I'm going to attack your post here. But I don't intend this as a personal attack; you may very well be arguing in good faith, just from outdated information. Being even a few years behind - and your citation of a decade-old book suggests you're farther behind than that - means you've missed out on a ton of new information about the practical scale of renewable power (did you know that worldwide we're installing almost 1.5GW of solar power EVERY DAY OF THE YEAR these days?).

> In the USA we've had decades of nuclear fission providing something like 20% of our electricity and with each closing of a nuclear power plant there's increased use of fossil fuels to replace them.

Coal use in the US has been plummeting (down 680TWh in the last decade), and rising natural gas use (up 460TWh in the last decade) is only offsetting about half that fall. You may be interested to read the EIA's annual report: https://www.eia.gov/electricit... (see, particularly, chapter 3). And renewables provide about 3x as much annual energy as nuclear plants do, per dollar spent (using un-subsidized prices: https://www.lazard.com/media/2...). So your argument that spending money on new nuclear plants is reducing carbon emissions is untrue.

> Uranium and thorium is stored energy, stores of energy upon which we can draw from as desired.

Also not true. Sure, uranium and thorium store quite a lot of energy. But we can't draw on them "as desired." We can draw on them with about three days' warning, assuming the plant's fueled, maintained, and waiting to start up. Nuclear power does not provide a backup to a renewable grid, unless that grid already has sufficient storage that you can forecast a need for a nuclear backstop 3+ days in advance. And if you think we're *not* going to have a renewable grid, you haven't been paying attention to how incredibly cheap renewables are. 94% of planned capacity additions to the US grid in 2024 are solar, wind, and batteries. Everything else combined is 6%. See EIA again: https://www.eia.gov/todayinene... But even that is an understatement of how renewables-dominated the grid pipeline is, because it ignores planned retirements. Coal, oil, natural gas, and nuclear power capacity are ALL forecast to fall this year as new capacity fails to offset retirements. New renewables are not only supplying the new energy required as grid demand rises, but are now displacing existing capacity.

And this year isn't a fluke, it's a continuation of a developing theme: cheap power sources get built. Check out the forecasts through 2027: https://www.eia.gov/electricit... Coal: down 32GW. Gas: up 5GW. Wind: up 30GW. Solar: up 100GW.

> This isn't because people hate clean air and a stable climate but because renewable energy cannot reliably provide energy when and where it is needed.

Why do you think that's true? We have ample modeling from research teams throughout the world all pointing to the same conclusion: there are tons of different ways to provide 100% supply/demand matching with different amounts of overbuilding and/or storage (and some amount of demand response would be even cheaper, but isn't technically necessary) - the "renewables work fine" conclusion is not sensitive to how the build-out proceeds. And we have practical examples of it working in the real world: e.g. South Australia which was 71% renewable-powered over the entire year last year, despite its tiny geographic footprint, weak interconnections with other states, and almost no storage at all. And they're targeting net 100% renewable electricity just THREE YEARS from now, enabled by a connection to NSW coming online, and a bit of new battery development. Fully one quarter of the time, the state's at or above 100% renewables, so clearly the predicted stability and grid control problems are surmountable: it's just a matter of building enough generation to supply the bulk energy, and then some combination of transmission and energy storage to match supply and demand.

And that's precisely what will happen. Any utility that wants new bulk energy will build wind and solar, because they can rely on 'free' load matching from gas plants ramping up and down, so it's extra cheap. The learning rate will drive down the cost of wind and solar even further, meaning they'll continue to be built, even after we've built so much of them that sometimes they get curtailed, because that will still be cheaper than building any of the alternatives. Then, during the day, everything non-solar will have to ramp down to make way for cheap solar power because it has the cheapest marginal cost and ample supply. And that means that plants that can't ramp much (coal) or at all (nukes) will either have to bid less than zero (see what's happening in Australia...) in order to be allowed to run so that they can supply energy into the evening peak, or they'll be forced to close (see...Australia). This happens gradually: their profitability keeps slipping year after year until their owners give up and quit. Note that it's already cheaper to build a brand new PV plant (in much of the world) to generate electricity than it is to simply keep operating an existing already-paid-for coal, gas, or nuclear plant (LCOE for new generation vs short-run marginal cost for coal, gas, nukes).

So now we have renewables and gas, but gas is increasingly expensive because its capacity factor keeps falling and the operators need to pay off their CAPEX. Enter: batteries (and CAES, PHS, normal hydro, etc.) which will increasingly cut further into gas's profitability by relegating gas plants to an increasingly marginal role providing extremely expensive peaking power, but little bulk energy. Which is TOTALLY FINE! A huge fleet of cheap-to-build, inefficient open cycle gas turbines operating for 100 hours a year to get us from 99% to 99.99% served energy is TOTALLY FINE. We just need to stop running them year-round to provide bulk energy.

I've presented this as a series of steps, but of course all these things will be - already are - happening gradually, and together, over the next few decades. The fossil fuel industry is mortally wounded but still alive; in a decade or so it'll be dead but still twitching; hopefully in two decades it'll be fully buried (and/or dug up: did you know that the oil and gas pipelines in the US have enough steel in them to satisfy two full years of nation-wide steel demand? More than enough to build all the wind turbine towers we'll need!).

