Yves here. Aside from the headline point that King Coal is not going away as quickly as some would like, further down this piece contains additional points that will not make climate change combatants happy. First is that coal is serving as a swing power source for many US utilities. Second is that (quelle surprise!) data center demand is resulting in electricity consumption exceeding many forecasts.
By Tsvetana Paraskova, a writer for Oilprice.com with over a decade of experience writing for news outlets such as iNVEZZ and SeeNews. Originally published at OilPrice
- Coal still holds about 16% share of electricity generation.
- Coal is declining, but not at the fast pace environmental campaigners and enthusiasts would have liked to.
- But coal-fired power generation could rise in the summer, especially if heatwaves hit areas where wind power cannot provide the additional power supply.
Despite a continuous decline in U.S. coal power generation, the share of coal in America’s electricity mix is still above 15%, more than any renewable energy source.
All renewable energy sources combined—wind, solar, hydropower, biomass, and geothermal—surpassed coal-fired generation in the U.S. electric power sector for the first time in 2022. But coal still holds about 16% share of electricity generation, more than wind’s share of around 11%, hydropower’s 6%, or solar power’s 4% share of the electric generation mix.
True, coal generation and share have been declining in recent years – thanks to a surge in renewables and a higher share of natural gas-powered electricity due to rising production and falling gas prices. But coal is still playing a role in providing reliable baseload electricity and while falling, its share and contribution to the U.S. power system isn’t negligible at all.
Coal is declining, but not at the fast pace environmental campaigners and enthusiasts would have liked to.
The Biden Administration aims to make the U.S. power grid zero-emission by 2035. But this would be difficult to achieve, considering that currently fossil fuels – mostly natural gas and coal – provide 60% of the total U.S. electricity generation. Last year, gas accounted for 43% and coal for more than 16%.
Early this year, coal’s share held above 15%, although coal power generation fell between January and April to the lowest level in four years, per LSEG data cited by Reuters columnist Gavin Maguire.
Coal consumption typically falls in the spring and autumn – the so-called ‘shoulder’ season – when demand for heating and cooling is at its lowest.
But coal-fired power generation could rise in the summer, especially if heatwaves hit areas where wind power cannot provide the additional power supply.
Moreover, operators have planned fewer coal capacity retirements this year, per EIA data. Operators plan to retire 5.2 gigawatts (GW) of U.S. electric generating capacity in 2024, with coal and natural gas jointly accounting for 91% of the planned capacity retirements in the United States this year. The total capacity planned for retirement would be 62% lower compared to last year, when 13.5 GW was retired, and the least in any year since 2008.
After 22.3 GW of U.S. coal-fired electric generating capacity retired over the past two years, coal retirements will slow down in 2024, the EIA said in February. The 2.3 GW of coal-fired capacity scheduled to retire accounts for 1.3% of the U.S. coal fleet that was in operation as of the end of 2023. Coal retirements are scheduled to rebound in 2025 when operators expect to retire 10.9 GW.
The U.S. is now retiring coal capacity every year, but some areas depend on coal for their power generation more than others, while the expected surge in electricity demand due to data centers to support AI technologies will also require a stable power supply.
Five U.S. states rely on coal for more than half of their electricity generation. These are North Dakota, Missouri, Kentucky, Wyoming, and West Virginia, Reuters’s Maguire notes.
Moreover, data centers have seen such explosive growth that they are taxing utilities beyond what soaring power demand is calling for.
Some utilities in the eastern and southern parts of the U.S. are proposing build-outs of new natural gas-fired capacity alongside renewables to support the growth in electricity consumption coming from data centers. Others have planned to delay the timeline for retiring coal-fired capacity to ensure grid reliability.
For example, Kansas City-based utility Evergy said in June 2023 that it would retire coal operations at its Lawrence Energy Center only in 2028, compared to earlier plans for end-2023 retirement.
“Our service area is experiencing some of its most robust electricity demand growth in decades, including very large projects like the Panasonic electric vehicle battery manufacturing factory and the Meta datacenter, as well as broad-based economic development in both Kansas and Missouri,” Evergy’s president and CEO David Campbell said last year.
21 comments
Comments are closed.
This article concentrates on coal meeting seasonal demand. However it will also be required for load following , renewables being unable to provide this service, plus system support services , which again renewables need but do not provide and batteries are insufficient. Gas-fired generation can take over these requirements , but diversity of fuel makes good security sense.
Its important not to confuse capacity with actual generation – coal plants can be kept in semi-mothballed condition for many years as a reserve in case of, for example, a particularly cold winter or a failure of other power sources. So, you have a theoretical capacity without necessarily using a lot of coal. China, for example, is significantly increasing coal generating capacity without significantly increasing coal use (the reasons are complex, but that’s a discussion for another day).
