I'd also note that most of what they propose could also be done with regolith + binder instead of ice, albeit giving up light transmission in the process.
Arbitrary saline water is not "excellent for electrolysis", you'll end up destroying your electrolysis cells. Look at all the trouble they've had with the Elektron systems on the ISS, and that's under perfectly controlled conditions. Screwups are not acceptable on Mars. You can't just guess that things will be okay. For any potential ice resource, you need to have it very well quantified (and not just a tiny surface sample - and not just the water, but all of the solid matter it's mixed in with), so that engineers on Earth can create an accurate testbed for their proposed hardware to operate on.
Re, boring: have you ever seen the size of a TBM? Don't get me wrong, nuclear-powered Martian backhoes aren't exactly a dime a dozen, but that sure sounds cheaper than martian TBMs.
I have to agree with weird_w - the simplest means of radiation shielding is to use loose regolith (in regions where it's available in a deep enough layer... which aren't exactly rare, although they're not universal). Whether that's via bagging, binding with cement, binding with materials from Earth (a thermoplastic, epoxy, water, etc, optionally plus reinforcing fibers), or just simple loose dumping over a form, they're probably your easiest bet.
If you are advancing to the point where you're going to be doing in-situ water harvesting for electrolysis and drinking, however, something like the ice house is probably worth consideration. It does provide for much better human factors via transmission of (and fresnel concentration of) light, and allows for some limited agriculture (without requiring vast amounts of power generation for artificial lighting). It's easy to want to ignore human factors, but they're very real. Having people live their lives inside a cramped windowless can isn't exactly good for mental health or morale.
However, IRSU water is not a given. Pretending that harvesting of water is just "you go there and dig it up" is a vast oversimplification. To the point that even a lot of IRSU propellant proposals call for sending the hydrogen for the fuel from Earth even while they get the carbon and oxygen from the atmosphere. The atmosphere is a fairly constant, reliable, predictable fluid feedstock. The ground... isn't.
(And yes, technically you can get water vapor from the atmosphere, but the quantities it's available in are so tiny that most analysis writes off the concept due to the amount of air you'd have to move through the system per unit water recovered, and the mass of the system you'd need to do so)
... I do have to admit, this one seems the best thought out (it's been covered here on Slashdot before). The level of detail that they went into on their work was impressive, on every front. Some of the unique concepts are rather interesting, such as having the outer ice shell shaped as a fresnel lens, thus concentrating sunlight to higher levels in the interior. I also like the nested aspects of it - providing a large uninsulated (but pressurized) staging yard (quite useful, particularly once you start ramping up ISRU and need room for lots of industrial systems and feedstock/output stockpiles), and an insulated greenhouse/courtyard around the primary shelter (nice thought toward human factors, as well as small scale agriculture). Having the primary shelter be constructed on Earth and simply landed (with its interior space initially filled up with the hardware needed to make the outer radiation protection / pressure shell) hits all the right buttons as well. Having the "printer" slide along grooves in the shell it sprays out is also a lot more elegant of a design than many other potential alternatives.
Still, there's a massive amount of engineering and testing that would be needed to make such a thing. And a lot of in-situ demo missions as well for each aspect of the technology, especially the (no hardware design given) vaporization-based water recovery system, but up to and including a small scale inflate-and-print testbed.
Yet in practice all the countries which use renewables have the most expensive electricity to the consumer in the whole world. Fancy that.
Nonsense. The most expensive country for electricity in the world is Italy, which is only 5% wind and solar (I assume you're not counting hydro and geo), about the same as the US, which has cheap power by the standards of the first world. Or break it down by US state. The top states in the US by percentage of the state's wind power generation are in order: Iowa, South Dakota, Kansas, Oklahoma, and North Dakota. All states with dirt cheap power. Oklahoma in fact has the cheapest commercial electricity rates in the nation.
Indeed. A nice impact on, say, Mars would suit. With us having several decades leadtime to properly instrument it with every last piece of data gathering equipment and camera we can get our hands on, surface and orbital.
It's crazy to imagine an impactor so power that even the heat from its reentering impact debris baked away most non-sheltered life on the opposite side of the planet.
I just told you the numbers for modern generator efficiency.
And I just told you the average difference in the real world. I don't give a rat's arse what the maximum theoretical efficiency of the latest top of the line pricey combined cycle NG baseload plant gets verses the crappiest old NG peaking plant you want to cite; I looked up what the actual running averages they're getting in California, and the peakers were running at about 80% the efficiency of the baseload plants.
It's behind a paywall
Then use SciHub, or just take my word for what it says. I'll sum up: using only solar, wind, NG peaking, and a HVDC grid - current technology, no storage - power is produced with guaranteed reliability at rates similar to today's but with vastly lower carbon.
I said it before use Whr or don't bother.
Please tell me that you're not so stupid as to be unable to divide $/Wh by capacity factor.
