U.S. ENVIRONMENTAL PROTECTION AGENCY
REGION IX
FACT SHEET AND
AMBIENT AIR QUALITY IMPACT REPORT
For a Clean Air Act
Prevention of Significant Deterioration Permit
Palmdale Hybrid Power Project
PSD Permit Number SE 09-01
August 2011
-------
This page left intentionally blank.
-------
PROPOSED PREVENTION OF
SIGNIFICANT DETERIORATION PERMIT
PALMDALE HYBRID POWER PROJECT
Fact Sheet and Ambient Air Quality Impact Report
(PSD Permit SE 09-01)
Table of Contents
Acronyms & Abbreviations i
Executive Summary 1
1. Purpose of this Document 1
2. Applicant 1
3. Project Location 2
4. Project Description 3
5. Emissions from the Proposed Project 7
6. Applicability of the Prevention of Significant Deterioration Regulations 9
7. Best Available Control Technology 12
7.1 BACT for Natural Gas Combustion Turbine Generators 16
7.1.1 Nitrogen Oxide Emissions 16
7.1.2 Carbon Monoxide Emissions 20
7.1.3 PM, PMio and PM2.5 Emissions 23
7.1.4 GHG Emissions 27
7.1.5 BACT During Startup and Shutdown 31
7.2. BACT for Auxiliary Boiler and Heater 32
7.2.1 Nitrogen Oxide Emissions 32
7.2.2 Carbon Monoxide Emissions 33
7.2.3 PM, PM10 and PM2.5 Emissions 34
7.2.4 GHG Emissions 37
7.3 BACT for Emergency Internal Combustion Engines 38
7.3.1 NOx, CO, PM, PM10, PM2.5, and GHG Emissions 38
7.4 BACT for Cooling Tower 40
7.5 BACT for Fugitive Road Dust 43
7.6 BACT for Circuit Breakers 45
7.6.1 GHG 45
8. Air Quality Impacts 46
8.1 Introduction 47
8.1.1 Overview of PSD Air Impact Requirements 47
8.1.2 Identification of PHPP Modeling Documentation 48
8.2. Background Ambient Air Quality 49
8.3 Modeling Methodology for Class II areas 50
8.3.1 Model selection 50
8.3.2 Meteorology model inputs 51
8.3.3 Land characteristics model inputs 51
8.3.4 Model receptors 52
8.3.5 Load screening and stack parameter model inputs 53
-------
8.3.6 Good Engineering Practice (GEP) Analysis 54
8.4 National Ambient Air Quality Standards and PSD Class II Increment Consumption
Analysis 55
8.4.1 Pollutants with significant emissions 55
8.4.2 Preliminary analysis: Project-only impacts 55
8.4.3 Cumulative impact analysis 56
8.5 Class I Area Analysis 62
8.5.1 Class I Increment Consumption Analysis 63
8.5.2 Visibility and Deposition in Class I areas 63
9. Additional Impact Analysis 65
9.1 Soils and Vegetation 66
9.2 Visibility Impairment 68
9.3 Growth 68
10. Endangered Species 69
11. Environmental Justice Analysis 70
12. Clean Air Act Title IV (Acid Rain Permit) and Title V (Operating Permit) 70
13. Comment Period, Hearing, Public Information Meeting, Procedures for Final Decision, and
EPA Contact 70
14. Conclusion and Proposed Action 73
li
-------
Acronyms & Abbreviations
Act
Clean Air Act [42 U.S.C. Section 7401 et seq.]
ACC
Air Cooled Condenser
AFC
Application for Certification
Agency
U.S. Environmental Protection Agency
AQMD
Air Quality Management District
bext
Light extinction coefficient
BA
Biological Assessment
BACT
Best Available Control Technology
BTU
British thermal units
CAA
Clean Air Act [42 U.S.C. Section 7401 et seq.]
CEC
California Energy Commission
CEMS
Continuous Emissions Monitoring System
CFR
Code of Federal Regulations
CO
Carbon Monoxide
CT
Combustion Turbine
CTG
Combustion Gas Turbine
DLN
Dry Low NOx
GE
General Electric
GHG
Greenhouse Gas (Greenhouse Gases)
g/hp-hr
grams per horsepower-hour
gr/scf
Grains per Standard Cubic Feet
EAB
Environmental Appeals Board
EPA
U.S. Environmental Protection Agency
ESA
Endangered Species Act
ESP
Electrostatic Precipitator
FWS
U.S. Fish and Wildlife Service
HHV
Higher Heating Value
HP
Horsepower
HRSG
Heat Recovery Steam Generator
HTF
Heat Transfer Fluid
IRIS
Integrated Risk Information System
ISO
International Organization for Standards
km
Kilometers
kW
Kilowatts of electrical power
kWhr
Kilowatt-hour
mg/L
Milligrams per liter
|ig/m3
Microgram per Cubic Meter
MMBTU
Million British thermal units
MW
Megawatts of electrical power
NAAQS
National Ambient Air Quality Standards
NESHAPS
National Emission Standards for Hazardous Air Pollutants
NMHC
Non-methane Hydrocarbons
-------
NO
Nitrogen oxide or nitric oxide
no2
Nitrogen dioxide
NOx
Oxides of Nitrogen (NO + N02)
NP
National Park
NSPS
New Source Performance Standards, 40 CFR Part 60
NSR
New Source Review
o2
Oxygen
PHPP
Palmdale Hybrid Power Project
PM
Total Particulate Matter
PM2.5
Particulate Matter less than 2.5 micrometers ([j,m) in diameter
PM10
Particulate Matter less than 10 micrometers ([j,m) in diameter
PPM
Parts per Million
PPMVD
Parts per Million by Volume, on a Dry basis
PSD
Prevention of Significant Deterioration
PTE
Potential to Emit
PUC
Public Utilities Commission
RATA
Relative Accuracy Test Audit
RBLC
U.S. EPA RACT/BACT/LAER Information Clearinghouse
SIL
Significant Impact Level
sf6
Sulfur Hexafluoride
SNCR
Selective Non-Catalytic Reduction
so2
Sulfur Dioxide
sox
Oxides of Sulfur
STG
Steam Turbine Generator
TDS
Total Dissolved Solids
TPY
Tons per Year
VV2
Victorville 2 (Hybrid Power Project)
WA
Wilderness Area
11
-------
Proposed Prevention of Significant Deterioration (PSD) Permit
Fact Sheet and Ambient Air Quality Impact Report
PALMDALE HYBRID POWER PROJECT
Executive Summary
The City of Palmdale has applied to EPA Region 9 (EPA) for authorization under the
Clean Air Act (CAA) Prevention of Significant Deterioration (PSD) program to construct
a new power plant that will generate 570 megawatts (MW, nominal) of electricity using
natural gas and solar energy. The power plant, known as the Palmdale Hybrid Power
Project (PHPP or Project), will be located in the town of Palmdale, in Los Angeles
County, California. EPA is issuing a proposed PSD permit for the PHPP, which is
consistent with the requirements of the PSD program for the following reasons:
¦ The proposed PSD permit requires the Best Available Control Technology
(BACT) to limit emissions of nitrogen oxides (NOx), carbon monoxide (CO), total
particulate matter (PM), particulate matter under 10 micrometers ([j,m) in diameter
(PMio), particulate matter under 2.5 ([j,m) in diameter (PM2.5), and greenhouse
gases (GHG), to the greatest extent feasible;
¦ The proposed emission limits will protect the National Ambient Air Quality
Standards (NAAQS) for nitrogen dioxide (N02), CO, PMio, and PM2 5. There are
no NAAQS for PM or Greenhouse Gases.
¦ The facility will not adversely impact soils and vegetation, or air quality, visibility,
and deposition in Class I areas, which are parks or wilderness areas given special
protection under the Clean Air Act.
1. Purpose of this Document
This document serves as the Fact Sheet and Ambient Air Quality Impact Report (Fact
Sheet/AAQIR) for the proposed PSD permit for the City of Palmdale's Project. This
document describes the legal and factual basis for the proposed PSD permit, including
requirements under the CAA, including CAA section 165 and the PSD regulations at Title
40 of the Code of Federal Regulations (CFR) section 52.21. This document also serves as
a Fact Sheet for the proposed PSD permit per 40 CFR section 124.8.
2. Applicant
The name and address of the applicant is as follows:
1
-------
City of Palmdale
38300 Sierra Highway, Suite A
Palmdale, CA 93550^
3. Project Location
The proposed location for the Palmdale Hybrid Power Project is 950 East Avenue M,
Palmdale, California 93550. It is located on an approximately 333-acre parcel west of the
northwest corner of Air Force Plant 42, and east of the intersection of Sierra Highway and
East Avenue M. The City of Palmdale is located within the Antelope Valley Air Quality
Management District (District).
The map below shows the approximate location of the proposed Project.
California
2
-------
4. Project Description
The City of Palmdale has submitted to EPA an application for a PSD permit for the PHPP.
The City of Palmdale's application materials for the PSD permit for the Project are
included in EPA's administrative record for EPA's proposed PSD permit. The PHPP will
be owned by the City of Palmdale and the development of the Project will be managed by
Inland Energy.
We note that the City of Palmdale also has submitted applications for State and local
construction approvals for the Project that are separate from EPA's PSD permitting
process. These applications are referred to as an Application for Certification (AFC)
submitted to the California Energy Commission (CEC) and an application for a
Determination of Compliance (DOC) submitted to the District. The District issued a final
DOC for the Project on May 13, 2010. The CEC issued its Final Commission Decision
approving the Project's Application for Certification on August 10, 2011 (08-AFC-09).
The PHPP is designed to use solar technology to generate a portion of the Project's
output. Primary equipment for the generating facility will include two General Electric
(GE) Frame 7FA natural gas-fired combustion turbine-generators (CTGs) rated at 154
megawatt (MW, gross) each, two heat recovery steam generators (HRSGs), one steam
turbine generator (STG) rated at 267 MW, and 251 acres of parabolic solar-thermal
collectors with associated heat-transfer equipment. The Project will have an electrical
output of 570 MW (nominal) or 563 MW (net). The GE CTG incorporates the "Rapid
Start Process" (RSP), which allows for shorter startup durations of the gas turbines. Table
4-1 lists the equipment that will be regulated by this PSD permit:
3
-------
Table 4-1: Equipment List
Equipment
Two natural uas-llreel
(ill 71'.\ Rapid Sum
Process conihusiion
liii hine generators
(CTCi) \\ilh I leal
Reco\ery Slcain
Generators (I IRS(i)
Description
• llach 154 M\V (gross) CT(i. with a maximum heal input
Auxiliary Boiler
llmeruency Diesel-lired
Internal Combustion
(IC) I-limine
11 ill e r ue n cy I) i ese I -1 ired
IC firewater Pump
I limine
Auxiliary I lealer
Coolinu Tower
Circuit Breakers
Maintenance Vehicle
Traffic Generating
ITmili\ e Road Dusl
rale of 1,7.i(-> MMBlu lir (IIIIV)
llt|uipped with nalural gas duel burners. rated at 5"()
MMBlu hi' (I III\') lor each tnrhine system
llach CTCi \ented to a dedicated Ileal Reco\ery Steam
(ieneralor (I IRS(i) and a shared 2(->7 \l\\ Steam Turbine
(ienerator (ST(.i)
I Emissions of M)-., and CO controlled h\ Dry l.ow-\()-.,
(1)1.\) Coml-iustors. Selecti\e Catalytic Reduction (SCR),
and an Oxidation Catalyst (Ox-Cat)
I l<) MMBlu lii' (I IIIV) with ultra low-\()-., burner. Ilred
on natural uas
2.1)00 |<\\ (Z.OS.i hp)
40 CI'R Part (->o. Subpart Mil emission standards
California Air Resources Board Tier 2 emission standards
IX2 hp (135 k\\ )
4o CI'R Part (->o. Subpart Mil emission standards
California Air Resources Board Tieremission standards
4o MMBlu lir (IIIIV) with ultra low-\()-., burner. Ilred on
natural uas
l.iO.ooo uallons per minute maximum circulation rale
Total dissoKed solids (IDS) concentration in makeup
water of 5.ooo ppm (5."51 mu |.)
Drill eliminator with drill losses less than or equal to
0.0005 percent based on circulation rate
Unclosed-pressure SI',, Circuit Breakers
o 5"ii (by weight) annual leakage rate
|o"„ (by weiuhl) leak detection system
Maintenance \ehicles generating lliuiti\e road dust when
lra\elinu on pa\ed and unpa\ed roadways in the solar field
with the Project
Project ITiuiti\e Dust Control Plan
Electricity will be generated by the combustion turbine generators when the combustion of
natural gas turns the turbine blades. The spinning blades will drive an electric generator
with the potential to generate up to 154 megawatts (MW) of electricity from each turbine.
4
-------
The facility will be operated in combined-cycle mode because each turbine will connect to
a dedicated heat recovery steam generator (HRSG), where hot combustion exhaust gas
will flow through a heat exchanger to generate steam. The facility will be equipped with
duct burners firing natural gas to increase steam output from the HRSG during periods of
peak demand.
The hybrid plant design will include a 251-acre solar field that will consist of parabolic
solar-thermal collectors and associated heat transfer equipment arranged in rows. The
heat transfer fluid will be circulated to a boiler to supply steam directly to the HRSGs to
increase electrical generation from the steam turbine. The fluid will then be recirculated to
the solar arrays. An auxiliary heater will be used to ensure that the heat transfer fluid does
not freeze and stays above 54 degrees F whenever the solar steam unit is off-line .
The Project will require periodic vehicle travel over the unpaved portions of the solar field
to perform routine maintenance including mirror washing, maintenance inspections and
repairs of the piping network, herbicide application and dust suppressant application.
Fugitive dust emissions are expected from maintenance vehicle traffic on the unpaved
areas in the solar fields.
The steam generated from each of the HRSGs will drive a 267 MW steam turbine. On
sunny days, the solar array is capable of providing 50 MW of the total electrical generation
from the steam turbine. Net power plant output, after subtracting electricity used on-site,
will be 563 MW.
Exhaust gas exiting the steam turbine will enter a condenser. Cooling water circulating
through the condenser will condense the steam into water, which will be circulated back to
each HRSG. The condenser cooling water will then flow through a mechanical draft wet
cooling tower, where the remaining heat will be dissipated to the atmosphere, and small
quantities of dissolved solids will become airborne as particulate matter.
The diagram on the following page shows a simplified diagram of the proposed Palmdale
Hybrid Power Project.
5
-------
i'L-w
•
B.H.
lmmII
-
!
2-6
t
I
PROCESS FLOW :IAGR*M
PALMDALE HtBRP POVEF PROJECT
F A i M l> A I I
'Wa. VT
fw^AH^' .*rv.
ASM
B.H.
KM*
n F«:*-r >.¦«
men ¦ i
S"AE
1/30/03 ' QH55-OD?-;>4£
X
6
-------
Air Pollution Control
The PHPP will use Selective Catalytic Reduction (SCR) to reduce NOx emissions from
the combustion turbine generators. The SCR will use aqueous ammonia as the reagent,
where the catalyst facilitates the reaction of the ammonia with NOx to create atmospheric
nitrogen (N2) and water. The PHPP will use an oxidation catalyst to reduce emissions of
CO and volatile organic compounds (VOCs). Although CO is regulated in this proposed
PSD permit, VOCs are regulated by the New Source Review (NSR) permit issued by the
District, as explained in Section 6 below. Pipeline quality natural gas fuel and good
combustion practices will be used to minimize particulate emissions. Thermal efficiency
will be used to minimize GHG emissions.
Additional equipment includes a natural gas-fired auxiliary boiler equipped with an ultra
low-NOx burner, a natural gas-fired auxiliary heater equipped with an ultra low-NOx
burner, a diesel-fired emergency generator and a diesel-fired emergency firewater pump
engine both fired with ultra-low sulfur diesel fuel and compliant with federal NSPS
requirements, and SF6 circuit breakers with leak detection systems.
Power Plant Startup
In a typical combined-cycle gas turbine power plant, components of the steam cycle
cannot withstand rapid temperature changes, limiting how fast the steam turbine may be
started. The "rapid start" design of the PHPP is expected to reduce the time required for
the steam cycle to start up. This is important to air quality for two reasons. First, the
exhaust gas temperature when the steam cycle is not operating is higher than the design
temperature window for the SCR and oxidation catalysts. Second, the plant will generate
more electricity for the amount of fuel burned when the hot gas turbine exhaust is used to
power the steam generator in combined cycle.
The auxiliary boiler is primarily designed to shorten the duration of startups as part of
GE's RSP technology, thus minimizing emissions during CTG startup.
5. Emissions from the Proposed Project
This section describes the pollutants that are covered by the PSD program within the
Antelope Valley Air Quality Management District (District), which is the area in which the
Project is proposed to be located.
The Clean Air Act's New Source Review (NSR) provisions include two preconstruction
permitting programs. First, the PSD program is intended to protect air quality in
"attainment areas,"1 which are areas that meet the National Ambient Air Quality Standards
(NAAQS). EPA is responsible for issuing PSD permits for major new stationary sources
emitting pollutants that are in attainment with (or unclassifiable for) the NAAQS, in
1 PSD also applies to pollutants where the status of the area is uncertain (unclassifiable) for NAAQS.
7
-------
general, and within the District.
