Supplement to the Non-Hg Case Study Chronic Inhalation Risk Assessment In Support of the Appropriate and Necessary Finding for Coal- and Oil-Fired Electric Generating


 W*°"*°" I"11
     rm,
Supplement to the Non-Hg Case Study Chronic
Inhalation Risk Assessment In Support of the
Appropriate and Necessary Finding for Coal-
and Oil-Fired Electric Generating Units

-------

-------
                                                                 EPA-452/R-11-013
                                                                   November 2011
Supplement to the Non-Hg Case Study Chronic Inhalation Risk Assessment In Support of
the Appropriate and Necessary Finding for Coal- and Oil-Fired Electric Generating Units
                        U.S. Environmental Protection Agency
                     Office of Air Quality Planning and Standards
                            Research Triangle Park, NC

-------

-------
Introduction

A previous document1 discussed the methods and results of the chronic inhalation risk
assessment of hazardous air pollutants (HAPs) other than mercury from coal- and oil-fired
electric utility steam generating units (EGUs) at sixteen case study facilities, which was
performed in support of the "appropriate and necessary" finding for coal- and oil-fired EGUs.
Several changes were made to the emissions estimates, dispersion modeling, and risk
characterization of these facilities in response to public comments on the proposed rule, and this
report documents those changes and their impact on the estimated risks from the case study
facilities.

1. Emissions

In response to comments on the proposed rule, the emissions data used in the case studies
were updated for several facilities in two ways. First, in response to comments that EPA's
methodology was not sufficiently refined, EPA used year-specific heat input for the modeled
years 2005 through 2009, rather than the average 2007-2009 heat input for all years. Second, in
response to comments that EPA's approach to emission factor development should use outlier
tests, EPA revised its calculations for emission factors to apply to those units that had not been
tested in the ICR. Only the arsenic, chromium, and nickel emissions were recomputed for use in
modeling because these pollutants were the key risk drivers for the case studies. Both of these
updates were made only to the case study facilities that had estimated cancer risks near 1 in a
million at proposal. The subsections below provide more information on these updates. In
addition, the detailed calculations for the case study emissions at the unit level are provided in
the spreadsheet "Case_Study_Emis_MATS_Final.xlsx," which is included in the docket (EPA-
HQ-OAR-2009-0234-2939).

1.1 Unit-Level Annual Heat Input

The preferred source of unit-level annual heat input data was EPA's Clean Air Markets
continuous emission monitoring (CEM) program. Unit-level annual heat input data for calendar
years 2002-2010 were obtained for all units that report these data to EPA. Heat input data are
important because emissions are proportional to heat input. The only facility without CEM data
was the HECO Waiau facility (ORIS 766). This facility was contacted directly to obtain actual
unit-specific annual heat input data. The year-specific heat input data used for the final rule are
an improvement over the approach used at proposal, which used a multi-year average heat input
for each modeled year.

Table 1 provides the total facility-level heat input for each year, as well as the average
value. While the average value was not used for any calculations, it is provided for comparison
to the value used at proposal. For three of the case study facilities remodeled for the final rule,
the revised heat input is higher (Amerenue-Labadie, 0.3%; James River, 2.5%; and Conesville,
24%). For the rest of the facilities, the heat input is lower (Dominion - Yorktown, 12%;
Chesapeake Energy Center, 8.8%, Heco Waiau, 2.5%; OG &E - Muskogee, 8.6%; PSHNH -
1 Non-Hg Case Study Chronic Inhalation Risk Assessment for the Utility MACT "Appropriate and Necessary"
Analysis, March 16, 2011. Docket ID No. EPA-HQ-OAR-2009-0234-2939. Available at www.regulations.gov.

-------
Merrimack, 16%; and TVA Gallatin, 3.2%).  While these facility-level changes give some
indication of the impact on risk, because each unit at a facility can contribute differently to the
risk, the unit-level heat input changes can give a somewhat different view of the changes. These
unit-level changes are available in the case study emissions spreadsheet discussed above.