.... the finalized source code / schematics / etc being highly delayed; that's not the regulators' fault. Beyond that: The concrete was poured wrong. The forgings were wrong and had to be recast. The containment structure was welded wrong and had to be redone. Pipes were welded wrong. One contractor was even running a protection racket for the Bulgarian mafia. A lot of the problems wouldn't have been a big issue for a conventional power plant, but even minor errors are not acceptable when you're dealing with the extreme level of toxicity and harsh conditions found within a nuclear core.
Oh, I'm sorry that after spending hundreds of billions of federal dollars on nuclear research over the past century that every last project under the sun doesn't get multi-billion-dollar prototype plants built at government expense. That poor mistreated industry.
They are not "structure designed around 1960's technology, essentially preventing all improvements except small refinements of the status quo", they're on generation 3, working on generation 4 reactors. It's purely financial factors that are keeping nuclear from being jumped at by investors. Whether you like it or not, nuclear power is damned expensive. You have a core of amazingly toxic materials, operating at high temperatures and pressures, creating literally every corrosive element known to man in every isotopic form, while heavily bombarding and weakening everything around it with an intense neutron flux, in a form that will stay hot for a good period of time even when you shut down all fission due to daughter products. Nuclear is very difficult to do right, and this equals expense and liability. Lots and lots of expense plus lots and lots of liabiltiy (even though the government lets them cheat on the worst of the latter).
Again, look at actual cost overruns in actual plants. They're not due to some sort of regulatory bullshit, they're due to people trying and failing to build very large, complicated, precision-demanding things correctly and on time.
You typically have to pay for the wind/solar, the backup peaking power plant generated power, and a subsidy to the peaking power plant (otherwise the cash from the generated power won't cover the costs of running the power plant) plus a wind/solar subsidy. Not to mention a more complex energy grid.
Forget a subsidy for the peaking plant - you can buy an entire peaking plant yourself covering the entire generation capacity of the renewable resource and it still comes out far cheaper. $1,50/W + $1/W = $2,50/W (plus the cost of the NG burned when peaking) for a total of 100% capacity factor. Versus say $9/W (plus the cost of operations and decommissioning) for nuclear with a 90% capacity factor.
In practice, of course, you don't do that. You diversity wind + solar across broad geographic areas with high power transmission, utilizing storage where appropriate, uprating hydro where appropriate, etc - and the amount of fossil peaking that you need to ensure a high statistically guaranteed uptime ends up as just a small fraction of the renewable nameplate.
Except that pumped hydro can't be built anywhere affordably - only in specific locations. A NG peaker is a general purpose solution. If we're going to consider "location limited" options, then even better than pumped hydro is simply uprating existing large hydro turbine houses. Very cheap versus how much peaking capacity it gives you.
A natural gas fired power plant which is run in a peaking scenario, with constant spooling up and down, won't be able to take advantage of combined cycle operation so it could be like 34% efficient. While if it was run at a constant level, i.e. in baseload mode, using combined cycle the same natural gas power plant would be like 50-60% efficient. So the idea that you can 'save natural gas' by using variable load wind, is uh, disingenuous to say the least.
While I don't care to look up the numbers yet again, the last time I was in such a debate I did; in the real world, in Californa, NG peakers get about 80% the average efficiency of a NG baseload plant. So if the peaker is run less than 80% of the time of a baseload plant, it saves gas. If it's only running 80% as much as a baseload plant it's not a peaker, it's a load follower at the worst; peakers run at very low capacity factors. So your comment pretending that you don't save gas with renewables and NG peaking is - to quote you - "disingenuous to say the least".
Or don't take my word for it; read the Nature study, which covers the CO2 emissions on a renewables + NG grid.
Only in an ideal scenario in a country which has a lot of hydropower capacity, with pumped storage, and a good wind or solar resource.
You know, the study was linked right in the post you're replying to. It wouldn't kill you to read it before asserting that a study in the world's most prestigious scientific journal is wrong.
No, it's about as expensive as coal, which is the cheapest form of power generation, unless the coal power plant is very close to the coal mines, like in the same state, in which case the coal will be cheaper.
Pure nonsense. Coal plants are in the rough ballpark of $1/W, give or take depending on the design. Nuclear plants in the western world average nearly $10/W nowadays.
Correct. In fact, given how frequently it's struck versus Earth, combined with how small impactors don't oxidize away, there are actually some proposals to mine iron on Mars simply by collecting and melting down meteorites.;) The MERs and Curiosity have found a good number of them just in their slow crawls across the surface.
Honestly, I don't expect anyone would ever approve of such a thing - pristine asteroids are just too valuable.
Cape York is estimated as the largest ever to hit the atmosphere with survivable fragments, although Hoba is the largest intact meteorite. The parent of Cape York is estimated to have been 200 tonnes. Assuming the average asteroid impact velocity of 17 km/s (it was probably less, given that pieces survived), that's an energy of 2,89e13 joules, or 7 kilotonnes. You can technically make a nuclear bomb that weak (and thus call it "nuclear force"), but you'd generally call that a fizzle.