Second, the nonattainment NSR program applies in areas where pollutant concentrations
exceed the NAAQS ("nonattainment areas"). The District implements the nonattainment
NSR program for facilities within its boundaries emitting nonattainment pollutants and
their precursors (e.g., volatile organic compounds and nitrogen oxides are precursors to
ambient ozone). Therefore, pollutants that are in nonattainment with the NAAQS within
the District are regulated under a separate nonattainment NSR permit issued by the
District.
Table 5-1 below describes the regulated pollutants that will be emitted by the Project and
their attainment status within the District.
Table 5-1: National Ambient Air Quality Standard Attainment Status for
Antelope Valley Air Quality Management District
Pollutant
Nitrogen Dioxide (\()•)
Sulllir Dioxide (SO.-)
Caii>on Monoxide (CO)
Paniculate Mailer (PM)
Paniculate mailer under l<>
micrometers diameter (PM|,,)
Paniculate Matter under 2.5
micrometers diameter (PM ¦.<)
Ozone
I .ead (Pb)
Sulfuric Acid Mist (11 SO|)
I Ivdrouen Sulfide (IIS)
Total Reduced Sulfur (TRS)
fluorides
(iivenhouse (uises (C ¦ 11(0
Attainment Status
Attainment I nclassiliaMe
Attainment I nclassiliaMe
Attainment
n a
I nclassiliaMe
Attainment I nclassiliaMe
Nonattainment"
Attainment1
n a
n a
n a
n a
n a
Permit Program
PSD
PSD
PSD
PSD
PSD
PSD
\.\-\SR
PSD
PSD
PSD
PSD
PSD
PSD
The PSD program (40 CFR § 52.21) applies to "major" new sources of pollutants for
which an area has been designated attainment or is unclassifiable. A fossil fuel-fired steam
There are no national ambient air quality standards (NAAQS) for PM, H2S04, H2S, TRS, fluorides, or GHGs.
However, in addition to other pollutants for which no NAAQS have been set, these pollutants are listed as regulated
pollutants with a defined applicability threshold under the PSD regulations (40 CFR § 52.21).
3 Because NOx is also a precursor to ozone in this area, it will also be regulated by the separate District ozone non-
attainment New Source Review permit in addition to this PSD permit.
4 Area has not yet been designated for lead and is therefore treated as an attainment area.
8
-------
electric plant with a heat input capacity of 250 MMBtu/hr or greater, such as the PHPP,
that emits or has the potential to emit (PTE) 100 tons per year (tpy) or more of any
pollutant regulated under the Clean Air Act5, is defined as a "major source."
6. Applicability of the Prevention of Significant Deterioration
Regulations
This section describes the PSD applicability thresholds, and our conclusion that N02, CO,
PM, PMio, PM2.5, and GHG will be regulated by EPA's proposed PSD permit.
The estimated emissions in Table 6-1 show that the PHPP will be a major source for NOx,
CO, PM, PM10, PM2.5 and GHG. The annual emission data in Table 3 (based on allowable
operation up to 8,760 hours per year) are based on the applicant's maximum expected
emissions, including emissions from startup and shutdown cycles. The applicant assumes
that all combustion-related emissions of PMi0 are of diameter less than 2.5 microns (i.e.,
PM2.5), which is a conservative estimate, as some particulate emissions may fall in the size
fraction between 2.5 and 10 micrometers.
Once a source is considered major for a PSD pollutant, PSD also applies to any other
regulated pollutant that is emitted in a significant amount. The data in Table 3 show that
emissions of sulfur dioxide (S02) will be less than the major source threshold and less than
the significant emission rate. Therefore, PSD does not apply for S02. Estimated emissions
of the PSD-regulated pollutants from each emission unit are listed in Table 6-1.
5 Other types of "source categories" are subject to either the same 100 tpy threshold, or else a 250 tpy threshold.
9
-------
Table 6-1: Estimated Emissions and PSD Applicability
Pollutant
CO
\().
I'M
I'M,..
I'M.,
SO.
Ph
11 SO,
11 S (incl
IRS)
Muoikles
(•I l(i (incl
COe)
Kslimaled Annual
Kmissions
(Ions/year)
25'i 2
I 14
7^ I
(•>2 5
5(i ()
S
i)
4
Major Source
Threshold
(Ions/year)
| DO
11)1)
11)1)
| 00
| 00
| 00
o (•>
7
ID
Significant
.mission Kale
(tons/year)
| oo
4i)
4i)
o (•>
7
ID
l.^l.i.DOO
| 00.000
75.000
Does PSI)
apply?
Yes
Yes
Yes
Yes
Yes
No
No
No
No
No
Yes
10
-------
Table 6-2: Estimated Emissions of PSD-Regulated Pollutants by Emission Unit
CO
NO\
PM
PM in
PM; ?
CMC (a)
CO;e (b)
Total Facility
250.2 tpy
1 14.9 tpy
79.1 tpy
62.5 tpy
56.0
1.913.376
1.913.000
ctc+iiusc (2)
24S ()
1 13 7
47 S
47 S
47.X
I.WS.074
| .IJ0X.000
Auxiliary Neater
o 74
o 22
o 15
i) 15
o 15
2.340
2.000
Auxiliary lioiler
1 <>l
() 30
o 2')
o 2<)
o 2<)
2.^2<)
.1.000
Knicrgency Diesel
Fngine
0
() 07
o 02
o o2
o 02
27.6
0
Kmcrgency Diesel
Firewater Pump
0 0.i
OO.i
o oo2
o.oo2
o oo2
441
0
Cooling Tower
n a
n a
7 13
7 13
7 13
n a
ii a
Circuit Breakers
n a
n a
n a
n a
n a
<¦) 5(i
0
.Maintenance
Vehicles (c)
n a
n a
2.i So
7 10
0.72
n a
n a
Notes:
(a) Represents all GHG emissions on a mass basis.
(b) Represents the carbon dioxide equivalent (C02e) of all GHG emissions, rounded to the nearest 1,000 tons.
(c) This category represents fugitive road dust emissions (e.g., particulate matter emissions) that are expected from maintenance
vehicle traffic on the unpaved areas in the solar fields.
11
-------
Best Available Control Technology
This section describes EPA's Best Available Control Technology (BACT) analysis for the
control of NOx, CO, PM, PMio, PM2 5, and GHG emissions from this facility. Section
169(3) of the Clean Air Act defines BACT as follows:
"The term 'best available control technology' means an emission limitation based on
the maximum degree of reduction of each pollutant subject to regulation under the
Clean Air Act emitted from or which results from any major emitting facility,
which the permitting authority, on a case-by-case basis, taking into account
energy, environmental, and economic impacts and other costs, determines is
achievable through application of production processes and available methods,
systems, and techniques, including fuel cleaning or treatment or innovative fuel
combustion techniques for control of each such pollutant. In no event shall
application of 'best available control technology' result in emissions of any
pollutants which will exceed the emissions allowed by any applicable standard
established pursuant to section 111 [New Source Performance Standards or
NSPS] or 112 [or NESHAPS] of the Clean Air Act."
See also 40 CFR 52.21(b)(12). In accordance with 40 CFR 52.21(j), a new major
stationary source is required to apply BACT for each regulated NSR pollutant that it
would have the potential to emit (PTE) in significant amounts.
EPA outlines the process it generally uses to do this case-by-case analysis (referred to as a
"top-down" BACT analysis) in a June 13, 1989 memorandum. The top-down BACT
analysis is a well-established procedure that EPA's Environmental Appeals Board (EAB)
has consistently followed in adjudicating PSD permit appeals. See, e.g., In re Knauf, 8
E.A.D. 121, 129-31 (EAB 1999); In re Maui Electric, 8 E.A.D. 1, 5-6 (EAB 1998).
In brief, under the top-down process, all available control technologies are ranked in
descending order of control effectiveness. The PSD applicant first examines the most
stringent technology. That technology is established as BACT unless it is demonstrated
that technical considerations, or energy, environmental, or economic impacts justify a
conclusion that the most stringent technology is not achievable for the case at hand. If the
most stringent technology is eliminated, then the next most stringent option is evaluated
until BACT is determined. The top-down BACT analysis is a case-by-case exercise for the
particular source under evaluation. In summary, the five steps involved in a top-down
BACT evaluation are:
1. Identify all available control options with practical potential for application to
the specific emission unit for the regulated pollutant under evaluation;
2. Eliminate technically infeasible technology options;
12
-------
3. Rank remaining control technologies by control effectiveness;
4. Evaluate the most effective control alternative and document results,
considering energy, environmental, and economic impacts as appropriate; if top
option is not selected as BACT, evaluate next most effective control option;
and
5. Select BACT, which will be the most stringent technology not rejected based
on technical, energy, environmental, and economic considerations.
The proposed Project is subject to BACT for NOx, CO, PM, PMi0, PM2.5, and GHG
emissions. A BACT analysis was conducted for each of the following emission units: the
two natural gas combustion turbines, the 40 MMBtu/hr auxiliary process heater, the 110
MMBtu/hr auxiliary boiler, the two diesel-fired internal combustion engines, the fugitive
road dust emissions, the cooling tower and the circuit breakers. Tables 7-1 and 7-2
provide a summary of the BACT determinations for NOx, CO, PM, PMio, PM2.5, and
GHG from the emission units listed above.
13
-------
Table 7-1: Summary of NOx, CO, PM, PMi0, and PM25 BACT Limits
and Requirements for Testing and Monitoring6
NO\
CO
PM. PManil P.M;5
Kcsirkiions on
I Sil«C
2 Coiiihusiiiin
•
1 1.55 lb/hr
•
5.74 lb/hr
•
4.7 lb/hr
n/a
I'u rhinos
•
1 -hr average
•
1 -hr average
•
3-hr average
(each, lit) duct
•
2.0 ppmvd, 15% (.);
•
1.5 ppmvd. 15% ():x
•
0.0027 lb/MMBtu
burning)
•
CI 'IMS
•
CI MS
•
PI JC natural gas (Sulfur
•
Quarterly and Annual
•
Quarterly and Annual
<0.20 gr/100 dscf on
RATA for CI Ms
RATA for CI Ms
12-month average and
not exceed 1.0 gr/dscf
at anytime)
•
Annual Performance
Testing
2 ( omhiisiion
•
14.6 lb/l:r
•
8.90 lb/hr
•
8.0 lb/hr
•
Total duct
I'u rhinos
•
1 -hr average
•
1 -hr average
•
3-hr average
burning (D3 &
(each, w ith
•
2.0 ppmvd. 15%();
•
2.0 ppmvd. 15% ():
•
0.0035 lb/MMBtu
D4) <2.000
duct burning)
•
PIJC natural gas (Sulfur
lirs/yr
<0.20 gr/100 dscf on
12-month average and
not exceed 1.0 gr/dscf
at anytime)
•
Annual Performance
Testing
2 ( onihusiiiin
•
Cold Start - 52.4 lb/hr.
•
Cold Start - 224 lb/hr.
n/a
•
Cold Start -1 10
I'u rhinos
96 lb/event
410 lb/event
minutes
(each, startup
•
Warm/I lot - 30 lb/hr.
•
W arm I lot - 247
•
Warm/I lot - 80
and shutdown)
40 lb/event
lb/lir. 329 lb/event
minutes
•
Shutdow n - 1 14 lb/hr.
•
Shutdown - 674 lb/hr.
•
Shutdown - 674
57 lb/event
337 lb/event
30 minutes
•
1 -hr average
•
1 -hr average
1 ll'illlT
•
9.0 ppni. 3% ();
•
50.0 ppm. 3% ();
•
0.3 lb/hr for 1 leater
•
1.000 lir/yr
40 MMBtu/hr
•
3-hr average
•
3-hr average
•
0.8 lb/lir for Boiler
•
Non-resettable
(111IV)
•
Initial Performance
•
Initial Performance
•
3-hr average
elapsed time
Testing and at least
Testing and at least
•
PIJC natural gas (Sulfur
meter
lioillT
every 5 years
every 5 years
<0.20 gr/100 dscf on
•
500 lir/yr
35 MMBtu/hr
12-month average and
•
Non-resettable
(IIIIV)
not exceed 1.0 gr/dscf
elapsed time
at anytime)
meter
6 PHPP must keep all records of all testing, fuel use, and fuel testing requirements for a period of five (5) years and must
report excess emissions to EPA semi-annually, except when: more frequent reporting is specifically required by an
applicable subpart; or the Administrator, on a case-by-case basis, determines that more frequent reporting is necessary
to accurately assess the compliance status of the source. .
7 During the initial 3-year demonstration period, the limit will be 7.65 lb/hr.
8 During the initial 3-year demonstration period, the limit will be 2.0 ppmvd, 15% 02.
14
-------
r.nu'r»i'iu'\
(•i'iH'1'iilur
2000 KW
(2.683 lip)
Hrensiier
Pump l.njiiiu-
135 KW (182
hp)
(ooliii" lower
130.000 gpm
N()\
6.4 g/KW-hr.
(4.8 g/hp-hr)''
3-hr average
Initial Performance
Testing
4.0 g/KW-hr.
(3.0 g/hp-hr)"1
3-hr tost average
Initial Performance
Testing
n/a
CO
3.5 g/KW-hr. (2.6
g/hp-hr)
3-hr average
Initial Performance
Testing
n/a
Circuit
lirciikers
na/
n/a
I'M. I'M illlll I'M;:
0.20 g/KW-hr. (0.15
g/hp-hr)
3-hr average
Inclusive use of ultra
low sulfur fuel, not to
exceed 15 ppmvd sulfur
fuel Supplier
Certification
Initial Performance
Testing
1.6 lb/lir (total PM)
< 0.0005% drift
eliminators
< 5000 pjini total
dissolved solids
Weekly water quality
testing
n/a
Keslriclions on
I S«I«C
• 50 hr/year
• Non-resettable
elapsed time
meter
• 50 hr/year
• As required for
lire testing
• Non-resettable
elapsed time
meter
n/a
n/a
Miiinlciiiincc
Vehicle
n/a
n/a
• Fugitive Dust Control
Plan
n/a
Emission standards for NOx in the New Source Performance Standard for stationary compression ignition internal
combustion engines (40 CFR Part 60 Subpart IIII) and the California Tier Emission Standards are based on the sum of
NOx and non-methane hydrocarbons (NMHC). For the NOx emission limits, the applicant assumes NMHC + NOx
emissions from the engine are 95% NOx.
10 Ibid.
15
-------
Table 7-2: Summary of GHG BACT Limits
and Requirements for Testing and Monitoring
2 Coiiihusiiiin
I'u rhinos
(each, lit) duet
burning)
2 Combustion
I'u rhinos
(each, with
duet burning)
2 (omhiislion
I'u rhinos
(each, startup
and shutdown)
(.IK.
774 lb C();/MWh
source-w ide net
output
1 17 lb C():/MMBlu
heat input, each at
IS() standard day
conditions
30-day rolling
average
Tcslin» iind
Monitoring
CI 'IMS
Ki'sirkiiuns on
I SilJ»C
n/a
• Total duct
burning (1)3 &
D4) <2.000
hrs/yr
• Cold Start-1 10
minutes
• Warm/1 lot - 80
minutes
I ll'ilUT
40 MMBtu/hr
(I IIIV)
lioillT
35 MMBtu/hr
(IIIIV)
Circuit
liri'iikiTs
Annual tune-lips
9.56 tpy C():e
0.5% maximum
annual leakage rate
• Non-resettable
elapsed time
meter
• Non-resettable
elapsed time
meter
• 10% leak
detection system
• Monthly pounds
of dielectric fluid
added
1.000 hr/vr
500 hr/vr
n/a
7.1 BACTfor Natural Gas Combustion Turbine Generators
The PHPP will have two combined-cycle, natural gas-fired combustion turbines (CTs). Each CT
has a maximum heat input capacity of 1,736 MMBtu/hr (at ISO conditions) and will have a
dedicated heat recovery steam generator (HRSG) with a 550 MMBtu/hr duct burner. Each duct
burner will be limited to 2,000 hours of operation per year. The CTs are subject to BACT for
NOx, CO, PM, PMio, PM2.5, and GHGs. A top-down BACT analysis for each pollutant has been
performed and is summarized below.
7.1.1 Nitrogen Oxide Emissions
Step 1 - Identify All Control Technologies
The following inherently lower-emitting control options for NOx emissions include:
• Low NOx burner design (e.g., dry low NOx (DLN) combustors)
16
-------
• Water or steam injection
• Inlet air coolers
The available add-on NOx control technologies include:
• Selective Catalytic Reduction (SCR) system
• EMx™ system (formerly SCONOx)
• Selective non-catalytic reduction (SNCR)
Step 2 - Eliminate Technically Infeasible Options
All of the available control options identified in Step 1 are technically feasible.
Step 3 - Rank Control Technologies
A summary of recent BACT limits for similar combined-cycle, natural gas-fired CTs is provided in
Table 7-3. There is one facility that was permitted with a BACT limit less than the limit
proposed by the applicant. The IDC Bellingham facility in Massachusetts was permitted in 2000
with a limit of 1.5 ppm. However, this project was cancelled, so this limit has never been
demonstrated as achievable. All recently issued permits indicate that a limit of 2.0 ppm based on a
1-hr average represents the highest level of NOx control. The available control technologies are
ranked according to control effectiveness in Table 7-4.
SCR and EMx™ for NOx Emissions
Selective catalytic reduction (SCR) is a well-demonstrated technology for NOx control and has
specifically achieved NOx emissions of 2.0 ppm on a 1-hr average on large CTs (greater than 100
MW).