Table 1: Final case study heat input (MMBtu) compared to heat input used at proposal.
Facility
Amerenue-
Labadie
City Utilities of
Springfield -
James River
Conesville
Dominion -
Yorktown
Dominion
Chesapeake
Energy Center
Heco Waiau
OG&E-
Muskogee
PSHNH-
Merrimack
TVA Gallatin
2005
168,514,660
16,452,496
96,719,546
40,366,974
41,070,023
14,673,660
106,454,236
34,045,565
73,113,053
2006
170,157,435
15,943,500
92,193,174
21,700,400
37,711,371
14,397,785
104,916,428
34,410,607
74,908,225
2007
182,401,603
13,610,535
108,230,769
29,638,959
40,777,823
14,935,405
87,737,665
36,317,955
77,474,644
2008
165,591,139
15,181,310
100,770,581
22,409,846
36,222,677
13,320,107
104,566,459
30,332,534
79,699,781
2009
166,226,274
11,968,020
66,087,859
19,286,905
35,255,050
14,884,983
92,434,936
25,319,652
64,123,359
Average
170,578,222
14,631,172
92,800,386
26,680,617
38,207,389
14,442,388
99,221,945
32,085,263
73,863,812
Proposed
Rule
171,126,775
14,997,208
115,092,253
23,447,200
34,853,246
14,080,684
90,739,465
26,821,184
71,515,151
1.2 Development of Emission Factors

       For several of the case study facilities, emission factors were used as the best available
alternative source of emissions information because appropriate facility-specific stack test data
were not available to compute emissions. The case study facilities using emission factors that
were not based on site-specific stack test data are Conesville (3% of the risk driver emissions),
PSFINH - Merrimack (35%), and all risk driver emissions at Chesapeake Bay Energy Center,
Dominion - Yorktown, OG&E - Muskogee, and Amerenue-Labadie. The risk driver emissions
are those emissions that contributed the most to the total facility risk. The emission factors used
in these cases were recalculated based on revised methods that had not been used  at the time of
proposal, including the use of well-established, robust outlier checks (Dixon or Rosner tests),
depending on the number of values and when more than three values were evaluated.
The complete documentation for the arsenic, chromium, and nickel emission factors is available
in the docket in separate documents: Coal_Fired_Utility_Boiler_Arsenic.pdf,
Coal_Fired_Utility_Boiler_Chromium.pdf, and Coal_Fired_Utility_Boiler_Nickel.pdf. The
same hexavalent chromium percentages of total chromium used at proposal were used for the
final rule analysis (12% for coal  units and  18% for oil units).

-------
1.3 Emissions Estimates
       Tables 2 through 4 give the estimated emissions for the facilities remodeled. These
emissions estimates reflect the changes in heat inputs and emission factors described above.
While the average values shown were not used in the modeling (the year-specific values were
used), they are provided for comparison to the emissions values used at proposal. The detailed
calculations and unit-level emissions are available in the case study emissions spreadsheet.

Table 2: Final case study arsenic emissions (tons/year) compared to emissions used at
proposal.
Facility
Amerenue-Labadie
City Utilities of Springfield
-James River
Conesville
Dominion - Yorktown
Dominion Chesapeake
Energy Center
Heco Waiau
OG&E -Muskogee
PSHNH -Merrimack
TVA Gal latin
2005
1.05E+00
1.96E-02
1.65E-01
2.29E-01
2.57E-01
2.53E-02
6.65E-01
2.76E-01
1.45E-02
2006
1.06E+00
1.90E-02
1.46E-01
1.32E-01
2.36E-01
2.50E-02
6.56E-01
2.79E-01
1.49E-02
2007
1.14E+00
1.63E-02
1.76E-01
1.73E-01
2.55E-01
2.56E-02
5.48E-01
2.94E-01
1.54E-02
2008
1.03E+00
1.82E-02
1.76E-01
1.35E-01
2.26E-01
2.30E-02
6.54E-01
2.46E-01
1.58E-02
2009
1.04E+00
1.43E-02
1.13E-01
1.17E-01
2.20E-01
2.57E-02
5.78E-01
2.05E-01
1.27E-02
Average
1.07E+00
1.75E-02
1.55E-01
1.57E-01
2.39E-01
2.49E-02
6.20E-01
2.60E-01
1.47E-02
Proposed
Rule
1.72E+00
1.78E-02
3.26E-01
2.09E-01
3.46E-01
2.46E-02
9.11E-01
2.33E-01
1.42E-02
Table 3: Final case study hexavalent chromium emissions (tons/year) compared
emissions used at proposal.
to
Facility
Amerenue-Labadie
City Utilities of Springfield
-James River
Conesville
Dominion -Yorktown
Dominion Chesapeake
Energy Center
Heco Waiau
OG&E -Muskogee
PSHNH -Merrimack
TVA Gal latin
2005
4.94E-01
5.55E-01
4.06E-01
1.51E+00
1.20E-01
2.86E-03
3.12E-01
3.69E-02
1.16E+00
2006
4.99E-01
5.42E-01
2.85E-01
2.84E-01
1.11E-01
2.89E-03
3.08E-01
3.73E-02
1.19E+00
2007
5.35E-01
4.66E-01
3.77E-01
7.78E-01
1.20E-01
2.86E-03
2.57E-01
3.94E-02
1.23E+00
2008
4.86E-01
5.24E-01
4.53E-01
3.67E-01
1.06E-01
2.60E-03
3.07E-01
3.29E-02
1.27E+00
2009
4.88E-01
4.13E-01
2.74E-01
2.51E-01
1.03E-01
2.91E-03
2.71E-01
2.75E-02
1.02E+00
Average
5.00E-01
5.00E-01
3.59E-01
6.37E-01
1.12E-01
2.82E-03
2.91E-01
3.48E-02
1.17E+00
Proposed
Rule
5.77E-01
4.99E-01
6.74E-01
4.25E-01
1.27E-01
2.87E-03
3.06E-01
4.77E-02
1.14E+00

