Rapid, large-scale deployment of solar photovoltaic energy generation technology is essential to the reduction of greenhouse gas emissions and by extension the preservation of the human way of...
Rapid, large-scale deployment of solar photovoltaic energy generation technology is essential to the reduction of greenhouse gas emissions and by extension the preservation of the human way of life. Or as the kids say, "We đź‘Ź stan đź‘Ź renewable đź‘Ź energy đź‘Ź oil and gas can eat my shorts."
In the 2024 Tildes Predictions thread, I predicted (among other things) that solar deployment would exceed expectations next year. I guess that's a bit of a paradox since now that I'm expecting it, it's no longer unexpected. Whatever. Some outlets are suggesting that we could see 1 TW of solar added to the grid annually by 2030. Outlandish... is it possible?
For reference, global installed (nameplate) electricity capacity right now is about 9 TW, "contributing about 20% of the total global primary energy." That means there's currently potential for at least ~50 TW of renewable nameplate capacity, and by the time that capacity is installed, at least another 10–15 TW.
The reason we would want to install even more than that is that the capacity factor of solar electricity is only about 10–29% (coal and natural gas are more like 55–65% and nuclear more like 90%). This would adequately if inefficiently account for variable demand peaks. Though if the technology becomes cheap enough (which seems plausible), it would be more efficient than relying on supposedly "stable" fuels like natural gas for variable peak demand. Obviously, battery storage is a more technically efficient alternative, but the economics are still being figured out on that one.
Recent assessments that indicated the energy transition was lagging are now being revisited in light of these developments. Released this summer, IRENA’s World Energy Transitions Outlook report showed that the 191 GW added in 2022 fell short of the Paris Agreement’s 1.5°C target. The report called for significant increases, targeting “551 GW annually by 2030, and 615 GW annually by 2050.”
Yet, with projections of 413 GW in 2023 and an expected 500 to 600 GW in 2024, it appears these ambitious targets may not just be achievable, but surpassed.
The IEA now projects that to meet global climate goals, approximately 633 GW per year will be needed by 2030, a target that could be achieved as soon as next year.
According to research from the Rocky Mountain Institute, “eight countries have already grown solar and wind generation faster than what’s needed to limit global warming to 1.5°C, proving that a rapid transition to renewable energy is possible.” According to their analysis, solar and wind are forecast to increase from approximately 12% of global electricity generation at the end of 2022, to over a third by 2030.
That's all to say that solar has plenty of room to grow, even accounting for competition from wind. I don't really give a hoot which one wins out because they're both renewable, but solar seems to be poised to take the lead. (BTW, nuclear fusion is so far off that it's not even worth discussing as part of the renewable pivot.) See also: solar panel recycling (Undecided with Matt Ferrell).
We're not hitting our 1.5 C targets, but 2 C seems plausible, especially if renewables continue to outperform expectations as they have this year. Recent government incentives like the Inflation Reduction Act in the United States are probably playing a major role, but the technology is inherently financially sustainable.
So, I made a critical error in my last math post, but the reason I made the post is the two things that I never see mentioned in the solar and wind deployment articles : Nameplate vs Real, and...
So, I made a critical error in my last math post, but the reason I made the post is the two things that I never see mentioned in the solar and wind deployment articles : Nameplate vs Real, and what is required annually for just covering new growth. We need to greatly exceed new growth to actually reduce greenhouse emissions.
At 50% I was getting 1,800TWh. Using the 10-20% nameplate for solar puts 2024's projected installs at 360TWh on the low bound and 720TWh on the upper bound.
Worldwide energy demand increases about 3-4% every year...has since before 1900. That's 870TWh, increasing every year.
So if solar installs are done in places that more reliably hit that 20% or better, we stand a good chance of eating away at fossil fuels with solar within a few years. But if they're only getting closer to that 10%, we're still a long ways off.
There are two big reasons we need to build many nuclear plants: So we can retire the ancient ones, because they are so old we should be looking to replace them ASAP....many are operating way past expected EOL. And because even with optimistic projections, even though solar is 5x cheaper to install than nuclear plants, nuclear plants product 5x more electricity and are going to be able to cover the gaps better, especially where winters days are short.
Not even factoring this is just about meeting existing, regular demand growth. If we start mass-deployment of plug-in hybrids, or want to greenly desalinate and generate hydrogen, we're gonna need to build as much capacity as we can.
And large storage mediums are still an unsolved problem. If 50% or more of grid power is to become wind/solar...this problem needs a solution soon.
The real usage figures make it seem like this is a more insurmountable issue than it actually is (9 TW and 1800 TWh sound far off—but they don’t measure the same thing). The shorthand that...
