this post was submitted on 11 Mar 2026
942 points (98.3% liked)
Climate
8493 readers
205 users here now
Discussion of climate, how it is changing, activism around that, the politics, and the energy systems change we need in order to stabilize things.
As a starting point, the burning of fossil fuels, and to a lesser extent deforestation and release of methane are responsible for the warming in recent decades:
How much each change to the atmosphere has warmed the world:
Recommended actions to cut greenhouse gas emissions in the near future:
Anti-science, inactivism, and unsupported conspiracy theories are not ok here.
founded 2 years ago
MODERATORS
you are viewing a single comment's thread
view the rest of the comments
view the rest of the comments
I don't know, man. What if its cloudy?
Me shouting the answer, but you can't hear it over the bombs exploding across the Straight of Hormuz
You can use batterys
I mean yes, but also then the investment gets a lot bigger too.
In my country (Estonia), if we did solar + batteries only, the batteries would have to be large enough to withstand electricity consumption being smaller than production for the entire summer (which at its peak has 18 or 19 hours of sunlight per day and most people don't have AC so our summer electricity usage is smaller than winter).
And also from about october to march, there's almost no sunlight, and electricity consumption is through the roof because heat pumps have been pretty common in new builds and renovations for like 2 decades now, replacing mostly solid fuel furnaces rather than resistive electric heaters.
Which is not to say we should abandon solar, but it'd be incredibly cost-prohibitive to go renewables-only here. In the summer our electricity prices often go negative already (still zero + network fees for consumers, not really negative prices -.-), but in winter I've seen 5 euros per kilowatthour at peak times.
Now I googled the cost of a terawatt hour of battery capacity and Google's AI was happy to report to me that a terawatthour is a million kilowatt hours and thus at ~80€/kWh it would be 80 million euros. That's peanuts! Just 640 million would get us enough battery capacity to store a year's worth of energy, that should surely get through a winter!
Trouble is, I was taught slightly different values for the SI prefixes and back when I went to school, tera was a billion kilos. So if it still functions that way, we're talking hundreds of billions instead. Our national budget for the year is 20 billion. But if every person with a job paid just a million extra euros in tax, we could afford to do it!
So obviously, solar alone + batteries won't do it at such a high latitude. Wind power helps a ton, but that's still unpredictable. And after everyone on a flexible-price plan saw a 5x increase on their power bill for january (1000+ euros being pretty common), I don't think the people will settle for "works most of the time". We actually need a nuclear power plant and we need it to be built before December 2025.
Till then we'll continue burning dirty ass coal and (yay, even worse) shale. Which I fucking hate, but the economic reality of our country is that this is what we can afford right now, with a gradual buildout of solar + wind.
But funnily enough, if we got the hundreds of billions worth of batteries magically out of thin air, the cost of buying enough solar panels to produce the entire country's annual electricity consumption every year... Would be in the hundreds of millions range or a bit over a billion at most if this meme/infographic is to be believed, even if adjusting the capacity factor, which is more like 10-15% here due to our nasty winter. Chump change pretty much for a country like ours.
This is the funny AI response that says both millions and billions for the cost of a terawatt hour of battery capacity. For my own calculations I actually went to the source at Bloomberg and took a number that was on the lower side, but not the minimum, of the range they provided for 2024.
I don't think we have to worry about AI developing the I part of AI anytime soon.
Also, in 2024 we roughly doubled our peak solar output from 600 MW to 1300 MW! (2025 unfortunately saw a LOT less new solar installation).
But our winter peak consumption is 1600 GW, so this is still a bit under 0.1% of that. And peak production is in the summer :/
You don't need 1twh of batteries to support 1gw of solar you need 2-4gwh depending on wanting 2 or 4 hours of overnight storage. Prices are dropping so fast, or so low now, that 6 hours is an easy option to choose. But for winter, see my other post on H2, or just don't nuke your legacy power from orbit, and keep them as backup/battery equivalent.
