Base generation was a cost optimization. Planners noticed that load never dropped below a specific level, and that cheapest power was from a plant designed to run 100% of the time rather than one designed to turn on and off frequently. So they could reduce cost by building a mix of base and peaker generation plants.
In 2025, that's no longer the case. The cheapest power is solar & wind, which produces power intermittently. And the next cheapest power is dispatchable.
To take advantage of this cheap intermittent power, we need a way to provide power when the sun isn't shining and the wind isn't blowing. Which is provided by storage and/or peaker plants.
That's what we need. If added non-dispatchable power to that mix than we're displacing cheap solar/wind with more expensive mix, and still not eliminating the need for further storage/peaker plants.
If non-dispatchable power is significantly cheaper than storage and/or peaker power than it's useful in a modern grid. That's not the case in 2025. The next cheapest power is natural gas, and it's dispatchable. If you restrict to clean options, storage & geographical diversity is cheaper than other options. Batteries for short term storage and pumped hydro for long term storage.
Solar and batteries aren't consumables, so they're not particularly vulnerable to supply chain disruption. If we lose our supply of batteries, we'll have ~10 years or so to find an alternate supply. We won't be able to do new installations during the disruption, but existing installations don't stop working.
Unlike a fossil plant when the supply of fuel is disrupted.
Ukraine is an excellent example of why centralizing your grid energy source is a bad plan... but not just for war situations. If you have an agile, adaptable modular grid you can recover for any form of disaster (natural or man made) very quickly and cheaply.
I really feel this is an under valued aspect of electrification and greening of the power sources we use.
Like nuclear, I believe geothermal has high capital cost and low running costs, suggesting that it isn't usefully dispatchable.
But that's too simplistic. A big limitation of geothermal is that rock has poor thermal conductivity. So once you remove heat it takes a while for it to warm up again. If you're running it 100% then you need a large area to compensate. OTOH, if you're running it at a lower duty cycle you likely need less area.
So if you know the duty cycle in advance, then you can likely significantly reduce costs. Yay!
But that also means that you likely can't run a plant built for low duty cycles continuously for 2 weeks during a dankelflaute. It's likely great for smoothing out daily cycles, but not as good for smoothing out annual cycles. That means it's competing against batteries, which are also great for smoothing out daily cycles, and are very inexpensive.
This "there is no base load" idea is a ridiculous myth trivially disproven: every grid on the planet has continuous demands on it and they're quite significant (5 GW is about 50% the day time peaks).
It doesn't matter what the cost is, because later this evening or tomorrow morning I can guarantee you the same thing: my state will need at least 5GW of power to literally keep the lights on.
When solar and wind produce at near-zero marginal cost, running inflexible baseload beside them just forces cheaper generation to switch off, driving up system costs.
What the grid needs is dispatchable capacity - batteries, hydro, gas peakers (if we must) and demand shifting - that can plug the gaps when cheaper forms of generation cannot.
What does it look like if you actively encourage people to use power when it is cheapest to produce now?
I guess we'll find out when 3 hours of free electricity at noon becomes a standard offer next year.
I think this abstraction is missing the elasticity of demand that can by unlocked by end-to-end dynamic pricing. Probably if the production was cut in half for some day, and hourly price hiked up until demand matches production, customers would still choose to keep most of the lighting while postponing some more energy intensive loads.
You are using long-term in an extremely vague way.
Pumped hydro is not a solution for seasonal storage or yearly storage. Seasonal variation can be a problem in higher latitudes.
For example we have a serious problem in New Zealand where our existing "green" hydro lakes are sometimes low and our economy is affected: creating national power crises during dry years. We use coal-burning Huntley and peakers to somewhat cover occasional low hydro generation.
Unfortunately our existing generators also have regulatory capture, and they prevent generating competition (e.g. new solar farms) through rather dirty tactics (according to the insider I spoke with).
Apparently much of our hydro generation is equivalent to “run-of-river” which requires the river to flow. Although the lakes themselves are large, they don't have enough capacity to cover a dry year.
NZ had planned a pumped hydro, but it was expensive: planned cost of 16 billion compared against total NZ export income of ~100 billion. https://www.rnz.co.nz/news/national/503816/govt-confirms-it-... So completely uneconomic risk (plus other problems like NIMBY).
