closing right now.
And say that the timing didn't line up and we closed it a year ago, and CA just burned a bunch of coal to meet its energy demand. Doesn't that look horrible for Democrats who are supposed to care about the environment? Shouldn't their finger be on the opposite side of the "margin" here?
Great article Matt! One more thing to comment about renewables to expound on your comment about the tradeoffs in building renewables and the incredible amount of space they take up. I recently listened to an excellent episode of the Ezra Klein Show where his guest was climate expert Jesse Jenkins of Princeton. When talking about the area needed provide all the electricity our country needs from wind and solar alone, Jenkins said, "The most cost effective of our net-zero scenarios spans an area that is equal to Illinois, Indiana, Ohio, Kentucky and Tennessee put together. And the solar farms are an area the size of Connecticut, Rhode Island and Massachusetts."
And a battery the size of Kansas.
Side note: is Farhad the worst NY Times columnist? I don't think I've ever read one of his columns and come away thinking it was well-reasoned.
I'll admit I don't read Farhad. The last time I did was when he was with Salon and I never thought much of his writing then.
>>California, probably the leading-edge state on renewables, actually built a small number of emergency gas generators that they turned on for the first time on September 4 of this year to meet rare peak demand and avoid blackouts.<<
The problem during California's big heat wave was actually a lot more daunting than this suggests. On Sept. 4, that 120M of emergency gas generation accounted for 0.5% (sic) of natural gas-generated electricity at the day's peak, and about 0.3% of total electricity demand at that time. And that wasn't the highest day, which came a couple days later.
The big heat wave proved unfortunately just how dependent California is still at this time on natural gas, not only because of its total size but because of its reliability. The heat wave put this big stagnant dome of air over the state, causing the winds to die down and wind-generated electricity to tumble. Smoke, ash and clouds from various fires caused solar-generated electricity to vary erratically. The sustained drought has emptied reservoirs making large hydro-generated electricity a much less important contributor. Battery power is increasing but only so far hitting a max of around 3% of demand. Nuclear power is a stable but small source (around 5% of demand) and can't ramp up for peak demand. Thus, natural gas.
More wind, solar and battery power has to be the way forward, but the package will always be subject to lower sustained reliability. We need to do everything obviously, including expanding nuclear, but the key factor will be linking into an integrated national network, so we can draw on solar, wind and battery power from surplus capacity around the nation. And that means passing the permitting legislation which environmentalists are partnering with Republicans to defeat. As Yakov Smirnoff would say, "What a country!"
This is only adjacent to the crux of the piece, but the "wasted electricity" part really seems like something we just kind of tolerate for no real good reason. Why does it only depend on energy storage? It seems like it should be somewhat easier to simply find uses for this energy - desalination seems like the most obvious opportunity but it's of course not plausible everywhere for various reasons (matching power generation to areas with water to desalinate or actual water problems, time to build out, etc).
Crypto mining to reduce the federal deficit!!!! \S
"Because massively overbuilding renewables would not only cost a lot of money but wastefully consume vast tracts of land, it seems like a better idea would be to use long-term batteries. If you had really big batteries that stored electricity for a long time, you could simply store surplus power in the high season and unleash it in the low season."
In the spirit of focusing the energy debate on numbers, not adjectives (https://bit.ly/3LJLFUZ), Iet's look at some of the literature we have on this trade-off.
Perez et. al, 2020 find that "while unconstrained, intermittent renewable generation will achieve very low [cost] targets-they are already below grid parity-transforming PV into the firm, effectively dispatchable resource needed by world economies will be very costly if done with storage alone, even when considering the most aggressive future cost projections for storage. Overbuilding renewables can reduce the storage requirements to the point where true below parity firm generation will be attainable." (https://bit.ly/3SE0YRq)
Concretely: in Minnesota, 0% overbuilding of panel requires so much storage to deal with seasonal intermittency that cost of power shoots up to 28 cents/KWh, compared to 5 cents/KWh today. By contrast, 50% overbuilding (implying you'd throw away a third of the summer surplus) drives the cost down below grid parity, 5 cents/KWh. (The mix assumed here is 55% solar / 40% wind / 5% natural gas). (https://bit.ly/3SE0YRq) They've observed the relationship between overbuilding and cost across other geographies (https://bit.ly/3ULxov9).
The reason this happens is that 1.5x overbuilding of panels allows you to use 10x less storage (https://bit.ly/3SDZ3fG). That's the tradeoff.
This is just one study. I'm it has limitations (it's not clear whether the growing marginal cost for summer-surplus panels is factored into the cost figures above), and I'd love for folks to chime in with other evidence. The key claim here that solving seasonal intermittency via storage requires such profound amounts of storage that it will never be cost-effective.
But I do think it's worth a response!