Coalbed Methane

OUTREACH PROGRAM

$
s

33

V

S>* <%>
&

PR0T^°

ro
z

LU
0
X"

£

U.S. Surface Mines
Emissions Assessment


-------
U.S. SURFACE MINES EMISSIONS
ASSESSMENT

U.S. Environmental Protection Agency

October 2005


-------
ACKNOWLEDGEMENT

This draft was prepared under U.S. Environmental Protection Agency Contract 68-W-00-092 by
Raven Ridge Resources, Incorporated.


-------
Executive Summary

The current surface mine methane emissions estimation protocol used by the U.S.
Environmental Protection Agency (U.S. EPA) involves the use of a Tier 2 approach by using
basin-specific gas contents, basin-specific coal production, and a nationwide emission factor.
The IPCC report Good Practice Guidance and Uncertainty Management in National
Greenhouse Gas Inventories states "It is not feasible to collect mine-by-mine Tier 3
measurement data for surface mines"; however, efforts have been made to develop a mine-
specific method closer to Tier 3 for determining emissions related to surface mining activities. In
order to make recommendations for improvements to the U.S. methodology, various aspects of
methodologies using a Tier 3 approach that were studied as part of other projects were
examined.

The purpose of this study was to look for ways to improve the surface mine methane (SMM)
emissions estimation methodology through a review of the current information used to
determine the SMM inventory. In order to assess the current methodology, the following data
was collected:

•	latest emissions factors used for other national inventories,

•	updated gas contents for several U.S. coal basins,

•	study of coal thicknesses and overburden depths at U.S. surface mines,

•	analysis of unmined coal seams adjacent to mined seams at surface mines,

•	recent surface mine emissions studies from Australia, Canada, and U.S. EPA,

•	surface mine emissions measurement attempts by Australians and U.S. EPA,

•	other applicable measurement technologies.

Based on the findings, areas of data improvement were identified and integrated into the
existing methodology in order to recalculate the SMM emissions estimate for comparison
purposes. In addition, various attempts at mine-specific methane measurements were analyzed
and reviewed to determine the possibility of combining or using other methods. Finally, the
location of the gassiest surface mines in the U.S., and mines where meaningful pre-drainage
(emissions avoided) could occur were identified.

i


-------
1.0 Background	1

2.0 Background Information on the U.S. Methodology	1

2.1	Sources of Fugitive Methane Emissions at Surface Coal Mines	4

2.2	Coal Gas Content	4

2.3	Emission Factor Used to Account for Over- and Underburden Coals	7

2.4	Post-Mining Emission Factors	9

2.5	Integrating Research Into the Surface Coal Mine Emissions Inventory	9

2.6	Comparison of U.S. Gas Contents with International Values	10

3.0 Review of Methane Measurement Technologies at Surface Coal Mines	12

3.1	Description of Open-Path FTIR Spectroscopy and Modeling Techniques -
U.S. EPA	12

Methods Used in the Study	13

Measurements and Results	14

3.2	Description of Combined Measurements Methodology - Australia CSIRO. 15

Gas Content Method	15

Direct Surface Emission Measurement Method	15

Borehole Measurement Method	16

Results	17

3.3	Other Possible Measurement Technologies	18

3.4	Applying Technologies to U.S. Surface Coal Mines	18

4.0 Identification of Opportunities for Methane Recovery and Use at U.S. Surface

Coal Mines	19

4.1	Sources ofSMM Emissions	20

4.2	Recovery	23

4.3	Outstanding Issues	24

4.4	Recommendations	26

5.0 Summary and Conclusions	26

6.0 References	28

ii


-------
1.0 Background

Currently, surface coal mines account for 67 percent of U.S. coal production, but constitute only
16 percent of the coal mine methane emissions. The primary reason for this is due to the
relatively low gas content of the coals that are extracted from surface mines. The low gas
content of these coal seams is likely related to the shallow depth of burial, and the fact that
some are lower rank with commensurately lower gas adsorption capacity. Unlike underground
mines for which degasification and ventilation emissions data is readily available, mine-specific
emissions measurements are generally not measured for surface mines because no
measurements are required for safety reasons due to the low risk of accidents resulting from
excessive methane concentrations. The current approach used for estimating surface mine
methane emissions is to apply a Tier 1 global average emission factor, or Tier 2 country or
basin-specific emission factor to the amount of coal produced. As a result, emissions from
surface mines (and post-mining activities) are calculated by multiplying basin-specific coal
production by a basin-specific gas content and then by the country-specific emission factor to
determine methane emissions. The emission factor currently used by the U.S. is based on 200
percent of the in-situ gas content of the coal. More accurate surface mine methane emissions
estimates are desired, as surface mining accounts for a larger fraction of coal produced.

2.0 Background Information on the U.S. Methodology

The first step used in estimating methane emissions from surface mining and post-mining
activities is to segregate the surface mines geographically by coal basin and by state. The
Energy Information Agency's (ElA) Coal Industry Annual reports state- and county-specific
underground and surface coal production by year. To calculate production by basin, the state
level data were grouped into coal basins using the basin definitions listed in Table A-119 of
Annex 3 of the Inventory of U.S. Greenhouse Gas Emissions and Sinks. For two states—West
Virginia and Kentucky—county-level production data was used for the basin assignments
because coal production occurred from geologically distinct coal basins within these states.
Table A-120 of the same publication presents the coal production data aggregated by basin.

Emission factors for surface mined coal were developed from the in situ methane content of the
coal in each basin down to a depth of approximately 250 feet. Revisions were recommended
for several of the gas contents used in the underground and surface coal mine inventories in a
memorandum to EPA on July 31, 2003, based on analysis of additional gas content data.
Additional publicly available gas content data such as raw data used in Evaluation and Analysis
of Gas Content and Coal Properties of Major Coal Bearing Regions of the United States,
EPA/600/R-96-065 and Coalbed Methane Resources of the United States, AAPG Studies in
Geology Series #17, as well as data from USGS and individual state geological survey
publications was uncovered and compiled. Analysis of this additional data resulted in the
following changes in surface mine gas contents for each basin:

l


-------
Table 2.0.1 - Recommended Revisions in 2003

Basin

Previous Gas Content

Revised Gas Content



(scf/t)

(scf/t)

Northern Appalachian

49.3

59.5

Central Appalachian

49.3

24.9

Black Warrior

49.3

30.7

Illinois

39.0

34.3

Rocky Mountain Basins*

15.5

--

Piceance

--

33.1

Green River

--

33.1

Raton

--

33.1

Uinta

--

16.0

San Juan

--

7.3

Northern Great Plains

3.2

5.8

Western Interior Basins**

3.2

--

Arkoma

--

5.4

Cherokee

--

34.3

Forest City

--

34.3

Gulf Coast***

--

33.1

Sources: Diamond et al., 1986, Kirschbaum et al., 2000, Rightmire et al. 1984 and Tewalt, 1986

*Rocky Mountain Basins: Due to discovery of additional data, gas contents were separated
from one value (15.5 scf/t) for all Rockies basins to separate values for each basin.

**Western Interior Basins: Geologic information revealed Arkansas, Missouri, Kansas and
Iowa coals are more similar to the Illinois Basin coals; therefore gas content values of these
coals were reassigned.

*** It was determined that Texas and Louisiana coals are more similar to Raton Basin coals,
therefore gas contents were reassigned.

As a result of the updates, the new gas content values were used for the 2002 and 2003
inventories. However, due to a lack of publicly-available data, several of the basins' gas
contents were still in question.

Based on an analysis presented in EPA's Anthropogenic Methane Emissions in the United
States: Estimates for 1990 (1993), surface mining emission factors were estimated to be from 1
to 3 times the average in situ methane content in the basin. Therefore, a mid-case emission
factor of 2 was applied to all in situ methane contents for surface mines. However, recent
research has found that the foundation for the assumptions used in the 1993 report were based
of two previous studies conducted by Environment Canada (1992) and U.S. EPA (Kirchgessner
et al. 1992). Both of these studies have since been updated by the authors and the emission
factor assumptions have changed.

Based on the current emissions inventory, approximately half, or 13, of the states with surface
mining activities account for -90 percent of the methane emitted to the atmosphere as a result
of surface coal mining activities. For this reason, the primary focus of this report was on coal
basins contained by these top emitting states. Table 2.0.2 shows the relative ranking of states
by coal production and methane emissions from 2000-2003.

2


-------
Table 2.0.2 - U.S. Surface Coal Mine Production and Emissions

Surface Coal Production by State (thousands of tons)



Surface Mine Emissions by State (mcf)

State

2000

2001

2002

2003 % of 2003 Total

State

2000

2001

2002

2003

% of 2003 Total



Wyoming

338,048

368,749

373,161

376,270

52%



Wyoming

3,786,138

4,129,991

4,179,403

4,214,224



20%

WestVA South

54,498

57,447

56,810

47,999





Texas

3,209,906

2,981,807

2,995,351

3,145,625





Texas

48,488

45,042

45,247

47,517





WestVA South

2,714,000

2,860,861

2,829,138

2,390,350





Kentucky, East

45,114

47,262

42,984

39,142





Kentucky, East

2,246,677

2,353,648

2,140,603

1,949,272





Montana

38,352

39,143

37,386

36,962 547,890

76%



Indiana

1,665,402

2,026,887

1,881,561

1,832,992





North Dakota

31,270

30,475

30,799

30,775





Pennsyl.

