While renewables will play a large role in decarbonizing electricity, there is also a need for clean firm generation. Advanced reactors are a promising technology to fill that gap, and in a piece today we take a look at economics of @NuScale_Powerthebreakthrough.org/issues/energy/…
1/12
To be competitive in the short-term, advanced reactors like NuScale need to be reasonably cost-competitive with natural gas – which currently fills the role of firm, dispatchable generation. We compare the two based on their levelized cost of energy (LCOE). 2/12
The LCOE of nuclear, it turns out, is very sensitive to the discount rate used, as it involves a very high upfront investment with very long-term returns over a ~60 year lifetime. Standard LCOE calculations – such as those from @Lazard – use a rather high 10% discount rate. 3/12
Here is how the LCOE of nuclear compares to natural gas across different discount rates. We look at three nuclear cost scenarios (stated NuScale cost, NuScale assuming 50% cost overruns, and conventional nuclear) and three gas scenarios (low, base, and high gas price): 4/12
Using NuScale's base cost estimate and the reference gas price, NuScale cost-competitive at discount rates less than 5%. This is 7.5% for high gas prices and 2.5% for low gas prices. 5/12
We can also look at the required subsidy – or implied carbon price – needed to make NuScale competitive with natural gas across different discount rates: 6/12
If advanced nuclear received a production tax credit of $25/MWh similar to wind – and, importantly, assuming it can be built on time and on budget – it would be quite competitive with natural gas at discount rates high enough (8%+) to attract significant private capital. 7/12
At the same time, there is a case to be made that infrastructure investments to decarbonize the economy should use a much lower discount rate than is common for private capital investments. After all, high discounting would lead to sub-optimal levels of decarbonization. 8/12
Using a government/public sector discount rate of 3% or so makes investing in nuclear much more attractive given the long lifetime of the projects. 9/12
This still depends on ability of advanced reactors to be built reasonably on time and on budget. Their small and modular nature should help overcome some of the cost overruns that have plagued massive bespoke reactor projects in recent years (e.g. Vogtle), but we shall see. 10/12
This study came from Andrew Fletcher's summer Breakthrough Generation fellowship. Details of the analysis can be found in the article, and in the table below: 11/12
Advanced nuclear is promising and fills an important decarbonization need – though it is also not a panacea. More federal support for early-stage demonstration and deployment will be critical to help the industry drive down costs and prove out their technology at scale. 12/12
• • •
Missing some Tweet in this thread? You can try to
force a refresh
The first quarter of the year is off to an exceptionally warm start, as I discuss in a new Q1 State of the Climate Report over at @CarbonBrief:
⬆️ Warmest Jan, Feb, March, and April (to date) by ~0.1C
⬆️ 2024 on track to be warmest or second warmest year carbonbrief.org/state-of-the-c…
Based on first three months of the year and the current El Nino / La Nina forecast, we expect 2024 to be similar to or slightly warmer than 2023. With only 3 months in we can't know precisely where the year will end up, but its virtually certain to be at least the 2nd warmest:
While the uncertainty bars have narrowed as each new month of data has come in, our @CarbonBrief projection of 2024 annual temperatures has remained mostly unchanged since the start of the year:
It wasn't only the warmest February on record – this past month saw the largest anomaly (change from the 1850-1900 preindustrial baseline) at 1.79C of any month on record, beating out December (1.77C) and September (1.73C) 2023.
The past 12 months are 1.56C above preindustrial.
Here are the temperatures for Februaries since 1940 in the ERA5 dataset, with 2024 highlighted:
Despite today's grim milestone with the world passing 1.5C over the past 12 months, I see some reasons for cautious climate hope.
We stand both on the brink of severe climate impacts, but also on the brink of a rapid energy transition away from fossil fuels.
A decade ago global emissions were skyrocketing, and many thought we were heading toward a particularly dark climate future where the 21st century would be dominated by coal and global emissions would double or triple by 2100.
But something began to change: rather than continue their meteoric rise, global emissions of CO2 from fossil fuels and land use began to flatten out.
Flat emissions still cause CO2 to accumulate and the world continuing to warm, but not as fast as we previously feared.
Over geologic time the amount of carbon in the atmosphere is determined largely by the balance of volcanism and weathering of silicate rocks (which largely come from volcanoes!).
Today around a billion tons of CO2 is removed from the atmosphere annually by natural weathering.
Enhanced rock weathering seeks to speed up this process, applying ultra-fine basalt rock dust (or other silicates) on farmland to achieve dissolution in years rather than centuries.
Global temperatures in October smashed the prior monthly record by 0.4C, and were ~1.7C above preindustrial levels.
It wasn't quite as gobsmacking as September, but still comes in as the second most anomalous month in what has been an exceptionally hot year already.
Here is a comparison of October 2023 compared to all prior Octobers in the two leading reanalysis products: ERA5 and JRA-55. Note that HadCRUT5 is used here to help estimate warming since preindustrial:
We can really see the past few months stand out if we look at absolute temperatures (rather than anomalies):
The claim by Hansen et al today that climate sensitivity is 4.8°C ± 1.2°C per doubling CO2 is just as plausible as the claim by Cropper et al four days ago that its 2.8°C ± 0.8°C.
Across hundreds of different studies, and our best estimate remains somewhere between 2C and 5C.
Given all the conflicting estimates, I'd strongly advise folks against glomming onto any single new study (particularly if it informs ones priors that sensitivity is high or low). Instead, we should synthesize all the different lines of evidence: agupubs.onlinelibrary.wiley.com/doi/full/10.10…
I should really update this at some point as its ~5 years out of date, but I put together a timeline of all the published studies on climate sensitivity in the literature here: carbonbrief.org/explainer-how-…