-------
Table 4: Final case study nickel emissions (tons/year) compared to emissions used at
proposal.
Facility
Amerenue-Labadie
City Utilities of Springfield
-James River
Conesville
Dominion - Yorktown
Dominion Chesapeake
Energy Center
Heco Waiau
OG&E -Muskogee
PSHNH -Merrimack
TVA Gal latin
2005
2.31E+00
3.36E+00
3.26E+00
7.26E+01
5.63E-01
3.96E+00
1.46E+00
1.55E-01
5.48E+00
2006
2.33E+00
3.27E+00
2.59E+00
1.18E+01
5.17E-01
3.93E+00
1.44E+00
1.56E-01
5.61E+00
2007
2.50E+00
2.80E+00
3.25E+00
3.63E+01
5.59E-01
4.01E+00
1.20E+00
1.65E-01
5.80E+00
2008
2.27E+00
3.15E+00
3.55E+00
1.59E+01
4.96E-01
3.60E+00
1.43E+00
1.38E-01
5.97E+00
2009
2.28E+00
2.48E+00
2.21E+00
1.04E+01
4.83E-01
4.02E+00
1.27E+00
1.15E-01
4.80E+00
Average
2.34E+00
3.01E+00
2.97E+00
2.94E+01
5.23E-01
3.90E+00
1.36E+00
1.46E-01
5.53E+00
Proposed
Rule
4.21E+00
3.03E+00
3.95E+00
1.84E+01
8.89E-01
3.87E+00
2.23E+00
2.85E-01
5.35E+00
Annual emissions estimates for all case study facilities are given in Table 5, including estimates
for facilities that have been revised as discussed above, and the same estimates used at the time
of proposal for facilities without revised emissions estimates.

-------
Table 5. Case study average annual emissions (2005-2009).
Facility
Xcel Bayfront
Cambria Cogen
SC&E Canadys
Dominion Chesapeake Energy Center*
Conesville*
Exelon Cromby Generating Station
TVAGallatin*
City Utilities of Springfield -James River*
Amerenue-Labadie*

roniMn iviernmacK
Monticello Steam Electric Plant
Location
Ashland, Wl
Ebensburg, PA
Canadys, SC
Chesapeake, VA
Conesville, OH
Phoenixville, PA
Gallatin,TN
Springfield, MO
Labadie, MO
Bow, NH
Mount Pleasant, TX
Unit Type
1 coal
2 coal
3 coal
4 coal
4 coal
1 coal
loil
4 coal
3 coal
4 coal
2 coal
3 coal
Unit ID
5
Bl
B2
CAN001
CAN002
CAN003
Unitl
Unit 2
Units
Unit 4
3
4
5
6
Unitl
Unit 2
1
2
3
4
3
4
5
1
2
3
4
1
2
1
2
As (TRY)

l.OOE-03
1.10E-03
5.10E-02
4.90E-02
2.20E-02
4.1E-02
4.4E-02
6.5E-02
9.0E-02
4.6E-02
4.7E-02
3.2E-02
3.0E-02
6.60E-02
2.40E-03
3.5E-03
3.4E-03
3.9E-03
3.8E-03
3.9E-03
4.2E-03
9.4E-03
2.4E-01
2.7E-01
2.7E-01
2.8E-01
7.8E-02
1.8E-01
6.10E-02
6.20E-02
Cr+6 (TRY)