The real usage figures make it seem like this is a more insurmountable issue than it actually is (9 TW and 1800 TWh sound far off—but they don’t measure the same thing). The shorthand that involves minimal math is just to multiply the nameplate capacity by the capacity factor to get the real capacity:
Solar capacity of 1 TW * 0.25 = 250 GW
Nuclear capacity of 1 TW * 0.9 = 900 GW
Something something units. I’m sticking with capacity (watts) rather than usage (watt-hours) because I can’t be bothered to translate, and I think real capacity does fine.
So yes, everything you’re saying is true and I think it’s reasonable to continue building and maintaining nuclear plants to make up for solar’s abysmal capacity factor. But the calculations there might change if solar continues to drop in price significantly. For utility-scale solar, I think that could happen. (But probably not for residential purposes where labor is the largest cost.)
As you remark, nuclear may be about equivalently generative as it is expensive right now, especially at northerly latitudes. The capacity factor could easily be 3x and perhaps 5x there, or as much as 9x. The levelized cost of electricity of nuclear is nearing $200 million/MWh compared to $50 million/MWh for solar (that’s about 4x). The economics favor nuclear in many situations, especially highly variable demand patterns in climates with little sunlight.
The observation I would make is that while solar has been dropping in price for decades (1/7 of the cost since 2009), nuclear has been going up (2x the cost). I can’t tell you the reason for the latter, but it’s a mature technology and I would be surprised if that trend reversed. At best, it could plateau. Whereas we have every indication that solar will continue dropping in price.
If the LCOE from wind can make up for nighttime generation, which given its track record (like solar, consistently falling prices over time in an emerging technology) it probably can, there would seem to be even less demand for new nuclear plants. That is, the “niche” of nuclear would be left to times when both solar and wind fail to meet peak demand. I see how that could be more of a concern in Germany than, say, the United States (which has a relative abundance of sunlight) or Scotland (which has a relative abundance of wind). So I see a place for nuclear in the energy mix, but in many places I’m not sure installing a significant amount of new nuclear capacity is exactly the optimal strategy. I’m not morally opposed, it just doesn’t seem all that economical.
Like you say, grid storage is probably the biggest issue here. I have been loosely following battery tech innovations for the past couple years and there seem to be promising developments. For example, sodium-ion batteries seem useful for grid deployments because their physical size is irrelevant to a large-scale battery farm. (Not so with an electric vehicle, but that’s neither here nor there.) Others too. Chemistry is a bit of a mysterious world to me, so I can’t comment on the details, but I think there is also a lot of progress happening here.
Part of the reason I converted to total TWh produced is that it was easier for me to find sources on that information, particularily for wind generation. Wind generation is especially tricky as...
Part of the reason I converted to total TWh produced is that it was easier for me to find sources on that information, particularily for wind generation. Wind generation is especially tricky as the current lack of storage is really killing its potential generation, best I can tell. I recall seeing several articles about needing to hit the brakes on various wind farms during huge production times because the grid couldn't accept the load.
IE, it doesn't matter nearly as much if there's still a 200GW (imaginary numbers) coal plant still out there if it's only turned on for a month and produces 1TWh.
Rapid, large-scale deployment of solar photovoltaic energy generation technology is essential to the reduction of greenhouse gas emissions and by extension the preservation of the human way of life. Or as the kids say, "We đź‘Ź stan đź‘Ź renewable đź‘Ź energy đź‘Ź oil and gas can eat my shorts."
In the 2024 Tildes Predictions thread, I predicted (among other things) that solar deployment would exceed expectations next year. I guess that's a bit of a paradox since now that I'm expecting it, it's no longer unexpected. Whatever. Some outlets are suggesting that we could see 1 TW of solar added to the grid annually by 2030. Outlandish... is it possible?
For reference, global installed (nameplate) electricity capacity right now is about 9 TW, "contributing about 20% of the total global primary energy." That means there's currently potential for at least ~50 TW of renewable nameplate capacity, and by the time that capacity is installed, at least another 10–15 TW.
The reason we would want to install even more than that is that the capacity factor of solar electricity is only about 10–29% (coal and natural gas are more like 55–65% and nuclear more like 90%). This would adequately if inefficiently account for variable demand peaks. Though if the technology becomes cheap enough (which seems plausible), it would be more efficient than relying on supposedly "stable" fuels like natural gas for variable peak demand. Obviously, battery storage is a more technically efficient alternative, but the economics are still being figured out on that one.
That's all to say that solar has plenty of room to grow, even accounting for competition from wind. I don't really give a hoot which one wins out because they're both renewable, but solar seems to be poised to take the lead. (BTW, nuclear fusion is so far off that it's not even worth discussing as part of the renewable pivot.) See also: solar panel recycling (Undecided with Matt Ferrell).
