A discussion from yesterday that probably got buried but I think is worth bringing to the forefront...

OilLearner on Friday November 17, 2006 at 11:28 AM EST Comments top

Most current railroad coal hopper cars carry between 100 and 120 tons with an average length of 56 feet. A mile long coal train would have about 85 such cars (plus engines).

The plant in the article burns 100 tons an hour to generate 250 MW, so a mile long train would provide about 85 hours of electricity production.

A 1,400 MW generating plant near here burns 550 tons an hour and would burn up a mile-long train load of coal in a bit over 15 hours.

[ Parent | Reply to This ]
------------------------------------

[new] Laurence Aurbach on Friday November 17, 2006 at 1:58 PM EST Comments top

Compare that to a 4600-acre solar thermal plant that produces 1,400 MW (using the land/power ratio demonstrated by the SEGS plant in Barstow, CA). How many acres of land are mined to obtain 85,000 tons of coal, every 15 hours?

The coal mine acreage is dependent on the thickness of the seam and how much other rock is between the layers. The estimate uses no calculation of the energy requirement of the mining machines or the comparative BTU content of the coal or its water content. Coal can also be mixed in a slurry and pumped in a pipeline. IMHO the figures are spurious.
Which figures do you believe to be false?

----------------------
Just some rough noodling with numbers to see what happens:

From here: Anthracite coal is 1506 kg/m^3

From a brief look around I'll assume an average coal seam to be approximately 4.57 meters

Which means that for every 1 m^2 of surface area, there are 4.57 m^3 of coal beneath, or 6882 kg of coal beneath for every square meter above.

From google: 1 acre = 4,046.85642 m^2

(4046 m^2/1 acre) X (6882kg/m^2) = (27844572 kg/acre)

or (61,386,773.3 lbs/acre) or (30,693 short tons/acre)
-- -- -- -- -- -- -- -- -- -- --
Now this is probably where you think the numbers are bogus, but I'll use them for now anyway... "A 1,400 MW generating plant near here burns 550 tons an hour and would burn up a mile-long train load of coal in a bit over 15 hours."

So let's say 1,400 MW consumes 13,200 tons of coal per day (550 X 24) or 4,818,000 tons per year.  That 1,400 MW plant would "destroy" roughly 157 acres per year. (Using the aforementioned assumptions.  My feeling is that I was optimistic on my assumptions.  If someone else would like to take the torch and punch out some more accurate numbers/correct any mistakes, please do so)

So, let's assume your calculations are in the ballpark. Over a 30-year period, coal mining "uses" the same amount of land as a 1980s-era solar thermal plant. The SEGS plant has a 30-year delivery contract; it has been in operation for 20 years and 15 years additional working life is expected.

In addition, the company that manufactured the SEGS plant claims that its latest-generation technology is 50 percent more efficient than the SEGS technology. That would imply a proportional reduction in the acreage requirement for an equal amount of electricity generation.

The tricky thing about that, is that after 30 years...the coal fired plant will continue to destroy land, whereas a solar plant will be fixed.  I also just did a "worst case" number crunch below and assuming an 18 inch coal seam would put the yearly land use at 1,568 acres per year as compared to 157 acres per year... or ten times as much.  So the coal plant may catch up to the land usage and begin surpassing that of the solar plant in as little as three years.
I'm don't know much about coal, but here's two points I've picked up:

Anthracite is now a quite small portion of the coal industry, and its primary deposit in eastern Pennsylvania has been mostly depleted.

Mountaintop mining often doesn't glean anywhere near that much - I seem to recall a National Geographic caption about removing a hundred feet of rock for an 18 inche thick seam.

I biased it towards the optimistic for two reasons in particular.  1: Since I probably couldn't come up with exhaustive numbers it's safer that way and would also help "fudge" in some things like underground mines...2:If even the optimistic case turns out pessimistic, you really know it's bad.
-- -- -- -- -- --
18 inches = 0.4572 meters

1 m^2 of surface area, yields 0.4572 m^3 of coal beneath, or 689 kg of coal beneath for every square meter above.

1 acre = 4,046.85642 m^2

(4046 m^2/1 acre) X (689 kg/m^2) = (2,787,694 kg/acre)

or (6,145,813 lbs/acre) or (3,073 short tons/acre)

1,400 MW consumes 4,818,000 tons per year.

So... 1,568 acres per year assuming a 0.4572 meter thick seam.  Ten times as much - ouch. (That, I imagine, should set the worst case boundry)

Based on this EIA DOE chart, the majority of coal production comes from mines with coalbed thickness around 4-6 feet. But there's a good 42% of production that comes from coalbeds 10 feet thick or greater.

Here's a record of some BLM sales that puts the coal/land ratio in Wyoming around 110 tons/acre. If your 30 tons/acre is true for 58% of the U.S., and 110 acres is true for 42% of the U.S., then the national average might be around 64 tons/acre.  

