US power sector carbon emissions intensity drops to lowest on record (because EPA forced coal power closure)

From the COLLEGE OF ENGINEERING, CARNEGIE MELLON UNIVERSITY and the “coal power plants saved your butt in the winter of 2017/18” department.

The former Indiana and Michigan Coal power plant (Now Tanners Creek generating station) on the Ohio river at Lawrenceburg, Indiana. 995 MWe nameplate capacity. Forced to close by EPA regulations, shut down on May 31, 2015. No replacement plant will be built. Photo: Anthony Watts

PITTSBURGH — (April 4, 2018) Mitsubishi Hitachi Power Systems (MHPS) and Carnegie Mellon University (CMU) today announced the release of the 2018 Carnegie Mellon Power Sector Carbon Index, at CMU Energy Week, hosted by the Wilton E. Scott Institute for Energy Innovation. The Index tracks the environmental performance of U.S. power producers and compares current emissions to more than two decades of historical data collected nationwide. This release marks the one-year anniversary of the Index, developed as a new metric to track power sector carbon emissions performance trends.

“The Carnegie Mellon Power Sector Carbon Index provides a snapshot of critical data regarding energy production and environmental performance,” said Costa Samaras, Assistant Professor of Civil and Environmental Engineering CMU. “We’ve found this index to provide significant insight into trends in power generation and emissions. In particular, the data have shown that emissions intensity has fallen to the lowest level on record, as a combination of natural gas and renewable power have displaced more carbon intensive coal-fired power generation.”

The latest data revealed the following findings: U.S. power plant emissions averaged 967 lb. CO2 per megawatt-hour (MWh) in 2017, which was down 3.1 percent from the prior year and down 26.8 percent from the annual value of 1,321 lb CO2 per MWh in 2005. The result for 2016 was initially reported as 1,001 lb/MWh, but was later revised downward to 998 lb/MWh. For the complete findings of the Index, please visit


“The power industry has made significant progress in reducing emissions for over a decade, as new technology, state and federal policies and market forces have increased power generation from natural gas and renewables, and decreased power generation from coal. As this Change in Power continues, the Carnegie Mellon Power Sector Carbon Index will not only report the results but also provide analysis of the underlying reasons for the changes we’re seeing,” said Paul Browning, President and CEO of MHPS Americas. “Our team at MHPS is proud to support this important work by Carnegie Mellon researchers.”

This year, CMU announced enhancements to the Index. New regional information from within the US will be reported, allowing for greater insight into the impact of regional trends on fuel types, usage, and emissions. In addition, the index will begin to incorporate emissions data from other countries across North and South America. As the Index continues to expand, it will serve as a source of objective insight regarding emissions trends across the Americas for policy makers, regulators, utilities, industry analysts and the public.


“We are excited to regionalize and internationalize our work on the Power Sector Carbon Index,” said Ines Azevedo, Professor of Engineering and Public Policy, and co-Director of the Climate and Energy Decision Making Center, “and to help educate and support the decision making process regarding emissions reductions.”


What is notable to me in the second graph, is the increase in natural gas to offset the loss of coal power. This is thanks to shale gas fracking. Solar and wind still remain just a fraction of the US total generating capacity, though wind has made some gains. Unfortunately, wind power is nowhere nearly as reliable as coal, hydro, and nuclear power – which enviros attack relentlessly.


newest oldest most voted
Notify of
Clyde Spencer

Do those prominent spikes in natural gas represent Winter heating? Basically, it appears that natural gas has just replaced the use of coal.

Tom in Denver

No those spikes are of electricity generated not of natural gas used.I would venture to guess that represents electric loading from AC usage in the summer


I did a quick lookup and couldn’t find the answer. As a former power-plant engineer, our system (VA, WV, OH, IN, KY) was a winter peaking area. The entire US — I don’t know.


If I read the chart correctly the spike seems to coincide with the end of the 4th quarter and beginning of the 1st quarter of each year. I was noticing how one could read the seasonal energy usages from the curves of coal energy production, yet none of the others seems to indicate seasonal spikes other NG which seems to infill the needed winter production needs.
Was the NG curve actual electrical production only, or was it all NG usage in the energy sector? If it was all NG usage then business and residential direct heating could account for the larger NG winter spike.


I’ll correct myself, after re-examining the chart I noticed I was reading it incorrectly. Tick marks indicating years would have been helpful. The spikes do appear to be summer loading.
Yet I am still intrigued as to why only look at CO2 production from the electrical power generation portion of the US energy usage? I know of no residential home that uses NG which incorporates CO2 capture or sequestration, and I use it for everything but my oven (I live near the beach in S. CA so no A/C is needed).


