[actinium226]

Well hello Mr. Malthus!

https://en.wikipedia.org/wiki/Thomas_Robert_Malthus

Back to seriousness - the author claims there are limits to nuclear, wind, and solar, but does not state what those limits are. The limit on nuclear is not clear to me - France gets something like 75% of its energy from nuclear. It seems the main limit on nuclear has been public sentiment, which must be weighed with public sentiment on climate change.

[1970-01-01]

Back in the 50s, nuclear was being treated as the magic that would solve all energy problems. There were ideas for making fission cars, trains, planes, etc. But since then, civilization has only benefited from nuclear reactors via power plants, naval ships, RTGs, and depending on how you see it, bombs. The magic of nuclear fission never provided the complete revolution it was setup to do. Just as we weren't responsible enough to plan for fission's partial failure back then, we are not able to envision the same partial failure with solar and wind. Fusion is the only thing I can think of that still has a chance to be the magic it is hyped to be. But that is only because its another 50 years away. And that is just too late to stop climate change in its accelerating state.

https://en.wikipedia.org/wiki/Ford_Nucleon

https://www.ans.org/news/article-109/army-offroad-nuclear-tr...

https://en.wikipedia.org/wiki/Convair_NB-36H

[aeternum]

The problem with nuclear is the slow iteration speed and extreme cost of failure.

Those are not good properties if you want to push technology forward quickly. The great thing about solar and wind is that we can iterate very quickly and catastrophic failure costs are nearly non-existent.

Nuclear still has great potential but the costs are just too high (and maybe they should be).

[actinium226]

Why do you need high iteration speed? Nuclear is pretty safe as it is. Chernobyl was using reactors that were obsolete in 1986. Fukushima was pretty fucked up in having the generators below sea level, but the result of that was 6 people got cancer.

[aeternum]

Fukushima may not have caused much loss of life, but you also need to consider the economic damage.

Natural disasters, terrorism, and unanticipated design flaws (generators below sea level) are important considerations. Those add costs that aren't always accounted for.

[actinium226]

Yes, exactly, consider the economic damage of NOT building nuclear plants and allowing the climate to continue to change! Surely that's a higher economic damage than that caused by Fukushima+Chernobyl.

[lm28469]

Anyone who argue against nuclear while failing to acknowledge the issues of solar and wind, which are 10 times more problematic, can't be truthful.

Unless we enter a _major_ era of degrowth there is no way we'll get out of fossil fuels without nuclear.

[aeternum]

What are the 10x more problematic issues with solar and wind?

[kinjba11]

Briefly: solar and wind energy requires sun or wind and tons of space relative to fossil/nuclear. Not every place has those.

"A reality check on renewables" https://www.youtube.com/watch?v=E0W1ZZYIV8o

[alexgmcm]

I think fusion is a lot closer than 50 years away.

We should know without a doubt whether it is workable by 2050, perhaps before then.

I strongly recommend the book "The Future of Fusion Energy"[1] as a good summary of the current state of the field. It's written by two fusion researchers and is legit (I have a Master's in Physics myself and almost did a Plasma Physics PhD so I am somewhat familiar with the field).

The book isn't dry either, it's honestly one of the best books I've read in years.

[1] https://www.amazon.com/Future-Fusion-Energy-Jason-Parisi/dp/...

[m463]

"It is not too much to expect that our children will enjoy in their homes electrical energy too cheap to meter, will know of great periodic regional famines in the world only as matters of history, will travel effortlessly over the seas and under them and through the air with a minimum of danger and at great speeds, and will experience a lifespan far longer than ours, as disease yields and man comes to understand what causes him to age." - Lewis Strauss, chairman of the United States Atomic Energy Commission 1954

https://en.wikipedia.org/wiki/Too_cheap_to_meter

[adrianN]

Fusion is not magic at all. It will most likely be a fairly expensive source of power. You might not need to spend a lot on fuel, but building large, technically very advanced installations is expensive.

[_acco]

Interesting, right?

If we believe the Titanic is sinking, why are we being picky about the lifeboat?

Anything beyond averting disaster is an over-optimization.

[ncmncm]

We just closed Diablo Canyon and Indian Point nukes, not a moment too soon. Money wasted propping up those ramshackle contraptions can now go into solar and wind construction, at a better marginal return on new construction than we got just operating them. But they will unfortunately drain another billion dollars just being decommissioned.

[byroot]

> France gets something like 75% of its energy from nuclear.