2,056,677

1,905,309

1,500,828

1,369,928





Indiana

24,277

29,546

27,428

26,720





Ohio

1,232,364

1,488,211

1,226,414

1,092,658

15,995,049

76%

New Mexico

27,320

28,937

27,163

20,499 625,884

87%



WestVA North

704,361

644,738

636,650

585,480





Arizona

13,111

13,418

12,804

12,059





Colorado

606,127

650,417

646,840

572,895





Pennsyl.

17,283

16,011

12,612

11,512





Virginia

480,769

511,496

471,357

516,476





Virginia

9,654

10,271

9,465

10,371





Montana

429,542

438,400

418,723

413,974

18,083,874

86%

Ohio

10,356

12,506

10,306

9,182





Illinois

260,817

389,057

437,874

387,933





Colorado

9,156

9,825

9,771

8,654 677,662

94%



North Dakota

350,224

341,325

344,949

344,680





Washington

4,270

4,624

5,827

6,232





New Mexico

398,872

422,487

396,580

299,285





Illinois

3,802

5,671

6,383

5,655





Kentucky, West

380,867

389,374

382,033

294,980

19,410,753

93%

WestVA North

5,919

5,418

5,350

4,920





Alabama

211,830

257,389

246,521

291,159





Alabama

3,450

4,192

4,015

4,742





Louisiana

244,146

245,942

251,759

266,654





Kentucky, West

5,552

5,676

5,569

4,300





Mississippi

49,243

37,075

141,527

226,873





Louisiana

3,688

3,715

3,803

4,028





Maryland

160,650

166,539

216,580

208,964





Mississippi

802

604

2,305

3,695





Arizona

191,421

195,896

186,938

176,061





Tennessee

1,213

2,003

2,081

1,907





Oklahoma

200,405

193,674

140,507

174,628





Maryland

1,350

1,399

1,820

1,756





Tennessee

60,407

99,759

103,634

94,969





Oklahoma

1,345

1,300

943

1,172





Washington

47,824

51,792

65,262

69,798





Alaska

1,641

1,514

1,146

1,081





Missouri

29,910

25,137

17,013

36,564





Missouri

436

366

248

533





Alaska

18,379

16,959

12,835

12,107





Kansas

201

176

205

154





Kansas

13,789

12,078

14,063

10,564





Utah

A

-

268

25





Arkansas

1,788

1,937

1,937

1,043





Arkansas

12

13

13

7





Utah

-

-

8,576

800





California

-

-

-

-





California

-

-

-

-





Iowa

-

-

-

-





Iowa

-

-

-

-





TOTAL

700,608

745,306

735,912

717,869

100%



TOTAL

21,752,535

22,838,183

21,895,457

20,980,937



100%

Source: EIA, 2003

3


-------
2.1	Sources of Fugitive Methane Emissions at Surface Coal Mines

There are three potential sources of fugitive methane emissions associated with surface coal
mining. These are:

•	Methane emitted by the coal excavated and processed during mining activities,

•	Methane emitted by the coal and other gas bearing strata in the overburden and/or
underburden exposed by mining activities, and

•	Methane emitted by the overburden coal excavated and stored on site in waste piles.

For methane emissions covered by the first point above, the available methane emitted by the
excavated and processed coal is the estimated total gas content of the material excavated. For
the second and third points above, the available methane is more uncertain as it depends on a
variety of factors such as gas content and thickness of the adjacent coal seams, permeability of
the coals and other strata found in the overburden and underburden, overburden thickness, and
the amount of disturbance to the mine floor and highwall as a result of mining.

The gas in coal and associated strata may be released during different stages in mining.
Excavated coal will release methane as it is broken and removed from the highwall face,
transported on site, and crushed and sized for transportation off-site. Overburden, inter-burden
and uneconomic coal is normally dumped together with non-coal material in waste piles. The
methane contained in these coals will be released as the material is excavated, broken,
dumped, and later used as backfill.

In addition, methane emissions will also migrate out of the floor and highwall of the surface
mine. The magnitude of the floor emissions will depend on several factors such as:

•	gas content of the unmined coal beneath the mine floor,

•	proximity of the coal seams to the mine floor,

•	extent of disturbance of the coal and the effect this has on its permeability,

•	amount of coal left in the floor, and

•	presence of water.

The magnitude of emissions from the highwall will similarly depend on:

•	gas content of the unmined coal remaining in the highwall,

•	extent of disturbance of the coal near the highwall and the impact this has on the
permeability, and

•	presence of water.

2.2	Coal Gas Content

In order to better accommodate the differences in coal properties occurring among the gassy
coal basins, it was determined that the gas contents of several of the coal basins needed
updating. Table 2.2.1 shows the recommended changes to gas contents used for surface mine
emission factors, based on recent research. In summary, it is recommended that changes to
gas contents of three coal basins and separating the Williston Basin coals in North Dakota from
the Wyoming-Montana group are necessary. The reason is that the coal mined in North Dakota
is lower rank (lignite) than Wyoming-Montana (bituminous, sub-bituminous). It is believed that
the gas content of 5.6 scf/t previously used for the entire region is appropriate for the North
Dakota coals only.

The coals in Washington and Alaska were also separated from the Wyoming-Montana group
since they are slightly higher rank, however, no information could be found on shallow gas
contents in these states. Recognizing the 5.6 scf/t value previously used for this region to be

4


-------
too low for bituminous coals, using an average western U.S. coal basin value of 16 scf/t
(Kirchgessner 2003), until further work can be carried out to refine this value, is recommended.

Similarly, coal mined in Texas and Louisiana was found to be borderline sub-bituminous
(Tewalt, 1986). It is thought that in many cases, these Eocene/Paleocene coals are similar to
the Paleocene coals found in the Powder River Basin as they were formed in similar
depositional environments. The previous gas content of 33.1 scf/t was said to be representative
of deeper, sub-bituminous and bituminous coals in Texas. During a USGS Resource
Assessment of Gulf Coast coals in 2000, two test wells were drilled into shallow coals in
northeast Texas to determine the gas contents of the low-rank Wilcox coals. The gas contents
of coals from these two test wells averaged 11.0 scf/t. For the time being, this content is more
representative of the coals mined in the Gulf Coast. However, further research is warranted to
refine this value.

Table 2.2.1 - Recommended Gas Content Changes to U.S. Coal Basins for Surface Mine

Emissions

Coal Basin

Inventory
Code

Major Coal Rank
Mined

2003 Revised Recommended
Gas Content New Gas
(cf/t) Contents (cf/t)

Comments

Northern App
Central App
Warrior
Illinois

S.West/Rockies
S.West (NM, AZ, CA)
Rockies (CO)

Rockies (UT)

N.Great Plains
Northern Rockies (MT,WY)
West Interior

Forest City, Cherokee
Arkoma (OK)
TX, LA
Northwest

NAB
CAB
WRB
ILB
WTB

NGP

WYM
WIN

NWB

Bituminous
Bituminous
Bituminous
Bituminous
Bituminous
Bituminous
Bituminous
Bituminous
Lignite
Sub-bituminous

Bituminous
Bituminous
Sub-bituminous
Sub-bituminous

59.5
24.9
30.7
34.3

7.3
33.1

16.0
5.6
5.6

34.3
74.5

33.1
5.6

59.5	Data compiled from USBM report

24.9	Data compiled from USBM and MRCP reports

30.7	Data compiled from USBM report

34.3	Data compiled from USBM and MRCP reports

7.3	Data compiled from USBM and MRCP reports

33.1	Data compiled from USBM and MRCP reports

16.0	Data compiled from USBM and MRCP reports

5.6	North Dakota mines lignite coal

20.0	Data compiled from USGS, and private sector

34.3	Arkansas, Missouri, Kansas, Iowa coals similar to Illinois Basin

74.5	Data compiled from USBM and MRCP reports

11.0	Texas & Louisiana mine borderline sub-bituminous coal

16.0	Washington, Alaska coals similar to Powder River Basin	

Sources: Diamond et al., 1986, Kirschbaum et al., 2000, Rightmire et al. 1984 and Tewalt, 1986

The gas content used for surface mine emissions factors in the Wyoming area coals has been
thought to be low, but the difficulty of obtaining more recent, publicly available data has
hindered the completion of any comprehensive study. This lack of available data still exists, but
sufficient data (from several different sources) during this latest attempt to justify increasing gas
content values for the region was discovered. Ninety percent of the coal mined in Wyoming is
from the Powder River Basin (PRB), which represents 47 percent of all surface coal production
in the U.S. This magnitude of coal mining increases the significance of gas content-based
emission factors from the PRB. The challenge of obtaining representative gas contents from
the PRB include:

•	most gas contents are taken at depths much deeper than 200 feet (overburden depth),

•	the methane occurring in the coals seams is considered biogenic and therefore gas content
is in flux, and

•	high permeability of coals allows for ease of gas migration once coals have been dewatered.