At the present state of things, you're definitely right.
I'm talking about winter, where you can count on solar panels producing... nearly nothing.
This is a company here in Estonia sharing customers' monthly production numbers. This is a company trying to sell you solar installations, so they have no reason to show any numbers as lower than reality. I clicked through several customer experience pages, and most have ~30x less energy generated in December vs May.
The Nebraska comparison in your other reply to me doesn't work out because Nebraska is way further south. In December, the sun doesn't "rise" here as much as it "drags its' rotting carcass across the horizon". Okay, we're not as far north as something like Svalbard, but the angle of the sun during solar noon on December 21 (shortest day of the year in the northern hemisphere) is around 7 degrees. In Nebraska it stays around 25 degrees. While we technically get up to 6 hours of daytime even in December, it's usually overcast so average sunshine per day is about 30 minutes over the winter. And if it's not overcast, you can expect it to get cold fast, driving up usage.
So to go full solar (which I'm discussing as a thought experiment, I don't actually know anyone who wants to go FULL solar), essentially all the energy needs to be generated in about 7-8 months each year, because once the days start getting shorter, they go short REALLY fast. That's going to be a lot of H2 to store.
That's a reasonable suggestion, it's just that we're not burning anything clean like coal here, we're burning shale. It's comparable to lignite (if not worse) in CO2, but way more ash. Yes, shale the actual rock, not shale gas.
It's super frustrating.
Believable for shallow roof angles. Steep angles make a large difference, but it's still definitely a challenge for winter peak demand, and huge summer surpluses.
In Estonia vs Nebraska, 1000 wh/watt/year vs 1800 is a signficant disadvantage, and as you say, December averages 15 minutes/day of solar energy.
I did pick Nebraska for relatively north and sunny location, with ethanol substitute land use. It has 9-10x Estonia's winter production, and so Estonia definitely seems like a shithole solar location.
The H2 system still works for Estonia. I made this for you:
This report outlines the technical and financial feasibility of a self-sustaining
125 kW Solar / 90 kW Electrolysis microgrid in Estonia. Optimized for the high-latitude constraints of the Baltics, this system leverages a summer hydrogen surplus to subsidize a 24/7/365 1 kW baseload datacenter requirement.
1. Core System Configuration
2. Financial & Cost Assumptions
3. Annual Capital & Operating Expense
| Expense Category | Amount (USD) | |
|
| | Total System CapEx | $139,792 | | Annual Debt Service (5%) | $12,383 | | Annual O&M (1%) | $1,397 | | Total Annual Cost (A) | $13,780 |
4. Energy Production & Hydrogen Revenue
Estonia receives ~950 Peak Sun Hours (PSH) annually. The 125 kW array generates ~118,750 kWh/year. After accounting for the 1 kW baseload (8,760 kWh), the remaining ~110,000 kWh is directed to the 90 kW electrolyzer.
5. Winter Reliability Analysis (The "Dark-Month" Stress Test)
Unlike the Nebraska model, the Estonia configuration faces extreme seasonal variance.
Average December Yield: ~30–35 kWh/day (Enough to cover the 24 kWh/day baseload).
Worst-Case "Deep Cloud" Day: ~6–8 kWh/day (
).
The Survival Buffer:
Operational Status: The 90 kW electrolyzer will be completely offline from late October to early March, as all available photons are prioritized for battery health and the 1 kW load.
6. Conclusion: The "Latitude Tax" Equilibrium
This system represents the Saturation Point for Estonia at $2/kg Hydrogen.
Does the wind not blow in Estonia?
Conveniently stops blowing when it's cold and electricity demand skyrockets, or at least that's the excuse they give for why the prices shoot to the moon.
There's also at least one major shale power plant in repair every time it gets proper cold lmao
analysis for Nebraska that would apply for Estonia or Canada as well with only a few parameters changed. Free 24/7 baseload solar electricity if Hydrogen can be sold for $2/kg (equivalent to 25c/liter gasoline in range). https://lemmy.ca/post/59615631
Nebraska actually gets like 5-10x the useable solar power in the winter months compared to Estonia. We essentially don't see the sun from about nov to mid feb.
