From Bagels to Coal Fires: An Unorthodox Economist Keeps Pushing for Change

You may remember Paul Feldman as the Bagel Man we wrote about in Freakonomics. You may also remember that he was an economist before he got into bagels, with an interest in agricultural, medical, and military issues.

He recently wrote to us about an environmental issue he’s been looking into: the abundance of underground coal fires in abandoned mines and other places that not only waste coal but contribute mightily to worldwide carbon dioxide emissions. According to Paul’s research — take a look at Web sites like this one, this one, and this one — underground fires in China alone contribute as much CO2 to the atmosphere each year as all the cars and light trucks in the U.S. (We touched on the issue of coal mining fatalities in the U.S. and China in a recent column about nuclear energy, but didn’t include this other important byproduct of widespread coal use.)

We asked Paul to write up a guest blog on the subject, which you’ll find below. Once you’ve read it, you’ll probably understand a bit better why this was the guy who could figure out how to make a handsome living by selling bagels on an honor system. It is a call to arms that understands the political and economic realities of the climate change debate. Here’s hoping that someone answers Paul’s call.

By Paul Feldman

There is growing acceptance that the earth’s climate is changing, and that action is needed to control the production of greenhouse gases. There is no consensus, however, on what should be done. Some proposals focus on using carbon taxes or carbon cap-and-trade arrangements to regulate industrial emitters. Other proposals are directed at re-designing individual emitters such as cars and trucks.

Such proposals are not likely to be accepted unless they are adopted worldwide. No nation wants to pay for emissions reductions while others continue to increase their emissions. Gaining international acceptance of broad policy actions, however, is a time-consuming and contentious process. For example, the Law of the Sea Treaty that is now being considered by the Bush Administration has been in the works since 1982, and it still fails to satisfy many parties to the negotiations. And the international treaty on climate change produced by the United Nations in 1992 and followed by the Kyoto Protocol of 1997 still lacks worldwide acceptance, because it excludes several of the most important emitters of greenhouse gases.

But while a worldwide policy to limit greenhouse gases will be difficult to achieve, not all interim steps to reduce emissions have to be contentious. One example concerns an issue that has escaped front-page attention: extinguishing underground coal fires that have been burning freely and putting massive amounts of carbon dioxide into the air for decades. Underground coal fires in China alone produce as much carbon dioxide annually as all the cars and light trucks in the United States. Fires in other countries, including the United States, are smaller but still add significantly to the total burden.

Although extinguishing these fires would be costly, it would reduce carbon emissions without the major disruptions to individual national economies mentioned above. And it would further benefit nations by eliminating the loss of their coal fields. As such, it might be possible to work out some international cost-sharing arrangement to attack those fires now, without waiting for all nations to agree on a wide-ranging treaty to limit carbon dioxide emissions.

Josh M

I'd like to hear more about the proposed methods of killing these fires. From what I understand, these are fires that are no longer contained within the mines that were exploiting the seam. The fire has progressed into the seam itself, spreading thousands of feet or more into the subsurface. This rules out the inert gas jet method from Isaac's post (#17).

Fred (#15) proposed following a model used in coal seam fires in Indonesia, but according to the article, these fires were at most 8 meters under the soil surface. The average depth for a coal mine in the US or China *must* be deeper than this. My quick searches showed an average depth of 300 meters for one seam in particular. Not exactly at a depth that makes direct excavation feasible.

The only method I saw mentioned through the given links is direct water or inert gas injection, which would necessarily involve tens of millions of US dollars just to drill one of the many required injection wells. The infrastructure costs alone must be enormous!



Hmmmmm, it couldn't possibly be as simple as that, could it?


The Smithsonian published an article about underground coal fires not too long ago:

It gives you an understanding of the complexity of Mr. Feldman's idea.


"Although extinguishing these fires would be costly..."

Costly? Isn't the problem not the cost of extinguishing these fires, but that it's essentially impossible? I mean, we LOST Centralia, Pa.

You know what we need? An international cost-sharing agreement to extinguish earthquakes, tsunamis, and hurricanes.

Sometimes, cost is not the issue.

Bagel man

WE lost Centralia? thefact is that the technology for extinguishing underground coal fires has advanced well beyond Centralia. There are firms that specialize in putting out underground fires and they could put out the Centralia fire now. What is holding back progress is that governments, are responsible (really, irresponsible). They'd rather spend the money on boondoggles like subsidies for ethanol and other farm supports.


I grew up not far from Centralia Pa. and can say there has been alot of money put into mine reclaimation (sic?) but not put into the putting out the fire. Just across route 61 from the fire the goverment is rebuilding a mountain that was leveled due to strip mining. So I don't see if they can put out the fire with new techniques money is the only reason?

Colin Toal

So Richard Branson has a bounty out from people who can scrub greenhouse gases from the atmosphere. Somebody should get a hold of him and ask him if he wants to pay to put some fires out.

Sounds to me like this is a no brainer.

Sean Samis

I did not see your column on nuclear power (The Jane Fonda Effect) so forgive me if I take this chance to ask some questions.

1. One of the technical difficulties with storing high-level nuclear waste is that it generates a great deal of heat; improperly stored it would melt the rocks around it. Has there been any recent work on reactor designs to take advantage of this residual heat to generate electricity? Nuclear power plants generate electricity by generating steam power, which any sufficiently hot source could do.

2. Related to the preceding, what percentage of economically available energy remains in nuclear fuel when it is regarded as "waste"?