As part of a USGS-BLM coalbed methane study in 2001, two test wells were drilled adjacent to
the Jacob Ranch Coal Mine in the PRB. The Jacob Ranch Mine is the fourth largest coal mine
in the U.S., producing 35 million tons in 2003. One well drilled 500 feet from the highwall
showed gas contents of 2 scf/t, while the second well drilled three miles from the face showed
12 scf/t.

5


-------
Personal communications with John Wheaton, Research Hydrogeologist of the Montana Bureau
of Mines and Geology1, indicated that the gas content of PRB coals in Montana average 30
scf/t. Depths of the coals that produce this average amount of gas were not discussed as his
research mainly focused on water quality issues. Gas contents at surface mine depths would
be lower than average.

Using a table on the Wyoming State Geologic Survey (WSGS) website showing a summary of
gas content values relative to coal 2depth, a linear regression plot was generated to estimate
the gas content at the mid-coal seam level for an average Wyoming surface coal mine
overburden depth of 180 feet (210 feet total depth). The results produced a gas content of 15.7
scf/t.

Figure 2.2.1 shows the plotted data, regression, and equation used for the estimate. In further
discussion with WSGS, it was agreed that the coals in southern Wyoming could contain higher
volumes of gas due to their higher rank and the coals tend to be buried more deeply.

Figure 2.2.1 - Wyoming Geologic Survey Gas Content Summary

Coal Depth (feet)

Source: WSGS, 2002

The parent company of a large surface mine operation was contacted about gas contents of the
coal mined in the Wyodak/Anderson coal seam. The overburden depths at the surface mine are
250 feet which is deeper than average for Wyoming mines. While the gas content data from the
coal in the CBM field adjacent to the mine is confidential business information, it was found to
be greater than 20 scf/t. Taking all this into account, the gas content for surface mines in the
Northern Rockies region (Wyoming and Montana) is estimated to be 20 scf/t.

1	Personal communication

2	Attempts to obtain the raw data used for the table from the WSGS were to no avail. The data originated from the
U.S. BLM and based on discussions with their Wyoming field office, the source data is considered to be proprietary
and would not be released

6


-------
2.3 Emission Factor Used to Account for Over- and Underburden Coals

As stated in Section 2.0, the current emission factor used to account for fugitive methane
emissions other than the coal being mined is a factor of two. The basis for this factor - studies
conducted by Environment Canada (1992) and U.S. EPA (Kirchgessner et al., 1992) - has
since been updated and the emission factor assumptions changed.

The original Environment Canada study estimated surface mining emission factors to be from
one to three times the average in situ methane content in the basin. Since that time,
Environment Canada has adapted a 1994 study conducted by Neil & Gunter, Ltd that quantified
methane emission rates from underground and surface coal mines throughout Canada (King,
1994). Using this Tier 2-3 hybrid approach, Canada applied emissions factors for mines
categorized by mine type, coal basin, and coal rank using mine-specific information gathered in
the 1994 study. The resulting average emissions factor at surface coal mines of 8.6 scf/t
includes a 50 percent increase of gas content data to account for unmined strata.

Attempts were made to compile surface mine data such as overburden depth, coal seam
thickness, and net thickness of coal seams in the overburden and underburden to develop a
matrix which could be used for basin-specific emission factors. It was determined that the data
was too wide-ranging among coal mines, and that using average basin values was not
necessarily an accurate assessment of surface mining activities. Overburden depths were
compiled from the 2004 Keystone Coal Industry Annual for all surface mines with coal
production greater than 200,000 tons. State-based average overburden depths were then
calculated. Figure 2.3.1 shows the results of the overburden study.

Figure 2.3.1 - Average Overburden Depth of U.S. Surface Coal Mines

200 -i

1 RH



I OU

160

— 140
a>

® 120

^ 100

a. 80
a>

Q 60
40
20
0

-

















































































































































































Source: Keystone, 2004

The average coal seam thickness of the major coal seams mined at U.S. surface coal mines
was also compiled. Information was used from the Department Of Energy's (DOE) EIA website
to determine the coal thicknesses at the surface mine locations. In several states, multiple
seams are mined and noted separately. Figure 2.3.2 shows the summary of the coal seam
thickness data.

7


-------
Figure 2.3.2 - Average Coal Seam Thickness at U.S. Surface Coal Mines





ou.u
70.0
60.0

a)
a)

50.0

V)

2 40.0
c

o 30.0

!E

>- 20.0
10.0













PI

	Dlinnnnnnnnnnnnn



Source: EIA, 2003

A desk study of rider coal seams located above or within 100 feet below the mined coal seams was
conducted. Rider seams are adjacent coalbeds that are often thinner than the mined seam. These
seams often contain methane contents similar to the mined coal seam. Quantifying the thicknesses
of the coals proved difficult for several reasons:

•	generalized stratigraphic columns are illustrative of the stratigraphic sequence and usually only
give a picture of relative thicknesses of coals,

•	the rider seams may pinch out laterally and are not always present at the surface mine locations,
or, they can thicken laterally, and

•	coal thickness can range widely, thus average thicknesses may not reflect conditions at the
mines.

What can be concluded is that net rider seam thicknesses may never approach the mined seam
thickness, thus providing some evidence that the factor of two currently used for the emissions factor
may be considered conservative (too large). It is believed that a Tier 2 (basin-specific) emissions
factor cannot be developed at this time due to a lack of data specific to the coal mines areas. The
current calculation method is to simply multiply the volume of coal produced by 200 percent, which is
supposed to account for 100 percent of the in situ content of the mined/ produced coal and any
immeasurable amount of the methane in the adjacent strata. As a result, the Tier 1 value of 200
percent appears to be very conservative, and a lesser value might be more appropriate. As a result,
further work needs to be carried out before a more accurate factor can be determined, and most-
likely, this factor will be basin or region specific. There are several reasons this is proposed:

•	The ratio of mined coal thickness to rider seam thickness varies greatly from basin to basin, or in
some instances, there are no seams in the overburden.

8


-------
•	Horizontal permeabilities of both rider coals and mined coals vary from basin to basin, which can
control the amount of gas that could potentially be emitted from the various seams as a result of
mining.

•	Overburden ratios (ratio of the thickness of overburden strata to the thickness of the mined seam)
vary greatly from basin to basin.

2.4	Post-Mining Emission Factors

Currently, the same post-mining emissions factor is used for both underground and surface coal
mines (32.5 percent of in-situ gas content of the coal). This was originally developed by a UK study
(Creedy, 1993) of British coals. The actual amount of gas that escapes into the atmosphere will be a
function of the methane desorption rate, the coal's original gas content, and the amount of time
elapsed before coal combustion occurs. Some limited studies have been conducted using United
States Bureau of Mines (USBM) gas content and desorption data and EIA route times for coal
transportation. Kirchgessner (EPA, 2001) estimates the post-emissions factor to be 55-90 percent of
in-situ gas content for underground coals mines and 72-78 percent for surface mines. Australia's
greenhouse gas emissions methodology uses a 20 percent factor based on a 1994 study (Williams
etal., 1993).

Due to the limited post-emissions data available and expert judgment, the current post-emissions
factor appears to be representative of typical bituminous coals in the U.S.

2.5	Integrating Research into the Surface Coal Mine Emissions Inventory

The surface coal emissions inventory for the year 2003 was recalculated in two ways. First, the
newly proposed gas contents from Table 2.2.1 were incorporated; and second, using the 150
percent emissions factor instead of the current 200 percent factor. Table 2.5.1 shows the inventory
by U.S. coal basin using the former and revised gas contents. The nearly 10 Bcf increase in the
Northern Rockies coal basins is partially offset by the 3 Bcf decrease in the Texas Gulf Coast coal
area. The net change is an increase of approximately 7.5 Bcf (from 24.4 to 31.9 Bcf).