What will likely happen is that people will decide that "99% is good enough", and use fossil generators to cover dankelflautes,
It was shocking to me to drive by many of the California lakes/reservoirs that were overfull in the spring of 2019 only to hear that they were basically running dry two years later, and realize that as substantial a water storage system as they are, they're not multi-year scale against the required water supply.
https://particulier.edf.fr/content/dam/2-Actifs/Documents/Of...
Western counties building nukes is so expensive it makes the cost of electricity go up.
Also, France can’t build new nuclear for cheap/fast anymore either. They have a program for new reactors, even if they go ahead the first one won’t come online till 2038 by the earliest. We can’t wait that long.
Flamanville 3 is 7x over budget and 12 years late on a 5 year construction program. The EPR2 program is in absolute shambles.
Currently they can’t even agree on how to fund the absolutely insanely bonkers subsidies.
Now targeting investment decision in H2 2026. And the French government just fell and was reformed because they are underwater in debt and have a spending problem which they can’t agree on how to fix.
A massive handout to the dead end nuclear industry sounds like the perfect solution!
not to say it is remains costlier than conventional sources, albeit not accounting for externalities.
A lot of mountainous places are dry, and a lot of wet places are flat.
Of the remaining places, some are so unique that they cannot be destroyed by industrial construction (National Parks etc.)
For example, the main ridge of Krkonoše (Riesengebirge) on the Polish-Czech border has a lot of wind and rain and deep valleys, but it is the only place south of Scandinavia with a Scandinavia-like tundra and many endemites surviving from the last Ice Age. Any attempt to construct pumped hydro there would result in a national uproar on both sides of the border.
But off river? The possibilities are vast.
We know that in North America, for example, significant energy use comes from transportation and heating requirements, and that at this time, very little transportation is powered by renewables, and not a whole lot of heat either (though both are growing).
On the other hand, the entire current residential electrical demand of the city of Santa Fe (about 82k people) can be met with a single relatively small PV+BESS plant (and might just be if it manages to get built).
Since you're comparing it to nuclear, I'm assuming you mean electricity production here, not energy production?
It's always worth remembering that electricity only accounts for ~20% of global energy consumption (in the US it's closer to 33%).
I suspect people confuse these two because in a residential context electricity plays a huge part of our energy usage, but as a whole it's a smaller part of total energy usage than most people imagine.
But any serious discussion of renewable energy should be careful not to make this very significant error.
The Lawrence Livermore National Laboratory publishes a great diagram of US energy use: https://flowcharts.llnl.gov/sites/flowcharts/files/2024-12/e...
So even in a residential context, electricity is only about 1/4 of the demand. Across the whole country it's less than 300TWh out of 1500TWh, under 20%.
That excludes "imported energy" though, as in goods which used energy to make but were then imported.
Great chart, by the way.
It's also quite hard to find suitably hot rocks suitably close to the surface.
Focusing on fusion .. I think that's a legacy of 60s SF, when the fission revolution was still promising "energy too cheap to meter".
That's basically it. Most geothermal plants today are in locations where there are hot rocks, maybe geysers, close to the surface. "Deep geothermal" gets talked about, because temperatures high enough for steam are available almost everywhere if you can drill 3,000 meters down. There are very few wells in the world that deep, not counting horizontal drilling runs.
The economics are iffy. You drill one of the most expensive wells ever drilled, and you get a medium-pressure steam line. Average output is tens of megawatts.[1]
[1] https://pangea.stanford.edu/ERE/db/GeoConf/papers/SGW/2020/A...
In a world where anyone could just YOLO any reactor into production with minimal red tape, consequences be damned, fission energy would actually be extremely cheap. Hence the optimism around fusion. The promise of fusion is an actualization of last century's idealistic conception of fission. It can be a silver bullet for all intents and purposes, at least once it's established with a mature supply chain.
https://www.withouthotair.com/c16/page_96.shtml
The problems are that rock isn't a good conductor of heat, so once you've cooled a bit down, you have to wait for it to warm up. Warming only happens very slowly at the rate of < 50mW / m² which limits the amount of power you can get out.