All of the H2 would have to be generated between spring and fall and stored for winter. Selling it and then buying it on-demand in the winter wouldn't work because fuels shoot up in price come winter. Cost of my wood briquettes tripled between July last year and February this year for an example, usually it at least doubles... And once I've seen them quadruple. Luckily it's a single house worth of solid fuel, it's easy to stockpile. I'm wondering how a couple of terawatthours worth of H2 storage would work.
To be clear, I'm not at all against solar or renewables in general, I just don't see any energy storage solutions that would work for my country if we tried to fix our shit as a nation. On an individual level it's doable, but payoff period is so long that it makes more sense to just keep using grid power.
analysis I replied with didn't require a separate heating solution, though heating 1000l or 2 of hot water in fall would be a great strategy for every home heating system. The reason H2 electrolysis (just sell it instead of using it for heat in winter, though that is also a solution) works even for "your solar shithole country" is the massive summer daylight. No H2 produced outside of the good months.
The issue with the H2 solution is that we still need electricity in the winter, especially as more heating is done with electricity than ever before. If you don't store it in the summer, you'll be buying it from other countries at 10-50x the summer price oftentimes.
The analysis I'm presenting provides 1kw or 1600mw of continuous 24/7 power at 0 electricity cost including 5% financing costs. Showing that this works in one of the most hostile places for winter solar in the world.
There are indeed a lot of practicalities that can improve upon this. Summer demand is typically 1/2 of winter peak demand. The H2 electrolysis system is there purely to monetize all electricity generated at 4c/kwh. Selling the massive summer surplus at 4c/kwh or more makes more money than H2. Selling the "free baseload" for anything at all makes money, or selling H2 for more than $2/kg.
Wholesale rates in estonia/regional market are 15c/kwh in winter. The model to provide 1600mw baseload takes about 200gw of solar (125kw per kw baseload), that still provides surpluses on average winter day, before accounting for less than 24/7 of peak usage, diverting heat to fall storage systems, using wind instead of all solar, trade with less hostile solar regions bidirectionally/seasonally, use EVs as even bigger battery buffer.
For one, the 1600mw peak isn't as relevant as the 27gwh/day peak = 1100mwh/hour baseload covered by battery = up to 30% smaller system. But even with original large system, there would be 10-20gwh/day in winter available to be sold at up to 15c/kwh profit. $1.5-$3M/day. The giant size of the solar would mean that electricity rates are 4c/kwh everywhere else in the region in other seasons, and H2 system is still needed to ensure 4c/kwh monetization even as it lowers rates everywhere around them. ie. a system this big forces permanent 4c/kwh wholesale electricity anywhere it can trade to for 3 seasons.
Even if there are much more profitable locations for solar than Estonia, the high cost of transmission still makes local systems pay off. Transmission links are a resilience option that is actually more expensive than more local solar, but pays off when neighbours don't adopt solar. OTOH, very small transmission lines work well with battery systems in that they can be trickle imported at 24 hour rate in anticipation of weather/demand/battery charge level rather than as a response to instant supply/demand imbalance surge.
The pricing is hourly and actually gets up to 5€/kwh which is the limit (this is where even modest sized batteries would help a lot) and is sometimes reached, but average was 19 cents in january and february. That does not include the transmission fee which is also several cents. Of course this is where batteries would help.
But mostly my musings about the required battery capacity were with the idea that we should produce our entire demand or more locally, all year round, because people get SUPER pissy when electricity prices go up (which they do any time we have to import electricity).