Chixulub, by contrast, was formed by an impact with about 100 *teratonnes* of energy. Over *10* orders of magnitude more.
You're not getting that it's physically impossible for a large object to survive. "Releasing enough energy to cause mass extinctions across the entire planet" and "remaining with large pieces intact" are mutually exclusive. The "immensity of forces" is precisely the problem. It's like expecting pieces of the casing to survive the detonation of an atomic bomb. Only many orders of magnitude less likely.
As for cracking the planet into separate plates, however, that's not that far fetched; there is a legitimate (although controversial) scientific hypothesis that such an impact weakened the crust there and helped allow for Antarctica to break off. And collisions are a leading, relatively non-controversial theory to explain axial tilts - although primarily collisions during formation and potentially the late heavy bombardment.
I live in Iceland. Apart from the increasing habitability to pests, and the aesthetic loss of our retreating glaciers, it's pretty nice up here. Tack on five degrees and we're Seattle.;)
My point however was that you give the wrong impression. You painted a picture of a steady cubic curve, when in reality it's a cubic curve only at low to moderate speeds, followed by a long plateau at moderate to high speeds, followed by a sudden dropoff to zero. You made it sound like turbines would yield a tiny capacity factor, when in reality they average over 30% of nameplate. I wanted to make sure people had the right impression.:)
"Endured" getting their catastrophic liability insurance provided to them for free by the federal government? What private company would insure against cleanups that can run into the hundreds of billions of dollars? Let alone with affordable premiums?
Nuclear has always had far more support on K Street than Wall Street. The cost overruns that have happened to the recent generation of nuclear plants have been overwhelmingly fabrication related, with the next highest portion of overruns being to address since-discovered safety issues.
It means that cost per kW isn't as good as nuclear.
Nonsense. Capital (no subsidy) on wind and solar in the US is dirt cheap. The US average for solar, for example, is now under $1,50/W and still falling at a pretty good clip. Average capacity factors are in the ballpark of 30% for both. Tack on a $1/W peaking plant and you have baseload. Or more efficiently, hook up both wind and solar to HVDC with geographic distribution and you only need peaking for a relatively small fraction of your renewable nameplate, with the grid coming in at about 0.3 cents per kWh, aka far less than the cost of the generation hardware you don't have to build.
New nuclear plants in the western world are averaging nearly $10W/W (just for construction, not counting operations or decommissioning - or freebies like government-provided catastrophic liability coverage), with an industry standard ~90% capacity factor. The cheapest western plant finished in modern times is Watts Bar (about $5,50/W), but most of that was done in the 70s and 80s when nuclear construction costs were cheaper (not just due to inflation, but also due to the industry's unfortunate "negative learning curve"). The modern completion of Unit 2 ended up nearly double its projected completion cost.
If you're willing to throw caution to the wind you can build cheaper. China might ultimately have some come in at around $4/W, maybe even cheaper (although China can probably do wind and solar at around $1/W - India's latest solar plant came in at around that, and the location wasn't optimal). But I wouldn't trust the low-end of Chinese nuclear power plants any further than I could throw them.
You're confused; you seem to think that wind turbines are designed to bear and generate from the maximum force winds that they experience. They don't. At high wind speeds they're feathered and/or braked. The nameplate capacity is met at about 25mph for a typical turbine. At very high speeds (for example, over 55mph) they outright shut off and don't generate anything, but between that range they generate at their nameplate capacity. At under the base speed (for example, below 25mph) they produce less - but not according to a cubic curve, but slightly steeper than that, as there's a base-level constant drag, which keeps them from turning at very low windspeeds. It's also important to realize that wind turbines experience wind at altitude, not surface winds; they're higher and steadier.
Average capacity factor for wind in the US is over 30% every year. You really do get a large chunk of the nameplate in terms of actual generation.
Tell it to Nature that it doesn't work. In the above paper (you can download it on SciHub if you don't have access) they model the creation of an optimal HVDC grid and cross-country solar, wind, and NG peaking plants (as well as one scenario with coal) and end up with reliable, low carbon power at lower costs than current grid rates - with no use of storage and no assumption of improved technologies.
Wind and solar tend to run counter to each other. Wind is strongest at night; solar only generates during the day. Wind is associated with low pressure zones; clear skies with high pressure zones. And while one system is moving off the US east coast, another one (or more) is moving in from the west. A HVDC grid also timeshifts loads - aka, the sun is still shining out west after sundown in the east, and so forth. It also spreads out peaking capabilities across the country. The grid costs about 0.3 cents per kWh to build/maintain but saves about 1.1 cents per kWh in generation hardware costs.
By adding in storage or allowing for tech improvements, the figures only get better. Indeed, the figures they use for solar pricing are already pessimistic. And they make no use of uprating existing hydro for storage (very cheap). Or no pumped hydro storage, which is already cheap in certain areas. No use of battery storage, which although marginal currently due to cost is expected to become vastly cheaper over the next decade. Etc.