Also, while coal’s contribution to electricity generation is in slow decline, its in much sharper decline for all other uses as gas and renewables have expanded. The real story with electricity generation in the US is how methane gas CCGT plants displaced other sources of generation due to fracking – gas plants used to be for peak demand, but lower costs meant they sharply displaced coal for baseline power too. This is not necessarily bad news for renewables as a gas/renewables mix works much better for load balancing than coal/renewables.
The increased use of electricity for data centres has little impact on coal as an electricity source (with a few exceptions due to local circumstances) – its mostly boosting methane use as gas plants are far more efficient as a form of peaking supply during emergencies. In this, its frequently displacing distillates, not coal as a source of electricity in localised circuits during extreme peaking load periods. Data centre developers usually prefer renewables as a source of power as its cheaper and their greater demand flexibility means they can utilise the most cost effective sources (the so called ‘duck curve‘) over a 24 hour cycle, which is now solar in most grids. Increasingly, solar is producing very cheap excess power during daylight hours (depending on region), which is proving a major boon to anyone with flexible demand. In other grids, wind power is producing similar duck curves, usually at night.
If Google needs coal-based electricity to run their data mining operations 24/7, then coal will be around for a long time.
A lot of people don’t want to hear it, but nuclear energy is the only way to generate enough electricity to meet our current needs while not pumping carbon into the air. We should be copying the French model where they get 70% of their energy from nuclear power. Nuclear power has its own set of risks, but it’s the cleanest dirty shirt right now. Climate change is the much greater emergency.
For some definition of “needs”.
(Not to pick on you individually. R. Bryce about blew my head up with his casual assumption in his piece this weekend that what passes for AI right now a) is necessary and naturally important, and b) should get power generation- indeed, societal- priority.)
So long as you have a magic wand to conjure up new reactors. No one in the west has been able to build new reactors on-time and on-budget in decades – Flamanville delayed by a decade, Olkiluoto by a decade-and-a-half, Hinkley Point C is on-track to bankrupt EDF, and Summer/Vogtle actually did bankrupt Westinghouse. Even China can’t build reactors fast enough to meet projections and are being left behind by renewables+storage buildout.
I think China has 20 reactors under construction with 44 more in the planning stages. If anyone can learn to mass produce nuclear reactors safely it would be China. They are using the Westinghouse design which is pretty safe. The US should contract with the Chinese to build reactors here. That would solve the cost overrun problem to a large degree, improve quality and safety as well. I do not want a surgeon to be operating on me who only does one operation ever five years. I do not think a nation should be doing nuclear if they do it once every 20 years.
Or the South Koreans. They built their nuclear fleet for about $2500/kW, which is about one-fifth the price of the Vogtle plant in Georgia. And they did it recently enough that most of the people who did the work are still around. [The people who built the bulk of the US and French nuclear fleets did their work decades ago. Most them retired many years ago, or have even died.]
The opportunity costs of new nukes makes their ability to provide significant electric power in the near future problematic at best. These include enormous up front capital costs, resource costs (including energy), and time constraints. IMO, society simply doesn’t have the time or resources to shift to a nuclear economy. My concerns also include what would happen to thousands (tens of thousands?) of irradiated fuel pools/cannisters when the lights go out.
Funny, how everyone knows about French reactors & new generation miracle reactors, here; promised since the 1960s. But it invariably results in gas peaking plants & bailing out dilapidated fission reactors & 1960’s vintage Foster-Wheeler coal fired plants?
https://www.catf.us/2023/07/2022-french-nuclear-outages-lessons-nuclear-energy-europe/
https://www.banque-france.fr/en/publications-and-statistics/publications/energy-balance-2022-crisis-nuclear-power-generation-came-worst-possible-time
Nuclear is crap for load balancing, even worse then wind. Neither can increase to meet peak demand, and if you need to load balance when having to much power you can safely shut off part of your wind parks. That is harder to do with nuclear plants.
While wind and solar varies with season and weather that can at least be predicted, nuclear varies with safety concerns which are to their nature unpredictable. France shut down much of its nuclear in 2022, losing 25 % over the year (282 TWh produced instead of an average of 395 TWh), which together with more expensive of gas made electricity prices in Europe go up, up, up. Ukraine’s current power situation also shows how reliant nuclear power is on other power sources to even be able to shut down safely, which given the increase in storms and extreme weather may become an issue.
And added to being unreliable, it is slow and expensive to build. You get more kWh per dollar if you build renewables, and you get them faster.
Of course using less tends to be even faster and cheaper. And done the right way makes for a better society. Banning crypto and AI as wasteful, useless crap would be a great first step.