3.60E-02
7.80E-03
1.90E-02
1.80E-02
2.40E-03
1.9E-02
2.1E-02
3.0E-02
4.2E-02
3.3E-01
1.1E-02
7.2E-03
6.9E-03
3.00E-03
6.00E-04
2.8E-01
2.8E-01
3.2E-01
3.1E-01
l.OE-01
3.8E-02
3.6E-01
1.1E-01
1.3E-01
1.3E-01
1.3E-01
l.OE-02
2.4E-02
1.16E-01
1.17E-01
Ni (TRY)

2.20E-01
7.20E-02
1.30E-01
1.30E-01
9.60E-02
8.9E-02
9.6E-02
1.4E-01
2.0E-01
1.8E+00
5.1E-01
3.5E-01
3.3E-01
1.80E-02
3.40E-01
1.3E+00
1.3E+00
1.5E+00
1.4E+00
6.3E-01
4.1E-01
2.0E+00
5.4E-01
5.9E-01
6.0E-01
6.1E-01
4.4E-02
l.OE-01
4.10E-01
4.10E-01

-------
Facility


OG&E-Muskogee*



Spruance Genco


PSI Energy- Wabash River



•



•

Location


Fort Gibson, OK



Richmond, VA


West Terre Haute, IN



\A/-J-,, |_| 1





Unit Type


3 coal



coal

1 coal-gas

coal


6«;i



2 coal
loil
Unit ID
3
4
5
6
GEN1
GEN2
GENS
GEN4
PG7221FA
4
6
W3
W4
W5
W6
W7
W8
Units 1&2
Units
As (TRY)
1.10E-01
2.0E-01
2.0E-01
2.2E-01
8.40E-04
9.00E-04
3.50E-03
2.40E-03
1.30E-03
1.10E-01
3.30E-01
2.0E-03
2.0E-03
2.8E-03
2.1E-03
8.5E-03
7.5E-03
1.2E-01
3.9E-02
Cr+6 (TRY)
1.98E-02
9.5E-02
9.3E-02
l.OE-01
3.10E-04
3.20E-04
5.60E-04
3.60E-04
7.50E-04
4.50E-03
1.40E-02
2.1E-04
2.1E-04
2.9E-04
1.2E-04
1.2E-03
7.9E-04
5.6E-02
5.8E-01
Ni (TRY)
1.70E-01
4.4E-01
4.4E-01
4.8E-01
2.40E-03
8.30E-03
6.20E-03
4.10E-03
5.00E-03
4.10E-02
1.20E-01
3.2E-01
3.0E-01
4.3E-01
3.2E-01
1.4E+00
1.2E+00
2.6E-01
2.9E+01
* Facility was remodeled for the final rule. Facilities not remodeled for the final rule have the same annual emissions estimate for each year.

-------
2. Dispersion Modeling
The methodology used for dispersion modeling of the case study facilities was discussed in
more detail in the previous document.2 The modeling was done using AERMOD, EPA's
preferred model for near-field dispersion. Table 6 provides the model scenario information used
for the case study facilities. This information is unchanged from the proposed rule. Table 7
provides the stack parameters used for the final rule modeling.