New nuclear plants, with their high price tags, have no place. It's nearly an order of magnitude more expensive than renewables per kWh generated, and it sucks for use as peaking (even if you use a plant which can ramp quickly - most can't - you fundamentally (by the nature of peaking) would cut the capacity factor severalfold, which directly corresponds to a severalfold increase in construction / operating / decommissioning costs per kWh generated.
1) You don't need baseload with intermittent renewables, you need peaking or storage. Nuclear makes for terrible peaking. It's literally the worst non-intermittent option available. Natural gas is the best.
2) With peaking and/or storage and/or a HVDC grid, intermittent renewables can make up the lion's share of the grid. The exact level of penetration depends on the details of the options chosen, but can in some cases even approach 100%. With current tech and current prices, penetrations of ~70-80% are reasonable - and very low carbon.
3) Renewables + peaking is already cheap. Renewables + storage is looking to be heading in that direction.
4) The problem with nuclear is the cost, not the waste. Nuclear plants are absurdly expensive per watt, even ignoring that they get their catastrophic liability coverage provided for free by the federal government.
5) Alpha radiation in free space is harmless, but alpha emitters are exceedingly dangerous. Ingested/inhaled alpha emitters are an order of magnitude more destructive to tissue than neutron, beta and gamma emitters. The fact that they can be inhaled or ingested if leaked into the environment (airborne dust, soil contamination, groundwater contamination) is precisely the reason that you have to contain them. Aka like the reason they've had to spend a fortune cleaning up Hanford Site (estimated ~115B remaining). Without proper containment and disposal procedures, every site would be a Hanford Site.
Cost of nuclear is out of control everywhere. Ever checked out, say, the Olkiluoto Nuclear Plant? In what way is the many-billion-euro overrun / decade-late reactor due to "red tape"? The answer is "virtually nothing". And it's the same story everywhere.
Beyond that, the last thing you want when dealing with nuclear is underregulation. Nuclear disasters aren't particularly deadly, but they're massively expensive. They're disasters in slow motion, like an advancing lava flow - you can run from it, but you can't ignore it. You think that billions of dollars to build a plant is expensive, check out the cost of cleaning up Fukushima. And that's hardly the worst case scenario. Imagine the cost of a Fukushima-scale disaster at, say, Indian Point. And please, no stupid "but there couldn't be a tsunami there" remarks, as if that is the only way in which a disaster can occur; unless you were out there saying "nuclear is safe except for in major tsunamis" before Fukushima, you have no grounds to consider yourself prophetic in what may or may not cause major disasters. The whole reason disasters happen at all is because people thought they couldn't - otherwise they would have taken the necessary steps to prevent them. Regulation of nuclear is rational. And IMHO in practice generally far too favorable to operators rather than the public.
Nuclear would be great if it wasn't absurdly expensive. It is. And it's not some "red tape" / NIMBY boogieman that makes it so. It's suffered from a negative learning curve over the years - the more we've learned, the more expensive it's gotten, not less. It's like buying a house and then discovering that there's termites in the walls and the foundation is cracked - only on the scale of an entire industry. The industry's solution was to try to launch a new generation of reactors to try to get the costs down. This much hyped "nuclear renaissance" sounded great but rapidly descended into an even worse financial quagmire.
To make it worse, nuclear is about the worst thing you can pair with the surge in now-super-cheap but still intermittent renewables. Nuclear's poor economics are already taking into account its high capacity factor. If you try to use it as peaking (which most plants aren't even capable of), you inherently greatly lower its capacity factor - and by doing that, you proportionally worsen its economics. If you go from a 90% capacity factor to a 30% one, it's as if your plant cost 3x as much to build. For peaking you use plants that are cheap to build but expensive to operate, not the other way around.
Most existing nuclear power plants will continue to operate for a good while. But there's little prospect of new nuclear plants filling in a relevant portion of new generation demand. Even if the economic picture radically changes, there won't be a sudden reversal, because nuclear plants take so long to build. It's yet another flaw - you have to forecast the energy market where you are (not just the total demand, but the type of supply needed) a decade or more in advance. And if you're talking plants that wouldn't come online until 2-3 decades from now (a decade or two for trends to shift, plus another decade for construction), you could well be competing against fusion, so long as ITER and DEMO aren't cut (the trends in that seem to be moving in the right direction, with improvements in magnets meaning you can get the same level of confinement from a smaller, cheaper plant). When you're talking so far into the future, who knows? Who in the 80/early 90s would have predicted wind and solar would cost only $1/W today?
For the foreseeable future, new demand will be predominantly filled by solar, wind and natural gas.
Yep - just like the Sudbury deposit. Probably a great spot to mine, if it wasn't buried beneath ice and in an area where mining is illegal. Large impact crater floors tend to stay molten for so long that they stratify, so you can find portions of the deposit rich in different minerals, such as nickel, copper, and precious metals.