Table 2.5.1 - Comparison of 2003 Surface Mine Emissions Inventory
Using Currently Used and Proposed Gas Contents3



Previous

















Gas

Surface Mine

Post-Mining

Total



Surface Mine

Post-Mining

Total



Content

Emissions

Emissions

Emissions

New Gas

Emissions

Emissions

Emissions



(cf/t)

(mmcf)

(mmcf)

(mmcf)

Content (cf/t)

(mmcf)

(mmcf)

(mmcf)

Northern App

59.50

3,257,030

529,267

3,786,297

59.50

3,257,030

529,267

3,786,297

Central App

24.90

4,951,066

804,548

5,755,614

24.90

4,951,066

804,548

5,755,614

Warrior

30.70

518,032

84,180

602,212

30.70

518,032

84,180

602,212

Illinois

34.30

2,515,905

408,835

2,924,740

34.30

2,515,905

408,835

2,924,740

S.West/Rockies (NM, AZ, CA)

7.30

475,347

77,244

552,591

7.30

475,347

77,244

552,591

S.West/Rockies (CO)

33.10

572,895

93,095

665,990

33.10

572,895

93,095

665,990

S.West/Rockies (UT)

16.00

800

130

930

16.00

800

130

930

N.Great Plains

5.60

4,972,878

808,093

5,780,971

5.60

344,680

56,011

400,691

Northern Rockies (MT,WY)

20.00

-

-

-

20.00

13,223,424

2,148,806

15,372,230

West Interior (Forest City, Cherokee)

34.30

47,128

7,658

54,787

34.30

47,128

7,658

54,787

West Interior (Arkoma)

74.50

175,671

28,547

204,218

74.50

175,671

28,547

204,218

West Interior (Gulf Coast)

33.10

3,412,279

554,495

3,966,774

11.00

1,133,990

184,273

1,318,263

Northwest

5.60

81,906

13,310

95,215

16.00

234,016

38,028

272,044



Total

20,980,937

3,409,402

24,390,339



27,449,984

4,460,622

31,910,606

3 Data used to develop this table was obtained during preparation of the annual Coal Mine Methane Emissions
Inventory.

9


-------
This net increase in the inventory is negated when an emissions factor of 150 percent rather than
200 percent is used to represent the additional emissions from overburden and underlying coal
seams. Table 2.5.2 shows the inventory by U.S. coal basin when using the new gas contents and
the 150percent factor. The net change is an increase of approximately 0.6 Bcf (from 24.4 to 25.0
Bcf).

Table 2.5.2 - Comparison of 2003 Surface Mine Emissions Inventory Using Currently Used
Gas Contents (with 200 percent emissions factor) and Proposed Gas Contents4

(with 150 percent emissions factor)



Previous

















Gas

Surface Mine

Post-Mining

Total



Surface Mine

Post-Mining

Total



Content

Emissions

Emissions

Emissions

New Gas

Emissions

Emissions

Emissions



(cf/t)

(mmcf)

(mmcf)

(mmcf)

Content (cf/t)

(mmcf)

(mmcf)

(mmcf)

Northern App

59.50

3,257,030

529,267

3,786,297

59.50

2,442,773

529,267

2,972,040

Central App

24.90

4,951,066

804,548

5,755,614

24.90

3,713,300

804,548

4,517,848

Warrior

30.70

518,032

84,180

602,212

30.70

388,524

84,180

472,704

Illinois

34.30

2,515,905

408,835

2,924,740

34.30

1,886,929

408,835

2,295,763

S.West/Rockies (NM, AZ, CA)

7.30

475,347

77,244

552,591

7.30

356,510

77,244

433,754

S.West/Rockies (CO)

33.10

572,895

93,095

665,990

33.10

429,671

93,095

522,767

S.West/Rockies (UT)

16.00

800

130

930

16.00

600

130

730

N.Great Plains

5.60

4,972,878

808,093

5,780,971

5.60

258,510

56,011

314,521

Northern Rockies (MT,WY)

20.00

-

-

-

20.00

9,917,568

2,148,806

12,066,374

West Interior (Forest City, Cherokee)

34.30

47,128

7,658

54,787

34.30

35,346

7,658

43,004

West Interior (Arkoma)

74.50

175,671

28,547

204,218

74.50

131,753

28,547

160,300

West Interior (Gulf Coast)

33.10

3,412,279

554,495

3,966,774

11.00

850,493

184,273

1,034,766

Northwest

5.60

81,906

13,310

95,215

16.00

175,512

38,028

213,540



Total

20,980,937

3,409,402

24,390,339



20,587,488

4,460,622

25,048,110

2.6 Comparison of U.S. Gas Contents with International Values

To further analyze the U.S. surface mine methane emissions inventory, a study of several
international surface mine methane gas contents was conducted for comparison with U.S. values.
Literature regarding gas contents of surface mined coal was researched for several countries and
coal was distinguished by rank. For each rank, the U.S. average emission factors appear to be close
to median. For example, the U.S. value for lignite falls between the average factor for Germany and
Russia (Figure 2.6.1). The U.S. average gas content value was also compared to the overall
international average value. The U.S. average falls close to but below the overall average factor for
each coal rank and appears low when plotted on the range of factors (Figure 2.6.2).

4 Data used to develop this table was obtained during preparation of the annual Coal Mine Methane Emissions
Inventory.

10


-------
Figure 2.6.1 - CH4 Gas Contents by Country and Coal Rank

X

o

(n

&
o
E
o
!a

D

o

4.5
4.0
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0.0

~	Lignite

¦ Sub-Bituminous

~	Bituminous

3

Germany Canada United Australia Poland United Russia South Ukraine India
Kingdom	States	Africa

Source: Izrael et al., 1997, KazNIIMOSK, 2002, Lloyd et al., 2005 and
Personal communications with UNFCCC

Figure 2.6.1 compares several average surface mining gas content values including the U.S.
average based on new gas content values. Most emission factor data was obtained from National
Communications to the UNFCCC. The range of values for South Africa was obtained from Lloyd, et
al. Data for Russia was obtained by applying the appropriate coal ranks to the gas contents found in
Izrael, et al. Kazakhstan information was taken from KazNIIMOSK, 2002.

Figure 2.6.2 - Range of Worldwide CH4 Gas Contents by Coal Rank

Sub-Bituminous

Lignite

U.S. Average
Overall Average

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5

Cubic meters CH4/ton coal

Source: Izrael et al., 1997, KazNIIMOSK, 2002, Lloyd et al., 2005 and Personal
communications with UNFCCC

11


-------
Figure 2.6.2 shows where the U.S. average surface mine gas contents fall within the range of
worldwide gas contents for each coal rank. For all three coal types, the U.S. gas content is lower
than the overall average. The ranges of gas contents by coal rank are shown in Table 2.6.1.

Table 2.6.1 - Ranges and Values of CH4 Gas Contents in m3 CH4 / ton of coal



Lignite

Sub-Bituminous

Bituminous

Germany

0.015





Canada

0.088

0.28

0.19-0.85

United Kingdom





0.49

Australia





CN
CO

1

o

Poland





2.5

United States

0.16-0.31

0.57

0.21-2.11

Russia

1.0

1.1 - 1.8

2.9-5.9

South Africa





0.002 - 0.064

Ukraine



1.35



India





1.8

Source: Izrael et al., 1997, KazNII

MOSK, 2002, Lloyd et al., 2005 and

Personal communications with UNFCC

3.0	Review of Methane Measurement Technologies at Surface Coal
Mines

This section summarizes two reported efforts made to develop methane emissions
measurement protocols, one by U.S. EPA in 1991 and the other by Australia's CSIRO in 2003.
Interestingly, the studies were vastly different in their approaches and conclusions,
demonstrating the difficulty of developing a transparent measuring methodology for surface
mine emissions on a site-specific basis.

3.1	Description of Open-Path FTIR Spectroscopy and Modeling Techniques - U.S. EPA

One previous effort to develop a methodology for estimating surface mine methane emissions
was carried out by the U.S. EPA involving the use of open-path Fourier Transform Infrared
spectroscopy (FTIR) and Gaussian-based plume dispersion modeling techniques. FTIR
technology has been applied in measurement of hazardous air pollutants, and is accepted by
EPA as one of the better technologies for measuring air pollutants in residential areas. Here, a
methodology has been developed for applying the technology to measuring surface mine
methane emissions. Kirchgessner et al. describe the results of the initial field trial of this
methodology.

Use of the FTIR spectrometer, a remote-sensing device, was chosen by the authors over point
sampling techniques because point sampling would present the need for an unreasonably large
number of samples (as well as give rise to potential errors from sample line leaks or loss or
production of gases in sampling containers). The open-path FTIR also accommodates the
sizable plumes being emitted from surface coal mines, which can be over 1000 meters in width.

12


-------
The initial mine site selected for measurement was the Caballo mine in Campbell County,
Wyoming, which is located in the Powder River coal basin. Of six sites considered in the
Powder River, Montana, and Illinois basins, Caballo was selected for its configuration and the
fact that the location exhibited properties such as flat terrain where Gaussian dispersion
modeling could be applied.

Methods Used in the Study

The methodology was developed using FTIR, meteorological measurements, and release of a
tracer gas at known rates. The tracer gas used in this study is SF6, which is non-reactive and
does not naturally occur; thus, there is no ambient concentration to be concerned with. The
open-path FTIR spectrometer directs a beam of infrared radiation along a path where it is
reflected back to the spectrometer with mirrors. Smoke releases show surface mine emissions
disperse in the direction of prevailing winds; thus, this study used a near ground-level
measurement taken with the open-path FTIR sensor downwind of the mine. The reflected beam
is subjected to absorption analysis to identify the gases present along the path and a path-
integrated concentration is determined. The concentration measurement at the point of the
FTIR path is then incorporated into a model which uses this value to estimate the emissions
from the entire area based on the plume dispersion characteristics.

The plume dispersion properties were determined by the simultaneous release of SF6 and CH4.
Using meteorological data (e.g. wind speed) obtained from a meteorological station located
near the designated FTIR path, any available site-specific plume characteristics, and a known
release of the tracer gas, standard Gaussian dispersion equations were applied to create a
plume dispersion model for the methane plume from the surface mine.