Table 6. Model scenario information.
Facility
Xcel Bayfront
Cambria Cogen
SC&E Canadys
Dominion Chesapeake
Energy Center*
Conesville*
Exelon Cromby
Generating Station
TVAGallatin*
City Utilities of
Springfield -James
River*
Amerenue-Labadie*
PSHNH-Merrimack*
Monticello Steam
Electric Plant
OG&E -Muskogee*
Spruance Genco
PSI Energy- Wabash
River
Heco Waiau*
Dominion -
Yorktown*
Downwash
No
Yes
No
No
No
No
No
No
No
No
No
No
No
No
Yes
Yes
Urban/rural
(population)
Rural
Rural
Rural
Urban (200,000)
Rural
Rural
Rural
Rural
Rural
Rural
Rural
Rural
Urban (200,000)
Rural
Urban (300,000)
Rural
Surface station
Ashland Kennedy
Memorial Airport, Wl
Johnstown Cambria
County Airport, PA
Charleston Intl.
Airport, SC
Norfolk Intl. Airport,
VA
Zanesville Municipal
Airport, OH
Philadelphia Intl.
Airport, PA
Nashville Intl. Airport,
TN
Springfield Regional
Airport, MO
St. Louis Lambert Intl.
Airport, MO
Concord Municipal
Airport, NH
Tyler Pounds Field, TX
Muskogee Davis Field,
OK
Richmond Intl. Airport,
VA
Terre Haute Hulman
Regional Airport ,IN
Honolulu Intl Airport,
HI
Newport News Intl.
Airport, VA
Upper air station
Minneapolis, MN
Pittsburgh, PA
Charleston Intl.
Airport, SC
Washington Dulles,
VA
Wilmington, OH
Washington Dulles,
VA
Nashville Intl.
Airport, TN
Springfield Regional
Airport, MO
Lincoln, IL
Albany, NY
Shreveport, LA
Norman, OK
Washington Dulles,
VA
Wilmington, OH
Lihue, HI
Washington Dulles,
VA
* Facility was remodeled for the final rule.

2 Non-Hg Case Study Chronic Inhalation Risk Assessment for the Utility MACT "Appropriate and Necessary"
Analysis, March 16, 2011. Docket ID No. EPA-HQ-OAR-2009-0234-2939. Available at www.regulations.gov.
7

-------
^able 7. Stack
Facility
Xcel Bayfront
Cambria
Cogen
SC&E Canadys
Dominion
Chesapeake
Energy
Center*
Conesville*
Exelon Cromby
Generating
Station
TVAGallatin*
City Utilities of
Springfield -
James River*
Amerenue-
Labadie*
PSHNH-
Merrimack*
parameter data.
Location
Ashland, Wl
Ebensburg, PA
Canadys, SC
Chesapeake, VA
Conesville, OH
Phoenixville, PA
Gallatin,TN
Springfield, MO
Labadie, MO
Bow, NH
Unit Type
Icoal
2 coal
3 coal
4 coal
4 coal
Icoal
loil
4 coal
3 coal
4 coal
2 coal
Unit ID
5
Bl
B2
CAN001
CAN002
CAN003
Unitl
Unit 2
Units
Unit 4
3
4
5
6
Unitl
Unit 2
1
2
3
4
3
4
5
1
2
3
4
1
2
Lat
46.5872
40.4748
40.4748
33.0646
33.0653
33.065
36.7705
36.7706
36.7709
36.7712
40.1648
40.1862
40.1856
40.1856
40.1524
40.152
36.3156
36.3156
36.3151
36.3151
37.1084
37.1084
37.1084
38.5626
38.5621
38.5614
38.5614
43.142
43.1418
Long
-90.9018
-78.703
-78.703
-80.6235
-80.6232
-80.6218
-76.3012
-76.3011
-76.3009
-76.3008
-81.9044
-81.8787
-81.8798
-81.8798
-75.5303
-75.5304
-86.4005
-86.4005
-86.4009
-86.4009
-93.2602
-93.2598
-93.2605
-90.8381
-90.8377
-90.8371
-90.8371
-71.4685
-71.4682
Stack Height
(m)
59.44
70.10
70.10
60.96
60.96
60.96
53.34
53.34
60.96
60.96
137.16
243.84
243.84
243.84
91.44
91.44
152.70
152.70
153.01
153.01
60.96
60.96
106.68
213.36
213.36
213.36
213.40
68.58
96.62
Stack
Temperature
(K)
371.48
466.48
466.48
415.37
412.04
413.71
430.35
429.25
407.05
427.05
416.48
416.48
324.82
324.82
388.71
388.71
406.48
406.48
395.37
395.37
427.59
427.59
433.15
436.21
424.21
413.54
446.93
391.48
422.04
Stack
Velocity
(m/s)
13.05
27.83
27.83
11.00
12.62
19.93
17.37
17.37
17.98
21.34
10.31
25.30
23.90
23.90
17.92
17.11
17.07
17.07
18.90
18.90
9.45
10.06
27.70
32.66
30.32
31.96
34.83
41.98
36.92
Stack
Diameter
(m)
1.86
2.29
2.29
4.88
4.88
4.88
3.97
3.97
3.97
4.27
5.33
7.93
7.93
7.93
4.27
4.27
7.62
7.62
7.62
7.62
3.65
3.65
3.12
6.25
6.25
6.25
6.25
2.62
4.42