> There aren’t gates of Hell just anywhere. A kilometre below ground in Kamchatka is considerably hotter than a kilometre below ground in Kansas. There is also readily accessible geothermal energy in Kenya (where it provides almost fifty per cent of the country’s energy), New Zealand (about twenty per cent), and the Philippines (about fifteen per cent)—all volcanic areas along tectonic rifts. But in less Hadean landscapes the costs and uncertainties of drilling deep in search of sufficient heat have curtailed development.
Ground-source heat pumps extract about 1000 times more power from ground loops, where does the difference come from?
The worst earthquake that was induced that way was 3.5, but given that one of the quakes happened in an area that had a catastrophic earthquake in the Middle Ages, some caution might be warranted: https://en.wikipedia.org/wiki/1356_Basel_earthquake
If economically viable fusion was "cracked" what would the nature of it's unreliability even be?
The reactor breaks because it's a large device operated at high stresses (power/area, neutron loading). There are many components and joints that can fail.
BTW, this means fusion will be expensive, because getting all those components to be reliable right off the bat becomes expensive. No tiny cracks in the welds means expensive quality control.
I like how David Hamel put it: We live in this thin sliver on the surface of the planet where it is reasonably peaceful. This is the tranquility! It's a good thing! If you go up or down by a mere few miles there is so much energy it kills you.
We have to see if and when any of them goes into production, but the technology seems very interesting
1. The host at our apartment encouraged us to leave the windows cracked and the heat on for good air circulation.
2. The hot water (at the taps) has a sulfer smell, because it's (also) piped geothermal water. My host explained they also had a water heater upstairs in their home because they preferred "heated cold water" over "hot water", which is a funny distinction to those of us who do not have the latter.
This is due to the hot water in those regions literally being pumped out of the ground and into homes, and on a completely separate plumbing system. Majority of other areas use heat exchangers with pristine cold water, thus no smell nor taste is transferred.
So if you are staying in any other municipality in the capitol, you can use the hot water in cooking directly without boiling cold water. It's the same.
Another way they've utilised geothermal energy is with large, sophisticated greenhouses which allow growing of many produce they would otherwise import. I only had the opportunity for a brief visit but a lot of it looked hydroponic with really interesting monitoring and control technology. (Plus the biggest bees this Antipodean has ever seen! These suckers were so big they didn't buzz, they rang the doorbell.)
Whoever was there before had left the shower running. We were the only people there, and hadn't seen anyone pass us on the (dead end) road, so it must have been on for quite a while.
Only when I went to for my pre-soak shower did I realize that it didn't actually have any kind of user-accessible way to turn it off.
On a per-minute basis that's comparable to highway driving in a smallish electric car.
"Geothermal energy" involves drilling down to hot rock to tap intense heat to run a turbine that produces electricity.
Seriously this would be such a dream!
Turns out that the best battery is literally 10 feet away* - and you don't even need to charge it!
*if you want to make steam its a few thousand, but for heating and cooling its literally just 10 feet!
However, I'm skeptical that geothermal energy can be economically competitive with solar without major innovations in heat engines, no matter how abundant the energy is and how easily you can get that energy to the surface.
https://www.eia.gov/analysis/studies/powerplants/capitalcost... outlines the estimated costs (five years ago) of a 650MW peak ultra-supercritical coal power plant without carbon capture; the total capital cost estimate comes out to US$2.4 billion, which is US$3.70 per peak watt. Of that, I think the only line item that wouldn't be the same in a 650MW peak ultra-supercritical geothermal plant is "Mechanical – Boiler Plant", which is US$905 million, leaving US$1.5 billion, US$2.30 per peak watt. (I'm not even sure you could eliminate even all of that US$905 million in a geothermal plant; some of it might be plumbing you'd also need to pass heat from your downhole heat exchange fluid with the ultra-pure deionized water you use to drive the delicate steam turbine. But let's suppose you could.) Of that US$1.5 billion, US$155.2 million is "Mechanical – Turbine Plant", so the turbine alone costs 24¢/Wp.