Why are we still having these conversations? Coal and gas cost too much for electricity generation. Combined-cycle gas is 50-100% more expensive than wind power in the US (source: https://www.lazard.com/perspec...). New-build solar or wind are already cheaper than continuing to operate an already-build coal plant (same source).

So what? So lots of solar and wind will continue to be built because they're the cheapest source of electricity. Lots of solar and wind will mean coal plants have to ramp at their maximum rates to keep up, which wears them out and makes them even more expensive to maintain; they're doomed and are going away. Gas plants generally can ramp faster but even they are getting stressed by the ramp rates in, e.g. Australia.

If lots of solar and wind will get built because they're cheaper, and as a side effect they drive up the prices / drive down the reliability of coal and gas, that just makes them EVEN CHEAPER by comparison, so even more will be built. Learning rates, though declining, will continue to depress prices, leading to overbuilding because they'll still be profitable even with increasing amounts of curtailment.

So what? So lots of short-duration energy storage (mostly batteries) will be built to take advantage of cheap daytime electricity. All sorts of flexible demand will shift from historically-cheaper nighttime electricity to now-cheaper daytime: EV charging, heat pump water heaters, heat pump space heaters, maybe even the electrolyzers the hydrogen-economy boosters hope for. Even with recently-cheap natural gas prices in the US, heat pump heating at my house (in Wisconsin) is cheaper than gas. It's slightly more expensive per unit of delivered energy to use electricity for heating, but not enough to make up the $20+/month cost of having gas service at all. Electric-only is cheaper year-round.

Gas plants, at best, can hope to be relegated to a peaker role: solar, wind, and hydro haven't been meeting demand for a while and batteries / pumped storage are getting low, so let's fire up some gas plants for 100 hours a year every January / February.

...and take your emojis with you :)

Both mobile messages I've grudgingly sent this year have been SMS messages.

No way any of the first four will happen, so it must be e).

This 'costs too much' lie has been repeatedly disproved, eg by Jacobson et. al. ( https://www.nrc.gov/docs/ML143... ) using 5-year-old prices. Nuclear and fossil fuel power sources have gotten more expensive since then (other than natural gas); wind and solar prices have continued to crash: down 55% for solar and 40% for wind over the past 5 years ( https://www.lazard.com/perspec... ).

Advocates of new nuclear plants (eg https://decarbonisesa.files.wo...) stress the technical attributes of fast reactors and the environmental advantages of nuclear power, particularly when using reprocessed used fuel from older designs. Which is all well and good (and we should probably be doing some of that), but it's a bit rich for someone advocating for nuclear reactors that will never be built at meaningful scale to cry "So come on back when you're ready to talk actual, implementable solutions." Wind and solar plants are actually being built, to the tune of ~100GW/yr for solar, and half that for wind, worldwide. If you're serious about low-carbon energy sources, put your effort behind solutions that are cheap, modular, and politically and socially acceptable, rather than waste your energy tilting at windmills. Huge amounts of effort might move the needle on nuclear power, but only moderate effort would be required to double the installation rate of wind and solar. The World Nuclear Association says wind and solar generated about half as much total energy in 2016 as nuclear ( https://world-nuclear.org/info... ). Nuclear generation was down slightly by 2018, while wind and solar had increased by well over 50% and are on track to surpass nuclear this year or next. Your advocacy on behalf of a new nuclear plant might move it from "not a chance in hell" to "still not going to happen" whereas advocating for new solar and wind might actually make the difference between new low-carbon generation being built or not. Spend your time making a positive difference!

Regarding storage:

With the cost to install brand new PV or wind plants (and, yes, replace them 25-30 years later) less than the operating costs of already-built coal and nuclear plants (and, soon, combined-cycle gas turbines too), they're unstoppable. The existing flexibility in most power grids can readily handle 30%+ variable renewable energy, and modest changes (tweaking what's there, not inventing anything new or building significant new infrastructure) can accommodate 60% or more VRE (about where South Australia is now, with 75% or more in the couple-years-from-now forecast: https://reneweconomy.com.au/wp...). A ~5x expansion in wind and solar will further lower costs; increased electrification of transportation and heating loads will provide cheap variable demand to match supply and demand within tens of hours (eg heat your water now while it's sunny and power is cheap, or accept a bribe to charge your car tomorrow instead of tonight), and more frequent zero-or-lower spot electricity prices will incentivize storage and other forms of dispatchable demand. Once that's done, then we can talk about storage at scale.

Nuclear fuel provides convenient and compact storage of huge amounts of energy, so it's easy to see a role for nuclear plants in the future, spooling up to provide reliable supply in the face of low-probability weather (either in the low-supply case of extended and geographically widespread calm cloudy weather, or in the high-demand case of extended atypically cold weather). But for bulk energy supply, it's going to be hard for anything to compete with cheap wind and solar. If the learning rates on solar and wind continue, they'll be under half the cost (LCOE) of merely continuing to run an existing competing power plant of any other current (thermal) technology long before we're at 80% renewable power. And from there, whatever we do to get the last 20% certainly won't be more expensive than just overbuilding, and overbuilding won't be that expensive.