It's one thing that Mars has over Earth in terms of mineral deposits. While Mars lacks relevant recent fluvial mineral concentration mechanisms, as well as those aided by life, by oxidation, and a number of other processes, it's also struck more often by large asteroids, and thus probably has more common stratified impact deposits.
Indeed. Large (dangerous) asteroids do not survive their descent. Ever. The largest single meteorite ever found is only 60 tonnes (Hoba meteorite), and it took exceptional circumstances for it to survive (an extremely shallow entry trajectory). If an impact is excavating a large (or even small) crater, it's turning to gas and/or plasma in the process.
Slashdot Mirror
I'd also note that most of what they propose could also be done with regolith + binder instead of ice, albeit giving up light transmission in the process.
Arbitrary saline water is not "excellent for electrolysis", you'll end up destroying your electrolysis cells. Look at all the trouble they've had with the Elektron systems on the ISS, and that's under perfectly controlled conditions. Screwups are not acceptable on Mars. You can't just guess that things will be okay. For any potential ice resource, you need to have it very well quantified (and not just a tiny surface sample - and not just the water, but all of the solid matter it's mixed in with), so that engineers on Earth can create an accurate testbed for their proposed hardware to operate on.
Re, boring: have you ever seen the size of a TBM? Don't get me wrong, nuclear-powered Martian backhoes aren't exactly a dime a dozen, but that sure sounds cheaper than martian TBMs.
I have to agree with weird_w - the simplest means of radiation shielding is to use loose regolith (in regions where it's available in a deep enough layer... which aren't exactly rare, although they're not universal). Whether that's via bagging, binding with cement, binding with materials from Earth (a thermoplastic, epoxy, water, etc, optionally plus reinforcing fibers), or just simple loose dumping over a form, they're probably your easiest bet.
If you are advancing to the point where you're going to be doing in-situ water harvesting for electrolysis and drinking, however, something like the ice house is probably worth consideration. It does provide for much better human factors via transmission of (and fresnel concentration of) light, and allows for some limited agriculture (without requiring vast amounts of power generation for artificial lighting). It's easy to want to ignore human factors, but they're very real. Having people live their lives inside a cramped windowless can isn't exactly good for mental health or morale.
However, IRSU water is not a given. Pretending that harvesting of water is just "you go there and dig it up" is a vast oversimplification. To the point that even a lot of IRSU propellant proposals call for sending the hydrogen for the fuel from Earth even while they get the carbon and oxygen from the atmosphere. The atmosphere is a fairly constant, reliable, predictable fluid feedstock. The ground... isn't.
(And yes, technically you can get water vapor from the atmosphere, but the quantities it's available in are so tiny that most analysis writes off the concept due to the amount of air you'd have to move through the system per unit water recovered, and the mass of the system you'd need to do so)
... I do have to admit, this one seems the best thought out (it's been covered here on Slashdot before). The level of detail that they went into on their work was impressive, on every front. Some of the unique concepts are rather interesting, such as having the outer ice shell shaped as a fresnel lens, thus concentrating sunlight to higher levels in the interior. I also like the nested aspects of it - providing a large uninsulated (but pressurized) staging yard (quite useful, particularly once you start ramping up ISRU and need room for lots of industrial systems and feedstock/output stockpiles), and an insulated greenhouse/courtyard around the primary shelter (nice thought toward human factors, as well as small scale agriculture). Having the primary shelter be constructed on Earth and simply landed (with its interior space initially filled up with the hardware needed to make the outer radiation protection / pressure shell) hits all the right buttons as well. Having the "printer" slide along grooves in the shell it sprays out is also a lot more elegant of a design than many other potential alternatives.
Still, there's a massive amount of engineering and testing that would be needed to make such a thing. And a lot of in-situ demo missions as well for each aspect of the technology, especially the (no hardware design given) vaporization-based water recovery system, but up to and including a small scale inflate-and-print testbed.
Nonsense. The most expensive country for electricity in the world is Italy, which is only 5% wind and solar (I assume you're not counting hydro and geo), about the same as the US, which has cheap power by the standards of the first world. Or break it down by US state. The top states in the US by percentage of the state's wind power generation are in order: Iowa, South Dakota, Kansas, Oklahoma, and North Dakota. All states with dirt cheap power. Oklahoma in fact has the cheapest commercial electricity rates in the nation.
Indeed. A nice impact on, say, Mars would suit. With us having several decades leadtime to properly instrument it with every last piece of data gathering equipment and camera we can get our hands on, surface and orbital.
It's crazy to imagine an impactor so power that even the heat from its reentering impact debris baked away most non-sheltered life on the opposite side of the planet.
See the reply below (when you break your responses into many comments like you're doing, you cause replies to get missed)
And I just told you the average difference in the real world. I don't give a rat's arse what the maximum theoretical efficiency of the latest top of the line pricey combined cycle NG baseload plant gets verses the crappiest old NG peaking plant you want to cite; I looked up what the actual running averages they're getting in California, and the peakers were running at about 80% the efficiency of the baseload plants.