EMx™ technology (formerly SCONOx) is a relatively newer technology that has yet to be
demonstrated in practice on CTs larger than 50 MW. The manufacturer has stated that it is a
scalable technology and that NOx guarantees of <1.5 ppm are available.11 As a result, EMx™ is
considered technically feasible for this facility. However, it is unclear what NOx emission levels
can actually be achieved by the technology.
We found only one BACT analysis that determined that EMx™SCONOx was BACT for a large
CT. However, the accompanying permit for the facility, Elk Hills Power in California, allowed
the use of SCR or SCONOx (the former name of EMx™) to meet a permit limit of 2.5 ppm, and
the actual technology that was installed in that case was SCR.
We also note that the Redding Power Plant in California, a 43 MW gas-fired CT, was permitted
with a 2.0 ppm demonstration limit using SCONOx. In a letter dated June 23, 2005 from the
Shasta County Air Quality Management District (Shasta County AQMD) to the Redding Electric
Utility, however, it was determined that the unit could not meet the demonstration limit and, as a
result, the limit was revised to 2.5 ppm. Based on these two examples, it appears EMx™ has
been demonstrated to achieve only 2.5 ppm and we are therefore evaluating it at this limit.
11 Information available at http://emerachemnew.ciplex.us/emx-product.html. See EMx White Paper 2008.
17
-------
Table 7-4: NOx Control Technologies Ranket
by Control Effectiveness
NOx Control Technology
Emission Rate (ppmvd
(a), 15% O2,1-hr average)
SCR with dry low NOx combustors and inlet air
coolers
2.0
EMx™ with dry low NOx combustors and inlet
air coolers
2.5
SNCR with dry low NOx combustors and inlet
air coolers
-4.512
Dry low NOx combustors and inlet air coolers
9
Water or steam injection
>9
Step 4 - Economic, Energy and Environmental Impacts
The applicant has proposed SCR, the top-ranked technology, as BACT. We have determined that
it is appropriate to consider the collateral environmental impacts associated with SCR. The SCR
system requires onsite ammonia storage and will result in relatively small amounts of ammonia slip
from the CTs' exhaust gases. Ammonia has the potential to be a toxic substance with harmful
side effects, if exposed through inhalation, ingestion, skin contact, or eye contact.13 Ammonia has
not been identified as a carcinogen. It is noted that the applicant will use aqueous ammonia, which
is considered the safer storage method. Additionally, we note that the California Energy
Commission's Presiding Member's Proposed Decision proposes to include Conditions of
Certification to ensure the safe receipt and storage of aqueous ammonia at the PHPP.14
Ammonia slip emissions for the proposed source are limited to 5 ppm by the nonattainment New
Source Review (NSR) permit issued by the District. The District conducted a Health Risk
Assessment (HRA) that included ammonia slip emissions. The results of the assessment showed
that the maximum non-cancer chronic and acute hazard indices were both less than the significance
level of 1.0 (0.0008 and 0.028, respectively).15
Considering the above factors, the possible risks associated with onsite storage and use of
ammonia do not appear to outweigh the benefits associated with significant NOx reductions.
Step 5 - Select BACT
Based on a review of the available control technologies for NOx emissions from natural gas-fired
combustion turbines, we have concluded that BACT for CTs is 2.0 ppm at 15% 02 based on a 1-
hr average. Additionally, we are adding a mass emission limit of 11.55 lb/hr without duct firing
and 14.6 lb/hr with duct firing based on a 1-hr average.
12 This is an approximate value that was estimated considering that the control effectiveness of SNCR has been
demonstrated to be between 40 and 60 percent.
13 Information is available from the Agency for Toxics Substances and Disease Registry at
http ://www. atsdr.cdc. gov/phs/phs. asp?id=9&tid=2.
14 This information is available at http://www.energv.ca.gov/201 lpublications/CEC-800-2011-005/CEC-800-2011-
005-PMPD.pdf. See conditions HAZ-1 through HAZ-6.
15 See Final Determination of Compliance for Palmdale Hybrid Power Project issued by the District on May 13, 2010,
Section 8.
18
-------
Table 7-3: Summary of Recent NOx BACT Limits for Similar Combined-Cycle, Natural gas-fired CTs
l-'sicililv
l.ocitlion
V)\ Limit
Aver:i»in»
Period
( onlrol
Permit Issuance
Source
Avenal Energy Project16
California
2.0 ppm
1-hr
SCR
May 2011
PSD Permit
Warren County Power Station
Virginia
2.0 ppm
1-hr
SCR/DLN
December 2010
PSD Permit
Carty Power Plant
Oregon
2.0 ppm
3-hr rolling
SCR
Draft December 2010
RBLC # OR-0048
Langley Gulch Power Plant
Idaho
2.0 ppm
3-hr rolling
SCR/DLN
Draft December 2010
RBLC # ID-0018
Live Oaks Power Plant
Georgia
2.5 ppm
3-hr
SCR/DLN
April 2010
RBLC #GA-0138
Colousa Generating Station
California
2.0 ppm
1-hr
SCR
March 2010
PSD Permit
Victorville II Hybrid Power Project
California
2.0 ppm
1-hr
SCR
February 2010
PSD Permit
Madison Bell Energy Center
Texas
2.0 ppm
24-hr rolling
SCR
August 2009
RBLC # TX-0548
Chouteau Power Plant
Oklahoma
2.0 ppm
1-hr
SCR/DLN
January 2009
RBLC #OK-0129
Kleen Energy Systems
Connecticut
2.0 ppm
1-hr
SCR/LNB
February 2008
RBLC # CT-0151
PSO Southwestern Power Plant
Oklahoma
9.0 ppm
-
DLN
February 2007
RBLC # OK-0117
FPL West County Energy Center
Unit 3
Florida
2.0 ppm
24-hr
SCR/DLN
July 2008
RBLC # FL-0303
FMPA Cane Island Power Park
Florida
2.0 ppm
24-hr
SCR
September 2008
RBLC # FL-0304
Blythe Energy LLC (Blythe II)
California
2.0 ppm
3-hr
SCR/DLN
April 2007
PSD Permit
Elk Hills Power
California
2.5 ppm
1-hr
SCR/DLN or
SCONOX
January 2006
PSD Permit
Modification
Rocky Mountain Energy Center
Colorado
3.0 ppm
1-hr
SCR/LNB
May 2006
RBLC # C0-0056
San Joaquin Valley Energy Center
California
2.0 ppm
1-hr
SCR/DLN
August 2006
PSD permit
Walnut Energy Center
California
2.0 ppm
1-hr
SCR
2004
California Energy
Commission
Donald Von Raesfeld Power Plant
California
2.0 ppm
1-hr
SCR
2003
California Energy
Commission
IDC Bellingham
Massachusetts
1.5 ppm
1-hr
SCR
2000
SCAQMD - project
cancelled
16 We note that this permit is currently the subject of an administrative appeal to EPA's EAB; however, the appeal does not pertain specifically to the BACT analysis
for NOx or the permit's emission limits for NOx.
19
-------
7.1.2 Carbon Monoxide Emissions
Step 1 - Identify All Control Technologies
The inherently lower-emitting control options for CO emissions include:
• Good combustion practices
The available add-on CO control technologies include:
• Oxidation catalyst
• EMx™
Step 2 - Eliminate Technically Infeasible
All of the available control options identified in Step 1 are technically feasible.
Step 3 - Rank Remaining Control Technologies
A summary of recent BACT limits for similar combined-cycle, natural-gas fired CTs is provided in
Table 7-5. The applicant proposed using oxidation catalyst with a limit of 2.0 ppm (with and
without duct burning) based on a 1-hr average. Currently, the lowest permitted limit for
oxidation catalyst is the Kleen Energy facility in Connecticut, which has a limit of 0.9 ppm (1.8
ppm with duct firing) based on a 1-hr average. The Kleen Energy facility has recently begun
commercial operation, but results from compliance demonstration testing are not available at this
time.17 The next most stringent permitted limit is the Avenal Energy Project in California, which
has a limit of 1.5 ppm following a demonstration period18 (2.0 ppm with duct burning) and also
uses oxidation catalyst. The Avenal Energy Project has not begun construction at this time.
Based on this information, oxidation catalyst is being evaluated at the most stringent control
option.
Oxidation Catalyst and EMx™
Oxidation catalyst is a well-demonstrated technology for large CTs. As discussed in the NOx
BACT analysis, it is clear that EMx™ is an available and technically feasible technology.
However, it is unclear what level of control would be achieved by the technology on a long-term
basis with a short (1-hr) averaging period. The manufacturer claims that emission rates below 1
ppm are achievable, but there is a lack of information that demonstrates this on large CTs. We
are not aware of any BACT determinations that have required EMx™ for CO emissions. Based
on the lack of information for similar units, EMx™ is conservatively being compared as equivalent
to oxidation catalyst.
17 See August 4, 2011 email from Louis Corsino to Lisa Beckham - "Kleen Energy - Middletown, CT".
18 This limit becomes effective after a 3-year demonstration period, during which the limit is 2.0 ppm. As noted above,
this permit is currently the subject of an administrative appeal to EPA's EAB; however, the appeal does not pertain
specifically to the BACT analysis for CO or the permit's emission limits for CO.
20
-------
The available control technologies are ranked according to control effectiveness in Table 7-6.
Table 7-6: CO Control Technologies Ranked by Control Effectiveness
CO Control Technology
Kmission Kale
(ppmvd a 15% O:. 1-
lir average, without
duct 11 ring)
Kmission Kate
(ppmvd a 15% ()2.
l-lir average, with
dud 11 ring)
Oxidation catalyst and good
combustion practices
0.9-2.0 ppm
2.0-2.4 ppm
EMx™ and good combustion
practices
0.9-2.0 ppm
2.0-2.4 ppm
Good combustion practices
8.0 ppm
8.0 ppm
Step 4 - Economic, Energy and Environmental Impacts
Although EMx™ is being considered equivalent to oxidation catalyst for controlling CO
emissions, it was determined to be inferior to SCR for controlling NOx emissions. Because
EMx™ would not ensure BACT is achieved for NOx, it is being eliminated in this step due to
environmental impacts. Overall, better and more reliable pollution control for NOx and CO will
be achieved for the Project with SCR and oxidation catalyst than with EMX™. We are not aware
of any significant or unusual adverse environmental impacts associated with good combustion
practices and an oxidation catalyst.
Step 5 - Select BACT
Based on the review of the available control technologies, we have concluded that BACT for CO
is good combustion practices and an oxidation catalyst with a limit of 1.5 ppm at 15% 02 based
on a 1-hr average without duct firing, and 2.0 ppm with duct firing. Additionally, we are adding a
mass emission limit of 5.74 lb/hr without duct firing and 8.90 lb/hr with duct firing based on a 1-
hr average. However, given the lack of long-term compliance data for the lower limits that would
apply without duct firing, we feel it is appropriate to include permit provisions establishing a
three-year demonstration period for those limits, during which time the limit will be 2.0 ppm at
15% 02 and 7.65 lb/hr based on a 1-hr average without duct firing.
Demonstration period permit provisions will require that, prior to construction, the permittee
submit design specifications as proof that the gas turbines were designed to achieve 1.5 ppm. The
permittee must also submit a plan that sets forth the measures that will be taken to maintain the
system and optimize its performance. The permittee must operate the gas turbines according to
the design specifications and within the design parameters, and consistent with the maintenance
and performance optimization plan. Following the first three years of commercial operation, the
limits of 1.5 ppm (1-hour average) without duct firing will take effect unless the emissions and
operating data collected by the applicant indicates that these limits are not feasible, and the
applicant submits an application to EPA no later than the end of the 3-year period requesting a
revision to the limit. If such a revision is requested but EPA determines that a revision is not
warranted, the lower emission limit will become applicable.
21
-------
Table 7-5: Summary of Recent CO BACT Limits for Similar Combined-Cycle, Natural gas-fired CTs
l.< iciilion
CO l.imil (( C)
l.imil with duel
llrinii)
.\M'rii»in» Period
Cunl ml
IVrmil Issuiinoc
Sou roc
Avenal Energy Project
California
1.5 ppm19 (2.0
ppm)
1-hr
Oxidation
catalyst
June 2011
PSD Permit
Warren County Power Station
Virginia
1.5 ppm (2.4
ppm with duct
burning)
1-hr
Oxidation
catalyst/GCP
December 2010
PSD Permit
Langley Gulch Power Plant
Idaho
2.0 ppm
3-hr rolling
Oxidation
catalyst/GCP
Draft December
2010
RBLC # ID-0018
Live Oaks Power Plant
Georgia
2.0 ppm
3-hr
Oxidation
catalyst/GCP
April 2010
RBLC # GA-0138
Colousa Generating Station
California
3.0 ppm
3-hr
Oxidation
catalyst
March 2010
PSD Permit
Victorville II Hybrid Power
Project
California
2.0 ppm (3.0
ppm)
1-hr
Oxidation
catalyst
February 2010
PSD Permit
Madison Bell Energy Center
Texas
17.5 ppm
1 -hr rolling
GCP
August 2009
RBLC # TX-0548
Chouteau Power Plant
Oklahoma
8.0 ppm
1-hr
GCP
January 2009
RBLC #OK-0129
Lamar Power Partners II
Texas
15 ppm
24-hr rolling
GCP
June 2009
RBLC # TX-0547
Patillo Branch Power Plant
Texas
2.0 ppm
3-hr rolling
Oxidation
catalyst
June 2009
RBLC # TX-0546
Cane Island Power Park
Florida
8 ppm
24-hr
GCP
September 2008
RBLC # FL-0304
Elk Hills Power
California
4.0 ppm
1-hr
Oxidation
catalyst
January 2006
PSD Permit
Modification
Kleen Energy Systems
Connecticut
0.9 ppm (1.8
ppm with duct
firing)
1-hr
Oxidation
catalyst
February 2008
RBLC # CT-0151
19 This limit becomes effective after a 3-year demonstration period. During the demonstration period, the limit is 2.0 ppm.
22
-------
7.1.3 PM, PMio and PM2 5 Emissions
Because the applicant has assumed that all particulate emissions from the turbines are PM2.5, the
BACT analyses for PM, PMi0 and PM2.5 have been combined. Additionally, the analysis evaluates
total particulate emissions - condensable and filterable.
Step 1 - Identify All Control Technologies
The following inherently lower-emitting control options for PM, PMio, and PM2.5 emissions
include:
• Low particulate fuels, low sulfur fuels, and/or pipeline natural gas (also referred to as
"clean fuel")
• Good combustion practices (including air inlet filter)
The available add-on PM, PMio, PM2.5 control technologies include:
• Cyclones (including multiclones)
• Wet scrubber
• Dry electrostatic precipitator (ESP)
• Wet ESP
• Baghouse/fabric filter.
Step 2 - Eliminate Technically Infeasible Control Options
All of the control technologies identified are technically feasible except for cyclones (including
multiclones). Although cyclones have been identified as being capable of marginal PM2 5
control20, the low grain loading makes them technically infeasible for this application. EPA's Air
Pollution Control Technology Fact Sheet for Cyclones (EPA-452/F-03-005) identifies typical
grain loading for cyclones as ranging from 1.0 to 100 gr/scf and being as low as 0.44 gr/scf.21 In
contrast, the grain loading for the CTs' exhaust stream would be about 0.0015 gr/scf based on the
applicant's proposed BACT limits. Cyclones are generally used in high dust applications where a
majority of the particulate emissions are filterable emissions. In contrast, the majority of
emissions from the CTs will be condensable particulate matter.
Step 3 - Rank Remaining Control Technologies
A review of other BACT limits for similar combined-cycle natural gas-fired CTs is provided in
Table 7-7. We note that many BACT determinations that were concluded prior to January 1,
2011 included limits only for filterable PM.22 Because our BACT analysis for the Project must
address total PM (filterable plus condensable), we did not further evaluate PM limits addressing
20 -Information available at
http://www.epa.gov/apti/Materials/APTI%20413%20student/413%20Student%20Manual/SM ch%204.pdf.
21 Information is available at http://www.epa.gov/ttn/catc/dirl/fcvclon.pdf.
22 See 40 CFR 52.21(b)(50) - On or after January 1, 2011, such condensable particulate matter shall be accounted for in
applicability determinations and in establishing emissions limitations for PM, PM2.5, and PMi0 in PSD permits.
23
-------
solely filterable PM, which would not be applicable here. The applicant proposed a total PM limit
of 12 lb/hr without duct firing and 18 lb/hr with duct firing. In order to compare these emission
rates to similar facilities, these limits were converted to lb/MMBtu - 0.0069 lb/MMBtu, and
0.0079 lb/MMBtu, respectively.
The most recently permitted units with total PM limits using lb/MMBtu are Warren County
Power Station in Virginia (Warren County) and the Chouteau Power Plant in Oklahoma
(Chouteau). Of these two facilities, only the Chouteau unit is operational and demonstrated to be
in compliance with its PM limits.23 The applicant's proposed emission rates appear to be
significantly higher on a lb/MMBtu basis when compared to Chouteau (0.0035 lb/MMBtu) and
Warren County (0.0027 lb/MMBtu without duct burning and 0.0040 lb/MMBtu with duct
burning). The results from the total PM testing at Chouteau showed total PM emissions to be
equivalent to 0.0029 lb/MMBtu (with a 99 MMBtu/hr duct burner).24 Therefore, we believe the
uncontrolled emission rates that should be evaluated are 0.0027 lb/MMBtu without duct burning
and 0.0035 lb/MMBtu with duct burning.