We're not hitting our 1.5 C targets, but 2 C seems plausible, especially if renewables continue to outperform expectations as they have this year. Recent government incentives like the Inflation Reduction Act in the United States are probably playing a major role, but the technology is inherently financially sustainable.
So, I made a critical error in my last math post, but the reason I made the post is the two things that I never see mentioned in the solar and wind deployment articles : Nameplate vs Real, and what is required annually for just covering new growth. We need to greatly exceed new growth to actually reduce greenhouse emissions.
At 50% I was getting 1,800TWh. Using the 10-20% nameplate for solar puts 2024's projected installs at 360TWh on the low bound and 720TWh on the upper bound.
Worldwide energy demand increases about 3-4% every year...has since before 1900. That's 870TWh, increasing every year.
So if solar installs are done in places that more reliably hit that 20% or better, we stand a good chance of eating away at fossil fuels with solar within a few years. But if they're only getting closer to that 10%, we're still a long ways off.
There are two big reasons we need to build many nuclear plants: So we can retire the ancient ones, because they are so old we should be looking to replace them ASAP....many are operating way past expected EOL. And because even with optimistic projections, even though solar is 5x cheaper to install than nuclear plants, nuclear plants product 5x more electricity and are going to be able to cover the gaps better, especially where winters days are short.
Not even factoring this is just about meeting existing, regular demand growth. If we start mass-deployment of plug-in hybrids, or want to greenly desalinate and generate hydrogen, we're gonna need to build as much capacity as we can.
And large storage mediums are still an unsolved problem. If 50% or more of grid power is to become wind/solar...this problem needs a solution soon.
The real usage figures make it seem like this is a more insurmountable issue than it actually is (9 TW and 1800 TWh sound far off—but they don’t measure the same thing). The shorthand that involves minimal math is just to multiply the nameplate capacity by the capacity factor to get the real capacity:
1 TW * 0.25 = 250 GW1 TW * 0.9 = 900 GWSomething something units. I’m sticking with capacity (watts) rather than usage (watt-hours) because I can’t be bothered to translate, and I think real capacity does fine.
So yes, everything you’re saying is true and I think it’s reasonable to continue building and maintaining nuclear plants to make up for solar’s abysmal capacity factor. But the calculations there might change if solar continues to drop in price significantly. For utility-scale solar, I think that could happen. (But probably not for residential purposes where labor is the largest cost.)
As you remark, nuclear may be about equivalently generative as it is expensive right now, especially at northerly latitudes. The capacity factor could easily be 3x and perhaps 5x there, or as much as 9x. The levelized cost of electricity of nuclear is nearing $200 million/MWh compared to $50 million/MWh for solar (that’s about 4x). The economics favor nuclear in many situations, especially highly variable demand patterns in climates with little sunlight.
The observation I would make is that while solar has been dropping in price for decades (1/7 of the cost since 2009), nuclear has been going up (2x the cost). I can’t tell you the reason for the latter, but it’s a mature technology and I would be surprised if that trend reversed. At best, it could plateau. Whereas we have every indication that solar will continue dropping in price.
If the LCOE from wind can make up for nighttime generation, which given its track record (like solar, consistently falling prices over time in an emerging technology) it probably can, there would seem to be even less demand for new nuclear plants. That is, the “niche” of nuclear would be left to times when both solar and wind fail to meet peak demand. I see how that could be more of a concern in Germany than, say, the United States (which has a relative abundance of sunlight) or Scotland (which has a relative abundance of wind). So I see a place for nuclear in the energy mix, but in many places I’m not sure installing a significant amount of new nuclear capacity is exactly the optimal strategy. I’m not morally opposed, it just doesn’t seem all that economical.
Like you say, grid storage is probably the biggest issue here. I have been loosely following battery tech innovations for the past couple years and there seem to be promising developments. For example, sodium-ion batteries seem useful for grid deployments because their physical size is irrelevant to a large-scale battery farm. (Not so with an electric vehicle, but that’s neither here nor there.) Others too. Chemistry is a bit of a mysterious world to me, so I can’t comment on the details, but I think there is also a lot of progress happening here.
Part of the reason I converted to total TWh produced is that it was easier for me to find sources on that information, particularily for wind generation. Wind generation is especially tricky as the current lack of storage is really killing its potential generation, best I can tell. I recall seeing several articles about needing to hit the brakes on various wind farms during huge production times because the grid couldn't accept the load.
IE, it doesn't matter nearly as much if there's still a 200GW (imaginary numbers) coal plant still out there if it's only turned on for a month and produces 1TWh.