jo melville
banner
jmelville.science
jo melville
@jmelville.science
climate tech (electrochemistry, industrial decarb, synthetic fuels, carbon removal, deep biogeochemistry, solar radiation management)
formerly: ARPA-E, Georgetown, MIT (PhD), UC Berkeley (BSc)
ask me anything: https://jmelville.science/ask/
🇺🇸→🇸🇬
this is a legacy of the exact kind of American supremacist thinking that assumes that if the US can't build it from scratch, it's not worth doing. I think the question I'm positing is something closer to, what does a world look like where the US is humbled enough to start stealing from others again
February 16, 2026 at 12:04 PM
ironically all the evidence I've seen is that China is also dramatically ahead of the US in advanced nuclear (certainly small modular reactors, and maybe soon nuclear fusion)
February 16, 2026 at 12:01 PM
I know some folks that are very bullish on dimethyl ether (DME) precisely because it's a fairly drop-in compatible fuel replacement for propane that can be synthesized directly from methanol, either from e- or biofuel sources!
February 16, 2026 at 11:59 AM
may be true now, but it won't be true forever. the US is well-resourced and at some point an intelligence agency with strategic vision will make a move.

easy to forget but for decades the *American* competitive advantage was that we had patriotic e.g. Chinese-Americans that would spy for the US!
February 16, 2026 at 11:55 AM
vanishingly few people seriously engage with the second-order implications of an American Century of Humiliation, one of which is being a global underdog for the first time in living memory
February 16, 2026 at 11:38 AM
a huge number of chemicals and reagents are byproducts of fossil fuel extraction and are implicitly subsidized by our fossil economy. not just plastic monomers but even non hydrocarbon things like sulfur from hydrodesulfurization or helium from fossil gas liquefaction!
February 16, 2026 at 8:53 AM
here in Singapore some local EPC firms are piloting containerized solid-state grid charging batteries to replace onsite diesel generators, which are the #1 source of operational emissions!
February 16, 2026 at 8:50 AM
capacity factor and storage cost are inversely related with an asymptote to infinity at 0%. (i.e. for EXISTENTIAL stuff, black start or ICU backup generators, you'll prob always have some diesel)

the 24/7 solar Ember Energy report actually explains this very well!

ember-energy.org/latest-insig...
Solar electricity every hour of every day is here and it changes everything | Ember
Batteries are now cheap enough to unleash solar energy’s full potential, getting as close as 97% of the way to delivering constant electricity supply 24 hours across 365 days cost-effectively in the s...
ember-energy.org
February 16, 2026 at 6:46 AM
my preferred framing for this truly last-mile backup fossil generation is that it's providing necessary optionality to backstop multiplicatively more VRE resources on the grid — a service it provides even if it never actually had to run! it's basically "blackout insurance".

bsky.app/profile/jmel...
even if you're burning fossil diesel, I'd love to see it reframed in terms of the renewable generation capacity enabled for every marginal unit of fossil backup used.

e.g. for every 1 MWh this diesel generator produces, it provides backup optionality for 100 (?) MWh of solar + wind on the grid
February 16, 2026 at 6:41 AM
it really, really depends on the specific capacity factor (implicitly, provided storage duration). a backup diesel generator that runs ~1% of the time is providing ~seasonal energy storage, unrealistic for batteries to meet regardless of freq management. 10% c.f. is weekly~monthly energy storage.
even if you're burning fossil diesel, I'd love to see it reframed in terms of the renewable generation capacity enabled for every marginal unit of fossil backup used.

e.g. for every 1 MWh this diesel generator produces, it provides backup optionality for 100 (?) MWh of solar + wind on the grid
February 16, 2026 at 6:41 AM
I think both that that's likely (in the context of total gas replacement) and that Kevin's question w/r/t gas power specifically is still valid in the context of e.g. frequency management on high-VRE grids (like the controversial Iberian peninsula blackout)
February 16, 2026 at 6:22 AM
—William 02:26
February 14, 2026 at 3:45 PM
193/ my problems not so bigdeal ✌️😺✌️
February 14, 2026 at 5:53 AM
Reposted by jo melville
It is both the case that the RCP8.5 emissions scenario is completely implausible and that we cannot preclude >4C warming by 2100 (though I’d argue SSP5-8.5’s 4.7C by 2100 is rather outlandish at this point). journals.sagepub.com/doi/10.1177/...
An assessment of current policy scenarios over the 21st century and the reduced plausibility of high-emissions pathways - Zeke Hausfather, 2025
The literature on current policy scenarios has become increasingly robust in recent years, with a growing consensus that the central estimate of 21st century wa...
journals.sagepub.com
February 12, 2026 at 4:19 AM
192/ wheat for sheep
February 12, 2026 at 9:34 AM
191/ 🐝🍯🏘️
February 12, 2026 at 9:34 AM
190/ 🐱🍎
February 12, 2026 at 9:33 AM
189/ survivor
February 12, 2026 at 9:32 AM