We were not able to identify any CT using add-on PM controls; however, such controls are
considered technically feasible and are therefore being further evaluated. Wet ESP has been
evaluated as the highest performing control option because all particulate emissions are expected
to be PM2.5 and wet ESP is expected to perform better in this range as compared to the other add-
on control technologies. The applicant eliminated the wet scrubber as an option due to possible
increases in PM emissions associated with the total dissolved solids (TDS) content of the water
available at the facility. However, it is not clear this has ever been demonstrated as a problem and
therefore we have conservatively included wet scrubber for further consideration in the BACT
analysis. We identified a control efficiency of 90% for this option based on the document used by
the applicant for the economic analysis - "Controlling Fine Particulate Matter Under the Clean Air
Act: A Menu of Options," prepared by the State and Territorial Air Pollution Program
Administrators (STAPPA) and Association of Local Air Pollution Control Officials (LAPCO)
(hereinafter "Controlling Fine PM'). 25 The applicant also conservatively assumed 99% PM2.5
control for baghouse and dry ESP.
23 See August 3, 201 lemail from Lisa Beckham, EPA Region 9, to Shirley Rivera, EPA Region 9 re: "Chouteau Power
Plant in Oklahoma".
24 See August 8, 2011 emails from Lisa Beckham, EPA Region 9, to Shirley Rivera, EPA Region 9 re: "Chouteau Power
Plant in Oklahoma".
25 Information is available at http://www.4cleanair.org/PM25Menu-Final.pdf.
24
-------
Table 7-7: Summary of Recent PM BACT Limits for Similar Combined-Cycle, Natural gas-fired CTs
l.< u'iilion
PM l imit (PM l imit
xWDncl l;irin»)
Tj|U'«.r PM -
l'ilU-r;il>k-( I').
Toliil(T)
A* crsi»in<>
Period
Colli ml
Pi-mi i I
IssllillH'O
Sou roc
Avenal Energy Project26
California
8.91 lb/hr (11.78
lb/hr)27
TPMio
12-month
rolling
Natural Gas
Fuel
June 2011
PSD Permit
Warren County Power
Station
Virginia
8 lb/hr (14 lb/hr)
TPMio, TPM2.5
3-hr
—
December 2010
PSD Permit
Warren County Power
Station
Virginia
0.0027 lb/MMBtu
(0.0040 lb/MMBtu)
TPMio, TPM2.5
3-hr
—
December 2010
PSD Permit
Carty Plant
Oregon
2.5 lb/MMscf
FPMio
—
Clean Fuel
Draft December
2010
RBLC # OR-0048
Langley Gulch Power
Plant
Idaho
No limit
FPMio
—
GCP
Draft December
2010
RBLC # ID-0018
Colusa Generating Station
California
13.5 lb/hr
TPM, TPMio
12-month
rolling
Natural Gas
Fuel
March 2010
PSD Permit
Victorville II Hybrid
Power Project
California
12.0 lb/hr (18.0 lb/hr)
TPM, TPM2.5
12-month
rolling
Natural Gas
Fuel
March 2010
PSD Permit
Chouteau Power Plant
Oklahoma
6.59 lb/hr, 0.0035
lb/MMBtu
TPMio
3-hr
Natural Gas
Fuel
January 2009
RBLC #OK-0129
Cane Island Power Park
Florida
2 gr S/100 scf
TPMio
—
Fuel Spec
September 2008
RBLC # FL-0304
FPL West County Energy
Center Unit 3
Florida
2 gr S/100 scf
PM/PMi0/PM2.5
—
Fuel Spec
July 2008
RBLC # FL-0303
Plaquemine Cogeneration
Facility
Louisiana
33.5 lb/hr, 0.02
lb/MMBtu
FPMio, TPM
—
Clean Fuel
July 2008
RBLC #LA-0136
Aresnal Hill Power Plant
Louisiana
24.23 lb/hr
FPM
—
GCP/Pipeline
NG
Mar-08
RBLC # LA-0224
Kleen Energy Systems
Connecticut
11 lb/hr (15.2 lb/hr)
FPMio
—
—
February 2008
RBLC # CT-0151
26 As noted above, this permit is currently under administrative appeal; however, the appeal does not pertain specifically to the BACT analysis for PMi0 or to the
permit's emissions limits for PMi0.
27 These limits are equivalent to 0.0048 lb/MBBtu without duct firing and 0.0049 lb/MMBtu with duct firing, based on the size of the CTs and duct burners.
25
-------
The available add-on control technologies are ranked according to control effectiveness in
Table 7-8.
Table 7-8: PM Control Technologies Ranked by Control Effectiveness
I'M Control Technologies
Emission Kale
(Ib/MMIilu. 3-hr
average)
Emission Kale
\v/l)ucl Burners
(Ih/M MIitii. 3-hr
average)
Wet ESP
0.00004
0.00004
Dry ESP/Baghouse
0.00004
0.00004
Wet Scrubber (Venturi)
0.0004
0.0004
Baseline (Clean Fuel)
0.0027
0.0035
Step 4 - Economic, Energy and Environmental Impacts
The applicant provided a cost analysis based on information provided in Controlling Fine PM. A
modified version of this analysis is provided in Table 7-9. The amount of PM2 5 removed is based
on the baseline (natural gas) emission rates in Table 7-8. Because add-on PM controls have not
been applied to CTs, the control efficiencies evaluated are considered conservative. With cost-
effectiveness values ranging between $109,000 and $193,000 per ton of PM2 5 removed, add-on
controls are considered cost-prohibitive for the PHPP.
Table 7-9: Cost Analysis for Add-on PM Control Technologies
W ei i:si>
l)rv I.SI>
li;ii>hoiisc
(pulse-jel
ck'illK'll)
Wei Scrubber
(\ enluri)
Flowrate (ft3/min)
946,777
946,777
946,777
946,777
Capital Costs ($/scfm)
$20
$10
$6
$3
Capital Costs ($)
$18,935,540
$9,467,770
$5,680,662
$2,366,942.50
Cost Recovery Factor
0.11
0.11
0.11
0.11
Annualized Capital Costs ($/yr)
$2,082,909
$1,041,454.70
$624,872.82
$260,363.68
O & M Costs ($/scfm)
$5
$3
$5
$4.40
O & M Costs ($/yr)
$4,733,885
$2,840,331
$4,733,885
$4,165,819
Total Annualized Costs ($/yr)
$6,816,794
$3,881,786
$5,358,758
$4,426,182
Removal Efficiency
99.1%
99%
99%
90%
Tons of PM2.5 Removed (TPY)
35.38
35.34
35.34
32.13
Cost Effectiveness ($/ton
removed)
$192,680
$109,830
$151,620
$137,760
Step 5 - Select BACT
After eliminating wet ESP, dry ESP, fabric filter, and wet scrubber due to economic impacts, we
26
Palmdale (SE 09-01)
Fact Sheet/Ambient Air Quality Impact Report
August 2011
-------
have determined that BACT is clean fuel, good combustion practices, a PM, PMi0, and PM2.5
limit of 0.0027 lb/MMBtu without duct burning and a limit of 0.0035 lb/MMBtu with duct
burning based on a 3-hr average. Additionally, we are setting mass emission limits of 4.7 lb/hr
without duct firing and 8.0 lb/hr with duct firing based on a 3-hr average. By "clean fuel" we
mean Public Utilities Commission (PUC)-quality natural gas. PUC-quality pipeline natural gas
shall not exceed a sulfur content of 0.20 grains per 100 dry standard cubic feet on a 12-month
rolling average and shall not exceed a sulfur content of 1.0 grains per 100 dry standard cubic feet,
at any time. This limit is lower than the limit proposed by the applicant. However, when
comparing the applicant's proposed emission rates to other recently permitted sources, the
applicant's values are in some cases twice as high. The applicant relied solely on the Victorville II
facility in California in proposing emission rates. While the two facilities are very similar, a BACT
analysis should be more comprehensive in evaluating proposed limits. A broader review of recent
BACT determinations demonstrates that BACT is lower than the limits proposed by the applicant.
7.1.4 GHG Emissions
Step 1 - Identify all control technologies
The inherently lower-emitting control options for GHG emissions include28:
• Use of new thermally efficient combined cycle gas turbines - A combined-cycle gas
turbine recovers the waste heat from the gas turbine using a heat recovery steam
generator (HRSG). The use of the HRSG allows more energy to be produced without
additional fuel use.
The add-on control options for GHG emissions include:
• Carbon capture and sequestration (CCS) - CCS is a technology that involves capture and
storage of C02 emissions to prevent their release to the atmosphere. For a gas turbine,
this includes removal of C02 emissions from the exhaust stream, transportation of the
C02to an injection site, and injection of the C02 into available sequestration sites.
Potential C02 sequestration sites include geological formations (including oil and gas
fields for enhanced recovery) and ocean storage.
Step 2 - Eliminate technically infeasible control technologies
CCS
As described briefly above, CCS involves three main components: capturing the C02 emissions
from the exhaust stream, transporting the captured C02to the sequestration site, and injection of
the C02 into a geologic reservoir for long-term sequestration. All three of these aspects are
relevant when determining whether CCS is technically feasible for a particular project.
28 In addition to the measures discussed here specifically for the gas turbines, we note that the project design includes 50
MW of potential solar thermal power generation, which represents an inherently lower-emitting technology for the
facility as a whole.
27
Palmdale (SE 09-01)
Fact Sheet/Ambient Air Quality Impact Report
August 2011
-------
The applicant proposed to eliminate CCS because C02 capture is not technically feasible for CTs.
The applicant identified three potential processes for capturing C02 from flue gas: solvent-based
processes, sorbent-based processes, and membrane-based processes. The applicant concluded
that these processes were not technically feasible due to limited experience in the energy industry
and lack of commercial demonstrations. However, commercial C02 recovery plants have been in
existence since the late 1970s, with at least one plant capturing C02 from gas turbines.29'30 The
applicant also identified as a hurdle that commercial demonstrations have only captured a fraction
of the C02 in flue gas. This consideration appears to be less of a technical feasibility issue than
one of cost, which would be more appropriately addressed in Step 4 of the BACT analysis. Based
on available information, we consider carbon capture from gas turbines to be technically feasible
for the Project.
In its application, the applicant identified several geological formations in the lower San Joaquin
Valley and Ventura County that could potentially provide a suitable site for geologic
sequestration; a map of those sites provided in the Project application is provided in Figure 7-1.
While geotechnical analyses have not been conducted to verify the suitability of these sites, other
proposals have been made to capture and sequester C02 emissions in the San Joaquin Valley; as a
result, there is a reasonable presumption that suitable sequestration sites do exist in these areas
despite the lack of extensive studies prepared for this Project. Nevertheless, the primary issue
with the feasibility of CCS in this case lies with the location of the PHPP in relation to the
sequestration sites and the surrounding geography. As shown in the figure above, significant
mountain ranges lie between the project location and the potential sequestration sites (oil fields,
gas fields, and ocean storage). Sequestration of C02 emissions from the Project would require
construction of C02 pipelines through these mountains. The offsite logistical barriers of
constructing such a pipeline (e.g., land acquisition, permitting, liability, etc.) make this technology
technically infeasible for the Project.
Because constructing a new C02 pipeline was determined to be technically infeasible, the
applicant also evaluated whether C02 pipelines were already available near the proposed Project.
The Technical Advisory Committee for the California Carbon Capture and Storage Review Panel
stated in an August 2010 report that there are no existing C02 pipelines in California.31 In
addition, based on a search of the California Environmental Quality Act (CEQA) State
Clearinghouse database maintained by the California Office of Planning and Research, there are
no C02 pipeline projects underway in California subject to CEQA. Last, the applicant also
contacted the Department of Oil, Gas and Geothermal Resources and facilities operating in Kern
County, and again, found no existing pipelines in California.
29 Herzog, H.J., "An Introduction to C02 Separation and Capture Technologies," Energy Laboratory Working Paper,
(1999). Available at http://seauestration.mit.edu/pdf/introduction to capture.pdf.
30 Johnson, D., Reddy, S., & Brown, J.H. (2009), Commercially Available C02 Capture Technology. Power. Retrieved
from http://www.powermag.com/coal/2064.html.
31 This information is available at http://climatechange.ca.gov/carbon capture review panel/meetings/2010-08-
18/white papers/Carbon Dioxide Pipelines.pdf.
28
Palmdale (SE 09-01)
Fact Sheet/Ambient Air Quality Impact Report
August 2011
-------
Figure 7-1 Potential CO: Sequestration Sites in Southern California
rt m /
- mt ¦<*>; „,
Jy-'Z-s?A'-r Ml p
W&^f T ' ¦¦
w&rtf'; t
* jSCT ; // r i
WkMp
' - v
Sfpl
r#jK
yggj}
Sr%
— » -is. & • * • » — i _L
Oil
Gas
Undetermined (Oil or Gas)
7a
•"•Mr A
3t./#-'s ft
PHPP
w_ - L '
V
Data source: National Energy Technology Laboratory, Department of Energy. 2010 Carbon
Sequestration Atlas of the United States and Canada, Third Edition
In sum, while we have determined that C02 capture and storage is technically feasible, we
conclude that transport of the captured C02 to the potential sequestration sites is not feasible. As
a result, CCS is not technically feasible for the Project and will not be considered further in the
BACT analysis. We note that evaluation of long-term C02 storage is an important part of the
29
Palmdale (SE 09-01)
Fact Sheet/Ambient Air Quality Impact Report
August 2011
-------
technical feasibility analysis. However, because transport of C02 is not technically feasible, it is
not necessary to evaluate the feasibility of C02 storage.
Step 3 - Rank remaining control technologies
After elimination of CCS as a potential control technology, the use of a thermally efficient
combined-cycle gas turbine and a combined-cycle facility are the only control methods remaining.
The expected emissions from a facility with these control options is compared with the emissions
from a simple-cycle gas turbine in Table 7-10. Currently, the only other similar facility with a
GHG BACT limit is the Russell City Energy Center, to be located in Hayward, California. The
PSD permit for this facility has a voluntary GHG limit of a heat rate not to exceed 7,730 Btu/kWh
for each CT and HRSG.
Table 7-10: GHG Control Technologies Ranked by Control Effectiveness
GIIG Control Technologies
Emission Kale
(lb ( ();/M\Vh)
New combined-cycle gas CT
774
Existing combined-cycle CTs32
824-996
Simple-cycle CTs33
1,319
Step 4 - Economic, Energy, and Environmental Impacts
The applicant has chosen the highest ranked control option for each unit, and we are not aware of
any significant or unusual adverse environmental impacts associated with the chosen technology.
Step 5 - Select BACT
Based on a review of the available control technologies for GHG emissions from natural gas-fired
combustion turbines, we have concluded that BACT for this source is the use of new thermally
efficient CTs and emission limits of 774 lb C02/MWh for source-wide net output, and 117 lb
C02/MMBtu heat input for each gas turbine and duct burner (both based on a 30-day rolling
average). The emission limits are based on the emission factor provided by the applicant of 53.06
kg/MMBtu, the 1,736 MMBtu/hr heat input of each CT operating 8,760 hours per year, and the
550 MMBtu/hr duct burner for each CT operating 2,000 hours per year.
A number of issues regarding these limits bear clarification. First, the pollutant that is subject to
regulation under the Clean Air Act for PSD permitting purposes is a group of six gases: carbon
dioxide, methane, nitrous oxide, hydrofluorocarbons, perfluorocarbons, and sulfur hexafluoride.
As a general matter, it may thus be appropriate to establish BACT limits on a C02e basis. In this
case, however, we have elected to establish the BACT limit for C02 specifically. The purpose of
this is to enable the use of C02 CEMS for monitoring purposes. Because the CEMS are required
for other regulatory purposes, they offer a cost-effective and reliable method for monitoring
32 These figures are based on GHG performance information provided by the applicant in Tables 3 and 4 to the PHPP
GHG BACT Analysis dated May 2011. These values are derived from 2008 data from the California Energy
Commission for similar facilities with energy output of at least 3,000 GWh per year.
33 These numbers are based on the proposed CTs operating in simple cycle with a gross output of 154 MW each.
30
Palmdale (SE 09-01)
Fact Sheet/Ambient Air Quality Impact Report
August 2011
-------
compliance. Using C02 as a surrogate for the total emissions on a C02e basis is appropriate in
this case because nitrous oxide and methane are emitted from CTs in minor amounts and the
majority of the GHG emissions actually are C02. For example, EPA's emission factors for C02,
methane, and nitrous oxide from the combustion of natural gas are 53.06 kg/MMBtu, 0.0059
kg/MMBtu, and 0.0001 kg/MMBtu, respectively. The emission factor for all GHGs on a C02e
basis is 53.21 kg/MMBtu. Thus, even after accounting for the global warming potential of
methane and nitrous oxide, the C02 emission factor accounts for 99.7% of the emission on a
C02e basis. Further, an emission limitation that limits C02 emissions from the combustion of
natural gas inherently limits the emission of methane and nitrous oxide. As a result, we believe
that for this particular source, formulating the emission limits and monitoring requirements in
terms of C02 rather than on a C02e basis is appropriate. The applicant has proposed a BACT
limit of 1,020,000 tons of C02 per year for each CT. However, a limit based on the amount of
C02 generated per MWh will ensure that the CTs are operating at peak efficiency. An input-
based limit is also necessary to ensure peak operating efficiency of the gas turbine because the
solar thermal operation will at times contribute to the electric output.
7.1.5 BACT During Startup and Shutdown
It is not technically feasible to use SCR and oxidation catalyst to control NOx and CO emissions
when the equipment is outside of the manufacturer's recommended operating temperature ranges.