-------
Facility
Monticello
Steam Electric
Plant
c\r*Q c
Muskogee*
Spruance
Genco
PSI Energy -
Wabash River
Heco Waiau*
Dominion -
Yorktown*
Location
Mount Pleasant,
TX
Fort Gibson, OK
Richmond, VA
West Terre
Haute, IN
Waiau, HI
Yorktown, VA
Unit Type
3 coal
3 coal
8 coal
1 coal-gas
2 coal
6 oil
2 coal
loil
Unit ID
1
2
3
4
5
6
GEN1
GEN2
GENS
GEN4
PG7221FA
4
6
W3
W4
W5
W6
W7
W8
Units 1&2
Units
Lat
33.0907
33.0914
33.0923
35.7618
35.7619
35.7621
37.4552
37.4555
37.4557
37.4559
39.5303
39.5274
39.5274
21.3891
21.389
21.3888
21.3887
21.3885
21.3884
37.2154
37.2152
Long
-95.0375
-95.038
-95.0378
-95.2886
-95.288
-95.2872
-77.4312
-77.4309
-77.4307
-77.4304
-87.4256
-87.4232
-87.4232
-157.9615
-157.9613
-157.9612
-157.961
-157.9606
-157.9603
-76.4622
-76.4612
Stack Height
(m)
121.92
121.92
140.21
106.68
106.68
152.40
76.20
76.20
76.20
76.20
68.58
137.16
137.16
42.09
42.09
41.91
41.91
41.91
41.91
98.80
149.05
Stack
Temperature
(K)
453.15
453.15
354.26
402.04
402.04
402.04
355.37
355.37
355.37
355.37
452.59
410.93
410.93
469.26
469.26
414.26
414.26
392.04
392.04
417.35
415.93
Stack
Velocity
(m/s)
24.69
24.69
26.52
14.02
14.02
25.13
17.03
17.03
17.03
17.03
19.17
34.26
34.26
12.25
12.25
12.25
12.25
16.12
16.12
22.60
33.52
Stack
Diameter
(m)
6.55
6.55
2.44
7.31
7.31
6.55
2.62
2.62
2.62
2.62
5.49
7.62
7.62
3.05
3.05
2.74
2.74
3.20
3.20
4.90
6.86
* Facility was remodeled for the final rule.

-------
The list below discusses the inputs used in the modeling of the case study facilities for the
final rule, and notes where there were differences compared to the proposed rule.

1. Each boiler or combination of boilers was modeled as an individual emission point in
AERMOD, unchanged from the proposed rule analysis.
2. Hourly emissions for 2005-2009 were modeled explicitly for each emission point for
arsenic, chromium (VI), and nickel. Five-year average concentrations were calculated
within AERMOD. For the proposed rule analysis, a unit emission rate (1 gram per
second ) was used, and the resulting hourly concentrations were scaled using hourly heat
input values to derive pollutant-specific 5-year average concentrations calculated outside
of AERMOD. This methodology difference does not change the concentration estimates.
3. Building parameterization and surface characteristics were unchanged from the proposed
rule analysis.
4. Some stack parameters changed because of new data received during the public comment
period.
5. Current versions of AERMINUTE (11059) and AERMET (11059) were used, and the
meteorological data used for the proposed rule analysis were reprocessed using these
versions. Beta versions of AERMINUTE and AERMET were used for the proposed rule
analysis.
6. Receptor locations (Census blocks within 20 km of the source) were unchanged from the
proposed rule analysis. The current version of AERMAP (11103) was used to calculate
source and receptor elevations, whereas version 09040 was used for the proposed rule
analysis.
7. The current version of AERMOD (11103) was used. A beta version of AERMOD was
used for the proposed rule analysis.

As noted above, updated versions of AERMAP, AERMINUTE, AERMET, and AERMOD
were used in the modeling of the case study facilities for the final rule. The changes between
versions 09040 and 11103 of AERMAP resulted in no differences in elevations of sources and
receptors. The meteorological data used for the proposed rule were reprocessed using versions
11059 of AERMINUTE and AERMET, and processed along with the proposal emission inputs
in version 11103 of AERMOD to compare differences due to AERMET and AERMOD changes.
The differences in the modeled concentrations were insignificant between the beta and current
versions of AERMET and AERMOD.