Correcton: ... BLM sales that puts the coal/land ratio in Wyoming around 110,000 tons/acre. If your 30,000 tons/acre is true for 58% of the U.S., and 110,000 acres is true for 42% of the U.S., then the national average might be around 64,000 tons/acre.  
wow...thanks for digging this up.  I'm surprised at the 110,000 tons/acre figure...no wonder Wyoming pumps out the coal.

Underground - 368,612 (thousand short tons)
Surface     - 762,190 (thousand short tons)

I'm guessing "surface" means strip mining?

Gotta go I'll take a closer look at this later

Sheesh, that was some correction. I'll try again.

... BLM sales that puts the coal/land ratio in Wyoming around 110,000 tons/acre. If your 30,000 tons/acre is true for 58% of the U.S., and 110,000 tons/acre is true for 42% of the U.S., then the national average might be around 64,000 tons/acre.

Note, however, that the Kentucky Geological Survey says that bituminous coal will yield 1,800 tons/acre foot. At a five foot coalbed thickness, that's 9000 tons/acre. Quite a range of estimates! I'm hoping someone with expertise in this field can weigh in.

A can of worms has definitely been opened.  This interests me for sure.  Only time will tell if it can battle my Americanized gnat-like attention span though.  I too hope someone with expertise will weigh in (aka TOD faeries).  But hope can be helped by giving it a chance, so I'll attempt to keep the motivation up, distill what's been learned today and expound a little, and probably drop it into Monday's drumbeat in the hopes of catching a wider audience.  With luck there'll be someone there with some insight.
Electricity - production: 3,892,000,000,000,000 Wh (2003)   https://www.cia.gov/cia/publications/factbook/geos/us.html

Coal is ~50% of the mix.  So 1,946,000,000,000,000 Wh attributed to Coal.

http://www.eia.doe.gov/cneaf/coal/page/acr/table4.html
68% of mines appear to be "surface" mines, so 1,323,280,000,000,000 Wh attributable to surface mined coal.

If a 1,400 MW (continuous duty plant) consumes roughly 13,200 tons of coal per day, which is (33,600 mWh/13,200 tons) or (2.54 mWh/ton).

(1,323,280,000,000,000 Wh) X (tons/2,540,000 Wh) = 520,976,378 tons of coal per year for "surface mined" coal.

Which best case 110,000 tons/acre: (520,976,378tons)X(acre/110,000tons)= 4,736 acres/year

Worst case 3,000 tons/acre (18" seam): (520,976,378tons)X(acre/3,000tons)= 173,659 acres/year

Vermont is 9,250 miles^2 or 5,920,000 acres.

Best case it would take 1,250 years to destroy a Vermont

Worst case it would take 34 years to destroy a Vermont

Most of the coal being hauled on trains to Texas is bituminous, sub-bituminous and lignite. Anathracite was nearly exhuasted in the middle of the last century. Coal goes down thousands of feet, although commercially mineable coal  coal seems to be above 500 ft subsurface. Coal contains varying ammounts of water which must be removed and disposed of to make the best fuel. I don't think we can easily calculate or predict commercial near surface deposit thickness or than guess its suitability for mining without some core drilling. I'm not saying this stuff can't be analysed. Its a job I be  I just don't know how without a bunch of data and time.
  This is great thought and an interesting idea. Thanks for bringing it up! I hope my criticisms helped clarify your thinking, not throw cold water on a good idea.
Anthracite is the best case in terms of tons/acre foot. Bituminous, sub-bituminous and lignite yield less coal per unit area, so Subtrate's figures are conservative in that respect. The Kentucky Geological Survey gives these figures:

Anthracite: 2,000 tons/acre foot
Bituminous: 1,800 tons/acre foot
Subbituminous: 1,770 tons/acre foot
Lignite: 1,750 tons/acre foot

Minor point.  Lignite has too low an energy value (and too dirty when burned) to be railed long distances.

Most lignite that is mined & burned is typically conveyer belted to a nearby power plant.

Alan

Hello Substrate,

To foster a desire in people to conserve electricity: at some carefully predetermined and constantly readjusted billing rate--money will not be accepted anymore, but physical labor will be required.  You pay your bill by helping shovel spilled coal, working at a recycling center or community food bank, picking up trash along a road, weeding in a community garden, helping insulate homes for the elderly and poor, mentoring at a school, etc.

If someone wants to burn alot of juice to heat their pool or power their McMansion, then they can plan ahead on required community service of some kind.  No exceptions or substitutions allowed!  Just another wild and crazy idea of mine-- I am full of them.

Bob Shaw in Phx,Az  Are Humans Smarter than Yeast?

This is a misleading calculation. You are assuming 100% load of the solar power plant (I'll ignore the storage and backup issues right now).

The typical real-world availability of solar power plant is 10-15% of the rated otput. At 15% you need to compare the SEGS solar plant to 0.15 x 1400 = 210 MW thermal power plant.

That's a good point about capacity factors and they are important to consider. The solar-only capacity factor of the SEGS plant is 22% and with its natural gas boiler assist it reaches 30%. With today's state of the art technology, solar-only capacity would be 29% and with natural gas assist would reach 40%. (Of course, when calculating land usage with natural gas assist, land used for gas wells must also be included. And natural gas has passed peak, contributes to global warming, etc.)