That's not quite correct. We do get 75% of our electricity from nuclear, but electricity is only ~25% of the energy consumed in France.

So in practice about 18% of our energy comes from nuclear, and way over 75% of it from fossil fuels.

[toast0]

Finding appropriate sites for generators would probably be the engineering limit for nuclear. I don't know if nuclear is easy or hard to get started in a black start condition; if they're hard to start, there's probably a max % of the grid you'd want to be nuclear.

But the will of the people and regulatory hurdles mean permitting and construction is slow and economics are not great. Also, there's probably a manufacturing capacity issue; if you wanted to get 100 new plants online in the next 5 years, and permitting and siting were a non-issue, getting the parts made would be.

[Gibbon1]

There are also technical skills and geopolitical issues with nuclear don't exist with solar. You can deliver solar panels to the Southern Congo on truck.

Also you have a choice, build a nuke plant that will produce power for 50 years. Or buid a plant to produce solar panels for 50 years. That's an accounting problem that would be interesting to see the results of.

[dredmorbius]

The limits to various forms of energy are well-known. It's a bit more than I can describe in an HN comment, though Tom "Do the Math" Murphy has a good guide, index to his posts here: https://dothemath.ucsd.edu/post-index/ Heinberg himself has written of them in earlier books and articles.

Essentially humans have access to the fluxes of solar, geothermal, and tidal energy, and the stores of fossil fuels, nuclear fission from naturally occurring uranium and plutonium, and potentially nuclear fission from hydrogen plus a few other essential light isotopes and/or elements.

All other energy sources are either carriers (as with hydrogen as a combustion fuel), or derivative. Notably hydroelectric, wind, biomass, and wave energy are all derivatives of solar flux. (Fossil fuels are derivatives of past solar flux.)

Solar is the most tractable large-scale power source. The raw rate of incidence is about 1 kW/m^2 at Earth's surface. This is reduced by a number of considerations, including land area, spacing factors, panel efficiencies, and losses through conversion (DC/AC), transmission, and storage. The net potential is perhaps 5% of the total incident quantity. And there's the small factor that all other life on Earth also competes for this resource.

Hydro is proven but largely exploited, and has environmental consequences now increasingly recognised and often untenable.

Total wind and wave power (the latter is effectively nil) are small fractions of total solar power. Wind power is attractive principally as in places where it HAPPENS to be prevalent, the capital costs are low relative to energy returned.

Geothermal, while independent of solar, is a small fraction of the latter, and is already largely utilised where available and practical, though there's significant undeveloped resource in Africa, and in the US in the Yellowstone Caldera, though official resource estimates exclude this due to its protected status as a National Park. (Pointed, the USGS utterly omits the Yellowstone Caldera in its geothermal resource survey of a decade or two back.) As baseload power, geothermal is attractive. Capital-intensive "enhanced" geothermal has proved disappointing to date (see Australia's Habanero project).

Tidal energy is worth mentioning only because it's independent of the usual solar/nuclear axis: tidal energy actually represents a tap on gravitational potential of the Earth-Moon-Sun system. It's slightly more viable than wave energy, but save for a few very limited local applications, not practicable. Tapping the entire tidal potential of, say, the San Francisco Bay would not even power the city of San Francisco at current electric utilisation, let alone full energy demands, or of the greater Bay Area. And this would require entirely damming the Bay.

Nuclear fission suffers from a fuel shortage problem: known nuclear reserves would power present human energy needs for about 15 years, total. At present rates of utilisation, that lifetime is extended, but still comes in at under a century. There's the standard bickering about definitions of reserves, and talk of seawater extraction (of uranium, other fuels not being salt-water soluable), breeding (of plutonium), or use of thorium, under either existing or novel reactor designs. All three options have significant limitations, though some may be technologically feasible. The resulting energy system and economy would be fragile and risk-prone.

Fusion is as it's always been, the power source of the future. And always will be, as the punch line goes.

That's the lineup. Murphy has a good overview of numbers. Vaclav Smil in numerous of his books (Energy and Civilization and Energy in World History, an earlier edition of the same book, though with somewhat different organisation, as well as others) takes a deeper dive into many of these issues.

Mind that solving the energy problem is only one of numerous stumbling blocks between now an a long-term viable technological human civilisation. Numerous others exist, and the fundamental fact remains that economic growth (and its concommittant and requisite resource and energy growth) simply cannot continue indefinitely.