3. Chronic Inhalation Risk Assessment

For chronic inhalation exposures, we used the 5-year average ambient concentrations of
HAP estimated from the refined dispersion modeling. The estimated ambient concentration at
each nearby census block centroid was used as a surrogate for the chronic inhalation exposure
concentration for all the people who reside in that census block. We assessed non-cancer health
effects from chronic exposures by comparing the chronic inhalation exposure concentration to
the Reference Concentration values. We calculated the maximum individual risk, or MIR, for
each facility as the cancer risk associated with a continuous lifetime (24 hours per day, 7 days
per week, and 52 weeks per year for a 70-year period) exposure to the maximum concentration at
the centroid of an inhabited census block. Individual cancer risks were calculated by multiplying

-------
the estimated lifetime exposure to the ambient concentration of each HAP (in micrograms per
cubic meter) by its cancer unit risk estimate (URE), which is an upper bound estimate of an
individual's probability of contracting cancer over a lifetime of exposure to a concentration of 1
microgram of the pollutant per cubic meter of air. We used URE values for arsenic and
hexavalent chromium from EPA's Integrated Risk Information System (IRIS), which is a human
health assessment program that evaluates quantitative and qualitative risk information on effects
that may result from exposure to environmental contaminants.  Unit risk estimates in IRIS have
undergone both internal and external peer review.

       In the preamble to the proposed rule, we discussed the reasons for using 65% of the IRIS
URE for nickel subsulfide for all nickel compounds. In July 2011, we completed an external
peer review (using three independent expert reviewers) of the methods used to evaluate the risks
from nickel compounds emitted by EGUs in a report titled, "Methods to Develop Inhalation
Cancer Risk Estimates for Chromium and Nickel Compounds."3 Based on the views of major
scientific bodies such as the International Agency for Research on Cancer (IARC),  the World
Health Organization (WHO),  and the European Union's  Scientific Committee on Health and
Environmental Risks (SCHER), and those of the expert peer reviewers  who commented on our
approaches to risk characterization of nickel compounds,  we consider all nickel compounds to be
carcinogenic  as a group and do not consider nickel speciation or nickel solubility to be strong
determinants  of nickel carcinogenicity.

       Based on this review, we decided to use 100 percent of the current IRIS URE for nickel
subsulfide, rather than assuming that 65 percent of the total mass of emitted nickel might be
nickel subsulfide, as used in previous analyses.  We used the IRIS URE value because IRIS
values are typically preferred for use in HAP risk assessments  performed in support of air toxics
regulations under the Clean Air  Act.4 The IRIS values are preferred because they are developed
in accordance with EPA risk assessment guidelines and because of the level of peer review IRIS
values receive. We used  100 percent of the IRIS value because of the concerns about the
potential carcinogenicity  of all forms of nickel raised by the major national and international
scientific bodies. Nevertheless,  taking into account that there are potential differences in toxicity
and/or carcinogenic potential across the different nickel compounds, and given that there have
been two URE values derived for exposure to mixtures of nickel compounds that are 2-3 fold
lower than the IRIS URE for nickel subsulfide 5, the EPA also  considers it reasonable to use a
value that is 50 percent of the IRIS URE for nickel subsulfide for providing an estimate of the
lower end of a plausible range of cancer potency values for different mixtures of nickel
compounds.

       The health reference values used in the assessment are given in Table 8.
3 Mercury and Air Toxics Standards Rule Docket, ID No. EPA-HQ-OAR-2009-0234. Available at
www.regulations.gov.
4 http://www.epa.gov/ttn/atw/toxsource/chronicpriority.html
5 Two UREs (other than the current IRIS values) have been derived for nickel compounds: one developed by the
California Department of Health Services (http://www.arb.ca.gov/toxics/id/summaryMckel_tech_b.pdf) and the
other by the Texas Commission on Environmental Quality
(http://www.epa.gov/ttn/atw/natal999/99pdfs/healtheffectsinfo.pdf).

                                            11

-------
Table 8. Health reference values used in the assessment.
Pollutant
Arsenic
Chromium (VI)
Nickel
HCI
CAS Number
7440382
18540299
7440020
7647010
URE
(l/ug/m3)
4.3E-03
0.012
0.00048

Source
IRIS
IRIS
IRIS

RfC
(mg/m3)
0.000015
0.0001
0.00009
0.02
Source
CalEPA
IRIS
CalEPA
IRIS
3. Results.