Didn’t want to spend much effort, but did find the below link. It does show monthly US electric usage for the past couple yrs. The highest months were July, August and January in that order, but not the huge “spikes” shown in this post’s graph. A number of pdfs there for viewing:

Please note from the graph that wind generation actually declines during peak load periods.

M Courtney

If peak load comes from very cold weather needing heating or very warm weather needing air conditioning then… that would be expected.
Both weather patterns are caused by stable weather systems (normally blocking high pressure areas).
And the thing about weather systems that don’t blow away is that they don’t have much wind.

Replacing coal with gas has provided virtually all of the reduction in U.S. CO2 emissons over the past 10 years.


All without CO2 taxation….

Tom in Denver

The reason Gas has replaced coal for electricity is not because of any Obama mandate, it is because the price of gas has dropped from $8/mcf to less than $3/mcf since the advent of shale gas development and multi-stage horizontal fracturing.

True, but there is a second reason also. The average age of the US coal fleet is 42 years, and the average retirement age is 48. About 1/3 of the coal fleet MW capacity was due for normal retirement between 2013 (when I last looked up the details in 2014) and 2025. The baseload replacement is all CCGT (except for troubled nuclear Voglte 3&4) not just because gas prices are low, but because CCGT is less than half the capital per MW, about half the time to bring on line, and from 1.5x (HELE USC coal) to 2x (regular coal) thermal efficiency. Gas wins even at about 2x its current price against even the best low cost Powder River Basin coal.


Coal has one advantage over natural gas in that it can be stockpiled onsite at most large gnerating plants. Natural gas storage usually comes from hundreds of miles away via pipeline. The Russian and Chinese intelligence agencies know that vulnerability of US electricity generation. In a cyberwar, their hackers will target not just the power grids but the pipeline infrastructure too.


You are partly to mostly correct, Tom, but the MATS rule promulgated by the last administration was responsible for a large wave of coal plant retirements even before the full impact of the cheap gas supplies hit. Another factor are the continued subsidies/mandates for renewables that, in spite of the relatively low amount of generation, lower the peak prices in the market.

“…critical data regarding energy production and environmental performance…” And then, they go on to talk about CO2 emissions. But, environmental performance is NOT the same as CO2 emissions – until they actually prove a link between CO2 and net negative impact of CO2 on the environment.

It beats me how they know the emissions from all that wind and solar capacity. And they know it to an accuracy of <1 lb. CO2 per megawatt-hour (MWh). Amazing accuracy for variable generators that have massive carbon footprints, with all that steel, concrete, rare earths etc and all the extra emissions from backup operation.


To make these cheerleading claims, only direct emissions are used. I these claims, if the total life cycle carbon footprint is used for wind, it’s close to carbon parity with natural gas. Which is why Google dropped the idea of using renewable power to exclusively run some of their server facilities, it simply costs more and doesn’t reduce the carbon footprint.

Bruce Cobb

The anti-coal carbonistas love to try to ignore Obama’s 8-year coal-killing reign of terror against coal for the big drop in coal use. Sure, it would have dropped some, but not nearly as much.

Tom in Denver

Bruce, that is simply not true, gas overtook coal as the primary electricity fuel long before Obama came out with his EPA initiative against coal. The reason is because the cost of natural gas dropped to 1/3 of what it used to be, because of shale gas development.

Bruce Cobb

You are missing the point. It is a given that the cost and availability of NG rose, which hurt coal. But Obama’s anti-coal policies acted as a double-whammy, punishing coal for “spewing carbon” into the atmosphere. The fact is that we do need both coal and gas. Coal is a natural competitor to gas, and helps keep gas prices from spiking, as they can do when demand is high.


Can’t quite agree with you, Tom. Coal/gas parity in electric generation did not occur until about 3 years ago.


I agree with Tom. Carbon dioxide aside, natural gas is a much better fuel than coal. It is more flexible, cleaner (no ash to dispose of), has a faster ramp up rate, etc. Gas generators doubled in efficiency over the last 10 years, while gas prices dropped 2/3. Utilities found that peaker plants using gas were suddenly cheaper to run than coal plants. Crazy. My utility actually retired an old gas plant with 10 years to go on its life span because it made more sense just to rebuild the thing with the new technology – it was that much more efficient and cost effective.
Did Obama had some effect? Sure, but that was just icing on the cake. Heck, if I can cut carbon dioxide emissions 60% and save money, why the hell not?
Coal will continue to decline as a part of the fuel mix in the U.S. for the next 20 years. I don’t see that changing no matter who is in the white house.


Oh, and to add one thing everyone forgets – demand destruction due to LEDs. We have had flat to declining electricity usage for years, meaning many plants are just superfluous.


A lot of people lost their jobs in the process…and for nothing more than policy error with coordinated regulatory and science distortion..