The following simplified relationship was derived from integrating the standard Gaussian
equation across the y direction and setting height equal to zero as the plume is a ground-level
source:

where,

Ccwi	= ground-level cross-wind-integrated concentration (g/m2)

Q	= emission rate (g/s)

u	= average wind speed (m/s)

oz	= vertical dispersion coefficient (m)

The equation can be used to assess dispersion characteristics by obtaining values of oz specific
to the site given 1) a measured tracer gas concentration (CCwi]) from an FTIR sensor, 2) a
measured value of u from the meteorological station near the FTIR path, and 3) a known
release rate Q from a tracer gas source. A number of oz values were determined based on
tracer gas releases conducted at different distances upwind of the monitoring path. Using these
values, a relationship of oz versus distance from the path was determined.

Also, a simpler method can be used to determine the plume's dispersion characteristics using
fewer tracer gas measurements. Given Q for the tracer gas, the release location, wind speed
(u), and wind direction, a plume dispersion model was used to predict CCwi for the tracer gas
plume. The model was run several times considering differing stability classes (Pasquill-Gifford

13


-------
atmospheric stability classes), producing a range of CCwi values. The predicted CCwi value
closest to the measured value was used to define the atmospheric stability class present when
the measurement was taken. CH4 measurements were collected at this time so the
atmospheric stability is applied to the CH4 plume.

The model was run to predict methane concentrations along the measured FTIR path, and the
predicted concentrations were compared with the measured values to calculate the actual
methane release rate once incorporated into the plume dispersion model. The following
relationship was used:

Q(actual) _ ConCGntration(measured)

Q(predicted)	C 0 l~l C6 l~l t rat i 0 l~l (p red icted)

Measurements and Results

Preliminary ambient measurements were collected at the site before the methodology was
applied. An organic vapor analyzer (OVA) was used to provide rough estimations of methane
emissions. These preliminary measurements indicated that disturbed coal areas, such as those
blasted, were likely to be the most significant source of CH4 at the mine. Ambient
measurements were also taken with the FTIR, upwind of the sampling area. The average
background concentration measured was 1.64 ppm (the global average is 1.7 ppm).
Background concentrations varied significantly, up to 0.5 ppm, between days 2 and 4 of
sampling.

Calibration cell measurements were also taken at the site to assess the performance of the
FTIR, and interpret results accordingly. The FTIR beam was passed through a chamber with a
known CH4 concentration. On average, the FTIR measurements appear to underestimate
actual CH4 calibration concentrations by about 20 percent. OVA measurements were also
taken during sampling in order to compare with the FTIR data. The OVA data estimated higher
concentrations than the FTIR measurements, when taken at the same time (7.0 ppm while the
FTIR measured 4.0 ppm), confirming the FTIR measurements to be low.

The plume from the Caballo mine was split into east and west sections to be measured by the
FTIR, as the maximum path length the FTIR sensor can measure is 650 meters which the
mine's plume exceeded. The path lengths ranged from 375 to 525 meters. Longer path lengths
are thought to have caused low FTIR measurement values due to the effect of light scattering,
diluting the overall signal.

Estimated emission rates for the mine range from 0.70 to 6.31 m3/min for the east side and 0.77
to 6.24 m3/min for the west side. The east side emissions are higher, as expected, due to the
presence of the coal blast area on the east side. The average east side emission rate was 1.85
m3/min and the average west side emission rate was 1.45 m3/min. Based on the average
values for each side, the estimated total annual emissions from the Caballo mine are 1.74
million m3/year (or 168 mcf/day). Emission factors reported in a later EPA study (Kirchgessner,
2001) range from 0.03 to 0.13 m3 CH4/metric ton of coal, with an average value of 0.09 m3
CH4/metric ton of coal (3.18 ft3CH4/ton of coal).

In development of this methodology, the authors determined open-path FTIR spectroscopy and
Gaussian based plume dispersion modeling to be a feasible approach for measuring methane
emissions from large surface mines. In the initial study, it was noted that additional work is
required primarily because methane concentration measurements determined by the FTIR were

14


-------
low (20 to 75 percent) based on preliminary measurements taken with an OVA. Kirchgessner
(et al.) note later that the techniques measured emissions often within 15 to 20 percent of the
known values obtained from validation by calibration cell measurements and usually within no
more than 30 percent (2001).

3.2 Description of Combined Measurements Methodology - Australia CSIRO

Another effort to develop a method for estimating fugitive surface mine emissions was
conducted for the Australian Coal Association (Saghafi et al. 2003). Briefly, the methodology
combined several measurements, including gas content measurement of coal samples, surface
emissions measured from exposed coal and interburden, and gas flow and gas composition
measured from a surface borehole. The development of this methodology was completed in
two coal basins in Australia; the Hunter Valley in NSW, and the Bowen Basin in Queensland.

Gas Content Method

The gas content of coal was measured based on three components in the following equation:
Qm = Qi + Q2 + Q3

where,

Qm = Measured gas content

Qi = Volume of gas lost during drilling

Q2 = Volume of gas desorbed during the period between measuring and
crushing the sample

Q3 = Volume of gas released after crushing

Coal samples were collected and sealed in leak tested canisters. The canisters can contain up
to 3.0 kg of coal. The Qi component is only pertinent to fresh bore coal samples and is
estimated by measuring the gas desorption rate over 20 to 30 minutes, fitting a desorption rate
equation, and calculating the gas desorbed back to zero time. The volume of gas desorbed
after drilling is typically measured over the time it takes to transport the sample to the laboratory
or one to two days. Finally, the coal is crushed to < 200|jm and the amount of gas desorbed
during and after crushing are measured. The sum of these three components yields the
measured gas content of the coal. For some samples, desorption of gas was monitored over a
6 week period to study the kinetics of desorption.

Direct Surface Emission Measurement Method

The second measurement in this methodology is the direct measurement of surface emissions
from exposed coal and interburden. This is done by placing a chamber (Figure 3.2.1) over a 4
m2 surface area and drawing ambient air through with a fan at a known rate to dilute the gas,
such that gas concentrations are maintained within the range measurable by gas analyzers
through which the gas is drawn. The methane content was measured with a Horiba
hydrocarbon analyzer, using a laptop to record the data. Carbon dioxide was also measured.
The surface emissions were calculated as emission fluxes expressed as volume of gas emitted
per unit time and unit area of ground surface. The emission fluxes were calculated using the
following expression:

q _ fi {Cs - Cb)

A

15


-------
Emission flux
Dilution air flow rate
Area of the chamber

Steady state concentration of the seam gas in the chamber
Concentration of the seam gas in the dilution air flowing through the chamber

Figure 3.2.1 - CSIRO Chamber

CSIRO'? special chamber on the surface of the spoil
measuring flux of greenhouse gas emissions

Source: Saghafi et al. 2003
Borehole Measurement Method

The final measurement in this methodology is the measurement of gas flow and gas
composition from surface boreholes that intersect coal seams. The study was carried out on a
borehole in the Cheshunt region that was drilled in the middle of an undisturbed area to be
mined through in approximately five years. Since it was desired to measure the gas flow from
individual coal seams, as well as the total gas flow from the borehole, two methods were used
to attempt isolating the zone for testing in order to measure gas flow from individual seams -
use of impermeable layers of bentonite to isolate seams within the borehole, and the use of
borehole packers.

Backfilling with bentonite proved unsuccessful due to its less than ideal sealing properties. The
borehole packer method for measuring gas flow from individual seams was also unsuccessful
where groundwater was present, as dewatering the boreholes with a pump was not possible
with the packer in place. Where dewatering is not an issue, the packer method may find
application. As a result, the total gas flow was measured using gas flow meters connected to a
cap fitted to the borehole casing. Due to the necessary dewatering procedure, the water

where,

Q =

fd =
A =

Cs =
Cb =

16


-------
extracted from the hole was tested for gas content; however, the gas lost with the water was not
significant.

Results

Gas content measurements were taken at both the Bowen Basin and Hunter Valley sites. At the
Moura mine in the Bowen Basin, gas contents ranged from 0.50 m3/t (C02 and CH4) to 1.0 m3/t
with -95 percent methane. At the other mines, Goonyella and Burton, in Bowen Basin, the
content of the gas was almost entirely C02, with gas contents in the range from -0.1 to 0.2 m3/t.
Back extrapolation of Moura mine gas content data showed that the Moura coal in Pit 5A could
have had a higher initial gas content of -1 m3/t, where values obtained from testing were around
0.48 m3/t. Extended time periods between uncovering the seam and mining could lead to
underestimation of coal gas content. The variability of the gas content data demonstrates the
importance of knowing length of time since the coal seams were uncovered.

Data from seven mines in Hunter Valley showed gas contents varying from -0.07 to 1.6 m3/t,
with compositions ranging from nearly 100 percent C02 up to 70 percent CH4. The larger gas
contents had higher CH4 compositions. At this site, gas desorption monitoring was performed,
which brings up another concern with gas content measurement as a methodology. For one
sample, 50 percent of the in situ gas was still present after 41 days with 10 percent still present
after 144 days. This indicates significant amounts of gas leaving the mine in the coal. Further
work is necessary to quantify this.