Then use SciHub, or just take my word for what it says. I'll sum up: using only solar, wind, NG peaking, and a HVDC grid - current technology, no storage - power is produced with guaranteed reliability at rates similar to today's but with vastly lower carbon.
Please tell me that you're not so stupid as to be unable to divide $/Wh by capacity factor.
.... the finalized source code / schematics / etc being highly delayed; that's not the regulators' fault. Beyond that: The concrete was poured wrong. The forgings were wrong and had to be recast. The containment structure was welded wrong and had to be redone. Pipes were welded wrong. One contractor was even running a protection racket for the Bulgarian mafia. A lot of the problems wouldn't have been a big issue for a conventional power plant, but even minor errors are not acceptable when you're dealing with the extreme level of toxicity and harsh conditions found within a nuclear core.
Oh, I'm sorry that after spending hundreds of billions of federal dollars on nuclear research over the past century that every last project under the sun doesn't get multi-billion-dollar prototype plants built at government expense. That poor mistreated industry.
They are not "structure designed around 1960's technology, essentially preventing all improvements except small refinements of the status quo", they're on generation 3, working on generation 4 reactors. It's purely financial factors that are keeping nuclear from being jumped at by investors. Whether you like it or not, nuclear power is damned expensive. You have a core of amazingly toxic materials, operating at high temperatures and pressures, creating literally every corrosive element known to man in every isotopic form, while heavily bombarding and weakening everything around it with an intense neutron flux, in a form that will stay hot for a good period of time even when you shut down all fission due to daughter products. Nuclear is very difficult to do right, and this equals expense and liability. Lots and lots of expense plus lots and lots of liabiltiy (even though the government lets them cheat on the worst of the latter).
Again, look at actual cost overruns in actual plants. They're not due to some sort of regulatory bullshit, they're due to people trying and failing to build very large, complicated, precision-demanding things correctly and on time.
Forget a subsidy for the peaking plant - you can buy an entire peaking plant yourself covering the entire generation capacity of the renewable resource and it still comes out far cheaper. $1,50/W + $1/W = $2,50/W (plus the cost of the NG burned when peaking) for a total of 100% capacity factor. Versus say $9/W (plus the cost of operations and decommissioning) for nuclear with a 90% capacity factor.
In practice, of course, you don't do that. You diversity wind + solar across broad geographic areas with high power transmission, utilizing storage where appropriate, uprating hydro where appropriate, etc - and the amount of fossil peaking that you need to ensure a high statistically guaranteed uptime ends up as just a small fraction of the renewable nameplate.
Are you incapable of dividing a cost by a capacity factor?
Except that pumped hydro can't be built anywhere affordably - only in specific locations. A NG peaker is a general purpose solution. If we're going to consider "location limited" options, then even better than pumped hydro is simply uprating existing large hydro turbine houses. Very cheap versus how much peaking capacity it gives you.
While I don't care to look up the numbers yet again, the last time I was in such a debate I did; in the real world, in Californa, NG peakers get about 80% the average efficiency of a NG baseload plant. So if the peaker is run less than 80% of the time of a baseload plant, it saves gas. If it's only running 80% as much as a baseload plant it's not a peaker, it's a load follower at the worst; peakers run at very low capacity factors. So your comment pretending that you don't save gas with renewables and NG peaking is - to quote you - "disingenuous to say the least".
Or don't take my word for it; read the Nature study, which covers the CO2 emissions on a renewables + NG grid.
You know, the study was linked right in the post you're replying to. It wouldn't kill you to read it before asserting that a study in the world's most prestigious scientific journal is wrong.
Pure nonsense. Coal plants are in the rough ballpark of $1/W, give or take depending on the design. Nuclear plants in the western world average nearly $10/W nowadays.
Correct. In fact, given how frequently it's struck versus Earth, combined with how small impactors don't oxidize away, there are actually some proposals to mine iron on Mars simply by collecting and melting down meteorites. ;) The MERs and Curiosity have found a good number of them just in their slow crawls across the surface.
Honestly, I don't expect anyone would ever approve of such a thing - pristine asteroids are just too valuable.
Cape York is estimated as the largest ever to hit the atmosphere with survivable fragments, although Hoba is the largest intact meteorite. The parent of Cape York is estimated to have been 200 tonnes. Assuming the average asteroid impact velocity of 17 km/s (it was probably less, given that pieces survived), that's an energy of 2,89e13 joules, or 7 kilotonnes. You can technically make a nuclear bomb that weak (and thus call it "nuclear force"), but you'd generally call that a fizzle.
Chixulub, by contrast, was formed by an impact with about 100 *teratonnes* of energy. Over *10* orders of magnitude more.
You're not getting that it's physically impossible for a large object to survive. "Releasing enough energy to cause mass extinctions across the entire planet" and "remaining with large pieces intact" are mutually exclusive. The "immensity of forces" is precisely the problem. It's like expecting pieces of the casing to survive the detonation of an atomic bomb. Only many orders of magnitude less likely.