The results of the assessment are given in Table 9. Based on estimated actual emissions,
the highest estimated lifetime cancer risk from any of the sixteen case study facilities was 20 in a
million, driven by nickel emissions from the one case study facility with only oil-fired EGUs.
For the facilities with coal-fired EGUs, there were five with maximum cancer risks greater than 1
in a million (the highest was five in a million), four driven by hexavalent chromium, and one
driven by nickel from an oil-fired EGU. There were also two facilities with coal-fired EGUs with
cancer risks at 1 in a million. All of the facilities had non-cancer target-organ-specific hazard
index (HI) values less than one, with a maximum HI value of 0.4 (also driven by nickel
emissions from the one case study facility with only oil-fired EGUs).

The cancer risk estimates from this assessment indicate that the EGU source category
would not be eligible for delisting under section 112(c)(9)(B)(i) of the CAA, which specifies that
a category may be delisted only when the Administrator determines "... that no source in the
category (or group of sources in the case of area sources) emits such hazardous air pollutants in
quantities which may cause a lifetime risk of cancer greater than one in one million to the
individual in the population who is most exposed to emissions of such pollutants from the
source...." We note that, since these case studies do not cover all facilities in the category, and
since our assessment does not include the potential for impacts from different EGU facilities to
overlap one another (i.e., these case studies only look at facilities in isolation), the maximum risk
estimates from the case studies may underestimate true maximum risks.
12

-------
Table 9. Chronic inhalation risk assessment results.
Facility
Xcel
Bayfront
Cambria
Cogen
SC&E
Canadys
Dominion
Chesapeake
Energy
Center
Conesville
Exelon
Cromby
Generating
Station
TVA Gallatin
City Utilities
of
Springfield -
James River
Amerenue-
Labadie
PSHNH-
Merrimack
Monticello
Steam
Electric
Plant
OG&E-
Muskogee
Spruance
Genco
PSI Energy -
Wabash
River
Heco Waiau
Dominion -
Yorktown
Proposed Rule Assessment
Max
Risk
4.0xlO"9
S.OxlO"7
e.oxio"7
S.OxlO"6
S.OxlO"6
S.OxlO"7
l.OxlO"6
S.OxlO"6
S.OxlO"7
l.OxlO"6
e.oxio"7
l.OxlO"6
S.OxlO"8
l.OxlO"7
l.OxlO"5
l.OxlO"6
Risk Driver
Formaldehyde
Chromium VI
Arsenic
Chromium VI
Chromium VI
Arsenic
Chromium VI
Chromium VI
Arsenic
Arsenic
Chromium VI
Arsenic
Arsenic
Arsenic
Nickel
Chromium VI
Max HI
0.005
0.003
0.009
0.05
0.01
0.008
0.006
0.04
0.006
0.01
0.003
0.01
0.007
0.001
0.4
0.02
HI
Driver
HCI
Nickel
HCI
HCI
Nickel
Nickel
Nickel
Nickel
Arsenic
Arsenic
Arsenic
Arsenic
HCI
Arsenic
Nickel
Nickel
Final Rule Assessment
Max
risk
Risk Driver
Max
HI
HI
Driver
Not Remodeled
Not Remodeled
Not Remodeled
2x10-6
2x10-6
Chromium VI
Chromium VI
0.05 a
0.008
HCIa
Nickel
Not Remodeled
2x10-6
5x10-6
7x10-7
1x10-6
Chromium VI
Chromium VI
Chromium VI
Arsenic
0.007
0.03
0.004
0.01
Nickel
Nickel
Arsenic
Arsenic
Not Remodeled
1x10-6
Chromium VI
0.008
Arsenic
Not Remodeled
Not Remodeled
2xlO-5b
2xlO-6b
Nickel
Nickel
0.4
0.03
Nickel
Nickel
 Although HCI was not included in the remodeling of this facility, HCI was the HI driver pollutant at proposal, and
this is carried through for the final rule assessment.
 Based on considering the emitted nickel to be 100% as potent a carcinogen as pure nickel subsulfide; if we
consider the emitted nickel to be 50% as potent a carcinogen as nickel subsulfide (see text), estimated risks would be
IxlO"5 for Heco Waiau and IxlO"6 for Dominion - Yorktown.
                                                 13

-------

-------
United States                             Office of Air Quality Planning and Standards             Publication No. EP A-452/R-11-013
Environmental Protection                         Research Triangle Park, NC                                      November 2011
Agency

-------