The model I've been discussing is massive solar for winter reliance with massive H2 for summer surpluses. Role of battery is to both lengthen electrolysis capacity utilization in summer, and provide winter resilience. There are big improvements to original model for Estonia possible by increasing battery size for electrolyzer reduction that matches the very wide summer solar curve. Translates to either 5.1% financing/ROI costs or $188/kw baseload profit at 5% financing. Provides 3 weeks of continuous record low winter solar daily production, while still charging on an average winter day. (Nebraska model benefits from more electrolyzers and less battery instead, due to more reliable winter)
By shifting the ratio to 125 kW Solar / 65 kW Electrolyzer / 363 kWh Battery (363 kWh is the 5.5-hour summer night requirement), you gain two massive structural advantages:
1. 24/7 Summer Electrolysis (The Profit Engine)
Previously, your 90 kW electrolyzer had to shut down at night because the 185 kWh battery was too small. Now:
Nightly Draw: 66 kW (Electrolyzer + Load)
×cross
×
5.5 hours = 363 kWh.
The Match: Your battery now perfectly fits the Estonian summer night. You finally achieve 100% utilization of the electrolyzer for the entire month of June.
Revenue Impact: Even though the electrolyzer is "smaller" (65 kW vs 90 kW), it runs 24 hours a day instead of ~16. Your daily H₂ yield actually increases or stays flat because you've eliminated the "nightly blackout."
2. Winter Resilience (The Survival Engine)
This is where the "Latitude Tax" starts working in your favor.
3. Updated Financial Estimates (with 35% Premium)
4. The "Zero-Cost" H₂ Check
Summary: The "Stability" Optimization
By trading 25 kW of electrolysis capacity for 178 kWh of battery storage, you have:
This is likely the most robust version of the Estonian model yet. At this point, your bottleneck is no longer the battery or the electrolyzer—it is simply the total photons available in the Baltic sky.
Not sure if you were taught wrong or misremembering, but giga is the standard notation for billion, and tera is trillion. Kilo, mega, giga, tera, ~~quad, quin~~, peta, exa.... They go on much farther than that, but at that point, just use exponential notation.
E: wrong notation form
If tera is trillion and kilo is thousand, tera is a billion times kilo.
I don't remember quad or quin, we had peta and exa here.
Derp, I knew that too.
Look at flowers and talk to people until it's sunny?
If it's cloudy, grains don't grow and you starve.
You can generate hydrogen from electrolysis.
Electrolysis efficiency is about 70% and you can store the hydrogen in pressurized underground caverns for a year or longer using another 0.12 kWh per kWh of hydrogen stored, which makes a total efficiency of around 0.6 kWh of hydrogen generation and storage for every kWh of electricity that you put in. (Source)
So if your electricity costs 6 ct/kWh (current LCOE of solar in many places), then hydrogen is gonna cost 10 ct/kWh to generate and store with current technology.
Currently, natural gas is around 5 ct/kWh, so solar would have to become a little bit cheaper to make it economically competitive.
Edit: to clarify, the 5 ct/kWh for natural gas is the gas alone; electricity from natural gas is more expensive than that (around 12 ct/kWh) and more expensive than solar.
What are you going to store hydrogen in to make this remotely viable? You lose like 60% of hydrogen within 7 days with current tanks and seals.
The new sodium batteries make this completely pointless from a cost and efficiency context
Hydrogen can be stored for years.
Hydrogen’s small molecular size facilitates permeation through many conventional materials, leading to leakage and potential embrittlement of metals. These characteristics create "unique" (read: expensive) engineering challenges.
And if you store it as a liquid that's when you enter the whole cryogenic problem, seeing it needs to be cooled to over - 225c
Even with advanced insulation, liquid hydrogen storage experiences unavoidable boil-off at rates of 0.3-3% per day, creating both economic and safety challenges.
And if you could make hydrogen via electrolysis, even with some uber wunderful unobtanium catalyst then you're still just waiting electricity that we can store far more efficiently in batteries (and with sodium batteries hitting the market there is going to be a huge revolution in battery economics and tech that will make lithium look like a drop in the ocean.
