4 ms·

Yep. Anyone who has bothered to look at the numbers would likely agree with you. To that end: - Here are my numbers on German solar & wind capacity factors (nu

by spangry 10y ago

Yep. Anyone who has bothered to look at the numbers would likely agree with you. To that end: - Here are my numbers on German solar & wind capacity factors (nuclear generally achieves 90% by comparison): https://gist.github.com/anonymous/f1a6d064890d67fbfc98d66dbdf610ac https://gist.github.com/anonymous/f1a6d064890d67fbfc98d66dbd... - Sketchier numbers for India: https://news.ycombinator.com/item?id=14085871 https://news.ycombinator.com/item?id=14085871 - The view of the Brookings Institute in 2014: http://www.brookings.edu/~/media/research/files/papers/2014/05/19-low-carbon-future-wind-solar-power-frank/net-benefits-final.pdf http://www.brookings.edu/~/media/research/files/papers/2014/... Their base-case assumption is a $50 per tonne price on co2 emissions, and they use a 'costs avoided' methodology so that things like cost of system stabilisation (e.g. grid batteries due to intermittent generation) are factored in as well. The paper concludes: Assuming that reductions in carbon dioxide emissions are valued at $50 per metric ton and the price of natural gas is not much greater than $16 per million Btu, the net benefits of new nuclear, hydro, and natural gas combined cycle plants far outweigh the net benefits of new wind or solar plants. Wind and solar power are very costly from a social perspective because of their very high capacity cost, their very low capacity factors, and their lack of reliability. For example, adjusting U.S. solar and wind capacity factors to take account of lack of reliability, we estimate that it would take 7.30 MW of solar capacity, costing roughly four times as much per MW to produce the same electrical output with the same degree of reliability as a baseload gas combined cycle plant. It requires an investment of approximately $29 million in utility-scale solar capacity to produce the same output with the same reliability as a $1 million investment in gas combined cycle. Reductions in the price of solar photovoltaic panels have reduced costs for utility-scale solar plants, but photovoltaic panels account for only a fraction of the cost of a solar plant. Thus such price reductions are unlikely to make solar power competitive with other electricity technologies without government subsidies. Wind plants are far more economical in reducing emissions than solar plants, but much less economical than hydro, nuclear and gas combined cycle plants. Wind plants can operate at a capacity factor of 30 percent or more and cost about twice as much per MW to build as a gas combined cycle plant. Taking account of the lack of wind reliability, it takes an investment of approximately $10 million in wind plants to produce the same amount of electricity with the same reliability as a $1 million investment in gas combined cycle plants.

Cthulhu_ 10y ago
Thanks for that summary - if I'm right, the summary is that solar is 29 times as expensive as fossil fuels, and wind is 10 times as expensive.
- spangry 10y ago
  No problemo. And yeah, that's how I read it, bearing in mind the usual caveats like: all costs (mining to site remediation) being accounted for, technological and price changes over the past 3 years etc. etc. I think 'cost avoided' is a good measure; LCOE is so sensitive to assumptions and accounting trickery that it's basically meaningless (see: Lazard). Brookings seem to have made a fair effort (best I've seen thus far). It makes sense that combined-cycle gas turbines would pip nuclear power (in most scenarios examined) as they produce highly dispatchable electricity (i.e. their output levels can be changed very rapidly in response to fluctuations in demand for electricity). You can forecast and 'plan ahead' with nuclear, but the ramping times are currently too slow. So you either overbuild, rely on a limited amount of grid-level storage or throw a more responsive form of generation into the mix (i.e. gas turbines). Most of the cost of nuclear is in building the plant, after which the variable cost of generation is tiny (i.e. you just run it full-bore all the time, since oversupplying doesn't carry a high cost). Coal probably has a similar cost profile (maybe with slightly higher relative variable costs), although on a smaller scale and in a relatively under-regulated market: the WHO estimates pollution from coal-fired generation, not taking into account climate-change, kills ~1m people per year, whereas nuclear containment buildings must be built to withstand a plane crashing directly into it. But the really big take-away for me is that, if we want to mitigate climate change efficiently (which I'm starting to think is the same as saying 'if we want to mitigate it at all'), we really badly need to place a price on carbon emissions (and their equivalents). It sucks that both extreme sides of the debate don't seem to want this: vested interests will obviously oppose it, and 'extreme greens' seem to think it's some sort of immoral conspiracy designed to allow financial markets to make money off climate change (if we're talking 'cap and trade'). Those who don't like market-based mechanisms such as 'cap and trade' should consider the other side of the coin: it creates the opportunity for direct action on your part. You could buy emissions permits on the open market and then sit on them, permanently lowering the emissions cap. You can finally 'put your money where your mouth is'.

by spangry 10y ago

Cthulhu_ 10y ago

Thanks for that summary - if I'm right, the summary is that solar is 29 times as expensive as fossil fuels, and wind is 10 times as expensive.

spangry 10y ago
No problemo. And yeah, that's how I read it, bearing in mind the usual caveats like: all costs (mining to site remediation) being accounted for, technological and price changes over the past 3 years etc. etc. I think 'cost avoided' is a good measure; LCOE is so sensitive to assumptions and accounting trickery that it's basically meaningless (see: Lazard). Brookings seem to have made a fair effort (best I've seen thus far). It makes sense that combined-cycle gas turbines would pip nuclear power (in most scenarios examined) as they produce highly dispatchable electricity (i.e. their output levels can be changed very rapidly in response to fluctuations in demand for electricity). You can forecast and 'plan ahead' with nuclear, but the ramping times are currently too slow. So you either overbuild, rely on a limited amount of grid-level storage or throw a more responsive form of generation into the mix (i.e. gas turbines). Most of the cost of nuclear is in building the plant, after which the variable cost of generation is tiny (i.e. you just run it full-bore all the time, since oversupplying doesn't carry a high cost). Coal probably has a similar cost profile (maybe with slightly higher relative variable costs), although on a smaller scale and in a relatively under-regulated market: the WHO estimates pollution from coal-fired generation, not taking into account climate-change, kills ~1m people per year, whereas nuclear containment buildings must be built to withstand a plane crashing directly into it. But the really big take-away for me is that, if we want to mitigate climate change efficiently (which I'm starting to think is the same as saying 'if we want to mitigate it at all'), we really badly need to place a price on carbon emissions (and their equivalents). It sucks that both extreme sides of the debate don't seem to want this: vested interests will obviously oppose it, and 'extreme greens' seem to think it's some sort of immoral conspiracy designed to allow financial markets to make money off climate change (if we're talking 'cap and trade'). Those who don't like market-based mechanisms such as 'cap and trade' should consider the other side of the coin: it creates the opportunity for direct action on your part. You could buy emissions permits on the open market and then sit on them, permanently lowering the emissions cap. You can finally 'put your money where your mouth is'.