> Batteries like this are almost 100% recycled into new batteries when they reach the end of their useful life.

That is not currently true, but is becoming more true as time goes on. Recycling rates for lead-acid batteries are quite high, and there's a reasonably robust supply chain from your local batteries plus or what have you, all the way back to building new batteries from reclaimed materials. The corresponding industry for the various LiIon chemistries is still in its infancy, but will grow rapidly as the supply of end-of-life lithium cells starts to boom in the coming years. We'll work out a reasonably optimal schedule of primary use, second-life use (eg EV battery packs refurbished / reused for stationary storage), and eventually recycling.

Lead-acid battery recycling has a fairly uniform product to be handled, both in terms of size and chemistry, whereas LiIon chemistries vary widely and individual cells range from the 0.1AHr range to the 100AHr range for big pouch cells. I'm not a battery expert so I'm not sure how big a problem that is from a recycling perspective. I suspect efficient and scalable battery disassembly is currently a larger problem than sorting the different electrodes for potentially-different processing afterward. Perhaps someone who actually knows something about this can comment further.

BNEF has access to good research and should have written a better article. Instead they've constructed a clickbait article full of gibberish that obscures rather than illuminates what data they do deign to present:

1. An average EV is less polluting, per mile, than even the best gasoline or diesel vehicle.
2. If that EV were built with dirty power, and charged throughout its life with dirty power, it would still be a net win, albeit a small one verging on a tie, on lifetime emissions.
3. We're projected to be be building a whole lot of new EVs.

And there's no mention of the obvious objections to this sort of facile analysis:
1. The average new EV probably displaces a purchase of an average new gasmobile, so the comparison with the most-efficient gasmobile is unrealistic. If the average new EV driver is particularly eco-conscious, and would otherwise be buying a highly-efficient gasmobile, that new driver is probably also sourcing the power from cleaner-than-average supplies, so calculating as if it were charged from the average local grid is unrealistic.
2. Grid carbon intensities are dropping worldwide, and the speed of this drop is accelerating as renewables get cheaper and cheaper relative to fossil-fueled plants. New renewables are cheaper than new thermal power plants almost everywhere, and we're only a few years away from new renewables being cheaper than continuing to fuel an already-built thermal plant in some parts of the world. Over a 15-year lifespan, EVs will keep getting cleaner per-mile, whereas gasmobiles will wear out and become less efficient.
3. While the article focuses on manufacturing emissions, their own graphs show that these correspond to only about 2 years worth of tailpipe emissions. A worthwhile target for reduction, for sure (and one that will happen naturally, as large manufacturers consistently seek to reduce their power costs by buying cheap renewable energy), but not the big target that we should be focusing on. The running costs dominate lifetime emissions, so we should tackle them first (especially as cleaning up electricity generation world-wide would also significantly reduce manufacturing emissions).

BNEF usually produces much better analysis than this. I'm disappointed in them.

The time has come to stop owning a car.

..for you. Which is great, and I'm glad for that.

And you're right that cars tend to be expensive, and people tend not to rationally think about those costs. However, it's not always so. Consider my annual vehicle expenses, which are about the same as your transportation expenses:

$1000 'depreciation'
$584 insurance
$159 electricity
$100 maintenance
$85 vehicle registration
$70 tires

I have a low-end Leaf, purchased at the end of the model year for about $14k after federal tax incentive. That's about $1k/yr for its expected lifetime. I drove 6225 miles last year at the cost of $159 worth of electricity, some wear on my tires, a few car washes, and new wiper blades. To do that driving legally added about $650 in registration and insurance as well.

Actually, yes. I can tell you exactly how much (electrical) energy I've been billed for every month for the past several years. In theory this tracks closely with actual energy consumption. I can further tell you that my energy consumption from gas for heating slightly exceeds my electrical energy consumption, but they're quite close (within 1% for the past 12 billed months). My condo association dues include hot water, so my true energy consumption is somewhat higher, and specifically underestimates gas.

I also burn on average 1.4gal/day of gasoline, or about 18.7MWh/yr, which exceeds the combined gas and electric usage for my home by a comfortable margin. There are further energy expenditures at work and of course in the production and transportation costs of the food and other materials I consume on an annual basis. Plus miscellaneous other expenditures of smaller magnitude.

If you're saving money by building machines yourself, you're paying too much for your computers. Which you knew, having asked the question in the first place.

Find a local computer store that builds whitebox machines. My workplace is lucky enough to be next door to such an outfit, and they build our machines for us. Basically, for the Newegg price of the components, we get an assembled, tested (they have a burn-in suite that runs for a day) machine with a 1-yr system warranty and our custom drive image installed. Customer support is great - walk over there and talk to a real person who isn't following a script and knows what he's talking about.

Your local computer store may be crap, but it's easy to tell, and if you have a good one it beats the hell out of buying from Dell.