California Air Resources Board has a related issue. They are considering a requirement that all trains in the state become all electric. That is not likely to succeed for numerous reasons. For example, look at how much the California economy depends on import and export carried by freight trains between ports and destinations around the country. It is one thing to have a smaller, visible urban network carrying people and running via pantographs, and quite another to have mile-long trains going over the Cajon Pass or elsewhere. The infrastructure development alone would be a monumentally
expensive tasklucrative grift and the goldenish state does not have a good recent track record.Take a look at the charts at EIA and decide for yourself how coal is doing. https://www.eia.gov/todayinenergy/detail.php?id=37692 The decline in consumption from ~2005 to the end of this chart is actually amazing given that were talking about massive power plants being taken offline and replaced with either gas or renewables. The current stable of 200+ coal plants in the US will take 20 years to retire, which seems like forever, but is shockingly fast given the process of phasing them out. I’m no fan of natural gas, but coal-generated power has to go ASAP.
The other good news is that renewables are the most popular option for new capacity. It makes sense given the low cost of wind and solar, but it’s heartening to see wind and solar coming online at utility scale. Massive declines in battery price will make PVs / Batteries cheaper than gas in the near future. If you look at wind and solar combined, they have outpaced coal already, and new capacity is overwhelmingly solar, batteries and wind. Take a look – https://www.pv-magazine.com/2024/02/19/us-utility-scale-solar-capacity-additions-hit-36-4-gw-in-2023/
Good charts here as well. It looks like wind power seasonality compliments solar nicely. That helps make it all work. – https://www.eia.gov/todayinenergy/detail.php?id=61242
I do wish we lived in a society that depended less on cars than we do, and generally lived in smaller, more energy-efficient housing, with less sprawl, but all that is academic if we burn the whole planet down. The good news right now is that the people working on renewables have won. Renewables have become cheaper than fossil fuels and the transition is underway.
Unfortunately, when you look at the coal picture globally, coal is doing “just fine”: https://ourworldindata.org/grapher/coal-consumption-by-region.
Decreases in coal consumption in the US and Europe have been more than offset by increases in Asia, with China leading the way. China is currently responsible for more than 50% of global coal consumption (per https://www.worldometers.info/coal/coal-consumption-by-country/), even though it only holds 18% of the world’s population. And India is still ramping up. We’ve got a LONG way to go before coal disappears.
“The Biden Administration aims to make the U.S. power grid zero-emission by 2035. But this would be difficult to achieve, considering that currently fossil fuels – mostly natural gas and coal – provide 60% of the total U.S. electricity generation. Last year, gas accounted for 43% and coal for more than 16%.”
As I read about California, there are big problems when weather/sunlight dependent renewables exceed 30% of the production. Hydro is good for balancing, but California needs to use reservoirs to go through multi-year droughts. Perhaps EV charging could be fashioned to balance the supply during day cycle, but not during seasonal cycles etc.
In the meantime, nuclear planning is lethargic. Timeliness and prices of RosAtom show what happens if you build those power plants on regular basis, the fact that Americans are not able to match it after decades of nuclear slumber is not a proof that what Russians and Chinese do cannot be reproduced. And if such cooperation is distasteful (electricity with a pungent odor of authoritarianism?), one should consider common projects with South Korea.
https://news.cgtn.com/news/2024-04-12/China-s-small-nuclear-reactor-starts-installing-digital-control-system-1sJs3fcLsGY/p.html
April 12, 2024
China’s small nuclear reactor starts installing digital control system
The world’s first land-based commercial small modular reactor (SMR) has started installation of its digital control system (DCS) in the country’s southernmost island province of Hainan.
The reactor, also known as the Linglong One, is a multi-purpose, small modular pressurized water reactor, which was domestically developed by the China National Nuclear Corporation.
The DCS is regarded as the “nerve center” of the operation of a nuclear power plant, and the reactor’s DCS adopts two domestically-developed platforms, said Deng Xiaoliang, deputy general manager of Hainan Nuclear Power Co., Ltd.
The reactor is an achievement of independent innovation after Hualong One, a homegrown third-generation nuclear reactor.
Compared with the bigger Hualong One, the 125 MW Linglong One reactor is designed for urban heating, urban cooling, industrial steam production or seawater desalination…
https://news.cgtn.com/news/2024-02-23/Construction-starts-on-second-phase-of-Zhangzhou-nuclear-power-project-1rqfeSxLMmk/p.html
February 23, 2024
China starts construction on second phase of Zhangzhou nuclear power project
China on Thursday commenced the construction on the second phase of the Zhangzhou nuclear power project using Hualong One reactors which are domestically developed third-generation reactors, in the city of Zhangzhou, east China’s Fujian Province.
The first four Hualong One units have been completed and put into operation, according to the China National Nuclear Corporation (CNNC).