In addition, thermal storage may be used to extend the solar thermal plant capacity. An example is the molten salt system being installed for the Solar Tres project in Spain. In that case, trough system capacity factors can increase to around 55%.

Here's a DOE presentation that gives the capacity factor figures, see especially slides 31, 33 and 41. The DOE Sandia Sun Lab states without qualification that solar thermal has the same footprint as a fossil-fired plant when mining is considered; I'll have to do some investigating to find out how they came to that conclusion.

The footprint problem and the comparisons based on it are a bit artificial IMO.

Thermal/PV solar plants will be mostly built in desert areas where the ecological impact will be miniscule, and may be even positive.

OTOH coal miners reclaim the land (enforced by law AFAIK) after the coal seam has been exhausted, allowing for its recovery in the long term. But still the polution from the tailings remains, potentially contaminating the ground water etc.

Either way looks like comparing apples to oranges to me.

"OTOH coal miners reclaim the land (enforced by law AFAIK) after the coal seam has been exhausted, allowing for its recovery in the long term. But still the pollution from the tailings remains, potentially contaminating the ground water etc.

Either way looks like comparing apples to oranges to me. "

Complicated, of course.  The way that strip mines are reclaimed now is by the "cheap ass" approach...which basically means the land is devastated.  The topsoil is pushed off into the valley and buried, and when the land is "reclaimed" there is only a rocky, sandy surface such that they use a special grass that manages to survive on it when nothing else does.  You're then talking about geologic time for the land to recover.  If they'd save the topsoil somewhere else, and replace it when they were done, it'd be a much less tragic thing.  But the way it's done now, you might as well consider the whole area toxic to life.  They can't even build homes or businesses on the reclaimed areas because they're too unstable.

I thought that pine and a few other trees could survive as well.

And the valley's full of fill are unstable for a century or so, but one build on the "flat top" mountain.

Alan

http://www.energylan.sandia.gov/sunlab/faqs.htm
Question:
Do concentrating solar plants require a lot of land?

Answer:
Relatively speaking, no. Consider the Hoover Dam. Lake Mead covers nearly 250 square miles. A Concentrating Solar Power system occupying only 10-20 square miles of land could generate as much power on an annual basis as the Hoover Dam did last year. Considering the land required for mining, concentrating solar power plants also use less land than coal power plants.

They actually say less.  Which, technically it should since coal will continue to use land, whereas solar will occupy fixed space for eternity.  That's really impossible to argue, but what we've been doing here is trying to figure out where the break even point is...that's a little trickier.

Solar Tres

Although the turbine will be only slightly larger than Solar Two's, the larger heliostat field and thermal storage system will enable the plant to operate 24 hours a day during the summer and have an annual capacity factor of approximately 65%.

Impressive.  This technology has always held the most promise to me.

(note: this isn't aimed at you but my frustration for this being so pesky)

There's a lot of assuming going on...it's what happens when you're not entirely sure where to begin.  Also, the TOD faeries haven't magically appeared with lots of data, so that's kind of a bummer.  I'd try over at www.TheCoalScuttle.com but it doesn't seem to exist.

Seriously though, that was an oversight.  On that note, coal fired plants don't operate 100% all of the time either.  They also lose efficiency as they reach their rated capacity.  Lots of real world contraints to deal with beyond the assumed steady state power delivery.  I would love to get some reliable numbers on all of this, or at least something in the ballpark...but for not seeing the forest due to all the trees getting in our way, the thing to take home at the end of the day is that coal will continue to devestate land as it operates, solar will take a big chunk and then stop.  For PV it can take already taken land (i.e. rooftops)

It dawns on me that PV peaking power could be grid-coupled with solar thermal to form a theoretically steady electricty supply (i.e. when PV is at full song, solar thermal plants can be devoting all gains to storage for the night)

Anthracite: 22 to 28 million Btu per ton - United States averages 25 million Btu per ton

Bituminous: 21 to 30 million Btu per ton - United States averages 24 million Btu per ton, on the as-received basis

Subbituminous: 17 to 24 million Btu per ton - United States averages 17 to 18 million Btu per ton, on the as-received basis

Lignite: 9 to 17 million - United States averages 13 million Btu per ton

On that note, coal fired plants don't operate 100%

The logical error was that you were comparing a 1400MW solar plant with a hypothetical 100% loaded 1400MW coal power plant  which you feed 24/7 with that coal trains. If the real world CPP is 80% utilised, this would mean the coal inputs will alse be 20% down. Anyway, no big deal.

I agree with your bottom line but I don't think you are using the correct arguments. Coal is not worse because it uses more land, it is worse because of the way it uses the land - this is the thing we need to concentrate.

Personally I am cautiously "for" solar power. What really concerns me are those capital costs and how much from them are predicated on cheap fossil fuels. If the end-to-end EROEI is as low as I think, any attempts to scale them up will look like a threadmill - running as hard as possible just to stay in place. PV looks like a better bet for potentially lower EREOEI in future than thermal, but both have a lot to prove yet.