As for cracking the planet into separate plates, however, that's not that far fetched; there is a legitimate (although controversial) scientific hypothesis that such an impact weakened the crust there and helped allow for Antarctica to break off. And collisions are a leading, relatively non-controversial theory to explain axial tilts - although primarily collisions during formation and potentially the late heavy bombardment.
I live in Iceland. Apart from the increasing habitability to pests, and the aesthetic loss of our retreating glaciers, it's pretty nice up here. Tack on five degrees and we're Seattle. ;)
Thanks :)
My point however was that you give the wrong impression. You painted a picture of a steady cubic curve, when in reality it's a cubic curve only at low to moderate speeds, followed by a long plateau at moderate to high speeds, followed by a sudden dropoff to zero. You made it sound like turbines would yield a tiny capacity factor, when in reality they average over 30% of nameplate. I wanted to make sure people had the right impression. :)
"Endured" getting their catastrophic liability insurance provided to them for free by the federal government? What private company would insure against cleanups that can run into the hundreds of billions of dollars? Let alone with affordable premiums?
Nuclear has always had far more support on K Street than Wall Street. The cost overruns that have happened to the recent generation of nuclear plants have been overwhelmingly fabrication related, with the next highest portion of overruns being to address since-discovered safety issues.
Nonsense. Capital (no subsidy) on wind and solar in the US is dirt cheap. The US average for solar, for example, is now under $1,50/W and still falling at a pretty good clip. Average capacity factors are in the ballpark of 30% for both. Tack on a $1/W peaking plant and you have baseload. Or more efficiently, hook up both wind and solar to HVDC with geographic distribution and you only need peaking for a relatively small fraction of your renewable nameplate, with the grid coming in at about 0.3 cents per kWh, aka far less than the cost of the generation hardware you don't have to build.
New nuclear plants in the western world are averaging nearly $10W/W (just for construction, not counting operations or decommissioning - or freebies like government-provided catastrophic liability coverage), with an industry standard ~90% capacity factor. The cheapest western plant finished in modern times is Watts Bar (about $5,50/W), but most of that was done in the 70s and 80s when nuclear construction costs were cheaper (not just due to inflation, but also due to the industry's unfortunate "negative learning curve"). The modern completion of Unit 2 ended up nearly double its projected completion cost.
If you're willing to throw caution to the wind you can build cheaper. China might ultimately have some come in at around $4/W, maybe even cheaper (although China can probably do wind and solar at around $1/W - India's latest solar plant came in at around that, and the location wasn't optimal). But I wouldn't trust the low-end of Chinese nuclear power plants any further than I could throw them.
You're confused; you seem to think that wind turbines are designed to bear and generate from the maximum force winds that they experience. They don't. At high wind speeds they're feathered and/or braked. The nameplate capacity is met at about 25mph for a typical turbine. At very high speeds (for example, over 55mph) they outright shut off and don't generate anything, but between that range they generate at their nameplate capacity. At under the base speed (for example, below 25mph) they produce less - but not according to a cubic curve, but slightly steeper than that, as there's a base-level constant drag, which keeps them from turning at very low windspeeds. It's also important to realize that wind turbines experience wind at altitude, not surface winds; they're higher and steadier.
Average capacity factor for wind in the US is over 30% every year. You really do get a large chunk of the nameplate in terms of actual generation.
Tell it to Nature that it doesn't work. In the above paper (you can download it on SciHub if you don't have access) they model the creation of an optimal HVDC grid and cross-country solar, wind, and NG peaking plants (as well as one scenario with coal) and end up with reliable, low carbon power at lower costs than current grid rates - with no use of storage and no assumption of improved technologies.
Wind and solar tend to run counter to each other. Wind is strongest at night; solar only generates during the day. Wind is associated with low pressure zones; clear skies with high pressure zones. And while one system is moving off the US east coast, another one (or more) is moving in from the west. A HVDC grid also timeshifts loads - aka, the sun is still shining out west after sundown in the east, and so forth. It also spreads out peaking capabilities across the country. The grid costs about 0.3 cents per kWh to build/maintain but saves about 1.1 cents per kWh in generation hardware costs.
By adding in storage or allowing for tech improvements, the figures only get better. Indeed, the figures they use for solar pricing are already pessimistic. And they make no use of uprating existing hydro for storage (very cheap). Or no pumped hydro storage, which is already cheap in certain areas. No use of battery storage, which although marginal currently due to cost is expected to become vastly cheaper over the next decade. Etc.
New nuclear plants, with their high price tags, have no place. It's nearly an order of magnitude more expensive than renewables per kWh generated, and it sucks for use as peaking (even if you use a plant which can ramp quickly - most can't - you fundamentally (by the nature of peaking) would cut the capacity factor severalfold, which directly corresponds to a severalfold increase in construction / operating / decommissioning costs per kWh generated.
This is incorrect.
1) You don't need baseload with intermittent renewables, you need peaking or storage. Nuclear makes for terrible peaking. It's literally the worst non-intermittent option available. Natural gas is the best.