The first (initial) reductions are the easiest, especially given the shale gas revolution that came inspite of Obama Admin resistance and not “because of.”
Further reductions without impacting grid reliability will require nuclear power ramp up.
As South Australia has demonstrated, going too far on the intermittent renewables, solarPV and wind, brings unreliable grid delivery. And without reliable grid power, industry leaves.

And this is as successful a strategy as having a city employee smash all the windows on Main St to spur economic activity and create jobs…

Gerry Cooper

It’s economics with dual fuel power stations burning the cheapest fuel – recently this is natural gas.

Stephen Singer

The categories, solar, other renewables, and other are so small they don’t deserve to be on this chart.


“In particular, the data have shown that emissions intensity has fallen to the lowest level on record, as a combination of natural gas and renewable power have displaced more carbon intensive coal-fired power generation.”
If , as shown in the last graph, renewables are about 1% of total output then it follows that the reduction of emissions they provided is 1% or less. Not a very good return for my subsidy dollar.

CD in Wisconsin

“…..In particular, the data have shown that emissions intensity has fallen to the lowest level on record, as a combination of natural gas and renewable power have displaced more carbon intensive coal-fired power generation.”
The U.S. Energy Information Agency updated its table last month for 2017 showing the electricity generation sources for the grid by percentage….
Wind is at 6.3% and solar remains at a paltry 1.3% (solar panels and solar thermal) according to the EIA. Somehow, despite the nuclear plant closings, it states that nuclear power still provides 20%. Not exactly impressive gains for wind and solar as far as I can tell.
I can’t say if 4th generation nuclear and/or nuclear fusion will ever find their way to commercialization someday, but I am hoping that will. If the environmentalists would think with their brains instead of their emotions (or whatever it is that fuels their belief systems), they would wake up and realize that wind and solar are probably not going to cut it as alternatives to fossil fuels given how long solar PV panels and windmills have been around. Nuclear does.
The environmentalists are hurting themselves and their movement by also opposing nuclear power. As long as they keep shooting themselves in the foot with their opposition to nuclear power, then Bill McKibben and company can keep banging their heads against a wall in frustration all they want. The concussion that they might eventually give themselves is nobody’s fault but their own.
BTW, Old Man Winter is keeping a tight grip on the Upper Midwest here in early April. We got dumped on with more of that white colored global warming stuff here yesterday and last night. High, yes high temperatures are going to be as much as 20 degrees F below normal this week according to forecasts. Looks and feels more like January. Winters getting shorter and warmer?? Ummmm. yea. Right.

Lockeed Skunkworks now has 4 very broad and detailed design patent application patents published for their compact high beta fusion appraoch, which went dark after the Solve for X talk at Google in 2013. All date to preliminary filings in 2013 just before that reveal. None of the published applications provide any experimental data (needed in utility but not design patents). The detailed design patents were really well lawyered, and suggest deep Navy involvement on the dark side.
I keep a close watch on that and on LENR (which is looking to be real physics based on Widom Larsen weak force theory, but not commercializable owing to insufficient net energy gain, best summary Steve Krivitz’s three book series [over two thousand pages] published just a few months ago). He gifted me a set as I had asked him to critique the LENR example in ebook The Arts of Truth.


Here in Northern Ohio today we are having gale-force winds and light white colored global warming. It has been a strange start to April.


Tanners Creek is/was in the same mother corporation that I worked in. Their board of directors any more seem to take orders from the Sierra Club….


Just to add, many of the now-defunct coal plants on this system had been well-maintained & would have run for more decades. Their capital cost had long ago been paid off and the cost of their electricity had been a bargain. What a waste of money and resources.

Cliff Hilton

The US CO2 reduction is not only due to reduction of coal utilization, we also continue to ship off manufacturing and jobs. No? World-wide CO2 continues, up, unabated! Get it?
Let us hope Trump and send back some of those manufactures and jobs. I hope to see an increase in CO2 in the next three years.

It would be interesting to draw this graph with total Gas, Coal, Nuclear and Hydro (traditional sources) combined, along with the overall Total line. The gap between that line and the Total line would represent the minuscule contribution of wind, solar, etc. I’ll look for the data to try that, but frankly, the message is that gas is increasing, coal is declining (meaning an overall CO2 reduction), and unfortunately, the two primary non-CO2 creating sources (nuclear and hydro) are flat. Isn’t this a strong case for more nuclear? Where are the alarmists on that issue? I know they hate hydro (all those dams clogging up the pristine wilderness), but do they seriously believe that this graph will ever change dramatically unless nuclear is exploited?