But SEIA last year published https://www.seia.org/research-resources/solar-market-insight.... They have a set of cost breakdowns for “turnkey installed price” for power plants, coming in at 98¢ per watt for “utility-scale fixed-tilt”, slightly higher than the previous year and almost half due to about 40¢ for the PV module itself. Residential is at 325¢, with about 20¢ for the PV module. That's even in the US, where the EIA report's estimates were sited, despite the US's prohibitive import tariffs on solar panels from China, which makes most of the world's solar panels.
Mainstream PV modules are now 12.3¢ per peak watt https://www.solarserver.de/photovoltaik-preis-pv-modul-preis... (except in the US), which would drop SEIA's cost estimates from 98¢/Wp to 70¢/Wp, even in the absence of any other cost optimizations in solar farm design.
Now, utility-scale fixed-tilt solar farms typically have a capacity factor of around 20%, depending on latitude, because the sun is below the horizon half the time and somewhat slanted and/or clouded most of the rest of the time, so 70¢/Wp is really about US$3.50 per watt, not counting the batteries. But geothermal typically only has a capacity factor of around 74% in the US https://en.wikipedia.org/wiki/Capacity_factor#Capacity_facto... so US$2.30/Wp is really US$3.10 per watt.
That leaves you 30¢/Wp (74% × ($3.50 - $3.10)) for geothermal exploration and drilling. And if you can reduce the 82% of the solar 70¢/Wp represented by the non-PV-module costs by a little bit, or if you're equatorial enough that your PV capacity factor is 23% or above, that's going to zero or negative. I think the average PV capacity factor in California is something like 29%, though that isn't fixed-tilt and therefore has slightly higher costs.
Also note that the PVXchange page I linked above lists "low-cost" solar panels as having fallen to €0.050/Wp this month, a new historic low, which is 5.9¢/Wp. That's a 50% price decline from two years ago.
Fundamentally I think it's just going to be very hard for 24¢/Wp steam engines to compete against 5.9¢/Wp solar panels. The steam engines have the additional disadvantage that, to get the price even that low, you need enormous degrees of centralization—on the order of a few thousand power plants for the whole population of the US. This requires long-distance electrical transmission lines as well as local distribution lines, which are both substantial costs of their own as well as wasting a double-digit percentage of the energy. Local electrical generation eliminates those costs; you can charge your cellphone or your angle-grinder battery directly from a 5.9¢/Wp solar panel with no more electronics than a couple of protection diodes, not requiring the rest of the 70¢/Wp in the utility-scale solar plant.
This cost analysis is completely indifferent to where the heat to boil the water comes from, so it applies equally well to nuclear power, except for Helion.
The exceptions would be in places where geothermal energy is available and solar energy is either unavailable or very marginal: the surface of Venus, the ocean floor, Antarctica, Svalbard, etc.
Does anyone have a trustworthy estimate of the costs of drilling? Even drilling into cold rocks (for oil) would be a good start, even if hot rocks are more expensive to drill into. The article says that Fervo has raised US$800 million in capital and drilled three appraisal and demonstration wells with it so far, which gives us a ballpark of US$200 million per well. This does not offer much hope that drilling costs will be a minor fraction of the costs of a geothermal plant.
The article unfortunately doesn't enter into this analysis at all.
I am somewhat skeptical of this figure:
> Geothermal energy production in the U.S. at that time [i.e., 02005] was around three or four thousand megawatts.
https://en.wikipedia.org/wiki/Electricity_sector_of_the_Unit... says that geothermal energy production in the US in 02022 was 16.09 billion kWh per year, which is 1825 megawatts. Does that mean that geothermal energy production fell by about half between 02005 and 02022? More likely Rivka Galchen got confused.
It's unfortunate that the article also confuses ground-source heat pumps (thermal energy storage) with geothermal energy sources. It's a common confusion, and it makes conversations about geothermal energy unnecessarily difficult.
Geothermal is a great fit for dispatchable power to replace coal and fossil gas today (where able); batteries are almost cheaper than the cost to ship them, but geothermal would also help solve for seasonal deltas in demand vs supply ("diurnal storage").
https://reneweconomy.com.au/it-took-68-years-for-the-world-t...
https://ember-energy.org/data/2030-global-renewable-target-t...
I also love geothermal for district heating in latitudes that call for it; flooded legacy mines appear to be a potential solution for that use case.