The problems are mostly solved already. You wouldn't use metals known for hydrogen embrittlement. Often times, you'd use something else, like HDPE or fiberglass that avoids this issue. Storage facilities can even be naturally occurring geological features, like salt caverns.
You would only use LH2 for specific cases, specifically where you are expected to use up the hydrogen quickly. But even this is changing, as self-refrigerating systems are being developed, allowing for very long-term LH2 storage.
We already can make hydrogen via electrolysis. This is a long-solved problem. Efficiency is not that relevant. The main limitation of batteries is that you simply cannot make enough of them. There are huge resource limitation problems. Meanwhile, hydrogen can be made from water and is effectively unlimited as a resource.
is home hydrogen a thing? i was wondering before, if it works in cars, why is it not in houses?
hydrogen scales well if you use big industrial setups, both for generation and for storage.
basically, bigger tanks are cheaper (consider higher volume/surface area ratio) and in fact the best tanks might simply be naturally occurring underground caverns. you can't have these at home.
There's a engineer that did it in his backyard. I'll see if I can find it when I get home.
https://www.youtube.com/watch?v=djg_l7cEtWg
https://hydrogenhouseproject.org/index.html
Technically it could work. However, traditional batteries make a lot more sense. Hydrogen makes some sense for a vehicle because it can be more energy dense (it actually only makes sense for large trucks). However, it has to be stored at cryogenic temperatures. In a place where you probably don't care about mass or space much, other battery technologies are far better, without the added cost of cryogenic cooling and having to deal with hydrogen, which leaks through anything.
You would store it as a pressurized gas in this scenario. You would only use liquid hydrogen in specific situations.
Hydrogen gas is really hard to store. It is tiny, so it's basically always leaking, no matter how good your seal is.
Underground caverns can store it for years. This is simply not true.
What do you mean it isn't true? It's a well known fact. It's just a proton and an electron, so it's absolutely tiny. There is almost no way to seal it perfectly, especially in gaseous form. It's always going to leak. Even for rockets this is an issue. You can make that amount relatively small, but it pretty much always has some loss.
Caverns may make sense for large-scale solutions, because the quantity is so large compared to the loss. Most people don't have massive caverns under their house though, nor do they have a need for that large of a quantity.
Even in tank form, you can store it for months. It is not much different than CNG.
Large-scale solutions matter too. The utility companies can utilize such a thing.
The question was specifically about home hydrogen. Yes, it makes sense for utility companies, as well as large vehicles, as I said before. It's a great solution to turn renewables into a shipable commodity. Home use though doesn't make sense. A regular battery has much better properties for home use.
The same applies for home hydrogen storage too. Compressed hydrogen is good for months. Another option would be metal hydrides which apparently last a long time too. The problem is that you simply cannot power your house entirely with batteries alone.
That sounds cheaper than battery storage (which at latitudes bigger than yours can get very expensive since there's little to no sun in the winter), and I'd assume more environmentally friendly than mining all that lithium as well.
How expensive is it to build out said caverns for this use, particularly if there aren't many natural ones available?
basically the caverns that are being considered/used for this are the same caverns that natural gas was extracted out of in the first place ... they clearly held some sort of gas fine for millions of years, so certainly they're gonna store a bit of hydrogen too.
Not to rain on your parade, but hydrogen and natural gas aren't really comparable for storage. The natgas molecule is 8x heavier and MUCH larger than a molecule of hydrogen. Just on the size alone, hydrogen can slip through just about everything and needs to be stored at cryogenic temperatures. I don't think rock is going to be as good of a storage media as you'd assume.
sure it's gonna leak but if the rate of leakage is slow enough, you can ignore it :)
Oh that makes sense.
We just don't have any natural gas production in Estonia lol. Perhaps the shale mines could be used. Unfortunately the biggest one had its permit extended till 2049 recently. Also I think they get filled with water naturally (they pump out a lot of dirty water), so I suppose the walls aren't actually completely sealed naturally.
yeah, geological availability might vary