For SCR and oxidation catalyst this occurs during turbine startup or shutdown. Therefore, BACT
is achieved by minimizing the time for startup and shutdown. The PHPP will have a 110
MMBtu/hr auxiliary boiler that will be used to reduce the startup time for each turbine. The
applicant has proposed the following NOx and CO emission rate limits for each event:
• Hot/Warm Startup: 40 pounds of NOx and 329 pounds of CO per turbine
• Cold Startup: 96 pounds of NOx and 410 pounds of CO per turbine
• Shutdown: 57 pounds of NOx and 337 pounds of CO per turbine
An evaluation of startup and shutdown emission limits for other similar sources found a wide
range of limits. In many cases, limits are based on pounds per hour or pound per event,34 and this
approach makes it difficult to compare BACT determinations because mass emission rates vary
based on the size of the unit. Other facilities have longer averaging periods (24-hr), which may
incorporate startup and shutdown emissions. Because the PHPP has short 1-hour averaging
periods, it is appropriate to set limits on a mass basis and limit the duration of startup and
shutdown events. Based on the available information, the emission rate limits and fast startup and
shutdown times for the CTs represent BACT for NOx and CO during startup and shutdown.
Therefore, we have determined that BACT during startup and shutdown for NOx and CO for the
PHPP is as described below in Table 7-11.
34 Recently issued permits with these types of limits include the permits for the Avenal Energy Project in California, the
Russell City Energy Project in California, the Victorville II Hybrid Power Project in California, and the Colusa
Generating Station in California.
31
Palmdale (SE 09-01)
Fact Sheet/Ambient Air Quality Impact Report
August 2011
-------
In addition, we have determined that the startup duration limits also constitute BACT for GHG
emissions, because the shorter startup time increases the overall thermal efficiency of the facility.
Therefore, BACT for the PHPP's GHG emissions during startup is 110 minutes for a cold startup
and 80 minutes for a warm/hot startup.
Table 7-11: Summary of NOx and CO BACT Limits During Startup and Shutdown
M>\
CO
Duration
Cold Startup
96 lb/event
410 lb/event
110 minutes
52.4 lb/hr
224 lb/hr
Warm/Hot Startup
40 lb/event
329 lb/event
80 minutes
30 lb/hr
247 lb/hr
Shutdown
57 lb/event
337 lb/event
30 minutes
114 lb/hr
334.6 lb/hr
7.2. BACT for Auxiliary Boiler and Heater
The applicant is proposing to construct a 110 MMBtu/hr boiler that will be used to start up the
CTs, and a 40 MMBtu/hr heat transfer fluid (HTF) heater as part of the solar array system. Both
units will be fired with natural gas. The boiler will be limited to 500 hours of operation per year
and the HTF heater will be limited to 1,000 hours of operation per year. The low hours of
operation and low emission rates proposed result in very low tons per year emission rates for each
unit. The boiler and HTF heater are subject to BACT for NOx, CO, PM, PMi0, PM2 5, and GHGs.
A top-down BACT analysis for each pollutant has been performed and is summarized below.
7.2.1 Nitrogen Oxide Emissions
Step 1 - Identify All Control Options
The following inherently lower-emitting control options for NOx emissions include:
• Low NOx burner design (e.g. low NOx burners, flue gas recirculation)
• Limited use of equipment (limits on the hours of operation)
The available add-on NOx control technologies include:
• Selective Catalytic Reduction (SCR) system
• EMx™ system (formerly SCONOx)
• Selective non-catalytic reduction (SNCR)
Step 2 - Eliminate Technically Infeasible Options
SCR, EMx™, and SNCR are considered technically infeasible control options. The applicant
estimated the exhaust temperature for each unit at 300°F. This is below the temperature
operating range for SCR, EMx™, and SNCR, which are all generally above 400°F.
32
Palmdale (SE 09-01)
Fact Sheet/Ambient Air Quality Impact Report
August 2011
-------
Step 3 - Rank remaining control technologies
The applicant proposed a NOx emission limit of 9 ppm at 3% 02 based on a 3-hr average using
ultra-low NOx burner design. With the proposed low NOx burner designs and limited hours of
operation the auxiliary boiler will emit up to 0.30 TPY of NOx and the heater will emit up to 0.22
TPY. A review of other BACT determinations was not performed because it is very unlikely that
a more detailed review would change the final determination due to the limited use and low ton
per year emission rates associated with the proposed limits.
Table 7-12: .NOx Control Technologies Ran
ted by Control Effect
NO\ Control Technologies
Emission Kale
(ppnivd jl, 3% 02)
Low NOx burners and limited use
9
Step 4 - Economic, Energy, and Environmental Impacts
The applicant has chosen the highest ranked control option for each unit, and we are not aware of
any significant or unusual environmental impacts associated with the chosen technology.
Step 5 - Select BACT
Based on the review of the available control technologies, we have concluded BACT is the limited
hours of operation, ultra-low NOx burners and an emission rate of 9.0 ppm at 3% 02 based on a
3-hr test average.
7.2.2 Carbon Monoxide Emissions
Step 1 - Identify All Control Technologies
The following inherently lower-emitting control options for CO emissions include:
• Good combustion practices
• Limited use (limits on the hours of operation)
The available add-on CO control technologies include:
• Oxidation catalyst
• EMx™ (formerly SCONOx)
Step 2 - Eliminate Technically Infeasible
Oxidation catalyst and EMx™ are considered technically infeasible control options. The applicant
estimated the exhaust temperature for each unit at 3OOF. This is below the temperature operating
range for oxidation catalyst and EMx™, which are generally above 400F.
Step 3 - Rank Remaining Control Technologies
The applicant proposed a CO limit of 50 ppm at 3% 02 based on a 3-hr average using good
combustion practices. With the proposed good combustion practices and limited hours of
operation, the auxiliary boiler will emit up to 1.01 TPY, and the heater will emit up to 0.74 TPY,
of CO. A review of other BACT determinations was not performed because it is very unlikely
33
Palmdale (SE 09-01)
Fact Sheet/Ambient Air Quality Impact Report
August 2011
-------
that a more detailed review would change the final determination due to the limited use and low
ton per year emission rates associated with the proposed limits.
Table 7-13: CO Control Technologies Ranked by Control Effectiveness
CO Control Technologies
Emission Kale
(ppmvd a 3% ():)
Good combustion practices and
limited use
50
Step 4 - Economic, Energy and Environmental Impacts
The applicant has chosen the highest ranked control option for each unit, and we are not aware of
any significant or unusual adverse environmental impacts associated with the chosen technology.
Step 5 - Select BACT
Based on the review of the available control technologies, we have concluded that BACT is the
limited hours of operation, good combustion practices and an emission rate of 50.0 ppm at 3% 02
based on a 3-hr test average.
7.2.3 PM, PM10 and PM2.5 Emissions
The applicant has assumed that all particulate emissions from the auxiliary boiler and process
heater are PM2 5. As a result, the BACT analyses for PM, PMi0 and PM2.5 have been combined.
Additionally, the analysis evaluates total particulate matter - filterable and condensable.
Step 1 - Identify All Control Technologies
The following inherently lower-emitting control options for PM, PMio, and PM2.5 emissions
include:
• Low particulate fuels, low sulfur fuels, and/or pipeline natural gas (also referred to as
"clean fuel")
• Good combustion practices (including air inlet filter)
• Limited use (limits on the hours of operation)
The available add-on PM, PMi0, PM2.5 control technologies include:
• Cyclones (including multiclones)
• Wet scrubber
• Dry electrostatic precipitator (ESP)
• Wet ESP
• Baghouse/fabric filter.
Step 2 - Eliminate Technically Infeasible Control Options
34
Palmdale (SE 09-01)
Fact Sheet/Ambient Air Quality Impact Report
August 2011
-------
All of the control technologies identified are technically feasible except for cyclones (including
multiclones). As evaluated for the CTs, the low grain loading associated with natural gas
emissions makes cyclones technically infeasible for this application.
Step 3 - Rank Remaining Control Technologies
We were not able to identify any CT using add-on PM controls; however, they are considered
technically feasible and are therefore being further evaluated. The available control technologies
are ranked according to control effectiveness in Table 7-14. This analysis is based on the PM,
PMio, and PM2.5 analysis for the CTs.
With the proposed good combustion practices and limited hours of operation, the auxiliary boiler
will emit up to 0.25 TPY of PM, PMio, and PM2.5 and the heater will emit up to 0.15 TPY. A
review of other BACT determinations was not performed because it is very unlikely that a more
detailed review would change the final determination due to the limited use and low ton per year
emission rates associated with the proposed limits.
Table 7-14: PM Control Technologies Ranked by Control Effectiveness
I'M Control Technologies
C011I rol
KITiciencv
Wet ESP
99.1%
Dry ESP/baghouse
99%
Wet Scrubber (Venturi)
90%
Clean fuel, good combustion
practices, and limited use
0% (baseline)
Step 4 - Economic, Energy and Environmental Impacts
The applicant eliminated the use of add-on PM controls for each unit because of the associated
economic impacts. The 110 MMBtu/hr auxiliary boiler is limited to 500 hours of operation per
year and has a potential to emit 0.2 TPY of PM, PMio, and PM25. The 40 MMBtu/hr heater is
limited to 1,000 hours of operation per year and has a potential to emit 0.15 TPY of PM, PMio,
and PM2.5. Due to the limited hours of operation and limited environmental benefit it would be
impractical to require add-on controls to remove less than 0.45 TPY of PM, PMio, and PM25.
However, the applicant also provided an economic analysis for add-on controls, which is provided
in Tables 7-15 and 7-16.
Palmdale (SE 09-01)
Fact Sheet/Ambient Air Quality Impact Report
August 2011
35
-------
Table 7-15: Cost Analysis for Add-on PM Control Technologies for the Auxiliary Boiler
Control Device
\\ ei i:si>
Drv ESI*
Pulse Jet
hihric Tiller
Wet
Scrubber
Flowrate (scfm)
28416
28416
28416
28416
Capital Costs ($/scfm)
$20
$10
$6
$3
Capital Costs ($)
$568,320
$284,160
$170,496
$71,040.00
Cost Recovery Factor
0.11
0.11
0.11
0.11
Annualized Capital Costs ($/yr)
$62,515
$31,257.60
$18,754.56
$7,814.40
O & M Costs ($/scfm)
$5
$3
$5
$4.40
O & M Costs ($/yr)
$142,080
$85,248
$142,080
$125,030
Total Annualized Costs ($/yr)
$204,595
$116,506
$160,835
$132,845
Removal Efficiency
99.1%
99%
99%
90%
Tons of PM2 5 Removed (TPY)
0.20
0.20
0.20
0.18
Cost Effectiveness ($/ton
removed)
$1,032,300
$588,400
$812,300
$738,000
Table 7-16: Cost Analysis for Add-on PM Control Techno
ogies for the HTF Heater
Control Device
\\ ei i:si>
Drv KM'
li;ii>housc
(pulse- jet
denned)
Wet
Scrubber
Flowrate (scfm)
10612
10612
10612
10612
Capital Costs ($/scfm)
$20
$10
$6
$3
Capital Costs ($)
$212,240
$106,120
$63,672
$26,530.00
Cost Recovery Factor
0.11
0.11
0.11
0.11
Annualized Capital Costs ($/yr)
$23,346
$11,673.20
$7,003.92
$2,918.30
O & M Costs ($/scfm)
$5
$3
$5
$4.40
O & M Costs ($/yr)
$53,060
$31,836
$53,060
$46,693
Total Annualized Costs ($/yr)
$76,406
$43,509
$60,064
$49,611
Removal Efficiency
99.1%
99%
99%
90%
Tons of PM2.5 Removed (TPY)
0.15
0.15
0.15
0.14
Cost Effectiveness ($/ton
removed)
$514,000
$293,000
$404,500
$367,500
Step 5 - Select BACT
Based on the review of the available control technologies, we have concluded BACT is the limited
hours of operation, good combustion practices, and clean fuel. By "clean fuel" we mean Public
Utilities Commission (PUC)-quality natural gas. PUC-quality pipeline natural gas shall not exceed
a sulfur content of 0.20 grains per 100 dry standard cubic feet on a 12-month rolling average and
shall not exceed a sulfur content of 1.0 grains per 100 dry standard cubic feet, at any time.
36
Palmdale (SE 09-01)
Fact Sheet/Ambient Air Quality Impact Report
August 2011
-------
Additionally, based on the PTE for each unit, we are setting a PM, PMi0, and PM2.5 limit of 0.8
lb/hr for the boiler and 0.3 lb/hr for the HTF heater based on a 3-hr average.
7.2.4 GHG Emissions
Step 1 - Identify all control technologies
The applicant generally assumed that the auxiliary boiler and HTF heater would incorporate the
newest designs that increase thermal efficiency, such as new burner technologies and modern
optimized instrumentation and controls.
The inherently lower-emitting control options for GHG emissions include:
• Conducting an annual boiler tune-up - this would ensure that optimal thermal efficiency
is maintained. Maintaining higher thermal efficiency reduces the amount of fuel
combusted, which helps to minimize GHG emissions.
The add-on control options for GHG emissions include:
• CCS - CCS is a technology that involves capture and storage of C02 emissions to prevent
their release to the atmosphere. For a gas turbine, this includes removal of C02 emissions
from the exhaust stream, transportation of the C02to an injection site, and injection of the
C02 into available sequestration sites. Potential C02 sequestration sites include
geological formations (including oil and gas fields for enhanced recovery) and ocean
storage.
Step 2 - Eliminate technically infeasible control technologies
CCS
The GHG BACT analysis for the CTs, discussed above, concluded that although C02 capture and
storage is technically feasible, transport of the captured C02 to the potential sequestration sites is
not technically feasible. Using this same analysis, CCS is also not technically feasible for the
auxiliary boiler and HTF heater and will not be considered further in the BACT analysis.
Step 3 - Rank remaining control technologies
After elimination of CCS as a potential control technology, the purchase of thermally efficient
units and annual boiler tune-ups are the remaining technologies. Both of these options will be
required.
Step 4 - Economic, Energy, and Environmental Impacts
The applicant has chosen the highest ranked control option for each unit, and we are not aware of
any significant or unusual adverse environmental impacts associated with the chosen technology.
37
Palmdale (SE 09-01)
Fact Sheet/Ambient Air Quality Impact Report
August 2011
-------
Step 5 - Select BACT
Based on a review of the available control technologies for GHG emissions from natural gas-fired
boilers and process heaters, we have concluded that BACT for this source is the purchase of
thermally efficient units, conducting annual boiler tune-ups on each unit, limiting the auxiliary
boiler to a heat input of 110 MMBtu/hr and 500 hours of operation per year based on a 12-month
rolling total, and limiting the HTF heater to 40 MMBtu/hr and 1,000 hours of operation per year
based on 12-month rolling total. Currently, there are no other facilities with GHG BACT limits
for limited use natural gas-fired boilers and process heaters.
7.3 BACTfor Emergency Internal Combustion Engines
The project includes a 2,862 HP (2134 kW) diesel-fired emergency generator and a 182 HP
(138kW) diesel-fired emergency fire pump engine. Each engine will be limited to 50 hours of
operation each year. The low hours of operation result in very low tons per year emission rates
for each unit. This equipment is subject to BACT for NOx, CO, PM, PMio, PM2 5, and GHGs. A
top-down BACT analysis has been performed and is summarized below.
7.3.1 NOx, CO, PM, PM10, PM2.5, and GHG Emissions
Step 1 — Identify all control technologies
The control options for NOx emissions from engines include SCR, NOx reducing catalyst, NOx
adsorber, catalyzed diesel particulate filter, catalytic converter, and oxidation catalyst.35 A
catalytic converter and oxidation catalyst are also control options for CO emissions. For PM,
PMio, and PM2.5 emissions, a diesel particulate filter/trap can be added on.
Unlike other combustion equipment (e.g., CTs and boilers), new engines are required to be
certified in compliance with NSPS requirements, including emission limits, upon purchase.
Different types of engines have different emission requirements based on the type of engine being
purchased (emergency engine, emergency fire pump engine, or non-emergency engine). Engine
manufacturers may need to employ some of the control technologies identified above in order to
comply with the NSPS emission limits, depending on the type of engine and the applicable limits.
The applicant is proposing to construct an emergency engine and an emergency fire pump engine.
As a result, to comply with NSPS the applicant must purchase engines that meet the emission
requirements for emergency engines and emergency fire pump engines. However, we note that the
applicant could purchase engines that meet the NSPS standards for non-emergency engines,
which have more stringent limits, and operate them as emergency engines. In addition, the
applicant must comply with California Air Resources Board (CARB) emission standards (Tier 2
standards for the emergency generator and Tier 3 standards for the emergency fire pump engine);
however, the CARB standards are the same as the applicable NSPS requirements. As a result,
this review identifies the control technologies to be:
35 The applicant discusses these control options in Section 8.4 of the "Supplemental Information for the Application for
PSD Permit" dated July 21, 2010.
38
Palmdale (SE 09-01)
Fact Sheet/Ambient Air Quality Impact Report
August 2011
-------
• NSPS-compliant emergency engine and NSPS-compliant emergency fire pump engine
• Engines that meets NSPS for non-emergency engines
• Limiting use (limits on the hours of operation)
Step 2 - Eliminate technically infeasible control options
All of the control technologies identified are assumed to be technically feasible.