Direct measurement of surface emissions was also conducted at both sites. At the Moura mine,
emission rates varied by more than a factor of ten, from -0.4 to -6 mg/s/m2 (almost 100 percent
CH4). The other Bowen Basin mines had much lower rates, high variability, and were almost
entirely C02. The Goonyella mine, for example, had emission rates vary from 0.02 to 0.45
mg/s/m2 over essentially the same surface.

The gas content and composition derived from the borehole data varied as expected with depth.
Higher gas content with greater CH4 concentration (up to 90 percent for the deepest seam)
resulted from measurement of deeper seams, while lower gas content with greater
concentration of C02 corresponded with shallower seams. Gas contents varied from -0.4 to
3.7 m3/t.

Coal sampling for gas content measurements and direct surface emission measurement does
not provide the data necessary for a Tier 3 methodology to determine emission factors for the
mines studied, as it was determined that coals sampled at the surface have already lost some
percentage of gas as a result of desorption processes once the coals were uncovered. Gas
content measurements and surface emission rates varied widely in C02/CH4 ratio as well as
total gas content. The results exhibited variability due to: 1) dependence on the initial gas
contained in the coal, 2) the time since the coal was disturbed, and 3) the mining method which
affects the permeability of the surface layer and the rate at which gas desorbs.

Rather than accepting a methodology to measure emissions from active mines using all three
measurements (gas content, direct surface emissions, and flow from boreholes), it is suggested
that further work be carried out on measurements from boreholes to develop a methodology for
estimating fugitive emissions after mine closure. Borehole emissions may approximate
emissions from a standing high wall on mine closure, as opposed to emissions from an exposed
coal seam that is measured with direct surface emission measurement.

17


-------
Of the three measurement methodologies applied in this study, an approach similar to that at
the Cheshunt borehole is required. The data obtained from a dedicated borehole, along with
data from an extension of the exploration drilling program to include a limited number of gas
content measurements could be used to develop a Tier 3 methodology.

3.3	Other Possible Measurement Technologies

Technologies that have been designed to measure methane leaks from natural gas pipelines
may also be considered for measuring methane emissions from surface mines. One such
technology is the duoThane® or active gas correlation radiometer (ACGR) pipeline monitoring
solution created by Ophir Corporation. This system is specifically designed for long-path,
perimeter measurements of methane and ethane from large area facilities such as landfills.
This product has been used previously for pre- and follow up surveys of coalbed methane sites,
and is considered applicable to measure surface mine plumes as well.

As an optically-based sensor, ACGR offers a solution to long-path perimeter monitoring. This is
because the optical method can integrate along a line-of-sight, detecting the total trace gas
concentration existing at any moment between the transmitter and the receiver. Trace gas
concentrations can be monitored in a continuous fashion, and flux measurements can be readily
achieved. When a facility or area is encircled with perimeter monitors, total emissions of the
trace gas under study can be determined (Ophir, 2005). It is believed that a weakness of the
attempt by EPA using FTIR was the use of only one monitor which may not have captured the
total emissions occurring at the mine and that utilization of a series of perimeter monitors as
applied here would be superior.

AGCR is a method of detecting trace gases using an active source and an optical correlation
detection method. The optical correlation hardware compares the spectra of the gas of interest
to that of the gas in the region under inspection. AGCR does not require laser sources, but
instead uses broadband illumination (Spaeth et al, 2003).

The technologies developed by Ophir have been used for path lengths up to 900m, and
representatives state that even longer lengths can be measured by dividing the area into
smaller sections much like what was done on the east side of the Caballo mine in
Kirchgessner's study. This system can measure trace concentrations as low as sub-ppm, as
well as higher concentrations more applicable to surface mine emissions, given the area is
divided into smaller sections as to avoid loss of the beam.

Other technologies have been explored for measuring natural gas pipeline leaks that may find
application measuring surface mine emissions as well. Physical Sciences Inc. (PSI) has
developed a passive infrared imaging system which combines passive infrared concepts similar
to FTIR with optical technology using an imaging sensor for remote detection of methane
(Cosofret et al, 2004). The sensor consists of an infrared focal plane array-based camera and
an interferometer. The interferometer functions as a filter which selects the wavelength
illuminating the focal plane array. The sensor generates methane images and the methane
column density at each pixel in the image is calculated using an algorithm. The algorithm
incorporates range-to-target together with ambient conditions (temperature and humidity).
Tests on this technology were done at 200m, but PSI note the system incorporates a wide field-
of-view for wide area coverage. Perhaps further study could find application for this technology
over larger areas for measurement of surface mine methane emissions.

3.4	Applying Technologies to U.S. Surface Coal Mines

18


-------
After reviewing the SMM measurement methodologies, important elements of each study that
could be combined to formulate a more robust methodology was found. The EPA measurement
technique appears to underestimate the total emissions originating from the mine. It is unclear if
the shooting of a single path using FTIR adequately represented methane emissions from all
sources at the coal mine such as the high wall, mine floor, and overburden coals. The results
were much lower than the gas contents of the coals, thus the duration in which the methane is
released from the coal may be a factor not considered. In other words, the temporal accounting
of methane released from the mine appeared not to consider methane emitting from the
highwall for weeks or months leading up to the study. A more comprehensive (larger temporal
and project boundary), perimeter-based, sample plan using updated equipment (i.e. Ophir) may
produce a more representative methane emission rate. Furthermore, the use of numerical
modeling techniques can provide information regarding the migration of methane from in-situ
conditions to its eventual release to the atmosphere.

The CSIRO study focused more on gas content sampling protocols rather than the emission flux
rate at the coal mines. The gas content data adjacent to a surface is of enormous value, but
again, a more strategic sampling plan (100, 200, 500, 1000, 5000 meters from the face) would
help facilitate more meaningful results. Procedures for measuring gas contents of coals are
already well established in the CBM industry in the U.S., however, accounting for lost gas with
high permeability coal (such as in the Powder River Basin) remains a source of high uncertainty.

In conclusion, the spatially-based gas content data should be measured and used in conjunction
with an optical-based measurement of methane flux rates in order to develop a more
representative measurement methodology for surface mine emissions. Since methane may be
emitted for months (or years) ahead of mining, a numerical model should also be integrated in
order to determine the temporal boundary of SMM emissions. Furthermore, since the ten
largest surface mines in the PRB produce nearly 50 percent of the U.S. SMM emissions
(discussed in more detail later in this report), a methodology geared to PRB mines and
conditions is recommended.

4.0 Identification of Opportunities for Methane Recovery and Use at
U.S. Surface Coal Mines

In addition to researching improvements to the U.S. surface mine methane emissions estimation
methodology by collecting data about surface mine emissions and researching alternate
methodologies, it was also desired to identify specific opportunities for methane recovery and
use at U.S. surface coal mines. This was done by identifying the top emitting surface mines by
utilizing the current methodology, and then conducting analysis of those mines in order to
identify a specific set of mines where recovery may be feasible and warrant further evaluation.
Methane recovery options were reviewed for specific mines and outstanding issues were
identified such as mineral ownership or gas quality at the mines.

19


-------
4.1 Sources of SMM Emissions

Based on the updated gas content data obtained and reported in Section 2.3, the new
recommended basin-specific gas contents in Table 2.2.1 were used to generate emission data
for individual surface mines.

Analysis of annual coal production showed that of the 716 active surface mines in the U.S., the
330 most productive mines (mines with production greater than 100,000 tons per year)
accounted for 98.56 percent of total production in 2003 (Table 4.1.1). Mine-specific emissions
for coal mines that produced less than 100,000 tons per year were not calculated. It was
concluded that due to the minute fraction of total production attributed to the remaining 386
mines, methane emissions from those mines are negligible. As a result, further analyses was
concentrated on the top coal (and therefore emissions) producers.

Source: EIA, 2003

Emission data were obtained by multiplying the basin-specific gas contents in Table 2.2.1 by
the national emission factor of 200 percent (to account for over- and underlying strata), and then
applying the result to the mines' annual coal production.

Underground mines with annual emissions greater than or equal to 100 MMcf/yr are considered
gassy. The same consideration was applied for this analysis of surface mines. The 50 top-
emitting surface mines in the U.S. had emissions greater than 100 MMcf/yr and accounted for
72.41 percent of total emissions. Figure 4.1.1 shows how those 50 mines are distributed with
regards to emissions. Approximately half of the 50 top-emitting mines produce 200 MMcf/yr or
less.

20


-------
Figure 4.1.1 - Frequency Distribution: 50 Gassiest Surface Coal Mines

in the U.S. (as of 2003)5

20 n

150 200 300 400 700 900 1000 2500 3000 More

CH4 Emissions (MMcf/year)

The 100 mmcf/yr threshold for underground mines triggers MSHA-directed engineering controls
and monitoring due to concern for miner safety. For diffuse emissions being emitted from
surface mines, the same threshold for methane emissions would not necessarily apply since
mine worker safety may not be threatened at those levels. Therefore a higher standard of
"gassiness" may need to be considered for surface mines. Figure 4.1.2 reveals that the high-
emission mines account for the most significant percentage of emissions, though in Figure
4.1.1 very few mines emitted more than 700 mmcf/year. For example, only 8 mines account for
over 40 percent of total emissions, while 27 mines account for less than 10 percent of
emissions.