2) With peaking and/or storage and/or a HVDC grid, intermittent renewables can make up the lion's share of the grid. The exact level of penetration depends on the details of the options chosen, but can in some cases even approach 100%. With current tech and current prices, penetrations of ~70-80% are reasonable - and very low carbon.
3) Renewables + peaking is already cheap. Renewables + storage is looking to be heading in that direction.
4) The problem with nuclear is the cost, not the waste. Nuclear plants are absurdly expensive per watt, even ignoring that they get their catastrophic liability coverage provided for free by the federal government.
5) Alpha radiation in free space is harmless, but alpha emitters are exceedingly dangerous. Ingested/inhaled alpha emitters are an order of magnitude more destructive to tissue than neutron, beta and gamma emitters. The fact that they can be inhaled or ingested if leaked into the environment (airborne dust, soil contamination, groundwater contamination) is precisely the reason that you have to contain them. Aka like the reason they've had to spend a fortune cleaning up Hanford Site (estimated ~115B remaining). Without proper containment and disposal procedures, every site would be a Hanford Site.
Cost of nuclear is out of control everywhere. Ever checked out, say, the Olkiluoto Nuclear Plant? In what way is the many-billion-euro overrun / decade-late reactor due to "red tape"? The answer is "virtually nothing". And it's the same story everywhere.
Beyond that, the last thing you want when dealing with nuclear is underregulation. Nuclear disasters aren't particularly deadly, but they're massively expensive. They're disasters in slow motion, like an advancing lava flow - you can run from it, but you can't ignore it. You think that billions of dollars to build a plant is expensive, check out the cost of cleaning up Fukushima. And that's hardly the worst case scenario. Imagine the cost of a Fukushima-scale disaster at, say, Indian Point. And please, no stupid "but there couldn't be a tsunami there" remarks, as if that is the only way in which a disaster can occur; unless you were out there saying "nuclear is safe except for in major tsunamis" before Fukushima, you have no grounds to consider yourself prophetic in what may or may not cause major disasters. The whole reason disasters happen at all is because people thought they couldn't - otherwise they would have taken the necessary steps to prevent them. Regulation of nuclear is rational. And IMHO in practice generally far too favorable to operators rather than the public.
Nuclear would be great if it wasn't absurdly expensive. It is. And it's not some "red tape" / NIMBY boogieman that makes it so. It's suffered from a negative learning curve over the years - the more we've learned, the more expensive it's gotten, not less. It's like buying a house and then discovering that there's termites in the walls and the foundation is cracked - only on the scale of an entire industry. The industry's solution was to try to launch a new generation of reactors to try to get the costs down. This much hyped "nuclear renaissance" sounded great but rapidly descended into an even worse financial quagmire.
To make it worse, nuclear is about the worst thing you can pair with the surge in now-super-cheap but still intermittent renewables. Nuclear's poor economics are already taking into account its high capacity factor. If you try to use it as peaking (which most plants aren't even capable of), you inherently greatly lower its capacity factor - and by doing that, you proportionally worsen its economics. If you go from a 90% capacity factor to a 30% one, it's as if your plant cost 3x as much to build. For peaking you use plants that are cheap to build but expensive to operate, not the other way around.
Most existing nuclear power plants will continue to operate for a good while. But there's little prospect of new nuclear plants filling in a relevant portion of new generation demand. Even if the economic picture radically changes, there won't be a sudden reversal, because nuclear plants take so long to build. It's yet another flaw - you have to forecast the energy market where you are (not just the total demand, but the type of supply needed) a decade or more in advance. And if you're talking plants that wouldn't come online until 2-3 decades from now (a decade or two for trends to shift, plus another decade for construction), you could well be competing against fusion, so long as ITER and DEMO aren't cut (the trends in that seem to be moving in the right direction, with improvements in magnets meaning you can get the same level of confinement from a smaller, cheaper plant). When you're talking so far into the future, who knows? Who in the 80/early 90s would have predicted wind and solar would cost only $1/W today?
For the foreseeable future, new demand will be predominantly filled by solar, wind and natural gas.
Yep - just like the Sudbury deposit. Probably a great spot to mine, if it wasn't buried beneath ice and in an area where mining is illegal. Large impact crater floors tend to stay molten for so long that they stratify, so you can find portions of the deposit rich in different minerals, such as nickel, copper, and precious metals.
It's one thing that Mars has over Earth in terms of mineral deposits. While Mars lacks relevant recent fluvial mineral concentration mechanisms, as well as those aided by life, by oxidation, and a number of other processes, it's also struck more often by large asteroids, and thus probably has more common stratified impact deposits.
Indeed. Large (dangerous) asteroids do not survive their descent. Ever. The largest single meteorite ever found is only 60 tonnes (Hoba meteorite), and it took exceptional circumstances for it to survive (an extremely shallow entry trajectory). If an impact is excavating a large (or even small) crater, it's turning to gas and/or plasma in the process.