Step 3 - Rank remaining control technologies
The available control technologies are ranked according to control effectiveness in Table 7-17.36
Table 7-17: Emergency Engine Control Technologies Ranked by Control Effectiveness
Kniiinc Tvpc
WIIIC +\()\
(»/k\Y-hr)
I'M
(»/k\Y-hr)
CO
(»/k\Y-hr)
NSPS-Non-emergency (for
135 kW)
0.0237
0.59
5.0
NSPS-Non-emergency (for
2000 kW)
1.0738
0.10
3.5
NSPS-Fire Pump Engines
(for 135 kW)
4.0
0.20
3.5
NSPS-Emergency (for
2000 kW)
6.4
0.20
3.5
Step 4 - Economic, energy and environmental impacts
Due to economic impacts and limited environmental benefit, the applicant eliminated add-on
controls for the engines. We agree that the top-ranked control technology (purchasing engines
that meet NSPS standards for non-emergency engines and operating them as emergency engines)
would be impractical in this case. This is illustrated in Table 7-18 by the potential emissions from
these units (based on 50 hours of operation per year and complying with the NSPS for emergency
engines and emergency fire pump engines). Requiring the additional reductions in emissions that
would be gained by use of engines that meet NSPS standards for non-emergency engines would
have very little environmental benefit, which would not justify the cost. While the potential C02e
emissions associated with this equipment are higher than those of the other pollutants, they still
represent less than 0.01% of source-wide C02e emissions. A review of other BACT
determinations was not performed because it is very unlikely that a more detailed review would
change the final determination due to the limited use and low ton per year emission rates
associated with the proposed limits.
36 CARB-compliant engines are not listed in the rankings because the emission limitations are the same as for NSPS-
compliant engines.
37 The actual applicable NSPS limits are 0.40 g/kW-hr for NOX and 0.19 g/kW-hr for NMHC. The tow limits were
added together in order to compare them to the other types of engines
38 The actual applicable NSPS limits are 0.67 g/kW-hr for NOx and 0.40 g/kW-hr for NMHC. The two limits were
added together in order to compare them to the other types of engines.
39
Palmdale (SE 09-01)
Fact Sheet/Ambient Air Quality Impact Report
August 2011
-------
Table 7-18: Summary o
Potential to Emit for Emergency Engines
l.iiiergciK'v
Kmergency l-'ire
Pollutant
(ieneralor
Pump Kngine
(TPY)
(TPY)
NOx
0.67
0.03
CO
0.39
0.03
PM, PM10, PM2,
0.02
<0.01
C02e
27.6
4.41
Step 5 - Select BACT
Based on the review of the available control technologies, we have concluded that BACT is the
limited hours of operation and the emission limits listed in Table 7-19 based on a 3-hour
average.39 The NSPS for engines does not currently regulate GHG emissions, but a separate
GHG limit is not being proposed. It is assumed that newly purchased engines would be the most
energy efficient available and that operating in compliance with NSPS requirements will ensure
that each engine is properly maintained and as efficient as possible.
Tab
e 7-19: Summary of BACT Emission Limits for Emergency Engines
llnginc
WIIIC +\(>\
(g/kW-hr)
PM
(g/kW-hr)
CO
(g/kW-hr)
135 kW Emergency Fire
Pump Engine
4.0
0.20
3.5
2000 kW Emergency
Engine
6.4
0.20
3.5
7.4 BACTfor Cooling Tower
The PHPP includes a 130,000 gallons per minute (gpm), ten-cell evaporative (wet) cooling tower.
Fugitive particulate emissions are generated from the cooling tower due to the total dissolved
solids (TDS) in the water. The cooling tower is subject to BACT for PM, PMio, and PM2.5. A
top-down BACT analysis has been performed and is summarized below. The applicant
conservatively assumed PM, PMi0 and PM2.5 emissions from the cooling tower were equivalent.
Step 1 - Available Control Technologies
The following inherently lower-emitting control options for PM, PMio, and PM2.5 emissions
include:
• Dry cooling - uses an air cooled condenser (ACC) that cools the steam turbine-
generators' exhaust steam using a large array of fans that force air over finned tube heat
exchangers. The exhaust from the steam turbine flows through a large diameter duct to the
ACC where it is condensed inside the tubes through indirect contact with the ambient air.
The heat is then released directly to the atmosphere.
39 These limits are the same as the applicable CARB Tier 2 and Tier 3 standards.
40
Palmdale (SE 09-01)
Fact Sheet/Ambient Air Quality Impact Report
August 2011
-------
• Wet-dry hybrid cooling - uses wet and dry cooling technologies in parallel, and uses all of
the equipment involved in both wet and dry cooling. Hybrid cooling technology divides
the cooling function between the wet and dry systems depending on the capabilities of
each system under different environmental and operational conditions.
The available add-on PM, PMi0, and PM2.5 control technologies include:
• Drift eliminators
Step 2 - Eliminate Technically Infeasible
All of the available control options identified in Step 1 are technically feasible.
Step 3 - Rank Remaining Control Technologies
The types of cooling towers are ranked according to control effectiveness in Table 7-20.
Table 7-20: Cooling Tower Control Technologies Ranked by Control Effectiveness
Control Technologies
Emission Kale
(TPY of
P.M/P.M in/PM; ?)
Dry cooling
0
Wet-dry hybrid cooling
3.640
Wet cooling with 0.0005% drift
eliminators
7.1
Step 4 - Economic, Energy and Environmental Impacts
The applicant eliminated the use of both a dry cooling system and wet-dry hybrid cooling system
due to the associated economic and environmental impacts. The use of a dry or hybrid wet-dry
system would reduce the overall efficiency of the facility, due to the additional energy
requirements for the wet and hybrid systems. The applicant also conducted an economic analysis
comparing the annual operation costs of wet and dry cooling systems. The applicant's analysis is
reproduced in Table 7-21.
Table 7-21: Wet and Dry Cooling Tower Cost Analysis Provided by the Applicant
Wei Cooling
l)rv Cooling
Tower
Tower
Required Power
Fan Power(e)
1,700 kW
6,350 kW
Circulating Pump Power
2,400 kW
OkW
40 The applicant did not estimate potential emissions from a wet-dry hybrid system. We have approximated emissions
from such a system to be one-half of those from a wet cooling system.
41
Palmdale (SE 09-01)
Fact Sheet/Ambient Air Quality Impact Report
August 2011
-------
Wei Cooling
l)rv Cooling
Tower
Tower
Power Loss Due to High Steam Turbine
Backpressure
OkW
536 kW
Water Treatment Power Consumption
(Zero Liquid Discharge)
850 kW
<200 kW
Total Net Power Loss Effect
12,798 kW
14,042 kW
Costs
Direct Capital Cost
$26,000,000
$59,000,000(e)
Water Pipeline Installation®
-$1,400,000
$0
Annualized Cost
Capital Recovery1-3-1
$1,940,000
$3,680,000
Equivalent Electrical Power Cost(b)
$16,816,500
$18,451,000
Treatment Chemical Addition(c)
$250,000
$0
Makeup Cooling Water(d)
$824,200
-$100,000
Total $/year
$19,830,700
$22,231,000
Notes:
a) Assumes a 30-year lifetime with a 5.75% interest rate.
b) Assumes the facility operates 8,760 hour/yr and a power cost of $0.15/kWh.
c) Assumes that water treatment chemicals would be needed in a wet tower to prevent
corrosion, bio-fouling, etc., but would not be needed for an ACC.
d) Estimated at $200/acre-foot and consumption of 4,121 acre-feet per year for wet
cooling.
e) Does not include additional costs required for a steam turbine that can be operated
at high back pressure.
f) Only includes the less than 2 miles of pipeline needed to connect to the regional
backbone system. Dry cooling costs are underestimated since some water is needed
even in a dry-cooled plant, which would still require a pipeline.
The cost effectiveness of using a dry cooling process to reduce 7.1 TPY of PM, PMi0, and PM2.5
is $338,000 per ton. The applicant estimated a hybrid cooling system would have direct capital
costs of $67 million and, as a result, would be even less cost-effective than a dry cooling system.
Based on this information, we agree that using dry or hybrid cooling systems in this case would
not be cost-effective and would contribute to a decrease in the overall energy efficiency of the
facility.
Considering collateral environmental impacts, the use of wet cooling has a potential impact
associated with additional consumption of water resources. However, the water being used for
the cooling tower is from the Palmdale Water Reclamation Plant and therefore wet cooling is not
expected to result in any significant adverse impact on water resources in the area.
Step 5 - Select BACT
The applicant proposed using a wet cooling tower with 0.0005% drift eliminators as BACT for
42
Palmdale (SE 09-01)
Fact Sheet/Ambient Air Quality Impact Report
August 2011
-------
the steam turbine cooling system. A comparison of the drift elimination rates for other recently
permitted cooling towers is provided in Table 7-22. Based on the available information, we have
determined that BACT for the cooling towers is 0.0005% drift eliminators. Additionally, we are
setting a mass emission limit of 1.6 lb/hr and TDS limit of 5000 ppm.
Table 7-22: Summary of Recent BACT Determinations for Drift Eliminators
l-":icilil v
Locution
Limit
IVrmil
IsMlillKT
Source
J.K. Smith
Generating Station
Kentucky
0.0005%
April 2010
RBLC # KY-0100
Chocolate Bayou
Facility
Texas
0.0020%
June 2009
RBLC # TX-0549
CPV St Charles
Maryland
0.0005%
November
2008
RBLC # MD-0040
John W Turk Jr
Power Plant
Arkansas
0.0005%
November
2008
RBLC # AR-0094
7.5 BA CT for Fugitive Road Dust
Fugitive dust emissions will occur as a result of maintenance vehicle travel on paved and unpaved
roadways in the solar field associated with the PHPP. Fugitive road dust is subject to BACT for
PM, PMio, and PM2.5. A top-down BACT analysis has been performed and is summarized below.
Step 1 - Available Control Technologies
The control technologies for fugitive roadway dusts include: paved roads, gravel roads, chemical
surfactants (also called "dust suppressants"), watering, and traffic speed controls.
Step 2 - Eliminate Technically Infeasible
All of the control technologies identified are technically feasible.
Step 3 - Rank Remaining Control Technologies
The available control options are ranked as follows:
• Paved roads
• Gravel roads
• Chemical surfactants, watering and traffic speed controls can result in various controls
efficiencies depending on how each technology is employed (e.g., rate of application,
specific speed limit)
Step 4 - Economic, Energy and Environmental Impacts
Paved roads - The applicant proposed to pave only the main access road to the plant because
paving other less traveled roads would only have minimal environmental benefits. The applicant
43
Palmdale (SE 09-01)
Fact Sheet/Ambient Air Quality Impact Report
August 2011
-------
noted that paving increases the amount of impervious surfaces, which increases storm water
runoff, and that the infrequent rainstorms in the desert can also erode the dirt out from under the
paved edges.
Gravel roads - The applicant eliminated gravel roads due to the potential for rocks to become
airborne and damage the parabolic mirrors in the solar field. This would result in additional costs
for repairing mirrors and a reduction in solar energy production.
Chemical surfactants, watering, and traffic speed controls - Surface watering and/or application
of surfactants can be supplemented with limiting vehicle speed and restricting traffic in the
unpaved areas. According to the applicant, experience in existing solar fields (e.g., the Solar
Energy Generating Systems (SEGS) facility near Kramer Junction and Harper Lake) shows that
use of a combination of the above methods is very effective in controlling fugitive dust. Use of
soil stabilizers during the first few years of operation of the solar facility, followed by application
of water and driving slowly in the solar field, leads to a very stable surface that yields only minor
amounts of fugitive emissions. In addition, after the solar facility is built, it is in the operator's
best interest to keep dust emissions to a minimum in order to reduce the amount of mirror
washing and loss of efficiency from dirty mirrors.
Step 5 - Select BACT
The applicant proposed BACT for fugitive road dust as:
• Paving the main access road into the plant site
• Developing a dust control plan that includes inspection and maintenance procedures
undertaken to ensure that the unpaved roads remain stabilized
• A durable non-toxic soil stabilizer will be applied through the solar field for dust control.
Additionally, unpaved roads within the solar field used by wash trucks that spray and clean
the mirrors will be treated with soil stabilizers periodically.
• Water will be applied by water trucks on regularly disturbed areas where soil stabilizers
are not as effective due to frequent use. The water used in the mirror washing will also
provide for some incidental dust control.
• Vehicle speeds will be limited to no more than 10 miles per hour on unpaved roadways,
with the exception that vehicles may travel up to 25 miles per hour on stabilized unpaved
roads as long as such speeds do not create visible dust emissions.
Based on the information provided, we have determined that the above measures represent BACT
for fugitive road dust, and the fugitive dust control plan must include, at a minimum, the
requirements listed above. This determination is consistent with other BACT determinations, as
illustrated in Table 7-23, for onsite operations that cause vehicle traffic.
44
Palmdale (SE 09-01)
Fact Sheet/Ambient Air Quality Impact Report
August 2011
-------
Table 7-23: Summary of Recent BACT Determinations for Fugitive Road Dust Emissions
Facility
Location
Control
Permit
Issuance
Source
V & M Star
Ohio
Water, sweeping, chemical
stabilization or suppressants
Draft
January
2011
RBLC #
OH-0344
Nucor Steel
Ohio
Water, resurfacing, chemical
stabilization, and/or speed reduction
Draft
December
2010
RBLC#
OH-0341
Flopam Inc.
Maryland
Paved where practical, precautions
taken to prevent dust from becoming
airborne
June 2010
RBLC#
LA-0240
Nucor Steel
Louisiana
Paved where practical, for unpaved
roads use water or dust suppressant
chemicals to reduce emissions and
15 mph speed limit
May 2010
RBLC#
AR-0094
John W. Turk Jr
Power Plant
Arkansas
Water/dust suppressing chemicals
November
2008
RBLC#
AR-0094
7.6 BACTfor Circuit Breakers
7.6.1 GHG
The circuit breakers are subject to BACT for GHG emissions. The only GHG emitted from
circuit breakers is sulfur hexafluoride (SF6). With the proposed control technologies, C02e
emissions are estimated at 9.56 TPY.
Step 1 - Identify all control technologies
The inherently lower-emitting control options for GHG emissions include:
• Use of dielectric oil or compressed air circuit breakers - these types of circuit breakers
do not contain any GHG pollutants.
• Totally enclosed SF6 circuit breakers with leak detection systems - these types of circuit
breakers have a maximum leak rate of 0.5% per year by weight and have an alarm
warning when 10% of the SF6 has escaped. The use of an alarm identifies potential leak
problems before the bulk of SF6 has escaped.
No add-on control options for GHG emissions were identified. Additionally, alternative gases to
SF6 are also currently not available.41
41 Information is available at http://www.epa. gov/electricpower-sf6/documents/new report final.pdf.
45
Palmdale (SE 09-01)
Fact Sheet/Ambient Air Quality Impact Report
August 2011
-------
Step 2 - Eliminate technically infeasible control technologies
Both control options are assumed to be technically feasible.
Step 3 - Rank remaining control technologies
The expected emissions from the two control options are compared in Table 7-24. Currently, the
only other similar facility with a GHG BACT limit is the Russell City Power Plant to be located in
Hayward, California. The PSD permit for this facility has a voluntary GHG requirement to install
the same leak detection system proposed for the PHPP.
Table 7-24
: Circuit Breaker Control Technologies I
tanked by Control Eff
(iIKi Control Technologies
C():c Kmission
Rate
(TPY)
Dielectric oil or compressed air circuit
breakers
0
Enclosed-pressure SF6 circuit breakers
with 0.5%) (by weight) annual leakage rate
and leak detection systems
9.56
ectiveness
Step 4 - Economic, Energy, and Environmental Impacts
The applicant eliminated the use of dielectric oil or compressed air circuit breakers because they
are an outdated technology and the SF6 circuit breakers are more reliable. Specifically the
applicant provides that according to the National Institute for Standards and Technology, SF6
"offers significant savings in land use, is aesthetically acceptable, has relatively low radio and
audible noise emissions and enables substations to be installed in populated areas close to the
loads."42 Dielectric oil or compressed air circuit breakers therefore have been eliminated based on
the potential adverse environmental and energy impacts. Additionally, we are not aware of any
significant or unusual environmental impacts associated with the chosen technology.
Step 5 - Select BACT
Based on a review of the available control technologies for GHG emissions from circuit breakers,
we have concluded that the applicant's proposed requirements are BACT for this source: the use
of enclosed-pressure SF6 circuit breakers with an annual leakage rate of 0.5% by weight, a 10%
by weight leak detection system, and 9.56 TPY of C02e based on a 12-month rolling total.
8. Air Quality Impacts
Clean Air Act section 165 and EPA's PSD regulations at 40 C.F.R. section 52.21 require
an examination of the impacts of the proposed PHPP on ambient air quality. The applicant
must demonstrate, using air quality models, that the facility's emissions of the PSD-
regulated air pollutants would not cause or contribute to a violation of (1) the applicable
42 Ibid.
46
Palmdale (SE 09-01)
Fact Sheet/Ambient Air Quality Impact Report
August 2011
-------
National Ambient Air Quality Standards (NAAQS), or (2) the applicable PSD increments
(explained below in Section 8.4). This section includes a discussion of the relevant
background data and air quality modeling, and our conclusion that the Project will not
cause or contribute to an exceedance of the applicable NAAQS or applicable PSD
increments and is otherwise consistent with PSD requirements governing air quality.