Figure 4.1.2 - 50 Gassy Surface Coal Mines in the U.S.: Distribution of Percent

of Total Emissions6

50

fk 40

LLI

5

30

o 20
o 10

5?

0

27 mines

12 mines

2 mines

1mine

100-200 200-400 400-700 700-900
CH4 Emissions (MMcf/year)

900+

5	Data used to develop this table was obtained during preparation of the annual Coal Mine Methane Emissions
Inventory.

6	Data used to develop this table was obtained during preparation of the annual Coal Mine Methane Emissions
Inventory.

21


-------
Of these 50 mines that account for approximately 72 percent of total SMM emissions, the top
ten emitting surface mines in the U.S. (listed in Table 4.1.2) account for 46.9 percent of all
surface mine emissions. The largest emitters, the North Antelope Rochelle Complex and Black
Thunder Mines, located in Wyoming, are estimated to contribute 19 percent of total surface
mine methane emissions alone. All of the mines listed in Table 4.1.2 are located in Campbell
County, Wyoming, except for the Antelope Coal Mine, which is located in nearby Converse
County, Wyoming. Locations of the ten mines as well as others in the area are shown on the
map in Figure 4.1.3. All are classified as being in the Northern Rockies Basin. The revised gas
content recommended earlier in Section 2.3 for the Northern Rockies Basin is 20 scf/ton.

Table 4.1.2 - Top Ten Emitting Surface Mines in U.S.7

Mine Name

2003
Production
(tons)

Gas
Content
cf/ton

Emission
Factor
(cf/ton)

Emissions
MMcf

Emissions
Tonnes
C02e

Coal Basin

% of Total
Emissions

Cumulative %
of Emissions

North Antelope Rochelle
Complex

80,083,444

20

40

3,203

1,295,750

NRB

10.71%

10.71%

Black Thunder Mine

62,620,417

20

40

2,505

1,013,198

NRB

8.38%

19.09%

Cordero Mine/Caballo
Rojo Mine

36,083,743

20

40

1,443

583,835

NRB

4.83%

23.91 %

Jacobs Ranch Mine

35,491,218

20

40

1,420

574,248

NRB

4.75%

28.66%

Antelope Coal Mine

29,533,072

20

40

1,181

477,845

NRB

3.95%

32.61%

Eagle Butte Mine

24,728,392

20

40

989

400,105

NRB

3.31%

35.92%

North Rochelle

23,923,145

20

40

957

387,076

NRB

3.20%

39.12%

Caballo Mine

22,743,284

20

40

910

367,986

NRB

3.04%

42.16%

Belle Ayr Mine

17,844,826

20

40

714

288,729

NRB

2.39%

44.55%

Buckskin Mine

17,539,156

20

40

702

283,784

NRB

2.35%

46.89%

7 Data used to develop this table was obtained during preparation of the annual Coal Mine Methane Emissions
Inventory.

22


-------
Figure 4.1

Bjuckekir
¦PawWkla

Dry Foik
fcUV ^ocljik

—— tochel le

Antelope

lJ__L	 T4Q N

V,	R63 W

INVERSE
OUNTY

.3 - Map of surface mines in Campbell and Converse Counties, Wyoming

T52 N

R73 W

Eagle
Fort

Gillette

Rocky
CAMPDEL.
COUNTY

Coal Crttfk

cobs Rsni h

ck Thunder
orth Rochille

Source: Molnia et. al, 1997

4.2 Recovery

It was determined that the ten surface mines listed in Table 4.1.2 emit enough methane to
warrant further evaluation as potential methane recovery projects. These ten mines are at the
eastern edge of the Powder River Basin (PRB) in Wyoming and account for nearly half of all
surface mine methane emissions in the U.S. Since the 1990s, the PRB has been the focus of
massive coalbed methane development efforts. Methane recovery at these mine sites would
make a significant contribution towards mitigating methane emissions from surface mines.

The PRB has estimated methane reserves of 25 trillion cubic feet. The coalbed methane
industry in the basin is flourishing as the number of producing wells has climbed to over 21,000
by the end of 2004, while in the mid-1990s, the basin had only 4,000 wells (Wilkinson 2005).
With the methane industry thriving, coalbed methane development in the form of surface mine
pre-drainage could make a sizeable contribution to methane recovery.

Realistically, the only feasible type of methane recovery to be deployed at surface mines is pre-
mine drainage. Because of their proximity to existing CBM production wells, any pre-drainage
wells placed in advance of the coal mining operations could be connected to an existing gas
pipeline infrastructure. Many of the surface mines in the PRB require dewatering wells in
advance of mining. It is also possible that some of the dewatering wells could be converted to
methane production wells once the water table has been drawn down ahead of the highwall.
Due to the high permeability of the PRB coals, only vertical CBM-type wells are feasible to use
to degas coals ahead of mining, since nearly all of the gas within 2000 feet of the highwall has
already been released.

23


-------
Maps of the CBM fields adjacent to the largest surface mines in the Campbell County area were
not available. A general map of CBM wells and mining operations in the PRB was located at
the Wyoming State Geological Survey; however, it only provides information as recent as 2002.
The map in Figure 4.2.1 shows CBM wells as of 2002 and their proximity to the major coal
mining operations in the PRB.

Active CBM wells are denoted by the black dots in the figure, which are located to the west of
the coal mines. This map also shows some of the pipeline structure present in the area
(denoted by red and yellow lines).

4.3 Outstanding Issues

The PRB has experienced a particularly dramatic increase in coalbed methane exploration and
development. It contains the largest coal reserves of any basin in the United States. Over 90
percent of the Basin's coal estate is in Federal ownership and accounts for one-third of all U.S.
coal production. About 45 percent of the oil and gas estate (including coalbed methane) in the
PRB is under Federal ownership. Conflict has surrounded the development of CBM resources
in the PRB in recent years (Fulton 2001).

A major clash has occurred between coal licensees and oil and gas developers. Commonly in
the PRB, resource ownership is a "split estate" issue where the surface owner may not own the
mineral rights below. Much of the mineral rights in the basin are owned by BLM and leased to
private companies. Most federal oil and gas leases in the PRB are senior to coal licenses;
however, at the time of overlapping licensure, extensive CBM development was not anticipated.
In the past, traditional oil and gas and coal conflicts generally involved oil and gas resources
contained in reservoirs much deeper than the coal, thereby allowing for development of coal
without loss of the oil and gas development. Since CBM is trapped within the coal seams and
was considered a valueless gas which escaped from coal, rather than part of the valuable coal
fuel itself, coal companies routinely vented the gas to the atmosphere. Rising interest in CBM
exploration and development as a result of new technology, a better understanding of the
resource and increasing energy demand has created a mineral conflict situation concerning
federal leases.

In 2001, the aforementioned conflict led to the introduction of federal legislation, HR 2952, the
Powder River Basin Resource Development Act, sponsored by Representative Barbara Cubin
(R -WY), which would have permitted the suspension of CBM operations in order to allow coal
production to continue while providing a means for the oil and gas lessee to be paid equitable
compensation. However, Congress did not enact this legislation and it did not become law.
BLM's current conflict resolution procedure involves ordering CBM drilling sooner than planned
if it might otherwise be vented during mining in order to avoid the waste of this resource.

24


-------
Figure 4.2.1 - Map of Powder River Basin Mines and CBM Wells


-------
Recently, environmental concerns have arisen regarding CBM production. A significant
consideration in any coalbed methane extraction project is the issue of by-product water
produced from the CBM wells. CBM operators now face added cost due to bonding
requirements for in-channel reservoirs used to hold water produced from their wells, as the
Wyoming BLM has extended bonding requirements to federal minerals. Operators have
previously only been required to post reclamation bonds for reservoirs on state and private
lands. Bonds may be $5,000 to $10,000 or more (depending on the nature of the site) and are
designed to protect the land from any known or unforeseen risks related to displacement of
millions of barrels of water onto the surface by providing funds for mitigation of detriments that
may occur later on. Efforts are also in progress to extend bond requirements to cover
downstream impacts to vegetation and soils. The added requirements pose significant added
costs to CBM operations in Wyoming (Bleizeffer, 2005a).

Another obstacle to CBM development in the PRB is a decision by the 10th Circuit Court of
Appeals which halted all leasing of federal gas in August 2004. The decision ruled that the BLM
had not addressed the effects of CBM development in earlier environmental impact statements
on which the decision to allow CBM leases was based. Also, BLM had not considered the
option of not issuing questionable leases. The court ruled that BLM must conduct an
Environmental Assessment (EA) specific to issues with coalbed methane extraction which were
not originally considered (e.g. by-product water) and consider not issuing leases. Until
completion of this EA, leasing of federal gas rights is on hold. However, the methane industry
already holds lease rights to 95 percent of federal land in the PRB, with about 3,000 new wells
being drilled annually (Bleizeffer, 2005b).