The Zhangzhou nuclear power project is designed to consist of six nuclear power units, each with a 1-million-kilowatt power generation capacity.
The first phase of the nuclear power project includes two units, with the first unit expected to be put into production this year and the second unit under preparation for cold tests.
Hualong One reactors were jointly designed by China General Nuclear Power Group and CNNC…
https://news.cgtn.com/news/2024-01-06/How-the-world-s-first-fourth-generation-nuclear-power-plant-works-1q8JzrGNrj2/p.html
January 6, 2024
How the world’s first fourth-generation nuclear power plant works
By Zheng Yibing
The world’s first fourth-generation nuclear power plant, Huaneng Shandong Shidao Bay Nuclear Power Plant in eastern China’s Shandong Province, went into commercial operation on December 6, 2023 and has been running well, according to officials at the plant.
The power plant has drawn global attention as it adopts High Temperature Gas-Cooled Reactor-Pebble-bed Module (HTR-PM), which is claimed to be able to steer away from a meltdown or leak of radioactive materials even in extreme conditions…
Early in the Industrial Revolution, steam engines consumed wood and exhausted the local forests. But coal saved the day.
Now, in the early 21st Century, the wonders and mysteries of ‘renewable’ energy have stepped in to help paint electric power Green, much like painting the white roses in the Queen of Hearts’s garden.
How much of the biomass displacing coal was obtained by converting trees, even sections of forests of old growth trees, into ‘renewable’ wood pellets burned in place of coal? The difference is that the forests exhausted to run early 21st Century electric generators need not be so local as in the early days of the Industrial Revolution, and there are machines to convert wood into convenient bags of pellets. Unfortunately the ‘harvest’, and transport of trees from forest to factory and some of the processes for converting trees into wood pellets, and transporting wood pellets across the oceans — uses fossil fuels. The correct Green thing to say is that wood pellets [misnomer?] are manufactured from ‘waste wood’ — slash left after ‘harvesting’ wood for lumber, sawdust, ‘waste wood’ ill-suited for lumber, and various sources of biomass from hemp, bamboo?, agricultural wastes and ‘other’ ‘renewable’ stuff.
Brett Christophers’ The Price is Wrong: Why Capitalism Won’t Save the Planet provides an in-depth account of why renewables are less profitable than conventional energy generators. In absolute terms solar panels and wind farms provide cheaper electricity than conventional sources, but several factors produce phantom costs for the “cheaper” options. These include the need for more land than a conventional power plant, grid connections to the remote areas easiest to install renewables, a perverse market-ization of generating power that can make the cost of electricity go negative from time to time. Etc.
Since the bulk of renewables’ costs are up front (the installation, the land, the grid connection) banks are essential to successful projects. Nothing makes a banker decline a loan faster than a low-profitability, uncertain project. Petroleum drilling demands less land and is significantly more profitable. Coal-fired plants already exist, and the cost of fuel is modest. Conventional, carbon-based energy has more certainty, and orders-of-magnitude more subsidies currently, too.
The bottom line: unless there is government support (guaranteed profits like those military contracts, feed-in tariffs, price guarantees, free grid connections, etc.) there will be no surge in renewable installations.
FALSE.
Read the last line of FERC Jan 2024 report:
Brief News Update & Analysis
LATEST FERC AND EIA REPORTS SHOW SOLAR
OFF TO A STRONG START IN 2024
INSTALLED SOLAR CAPACITY NOW EXCEEDS
NUCLEAR POWER AS WELL AS HYDROPOWER
SOLAR ELECTRICAL OUTPUT GREW 20.5% IN JANUARY
Contact: Ken Bossong, 301-588-4741
Washington DC – A review by the SUN DAY Campaign of data recently released by the Federal Energy Regulatory Commission (FERC) and the U.S. Energy
Information Administration (EIA) confirms that 2024 has begun with strong growth by solar in both capacity additions and electrical generation. This is
projected to continue for at least the next 2-3 years.
Federal Energy Regulatory Commission:
In its latest “Energy Infrastructure Update” (with data through January 31, 2024), FERC reports that solar accounted for 2,527 megawatts (MW) of new generating capacity in the first month of this year – or 87.3% of the total. That is the second highest monthly total ever reported for solar, behind only the prior month when 4,979-MW were added.
The new solar capacity reported for January was accompanied by 320-MW of new wind capacity as well as 44-MW of natural gas, 2-MW of oil, and 3-MW classified as “other.” Thus, renewable sources accounted for 98.3% of capacity additions in January.
The new solar and wind brought the total available installed generating capacity of renewable energy sources (i.e., also including hydropower, biomass, and geothermal) up to 376.33 gigawatts (GW) or 29.17% of all U.S. generating capacity. That is more than the installed capacity of either coal (207.15-GW) or nuclear power (103.27-GW). [1]