[XorNot]

> economic growth (and its concommittant and requisite resource and energy growth) simply cannot continue indefinitely.

Economic growth occurs in any situation where you increase productivity. But this means you can increase productivity in anyway, including by efficiency improvements.

Economic growth can't be sustained at the same rate it has been, but it can be sustained because while 100% efficiency is an asymptote it is approachable.

[dredmorbius]

The assertion that economic growth can occur without an increase in resource consumption is one that's been made repeatedly, but that flies in the face of all evidence.

The story is somewhat complicated by the fact that primary consumption of energy and materials can be outsourced, giving the appearance of decoupling of growth from resource use. Once net imports and resource consumption at point-of-origin are accounted for, the connection is resumed.

Global GDP growth to date has occurred in lockstep with increased material and energy resource use.

I've looked at the relations myself simply using national GDP and energy use through about 2010, see: https://old.reddit.com/r/dredmorbius/comments/1vlksg/economi...

And more robustly:

"The material footprint of nations ", Thomas O. Wiedmanna, Heinz Schandl, Manfred Lenzenc, Daniel Moranc, Sangwon Suhf, James Westb, and Keiichiro Kanemotoc. doi: 10.1073/pnas.1220362110. PubMed ID24003158. http://www.pnas.org/content/early/2013/08/28/1220362110

"The true raw material footprint of nations ", September 3, 2013. "The study, involving researchers from UNSW, CSIRO, the University of Sydney, and the University of California, Santa Barbara, was published today in the US journal Proceedings of the National Academy of Sciences. It reveals that the decoupling of natural resources from economic growth has been exaggerated." https://web.archive.org/web/20130906063246/https://newsroom....

[toecutter]

"wave power (the latter is effectively nil) "

~30 terawatts, about double the electrical consumption of the whole planet, is effectively nil?

" [tidal is] slightly more viable than wave energy"

Why do you say this?

[dredmorbius]

Distributed over the length of all coastlines of all continents. The per-linear-meter power density is extraordinarily low. Wave power doesn't concentrate. Capital is extraordinarily expensive, and salt-water environments are hell on materials.

See Tom Murphy's analysis for more: https://dothemath.ucsd.edu/2012/01/the-motion-of-the-ocean/

(He also mentions currents and thermal gradient power from the ocean. OTEC might be more viable IMO.)

[toecutter]

I read his analysis, I found it roughly to be back of the envelope math and I did not think it was authoritative especially when he uses phrases like "my crude estimate" and "my stupid calculation".

The "per-linear-meter power density" what some would call "wave energy flux" represents an order of 5x greater energy density than wind, which is roughly 10x more dense than solar. When T Murphy describes "third string solar" he highlights that a quantity of energy is lost in each conversion, yes but also the density is increased. Similar to following energy starting with biomass, fermenting to dilute alcohol, and distilling to pure ethanol. In this case there is no effort or external energy required for incident solar energy to generate a lesser amount of denser wind energy, and for wind energy to create a smaller still amount of high density wave energy. However, this quantity is still large enough that even a fraction of it converted to electricity represents an quantity of power that I strongly reject to being called "nil" or "puny".

Following a thorough analysis and R&D phase, the important metric, levelized cost of energy, relates to the quantity of material required to construct a device that interacts with a given quantity of power. Operating in a more power dense medium favors lower LCOE. There are challenges with the salt-water environment, that fact doesn't preclude the existence of industries such as trans-oceanic shipping, offshore oil and gas, navigational and observational buoys, and other such endeavors.

Personally I consider it a good thing for an energy technology to be distributed throughout the world, I think it is preferential than having the entire energy resource concentrated in one part of the world. Another benefit it offers is that its availability is decoupled from wind and solar, the waves don't stop at night, and once established continue traveling without wind.

There is no silver bullet and wave energy is no exception, there will always be a finite quantity recoverable, and a certain cost to recover it. However, to determine those specific numbers would require a herculean effort to thoroughly analyze all of the possible wave energy converter designs - which consist of a number of major of typologies, and within each topology an even greater number of specific designs and sizes, each with their own cost and performance, which also varies depending on seastate - you would need to analyze every possible device not just for power converting performance, but for an estimate of suitable materials and construction techniques, and their costs. Only then could you answer the important question, can any quantity of wave energy be economically recovered at a cost competitive with other leading renewable energy sources.

[jfengel]

That's "Rev. Malthus". He was an ordained curate of the Church of England.