Flooded UK coalmines could provide low-carbon cheap heat 'for generations' - https://news.ycombinator.com/item?id=45860049 - November 2025
We deploy solar PV capacity, this doesn't mean we actually get that much power from the deployments. Nuclear fission provides reliable, baseload power, and doesn't require huge battery arrays to compensate for the sun setting or winds calming.
https://particulier.edf.fr/content/dam/2-Actifs/Documents/Of...
Geothermal is not nuclear fission. The heat comes from a combination of primordial heat (from the gravitational energy turned to heat as the Earth formed) and radioactive decay (which is some combination of alpha and beta decays; spontaneous fission is extremely rare.)
Maybe SMR's, thorium, 4th gen, etc will work out, but maybe not.
The EU also forgot how to build airports and train stations on budget and on time.
Should we stop building airports and train stations?
As for nuclear power plants: Germany and France built most of their reactors on budget and on time.
Having smaller scale local power generation, whether it’s SMRs, solar, wind or geothermal, there’s a huge advantage in terms of economy, investment, and politics.
installs: https://www.pv-magazine.com/2025/01/13/the-fastest-energy-ch...
costs: https://www.reddit.com/r/energy/comments/11q58pe/price_trend...
The renewables are so cheap and quick to provision it's hard to see how fission can compete.
The LED bulbs I have access to (whatever's in the aisles at Home Depot, Costco, etc.) fail much more frequently than the incandescent bulbs I used to buy, and produce an uglier light that is less warm even on the softest/warmest color settings.
My suspicion is that incandescents were at the "end" of their product lifecycle (high quality available for cheap) and LEDs are nearing the middle (medium quality available for cheap), and that I should buy more expensive LED bulbs, but I still think that there are valid "complaints" against the state of widespread LED lighting. I hope these complaints become invalid within a decade, but for now I still miss the experience of buildings lit by incandescent light.
The other thing with AI--the LED revolution was led on this idea that we all need to work as hard as we can to save energy, but now apparently with AI that's no longer the case, and while I understand that this is just due to which political cabals have control of the regulatory machinery at any given time, it's still frustrating.
LED lamps work just fine, you just need to pay more attention when you’re buying them. Philips makes decent LED lamps.
Make sure you’re buying lamps with 90+ CRI, that will help with the quality of light. 2700K is a good color temp for indoor living room/dining room/bedroom lighting, 3500-4000K for kitchen/garage/task lighting.
You also need to buy special lamps if you put them in an enclosed fixture, look for ‘enclosed fixture’ rated lamps. Regular LED lamps will overheat in an enclosed fixture.
I figured out why this happens.
The light color they call "daytime" is around 5000K, so I expected it to look like being outside in the sun; but instead I got a cold blueish vibe. The problem? Not enough power! I got the equivalent of a moonlit room.
So I got this 180W LED lamp (that's actual 180W, not 180W equivalent) [1]. It's so bright I couldn't see for 5 minutes. I put two in my office on desk lamps. The room now looks like being outside, without the "ugly blue" tint, even though the product says it's 6000K. The days of my SAD suffering are over!
None of this will change the CRI.
citation needed
> “West Virginia has numerous coal plants that have powered this country for decades. We need these plants to remain operational,” [WV Governor] Morrisey said. “… We will never turn our backs on our existing coal plants and we will work with the federal government to pursue new coal-fired generation.”
https://westvirginiawatch.com/2025/09/11/morrisey-shares-new...
https://wvpublic.org/story/energy-environment/data-center-bi...
https://www.wvlegislature.gov/Bill_Status/bills_text.cfm?bil...
Also, due to solar not panning out at scale.[1]
More seriously, coal is just cheaper and, with incentives being removed for green energy, it's the cheapest and fastest option to deploy. It's dead simple and well understood reliable power.
[1]https://apnews.com/article/california-solar-energy-ivanpah-b...
That solar plant you linked is an obsolete experimental technology. Obsolete because regular PV became so much cheaper.
Direct solar continues to be installed at greater amounts every year and coal is economically uncompetitive with basic anything (which is why it is collapsing), and especially against natural gas.
Assuming zero growth in energy consumption (hello AI), extracting even half of that seems like it would be consequential.