8.1 Introduction
8.1.1 Overview of PSD Air Impact Requirements
Under the PSD regulations, permit applications for major sources must include an air quality
analysis demonstrating that the facility's emissions of the PSD-regulated air pollutants would not
cause or contribute to a violation of the applicable NAAQS or applicable PSD increments. (A
PSD increment for a pollutant applies only to areas that meet the corresponding NAAQS.) The
applicant provides separate modeling analyses for each criteria pollutant emitted above the
applicable significant emission rate. If a preliminary analysis shows that the ambient concentration
impact of the project by itself is greater than the Significant Impact Level (SIL), then a full or
cumulative impact analysis is required for that pollutant. The cumulative impact analysis includes
nearby pollution sources in the modeling, and adds a monitored background concentration to
account for sources not explicitly included in the model. The cumulative impact analysis must
demonstrate that the Project will not cause or contribute to a NAAQS or increment violation.
Required model inputs characterize the various emitting units, meteorology, and the land surface,
and define a set of receptors (spatial locations at which to estimate concentrations, typically out to
50 km from the facility at issue). Modeling should be performed in accordance with EPA's
Guideline on Air Quality Modeling, in Appendix W to 40 CFR Part 51 (GAQM or Appendix W).
AERMOD with its default settings is the standard model choice, with CALPUFF available for
complex wind situations.
A PSD permit application typically includes a Good Engineering Practice (GEP) stack height
analysis, to ensure a) that downwash is properly considered in the modeling for stacks less than
GEP height, and b) that stack heights used as inputs to the modeling are no greater than GEP
height, so as to disallow artificial dispersion from the use of overly tall stacks. The application
may also include initial "load screening," in which a variety of source operating loads and ambient
temperatures are modeled, to determine the worst case scenario for use in the rest of the
modeling.
The PSD regulations also require an analysis of the impact on nearby Class I areas, generally
those within 100 km, though the relevant Federal Land Manager (FLM) may specify additional or
fewer areas. The analysis includes the NAAQS, PSD increments, and Air Quality Related Values
(AQRVs). AQRVs are defined by the FLM, and typically limit visibility degradation and the
deposition of sulfur and nitrogen. CALPUFF is the standard model choice for Class I analyses,
since it can handle visibility chemistry as well as the typically large distances (over 50 km) to Class
I areas.
47
Palmdale (SE 09-01)
Fact Sheet/Ambient Air Quality Impact Report
August 2011
-------
Finally, the PSD regulations require an additional impact analysis, showing the Project's effect on
visibility, soils, vegetation, and growth. This visibility analysis is independent of the Class I
visibility AQRV analysis. The additional impact analysis for the PHPP is discussed in Section 9
below.
8.1.2 Identification of PHPP Modeling Documentation
The PSD modeling analysis for the PHPP went through several stages, reflecting the regulatory
requirements and guidance clarifications that came into effect over time, as well as discussions
between the applicant and EPA about the appropriate methodologies for impact assessment. In
general, the latest analyses submitted by the applicant are discussed in this AAQIR, with some
references to earlier work.
The PHPP modeling analysis comprises the eight documents listed in Table 8-1 below. The Class
I and Class II Modeling Protocols (July 2008) describe the methods to be used for the air quality
impact analyses, including choice of model and the preparation of model inputs such as
meteorological data. The PSD Application (March 2009) contains the results of the modeling.
After the application submittal, EPA policy changed so that the PMioNAAQS could no longer be
used as a surrogate for the PM2.5 NAAQS, and EPA promulgated the 1-hour N02 NAAQS;
neither PM2.5 nor 1-hour N02 these was addressed in the original modeling. The applicant
submitted Supplemental Information (June 2010) to update its modeling analysis by providing a
PM2.5 analysis and a 1-hour N02 analysis considering the Project and background concentrations;
it also upgraded the additional impact analysis discussed in Section 9 below. The applicant's NQ2
Memo #1 (October 2010) provides a cumulative 1-hour N02 analysis, which includes nearby
sources in addition to the Project itself. Finally, the Updated Analyses Memo (March 2011)
revises the PM2.5 and 1-hour N02 analyses to account for corrected hourly emissions estimates for
the nearby U.S. Air Force Plant 42, and to use a more conservative estimate of the N02
background concentration. The applicant also submitted additional documentation in NQ2 Memo
#2 (December 2010), and the NQ2 Background Memo (July 2011), providing additional
justification for the approaches taken for the applicant's 1-hour N02 analysis.
Table 8-1: Modeling Documentation for Palmdale Hybrid Power Project PSD Application
Short name
Citation
Class I Modeling
Protocol
"Class I Area Dispersion Modeling Protocol for the Proposed Palmdale Hybrid
Power Project", ENSR Corporation (document 10855-002-040C1MP), July 2008
(file "PHPP Class I Modeling Protocol.pdf'
Class II Modeling
Protocol
"Class II Area Dispersion Modeling Protocol for the Proposed Palmdale Hybrid
Power Project", ENSR Corporation (document 10855-002-040C2MP), July 2008
(file "PHPP Class II Modeling Protocol.pdf')
Original PSD
Application
"Application for Prevention of Significant Deterioration Permit for Palmdale Hybrid
Power Project", AECOM Environment (document 10855-002-040 PSD), March
2009
(file "Palmdale PSD Application.pdf')
48
Palmdale (SE 09-01)
Fact Sheet/Ambient Air Quality Impact Report
August 2011
-------
Supplemental
Information
"Palmdale Hybrid Power Project PSD Application, Supplemental Information",
AECOM, June 2010
(file "Supplemental PSD Submittal 072010.pdf')
N02 Memo # 1
"Response to EPA Comments on AECOM 1-hour N02 NAAQS Analysis for
PHPP", Memorandum from Richard Hamel, AECOM, to Scott Bohning, EPA,
October 7, 2010
(file "Response to EPA Comments on N02 Modeling.pdf')
N02 Memo #2
"Response to EPA Additional Comments on AECOM 1-hour N02 NAAQS
Analysis for Palmdale Hybrid Power Project", Memorandum from Richard Hamel,
AECOM, to Scott Bohning, EPA, December 14, 2010
(file "Response to 2nd set of EPA Comments on N02 Modeling.pdf')
Updated Analyses
Memo
"Final Update to 1-hour N02 and 24-hour PM2.5 NAAQS Analyses for Palmdale
Hybrid Power Project", Memorandum from Richard Hamel, AECOM, to Scott
Bohning, EPA, March 30, 2011
(file "Updated N02 and PM2.5 Modeling Analyses for PHPP 03301 l.pdf')
N02 Background
Memo
"Justification of the use of the 3-year average 98th percentile ambient background
concentration for PHPP 1-hour N02 NAAQS Modeling", Memorandum from
Richard Hamel, AECOM, to Scott Bohning, EPA, July 21, 2011
(file " 1-hour N02 Ambient Background Justification for PHPP NAAQS Modeling
07211 l.pdf)
8.2. Background Ambient Air Quality
The PSD regulations require the air quality analysis to contain air quality monitoring data as
needed to assess ambient air quality in the area for the PSD-regulated pollutants for which there
are NAAQS that may be affected by the source. In addition, for demonstrating compliance with
the NAAQS, a background concentration is added to represent those sources not explicitly
included in the modeling, so that the total accounts for all contributions to current air quality.
For background concentrations, PHPP chose the Lancaster Division Street monitor, which is the
nearest available, except for S02, for which the Burbank West Palm Avenue is nearest. The most
recent three years of data available at the time of the application are 2005-2007. (PSD
Application p.6-2 pdf.47; see also Class II Modeling Protocol p.2-19 pdf.24) Based on their
siting at more urbanized locations than the Project site, these monitors provide conservative
estimates of background concentrations. The S02 monitor at Burbank West Palm Avenue is 34
miles away, but is in the eastern portion of urbanized Los Angeles with its many pollution
sources, and therefore it provides a conservative estimate of the S02 background. The Lancaster
Division Street monitor is just 2.5 miles from the PHPP power block; it is within the city of
Lancaster, which has a population of some 150,000, and is near several roads; it is thus
conservative for most pollutants. This site is discussed further below in the section on N02-
specific issues.
Table 8-2 below describes the maximum background concentrations of the PSD-regulated
pollutants for which there are NAAQS that may be affected by the Project's emissions, and the
corresponding NAAQS.
49
Palmdale (SE 09-01)
Fact Sheet/Ambient Air Quality Impact Report
August 2011
-------
Table 8-2 Maximum background concentrations and NAAQS
NAAQS
pollutant &
averaging time
Background
Concentration, jig/m3
NAAQS, jiig/m3
CO, 1-hr
3,680
40,000 (35 ppm)
CO, 8-hr
1,840
10,000 (9 ppm)
N02, 1-hr
77.1
188 (100 ppb)
N02, annual
28.2
100 (53 ppb)
PM10, 24-hr
86
150
PM2.5, 24-hr
16.3
35
PM2.5, annual
7.6
15
Note: The PM2.5 24-hr value is 98 percentile rather than maximum
8.3 Modeling Methodology for Class II areas
The applicant modeled the impact of PHPP on the NAAQS and PSD Class II increments using
AERMOD in accordance with EPA's GAQM (Appendix W of 40 CFR Part 51). The modeling
analyses included the maximum air quality impacts during startups and shut-downs, as well as a
variety of conditions to determine worst-case short-term air impacts.
8.3.1 Model selection
As discussed in the modeling protocol (Class II Modeling Protocol sec. 2, p.2-1 pdf.6; also PSD
Application p.6-1 pdf.46), the model that the applicant selected for analyzing air quality impacts in
Class II areas is AERMOD, along with AERMAP for terrain processing and AERMET for
meteorological data processing. This accords with the default recommendations in EPA's
GAQM, section 4.2.2 on Refined Analytical Techniques.
Palmdale (SE 09-01)
Fact Sheet/Ambient Air Quality Impact Report
August 2011
50
-------
8.3.2 Meteorology model inputs
AERMOD requires representative meteorological data in order to accurately simulate air quality
impacts. For surface air data, PHPP selected 2002-2004 data from the Palmdale Regional
Airport. Other nearby meteorological sites were examined, but the Palmdale Airport had better
data completeness, is the closest, and has the same surface characteristics as the Project site. It is
at or barely below 90% completeness for every quarter; it is within 2 miles, just on the other side
of the airport's airstrip; and it is on flat, desert scrub land, with no intervening high ground
between the Project and the meteorological tower (Class II Modeling Protocol p.2-4 pdf.9 and
Figure 2-2, p.2-5 pdf.10).
The applicant made additional comparisons of land surface characteristics of the Project and
meteorological sites, in terms of surface roughness in each radial direction, concluding that
because of the sites' proximity and essentially identical characteristics, the Palmdale Airport data
should be considered "site specific" (or "on-site") data (N02 Memo #2 p.9ff pdf.9). Normally
GAQM would require 5 years of airport data for modeling, but if on-site data is used, then a
single year or those years available, may be used (GAMQ 8.3.3.2). In this case, additional data
were available for 2005-2006, but the corresponding upper air data had a substantial amount of
missing data (N02 Memo #2 p. 10 pdf. 10). In any case, the wind roses for the various years are
virtually indistinguishable, evidence that the 2002-2004 data are adequately representative of the
meteorological conditions at the site. EPA believes that the chosen 2002-2004 Palmdale Regional
Airport data is amply representative for the PHPP analysis.
For upper air data, the applicant selected Mercury Desert Rock Airport in Mercury, Nevada, as
being the most representative site available that had data complete enough to use (Class II
Modeling Protocol p. 2-4 pdf.9). PHPP later elaborated on the representativeness of the Mercury
Desert Rock Airport Data, noting that Vandenberg AFB in Lompoc, CA and the Marine Corps
Air Station in Miramar, CA, near San Diego are near the ocean and have a very different climate
than the high-altitude, desert Palmdale location (N02 Memo #1 p.2ff pdf.2). EPA agrees that it is
appropriate to use the Mercury Desert Rock Airport upper air data for the PHPP analysis.
8.3.3 Land characteristics model inputs
Land characteristics are used in the AERMOD modeling system in three ways: 1) via elevation
within AERMOD to assess plume interaction with the ground; 2) via a choice of rural versus
urban algorithm within AERMOD; and 3) via specific values of AERMET parameters that affect
turbulence and dispersion, namely surface roughness, Bowen ratio, and albedo.
The applicant used terrain elevations from United States Geological Survey (USGS) Digital
Elevation Model (DEM) data for receptor heights for AERMOD, which uses them to assess
plume distance from the ground for each receptor. The elevations were also used within the
AERMAP preprocessor to determine hill height scales for each receptor, used by AERMOD to
51
Palmdale (SE 09-01)
Fact Sheet/Ambient Air Quality Impact Report
August 2011
-------
determine whether the plume goes over or around the hill.
For rural versus urban algorithm within AERMOD, the applicant classified land use within 3 km
of the project using the 12-category Auer procedure, one of the methods recommended by EPA
(GAQM 7.2.3(c)). Since desert scrub land is more than 50% of the area, it is classified as "rural"
for choosing dispersion algorithms within AERMOD (Class II Modeling Protocol p.2-2 pdf.7, and
Figure 2-1, p.2-3 pdf.8).
The applicant followed EPA's "AERMOD Implementation Guide" (2008 version) in using EPA's
AERSURFACE processor with the National Land Cover Data 1992 archive to determine surface
characteristics for AERMET (Class II Modeling Protocol p.2-9 to 2-14 pdf. 14 to 19). A 2005
satellite image shows no significant change in land use since the 1992 data was compiled, so it
remains appropriate. Land use cover categories were translated by AERSURFACE into monthly
parameter values used in AERMET's stage 3 input files. The AERSURFACE determination of
surface roughness length used land cover in 2 radial sectors, desert scrub and the airport's airstrip,
which appears reasonable. The Bowen ratio (ratio of sensible to latent heating, i.e., direct
temperature change versus air heating via evaporation), and albedo (reflection coefficient) affect
heat-driven turbulence and dispersion under daytime convective conditions. Seasonal Bowen
ratio for the surrounding 10x10 km area was estimated by AERSURFACE using three surface
moisture categories and the amount of precipitation relative to the 30-year climatological record.
Seasonal albedo was also supplied by AERSURFACE for the 10x10 km area based on land cover.
All of these are the standard EPA-recommended procedures for AERMOD inputs.
8.3.4 Model receptors
Model receptors are chosen geographic locations at which the model estimates concentrations.
The receptors should have good area coverage and be closely spaced enough so that the
maximum model concentrations are be found. At larger distances, spacing between receptors may
be greater than it is close to the source since concentrations vary less with increasing distance.
The spatial extent of the receptors is limited by the applicable range of the model (roughly 50 km
for AERMOD), and possibly by knowledge of the distance at which impacts fall to negligible
levels. Receptors need be placed only in ambient air, that is, locations to which the public has
access, and not inside the project fence line. In addition, to avoid overly conservative estimates
when multiple sources are being modeled, separate modeling runs may be needed for different
subsets of receptors, so that a given source's emissions are not counted toward concentrations
within its own fence line.
The applicant used receptors every 50 m along the project fence line, together with a Cartesian
grid (rectangular array) of receptors, starting with 100 m spacing out to 3 km distant, and with
progressively larger spacing, with 1000 m spacing between 10 and 20 km distant (PSD
Application p.6-3 pdf.48). The applicant supplied a rationale for limiting the grid extent to 20
km, as opposed to 50 km. It found that short-term impacts were caused mainly by the ancillary
52
Palmdale (SE 09-01)
Fact Sheet/Ambient Air Quality Impact Report
August 2011
-------
equipment, such as the emergency generator, rather than the main combustion turbines, and that
maximum impacts were on the fence line or within 100 m, and likely driven by downwash effects.
The applicant conducted additional modeling to compare distance impacts to those within the 20
km grid, and found that the maximum impacts within 20 km are 2 to 50 times higher than those
outside, depending on averaging time (Supplemental Information p.6-1 pdf.41). EPA agrees that
the receptor spacing and 20 km spatial extent are adequate for analysis of PHPP impacts.
8.3.5 Load screening and stack parameter model inputs
The applicant performed initial "load screening" modeling, in which a variety of source operating
loads and ambient temperatures were modeled, to determine the worst case stack parameter
scenario for use in the rest of the modeling. It modeled 100% load, 100% with duct burners
operating, 75% load, and 50% load. For annual averages, it used 100% load with a
conservatively low temperature of 64°F (lower than actual annual average). (PSD Application
Table 6-3, p. 6-4 pdf.49) The choice of "worst case" is different for each pollutant, since different
pollutants' emissions respond differently to temperature and flow rate. Worst case for CO and
N02 was 100%) with duct burners operating; for PMi0 and PM2.5 it was 50% load (PSD
Application p.6-6 pdf.51). The corresponding stack parameters were used in the remainder of
the modeling to provide conservative estimates of PHPP impacts.
Palmdale (SE 09-01)
Fact Sheet/Ambient Air Quality Impact Report
August 2011
53
-------
Table 8-3: Load screening and stack parameters
Parameter
Value
North Stack
South Stack
UTM Coordinate East (m)1
398680.2
398679.8
UTM Coordinate North (m)1
3833520.8
3833479.7
Stack Base Elevation (ft)
2,517
2,517
Stack Height (ft)
145
145
Stack Diameter (inches)
218
216
Load
100'%
w/DB
100%
75%
50%
Annual
Avg.2
Exit Temperature {CF)
172.9
176.5
166.7
106.9
174.1
Exit Velocity {ft'sec)
62.0 t
81.08
46.28
39.7
64.9
Pollutant
Emissions Per
Conbustion
Turbine (Ib/hr)
NO.