4.4 Recommendations

In analyses of mine-specific surface mine methane emissions in the U.S., it has been
determined that emissions from the mines of the PRB are most significant and may warrant
further evaluation as candidates for methane recovery. The ten highest emitting mines in the
PRB account for nearly half of all surface mine methane emissions in the U.S. and could be
considered for pre-drainage projects with connection to existing pipeline infrastructure.

CBM development in this area is flourishing at present; however, any methane development in
this area will be subject to stipulations brought about by the conflicting gas ownership issues, as
well as consideration for environmental issues especially by-product water disposal. Even with
obstacles, continued CBM development in the PRB will result in an estimated 40,000 new wells
being drilled over the next decade (Jackson, 2003).

5.0 Summary and Conclusions

Research was conducted in order to assess and improve the current U.S. surface mine
methane emissions inventory. Improvements were made to the SMM emissions inventory in
2003, with the compilation of additional gas content data that led to the use of more
representative gas content values for several coal basins (Table 2.0.1). To further this
improvement, geological research was performed, providing information leading to further
distinction among the gassy coal basins and additional revisions to gas content values used in
emission calculations (Table 2.2.1). Research was also conducted on overburden and coal
seam thicknesses occurring at surface mines as a basis for assessing the current emission
factor of 200 percent used for calculating SMM emissions in the U.S. Comparison of adjacent
seams and strata to the mineable seam has yielded the conclusion that the current 200 percent
emission factor is likely too large, and a factor of 150 percent may be more appropriate. The
current post-mining emission factor was assessed, and it was concluded that without additional

26


-------
data there is no reason for it to be changed. The newly proposed coal basin gas content values
and revised basin distinctions were applied, as well as the suggested emission factor for
comparison with the previous values used. When the new gas contents were applied without
the suggested emission factor, a net increase of 7.5 Bcf was calculated.

It was concluded from the study of international gas contents that the values being used in the
U.S. methodology fall within a reasonable range of international values, and in some instances
are somewhat lower than the worldwide average (Figure 2.6.2).

In order to assess potential improvements to the current methodology and explore methods
closer to Tier 3, research was conducted on several technologies and methods proposed for
measurement of surface mine emissions. It was concluded from literature research that the
spatially-based gas content data should be collected and used in conjunction with an optical-
based measurement of methane flux rates in order to develop a more representative
measurement methodology for surface mine emissions.

Finally, the SMM emissions estimation methodology was applied to the highest producing (more
than 100,000 tons/year) surface mines in the U.S., and the gassiest mines which could be
considered for methane recovery were identified. These mines of interest are in the Powder
River Basin. Though there are several outstanding issues with gas projects in this region, CBM
development is flourishing and recovery looks promising.

27


-------
6.0 References

Australian Methodology for the Estimation of Greenhouse Gas Emissions and Sinks: 2003.
Energy, Fugitive Fuel Emissions. Australian Greenhouse Office, Department of the
Environment and Heritage. May 2005.

Bleizeffer, Dustin. 2005a. CBM operators must bond ponds, www.casperstartribune.net.

August 29, 2005.

Bleizeffer, Dustin. 2005b. BLM on methane leasing: We had it right all along.
www.casperstartribune.net. August 23, 2005.

Canada's Greenhouse Gas Inventory 1990-2002, Energy Section 3.2, Fugitive Fuel Emissions.
Environment Canada, Greenhouse Gas Division. May 2004.

Cosofret, B.R., W.J. Marinelli, T. Ustun, C.M. Gittins, M.T. Boies, M.F. Hinds, D.C. Rossi, R.
Coxe, S. Chang. 2004. Passive infrared imaging sensor for standoff detection of methane
leaks. Presented at SPIE Optics East Chemical and Biological Standoff Detection II,
October 25-28, in Philadelphia, PA.

Creedy, D.P. 1993. Chemosphere. Vol. 26, pp. 419-440. 1993.

Diamond, W.P., LaScola, J.C., and Hyman, D.M. 1986. Results of Direct Method Determination
of the Gas Content of U.S. Coals: U.S. Bureau of Mines Information Circular 9067, 95 p.

Division (PNNL) and U.S. Environment Protection Agency (U.S. EPA), http://www.pnl.gov/

Energy Information Administration (EIA). 2003. Annual Coal Report 2003.
http://www.eia.gov/coal/annual/

Environment Canada, 1992. Canada's Greenhouse Gas Emissions Estimates for 1990. April 1992

Fulton, Tom. 2001. Statement on HR 2952, "The Powder River Basin Resource Development
Act". Presented before the House Resources Subcommittee on Energy and Natural
Resources, October 11, 2001, in Washington, D.C.

IPCC. 2000. Good Practice Guidance and Uncertainty Management in National Greenhouse
Gas Inventories. IPCC, 2000.

Izrael, Yu. A. and S. I. Avdjushin. 1997. Russian Federation Climate Change Country Study -
Final Report - Volume 1 - Inventory of Technogenic GHG Emissions. U.S. Global Change
Research Information Office, http://www.gcrio.org/

Jackson, Hugh. Drilling for tax credits. Citizen.org. July 2, 2003.

KazNIIMOSK. 2002. Kazakhstani GHG Emissions Inventory from Coal mining and Road
Transportation. Prepared by the Kazakh Research Institute for Environment Monitoring and
Climate (KazNIIMOSK) with the support of Battelle Memorial Institute, Pacific Northwest
Division (PNNL) and U.S. Environment Protection Agency (U.S. EPA), http://www.pnl.gov/

28


-------
Keystone Coal Industrial Manual (Keystone). 2004. Published by Coal Age, a Mining Media
publication.

2004.

King, Brian. 1994. Management of Methane Emissions from Coal Mines: Environmental,

Engineering, Economic and Institutional Implication of Options. Neil and Gunter Ltd., Halifax,
March 1994.

Kirchgessner, D.A., S.D. Piccit, and J.D. Winkler. 1992. Estimation of Global Emissions from Coal
Mines. U.S. EPA, 1992

Kirschbaum, M.A., Roberts, L.N.R., and Biewick, L.R.H.(eds.), 2000. Geologic Assessment of
coal in e Colorado Plateau: Arizona, Colorado, New Mexico, and Utah: U.S. Geological
Survey Professional Paper 1625-B, Discs 1 and 2. Version 1.0.

Lloyd, P. and Alan Cook. 2005. Methane Release from South African Coal Mines. Publications
of the University of Cape Town, Energy Research Centre.
http://www.ere, uct.ac.za/Proiects/Coaltech 2020.htm. March 2005.

Molnia, Carol L., Laura R. H. Biewick, and Dorsey Blake. 1997. Coal availability in the Hilight
guadrangle. Powder River Basin. Wyoming: a prototype study in a western coal field.
www.USGS.gov, 1997.

Ophir Corporation. 2005. OPHIR Corporation Recent Research Activities.

Rightmire, C.T., Eddy, G.E., and Kirr, J.N. (eds). 1984. Coalbed Methane Resources of the
United States. AAPG Studies in Geology, Series #17. 1984.

Saghafi A., S Day, DJ Williams, DB Roberts, A Quintanar and JN Carras. 2003. Toward the
Development of an Improved Methodology for Estimating Fugitive Seam Gas Emissions
from Open Cut Mining. Australian Coal Association Project 9063. February, 2003.

Saghafi, A., S. Day, D.J. Williams, D.B. Roberts, A. Quintanar, and J.N. Carras. 2003. Toward the
Development of an Improved Methodology for Estimating Fugitive Seam Gas Emissions From
Open Cut Mining. CSIRO Energy Technology, 2003.

Spaeth, Lisa and Martin O'Brien. 2003. An Additional Tool for Integrity Monitoring. Pipeline and
Gas Journal. March 2003.

Tewalt, Susan J, 1986. "Chemical Characterization of Texas Lignite." The University of Texas
at Austin, Bureau of Economic Geology Geological Circular 86-1.

U.S. EPA, 2001. Kirchgessner, D.A., S.D. Piccot, and S.S. Masemore. An Improved Inventory of
Methane Emissions from Coal Mining in the United States. U.S. EPA , 2001

U.S. EPA, 1996. Masemore, S., S. Piccot, E. Ringler, and W.P. Diamond. Evaluation and
Analysis of Gas Content and Coal Properties of Major Coal Bearing Regions of the United
States. EPA/600/R-96-065. U.S. EPA, 1996.

U.S. EPA. 1991. Kirchgessner, D.A., S.D. Piccot, and A. Chadha. Estimation of Methane
Emissions from Surface Coal Mines Using Open Path FTIR Spectroscopy and Modeling
Technigues. U.S. EPA, 1991.

29


-------
Wilkinson, Todd. 2005. Letter from Wyoming: A Sense of Deja-Boom. USNews.com. August
15, 2005.

Williams D.J., Saghafi A., Lange A., & Drummond, M.S. 1993, Methane Emissions from Open
Cut Mines and Post-Mining Emissions from Underground Coal, CSIRO Investigation report
CET/IR173.

Wyoming State Geological Survey (WSGS). 2002. http://www.wsgs.uwyo.edu/

30


-------