3. I have heard that when the Three-Mile Island disaster occured, STATISTICALLY most of the casualties caused by the accident are or will be people injured by the NORMAL operation of the non-nuclear power plants which took over for the damaged nuclear generator. I.E., non-nuclear plants in normal operation are more dangerous than nuclear plants suffering a TMI-type "disaster". Is this true?

sean s.


Kevin C

Response to Sean Samis: Someone should do a calculation of fatalities per Watt delivered.

Note how many times there are fatal accidents at oil refineries, along oil or gas pipelines, or even homes exploding from natural gas or at filing stations. Fairly high nominal totals, but fossil also delivers the vast majority of Watts if you include autos.

My understanding is that the whole nuclear cycle from mining to generation to (short term) storage is very low-accident, but since nuclear energy is delivered as electricity, some electric distribution fatalities (guys on aluminum ladders cleaning out their gutters)would have to be allocated back to nuclear.


Some quick facts/answers about nuclear power:
-In terms of fatalities per man-hour worked, nuclear power plants are only a fraction as dangerous as your typical office.
-Sean's point 3 is correct. The worst possible exposure during TMI was calculated at about 25 millirems: this is about 2.5 chest x-rays, 8 cross-country flights, or one month of background radiation. Using the LNT model, this yields an additional 1-in-50,000 chance of getting cancer.
-Removing waste heat is not particularly viable for several reasons. Foremost among them: for safety and security, high-level waste is contained in a Matryoshka doll of heavy casks, and it's not feasible to remove it from them; also, unlike a normal nuclear reactor, where one can "shut off" the radioactivity by shutting down the reactor, one cannot stop the radioactivity from the waste (making installation, maintenance, and repair of equipment dangerous or impossible).
-About the only way to take "waste" and make it viable for fuel is reprocessing; this is done by several other countries, but is not done in the US for political reasons (thank Jimmy Carter for this one).
As for the coal fires: If one insists that carbon emissions must be limited, which would be cheaper? Increasing the efficiency of every car in America by 100%, or putting out half the coal fires in China?



"Underground coal fires in China alone produce as much carbon dioxide annually as all the cars and light trucks in the United States."

Not that I don't believe this, but where is the research on this? I think it would be an interesting read.


So how much would it cost to put these fires out?


Putting out the fires in China would be great - and that would be a great thing to do for the environment. But while we are thinking about fires in China caused from Coal mining - are we seriously thinking about letting places like China have nuclear power - you have to be kidding! I live as an expat in a third world country - if you think there is any regulation even close to what you experience in the US, then you are kidding yourself - disaster will strike, but it will be a lot more serious than some coal fires.


Anthony said "Not that I don't believe this, but where is the research on this?"

HELLO! How about clicking through the links provided in Dubner's intro to start with.

As for Centralia; Nobody ever said it was impossible to put the fire out. The government simply decided that evacuating the residents and ignoring the fire was the most economical (i.e. least expensive) solution. That, of course, was before anyone considered the cost of carbon emissions.

Assume the ridiculous for a moment i.e. that a global cap and trade regime for carbon emission has been implemented and coal fires consume their fair share of a countries carbon emissions. In this scenario there would be lots of incentive to research new techniques for reducing the cost of extinguishing coal fires and the benefit of extinguishing Centralia, amortized over the expected life of the fire times the carbon emissions saved, could be huge.



I suspect it would be easier to get China to foot some of the bill for putting out half its coal fires, while it would probably not be willing to pony up a dime to pump our cars' fuel-efficiency.


The key is apparently to find a fire-fighting group of locals who will work for non-Western wages; this says the cost of putting out 60+ small coal fires in an Indonesian rain forest was about $21,000.


I wonder what kind of penalties could be imposed on large CO2-emitting firms that wouldn't create the same effect as the day care experiment.


A year ago I watched an item about an innovation that uses a jet engine to create an inert gas to extinguish fires very rapidly. It's called the Steamexfire and they say it's especially suitable for tunnel fires and mine fires.
There is not much information there, so I don't know if it could be used to put out all of those existing fires, but if this could be used to put out new fires at an early stage, that would be a good start.

Sean Samis

Matt, Thanks for your response re. the viability of using nuclear "waste" as a fuel source. Although I am no expert on this, I am pretty certain that high-level waste is not moved directly from the reactor to the long-term storage casks; or certainly need not be. A properly designed "secondary" reactor could store the waste just as safely and securely as the casks, yet allow it to generate sufficient heat to generate electricity. So long as the value of the electricity generated offsets the cost of the "secondary reactor", it could be economically viable.

I believe that the purpose of reprocessing fuel is to re-use it in reactors of the current "standard" design. I suppose I'm wondering about a different design intended to leverage the residual heat of high-level "waste" as is.

Frank C

In these situations, this seems like a simple answer to a difficult problem. I don't see any chance that China would devote any resources to putting out underground fires creating massive amounts of CO2. They have super high levels of pollution, cancer and illness now. And, the government has done nothing to reduce them because to reduce them would result in slower economic growth. Their whole social structure is currently built on massive year over year growth to provide income and jobs to their notoriously poor population.
Until the US as their largest trading partner takes the steps to reduce imports from China with specific conditions, and until the US government devotes itself to concerted regulation and economic incentive to reduce carbon emissions and pollutions within our own country, why would the Chinese government change anything?

This is one of those situations where the solution seems simple but is practically impossible. I don't see any hope of the human race reducing carbon emissions and carcinogenic pollutants in the future to have any impact other than the track we're on. Unless the wealthiest countries in the world reduce massive per capita consumption (requiring massive energy usage), we are going to stay on this curve into the foreseeable future. Plus, we have to worry about emerging countries coming on-line with their own population consumption and emissions.