J Mac

Fracking for enhanced natural gas and oil production is economically the single most important improvement to energy production and reduced energy costs on Planet Earth. The beneficial release of CO2 from combustion based power plants enhances the productivity of the worldwide biosphere, through enhanced plant growth. The correlated ‘greening’ of the planet demonstrates the positive benefits of greater atmospheric CO2.
The Carnegie Mellon Power Sector Carbon Index tracks and applauds reductions in the beneficial release of CO2 from combustion based power plants. The harmful impacts of their entrenched irrational perspectives on CO2 cannot be stressed enough. A significant portion of national wealth has been and is being wasted on further reducing production of already low atmospheric CO2, the single most important nutrient for all plant growth. To continue wasting precious national treasure on this environmental schizophrenia is frankly ‘nuts’!

Combined cycles at 60% efficiency and cheap NG doomed many of the old inefficient coal units. There are several hundreds still humming along – in spite of EPA.

Steve Zell

Burning one mole of carbon (as graphite) releases about 94 kcal of heat, while burning one mole of methane (natural gas) releases 213 kcal of heat, and both emit exactly the same 44 grams of CO2. So, for a coal plant and a natural gas plant of equal efficiency, the natural gas plant will emit less than half the CO2 of the coal plant. Replacing coal-fired power plants by natural gas-fired power plants with the same output will reduce CO2 emissions. However, proven reserves of coal would last much longer than proven reserves of natural gas, so there could be a future for coal if some of the fracking sites start running out of natural gas.
Taylor Pohlman has a valid point that power output from nuclear power plants is flat, and it doesn’t make sense that global-warming activists don’t push nuclear power, since it produces no CO2 emissions. Nuclear power plants are more expensive to build than coal- or gas-fired power plants for the same power output, but their operating costs are much lower, and the fuel supply is virtually unlimited. If coal and/or natural gas started becoming scarce in future centuries, nuclear power could become the power supply of the future. But well-publicized incidents (such as those at Three Mile Island, Chernobyl, and Fukushima) have instilled fear of nuclear power plants in the public.
Hydro is another clean and renewable source of electricity, although its use is limited to hilly areas with heavy rainfall. But “all those dams clogging up the pristine wilderness” could have a secondary benefit, such as saving water from winter storms and spring snowmelt to provide a reliable irrigation source during dry summers. Here’s looking at you, California–it might be worth sacrificing a few delta smelts to keep the Central Valley fertile and productive!


Steve Zell April 4, 2018 at 4:20 pm
Burning one mole of carbon (as graphite) releases about 94 kcal of heat, while burning one mole of methane (natural gas) releases 213 kcal of heat, and both emit exactly the same 44 grams of CO2. So, for a coal plant and a natural gas plant of equal efficiency, the natural gas plant will emit less than half the CO2 of the coal plant.

Coal is not graphite, it’s approximately CH.


Coal is not graphite, it’s approximately CH.

I’ll disagree with you there: Coal is near-pure carbon. The hydrogen mass is very, very low, mostly from the water contaminating the rock due to cleaning, washing and open-top rail cars.
Depends, of course, if you count molecular weight percents or number of atoms.

The composition of a bituminous coal by percentage is roughly: carbon [C], 75–90; hydrogen [H], 4.5–5.5; nitrogen [N], 1–1.5; sulfur [S], 1–2; oxygen[O], 5–20; ash, 2–10; and moisture, 1–10.)

Read more:


RACookPE1978 April 10, 2018 at 10:17 pm
Coal is not graphite, it’s approximately CH.
I’ll disagree with you there: Coal is near-pure carbon. The hydrogen mass is very, very low, mostly from the water contaminating the rock due to cleaning, washing and open-top rail cars.
Depends, of course, if you count molecular weight percents or number of atoms.

We’re talking chemistry of course which is why I represented it as CH, that is a equal proportion of H atoms and C atoms (~8% by mass), here’s an example of the typical molecular structure of coal, in this case it’s C70H41. The hydrogen mass is low because the atomic wt of H is 1 compared with 12 for C. It’s the breaking of the C-H bonds that contribute to the heat released by burning the fuel which the original poster referred to, 0 for graphite, 4 for methane and ~1 for coal (per C atom).
Have you heard of ‘coal gas’, produced by the destructive distillation of coal, major component is H2. Up through the 1960s the major gaseous fuel supplied to homes and industry.

Yes. Anthracite, the purest (and hardest and highest-temperature) coal has much less H than bituminous, a softer cooler-burning coal with more slag. But cheaper, more readily available.
Lignite (derisively) is “flammable dirt” .

Michael Jankowski

More evidence that we had nothing worth committing to in Paris.

ken morgan

in Australia we are going down the gurgler fast we are the biggest coal exporter to china power stations we have to pay more for coal and gas then they do and the government here are hell bent on wind and solar power our base is getting unreliable