16.80
13.47
10.97
8.73
13.0
CO
15.16
8.20
8.68
5.3'
28.8
PM '0/PM2.5
18
12
12
12
".3.4
Coord nates for UTM Zone 11 referenced to Datum NAD27,
: Anrr-a: average emissions include norrral opera*Jors as we'l as startup/shutdown. E'.t ".errpe^ture and veoc ty
are the 100 oercert ,oad case a: 64: F.
Notes:
rr = nete-5
^t. = feel
Source: PSD Application Table 6-3, p.6-4 pdf.49
8.3.6 Good Engineering Practice (GEP) Analysis
The applicant performed a Good Engineering Practice (GEP) stack height analysis, to ensure a)
that downwash is properly considered in the modeling for stacks less than GEP height, and b) that
stack heights used as inputs to the modeling are no greater than GEP height, so as to disallow
artificial dispersion from the use of overly tall stacks. As is typical, the GEP analysis was
performed with EPA's BPIP ( Building Profile Input Program) software, which uses building
dimensions and stack heights. The analysis found that GEP stack height for the main combustion
turbines was 83.8 m, greater than the planned actual height of 44.2 m. GEP stack height for the
other equipment was similarly greater than the planned heights. So, for all emitting units, the
AERMOD modeling used the planned actual stack heights, and included wind direction-specific
Equivalent Building Dimensions to properly account for downwash. (PSD Application p.6-5
pdf.50)
54
Palmdale (SE 09-01)
Fact Sheet/Ambient Air Quality Impact Report
August 2011
-------
8.4 National Ambient Air Quality Standards and PSD Class II Increment
Consumption Analysis
8.4.1 Pollutants with significant emissions
An air quality impact analysis is required for each PSD-regulated pollutant (for which there is a
NAAQS) that is emitted in a significant amount, i.e., an amount greater than the Significant
Emission Rate for the pollutant. Applicable PHPP emissions and the Significant Emission Rates
are shown in Table 8-4 (derived from PSD Application Table 1-1, p.8 pdf.8). PHPP emissions of
S02 are not significant. However, PHPP emits significant amounts of CO, NOx, PMio, and PM2.5,
so air impact analyses are required for CO, N02, PMio, and PM2 5.
Table 8-4: PSD Applicability to PHPP: Pollutants Emitted in Significant Amounts
PIIPP Emissions.
Significant Emission
Criteria Pollutant
tons/year
Kate. tons/year
PSI) applicable?
CO
254.6
100
Yes
NOx
114.9
40
Yes
PMio
131.8
15
Yes
pm2,
125.3
10
Yes
so2
8.9
40
No
Source: PSD Application Table 1-1, p.8 pdf.8
8.4.2 Preliminary analysis: Project-only impacts
EPA has established Significant Impact Levels (SILs) to characterize air quality impacts. A SIL is
the ambient concentration resulting from the facility's emissions, for a given pollutant and
averaging period, below which the source is assumed to have an insignificant impact. For
maximum modeled concentrations below the SIL, no further air quality analysis is required for the
pollutant. For maximum concentrations that exceed the SIL, a cumulative modeling analysis,
which incorporates the combined impact of nearby sources of air pollution, is required to
determine compliance with the NAAQS and PSD increments.
The results of the preliminary or Project-only analysis are shown in Table 8-5. PHPP impacts are
significant for 1-hour N02, 24-hour PMio, 24-hour PM2.5, and annual PM2.5, so cumulative impact
analyses are required for these pollutants.
Table 8-5: PHPP Significant Impacts, Normal Operations
NAAQS pollutant &
a\craging lime
Project-only
Modeled Impact
Significant Impact
Ley el (SI 1.). fig/m'
Project impact
significant?
CO, 1-hr
369.6
2000
No
CO, 8-hr
20.4
500
No
N02, 1-hr
106.9
7.5 (4 ppb)
Yes
55
Palmdale (SE 09-01)
Fact Sheet/Ambient Air Quality Impact Report
August 2011
-------
N02, annual
0.98
1
No
PMio, 24-hr
12.7
5
Yes
PM2.5, 24-hr
12.57
1.2
Yes
PM2.5, annual
1.2
0.3
Yes
Sources:
Impacts (except for 1-hr N02 and PM2.5): PSD Application p.6-7 pdf.52
N02 1-hr: Supplemental Information p3-2. pdf.22
PMi0: PSD Application Table 6-7, p.6-8 pdf.53
PM2.5: Updated Analyses Memo Table 9, p. 15 pdf.15
8.4.3 Cumulative impact analysis
A cumulative impact analysis includes nearby sources in addition to the Project itself. For
demonstrating compliance with the PSD increment, only increment-consuming sources need be
included, since the increment concerns only changes occurring since the applicable baseline date.
However, a conservative and sometimes easier approach is simply to model all nearby sources;
this was the approach taken by PHPP. For demonstrating compliance with the NAAQS, a
background concentration is added to represent those sources not explicitly included in the
modeling, so that the total accounts for all contribution to current air quality.
8.4.3.1 Nearby source emission inventory
For both the PSD increment and NAAQS analyses, there may be a large number of sources that
could potentially be included, so judgement must be applied to exclude small and/or distant
sources that have only a negligible contribution to total concentrations. Only sources with a
significant concentration gradient in the vicinity of the source need be included; the number of
such sources is expected to be small except in unusual situations. (GAQM 8.2.3)
The applicant identified two sources nearby for inclusion in the emission inventory for the
cumulative analysis, based on discussions with the Antelope Valley Air Quality Management
District (District) (PSD Application p.6-7 pdf.52). These are Lockheed Martin Aeronautics and
Northrop Grumman, both within or adjacent to U.S. Air Force Plant 42 near the Palmdale airport.
These sources had a large number of individual emitting sources (284), most of which had very
low emissions. For practicality of modeling some of these were combined in a conservative way:
emitters with less than 5% of total had their emissions added to the largest emitters.
In support of limiting the inventory to these sources, the applicant quoted a statement from Mr.
Chris Anderson, Air Quality Engineer, and Mr. Alan De Salvio, Supervisor of Air Quality
Engineering, of the District: "Minor facilities located within the 6 mile radius are expected to be
included in the background monitored at the AVAQMD [District] air monitoring station which is
located in close proximity (approximately within 2 miles) of the PHPP site." (N02 Memo #2 p. 11
pdf.ll)
The applicant also documented discussions with the District, Mojave Desert Air Quality
56
Palmdale (SE 09-01)
Fact Sheet/Ambient Air Quality Impact Report
August 2011
-------
Management District (AQMD), Kern County Air Pollution Control District, and South Coast
AQMD showing that there are few substantial PM2.5 sources nearby; however, Granite Rock
Construction and Robertson's Ready Mix were included in the modeling, both about 15 km (9
miles) from PHPP (Supplemental Information p.2-1 to 2-2 pdf.9 to 10, and Figure 2-1 p.2-3
pdf.ll).
Also, recent EPA N02 guidance clarification states that the nearby source inventory "should focus
on the area within about 10 kilometers of the project location", which suggests that the PHPP
inventory is adequate for N02 analyses (p. 16 of "Additional Clarification Regarding Application
of Appendix W Modeling Guidance for the 1-hour N02 National Ambient Air Quality Standard",
Memorandum from Tyler Fox, EPA Air Quality Modeling Group to EPA Regional Air Division
Directors, March 1, 2011).
Nevertheless, the applicant also performed a "Q/D" analysis, which provides another factor for
consideration in determining whether sources with small emissions (Q) and/or at large distances
(D) would be reasonable to exclude from the analysis. The applicant proposed that sources with a
km distance greater than the NOx emissions in tons per year divided by 20 would be eligible for
exclusion. (Updated Analyses Memo p.6 pdf.6, citing "Screening Method for PSD" developed by
the North Carolina Air Quality Section of the North Carolina Department of Natural Resources,
in file "NC 20D Letter to EPA.pdf'). The only sources to pass this initial screen were those
within US Air Force Plant 42, already included in the cumulative modeling, and Bolthouse Farm
emissions. In addition to being mostly downwind (east) of the project, the emissions of Bolthouse
Farm are widely distributed throughout the area, and therefore are dispersed enough that they
would have a negligible contribution to maximum concentrations (Updated Analyses Memo p.8
pdf.8). The Q/D analysis provides additional evidence that the source inventory is adequate for
the cumulative impact analysis.
EPA believes that the combination of a conservative background monitored concentration
expected to include the effect of most nearby sources, EPA guidance clarification focusing on
sources within 10 km, and the Q/D analysis are sufficient justification for the inventory used in the
cumulative analysis.
8.4.3.2 PM2.5-specific issues
The applicant originally relied on the PMi0 NAAQS as a surrogate for the PM2.5 NAAQS, which
was allowed under previous EPA policy. However, EPA repealed this policy (proposed February
11, 2010; final May 18, 2011), so that PM2.5 itself must be modeled. EPA also issued guidance
clarification on how to combine modeled results with monitored background concentrations
("Modeling Procedures for Demonstrating Compliance with PM2.5 NAAQS", memorandum from
Stephen D. Page, Director, EPA OAQPS, March 23, 2010).
Accordingly, the applicant replaced the original analysis with a new cumulative PM2 5 analysis.
The applicant still conservatively used PMi0 emissions as input to the modeling, so actual PM2 5
impacts may be lower than those indicated in the model results. Maximum model results were
57
Palmdale (SE 09-01)
Fact Sheet/Ambient Air Quality Impact Report
August 2011
-------
correctly added to the ninety-eighth percentile of the monitored background concentration, as
called for in the EPA guidance clarification. (Updated Analyses Memo p. 12ff pdf. 12)
The PHPP application has little discussion of secondarily formed PM2.5 (as distinguished from
directly emitted primary PM2.5). However, the applicant does cite an earlier AECOM analysis
showing that that near the source, primary PM2.5 emissions dominate the modeled impacts
(Supplemental Information, p.2-10 pdf. 18). EPA notes that, due to the time needed for chemical
formation, secondary PM2.5 impacts are likely to occur much farther downwind than the
significant primary impacts, which occur within 400 m of the project (Updated Analyses Memo
p. 12 pdf. 12), and so are likely to be small and not overlapping with the impacts estimated in the
application.
8.4.3.3 NO2-specific issues
The applicant used the Ozone Limiting Method (OLM) option in AERMOD, in which ambient
ozone concentrations limit the amount of emitted NO that is converted to N02 (after an initial
10% conversion). In addition to requiring monitored ozone, the method requires specification of
an in-stack N02/N0X ratio. EPA believes the OLM method is justified in this area because while
it has substantial ozone, most of that is due to transport from outside the area, rather than to
photochemistry operating on VOC and NOx emiossions from sources within the area. Therefore,
the alternative mechanisms for conversion of NO to N02 by the hydroxyl and peroxy radicals are
likely to be less important than the ozone conversion mechanism, and so the conversion is ozone-
limited.
A. In-stack NO2/NOx ratio
The applicant notes that since the Project would be located in an ozone nonattainment area, ozone
concentrations are generally high, so that the initial in-stack N02/N0X ratio is of less importance
than would otherwise be the case, since plentiful ozone is available to convert NO to N02 (N02
Memo #2 p.3 pdf.3).
GE Power and Water, the vendor of the GE7FA turbines planned for PHPP, provided an in-stack
N02/NOx ratio of 0.10 to 0.15 based on its review of available N02 emission data; the Selective
Catalytic Reduction (SCR) planned for PHPP would make this ratio even lower (N02 Memo #1
p.8 pdf.8; N02 Memo #2 p.3 pdf.3). Since little data is available for the ratio during startup and
shutdown conditions, the applicant relied on a 0.4 ratio as recommended by the San Diego
County Air Pollution Control District for a project with similar turbines, despite some evidence
that the actual ratio could be lower for both startup and shutdown events. The short duration of
these events implies that that actual ratio would be closer to the 0.10 used for normal operations
(N02 Memo #1 p.9 pdf.9).
58
Palmdale (SE 09-01)
Fact Sheet/Ambient Air Quality Impact Report
August 2011
-------
B. NO2 monitor representativeness/conservativeness
As mentioned above, the applicant chose the Lancaster Division Street monitor for background
N02 concentrations. This monitor is just 2.5 miles from the PHPP power block, and is near the
Sierra Highway (110 m), the Antelope Valley Freeway (SR-14) (4 km), commute traffic on
Division Street (50 m), and the Southern Pacific Railway (80 m). EPA agrees with PHPP that
this location is quite conservative for providing N02 background concentrations.
C. O3 background monitor representativeness
The applicant notes that since 03 is a regionally formed pollutant, the nearness of the monitoring
site to the project is the most important criterion for representativeness (N02 Memo #1 p. 10
pdf. 10). The Lancaster Division Street monitor is just 2.5 miles away from the PHPP power
block, and EPA agrees that it is adequately representative.
D. Missing O3 data procedure
The applicant filled in missing ozone data using a procedure to ensure that NO to N02 conversion
is not underestimated. When 1 or 2 hours are missing, the higher of the two endpoints are used
for the missing hours. When 3 or more hours are missing, the higher of the two end points and of
the corresponding hours from the two neighboring days are used for the missing hours. (N02
Memo #2 p. 8 pdf. 8) Under this procedure, professional judgement is applied to ensure that the
data from the neighboring days are not anomalously low.
The applicant provided an example of the application of this procedure (Updated Analyses Memo
p.3 to 4 pdf.3 to 4), as well as details of the full calculations (file "PHPP Ozone Filling
Analysis.xlsx" from July 2011).
EPA believes that the applicant followed a reasonable and conservative procedure for filling in
missing ozone values.
E. Combining modeled and monitored values
Originally, the applicant combined each modeled concentration with the background
concentration from the corresponding hour ("hour-by-hour" approach). The applicant later
switched to a variant of EPA's March 2011 memo's43 "first tier" approach: it used the 98th
percentile of all monitored values, though only for model receptors outside the USAF Plant 42
boundary; the hour-by-hour approach still applied to other receptors. (The EPA March 2011
memo's "first-tier" approach uses the 98th percentile from among only the daily maxima, whereas
43 "Additional Clarification Regarding Application of Appendix W Modeling Guidance for the 1 -hour N02 National
Ambient Air Quality Standard", Memorandum from Tyler Fox, EPA Air Quality Modeling Group to EPA Regional Air
Division Directors, March 1, 2011. http://www.epa.gov/ttn/scram/Additional Clarifications AppendixW Hourlv-NQ2-
NAAQS_FINAL_03-01-201 l.pdf
59
Palmdale (SE 09-01)
Fact Sheet/Ambient Air Quality Impact Report
August 2011
-------
the applicant's variant uses the 98th percentile from among all hourly values.) While the
applicant's approach is less conservative than EPA's first-tier approach, we believe that it remains
conservative given the very conservative background monitor that is being used (N02
Background Memo). The maximum values coincide with morning and evening commute traffic,
due to the several roads near the monitor.
A key concern expressed in EPA's March 2011 memo about the hour-by-hour approach is that it
implicitly assumes concentrations are spatially uniform, i.e., that the background monitor is
representative of all locations44. Since this is not generally true, some degree of temporal
conservativeness is warranted, as in the memo-recommended 98th-percentile of the available
background concentrations by season and hour-of-day. However, for PHPP, the background
monitor appears to be very conservative, so that the implicit spatial uniformity assumption of the
hour-by-hour approach is actually a conservative assumption in this case. If the memo-
recommended procedure were to be used in this case, then a single unusually high morning
commute hourly concentration would be assumed to apply to every day of the season; a single
N02 exceedance would then become 90 exceedances, thus possibly causing an erroneous
prediction of a 1-hour N02 violation, an overly conservative approach.
In addition, the applicant's modeling included some intermittent sources (PHPP's emergency
generators) that may not need to be included, per EPA's March 2011 memo45 on hourly N02
modeling, further adding to the conservativeness of the analysis.
EPA believes that the applicant's overall approach to the 1-hour N02 analysis for the PHPP,
including the emission inventory, background concentrations of N02 and 03, and method for
combining model results with monitored values, is adequately conservative.
8.4.3.4 Results of the cumulative impacts analysis
The results of the PSD cumulative impacts analysis for PHPP's normal operations is shown in
Table 8-6. The analysis demonstrates that emissions from PHPP during normal operations will
not cause or contribute to exceedances of the NAAQS for 1-hour N02, 24-hour PMio, 24-hour
PM2 5, or annual PM2 5 or applicable PSD increments. As discussed above, PHPP's maximum
modeled concentrations are below the SILs for annual N02, 1-hour CO, and 8-hour CO;
therefore, a cumulative impacts analysis was not required to demonstrate compliance for these
pollutants/averaging times.
AAIbid.,y.2\.
A5Ibid.,v.\0.
60
Palmdale (SE 09-01)
Fact Sheet/Ambient Air Quality Impact Report
August 2011
-------
Table 8-6: PHPP Compliance with PSI
) Increments ant
NAAQS, Normal Operations
NAAQS
pollutant &
averaging
lime
All Sources
Modeled
1 in pad
PSI)
Increment
Background
Concentration
Cumulative
impact w/
background
NAAQS
N02, 1-hr;
USAF
106.9
NA
(hourly)
175.3
188 (100 ppb)
N02, 1-hr;
other
108.2
NA
77.1
185.3
188 (100 ppb)
PMio, 24-hr
12.9
30
86
98.9
150
PM2.5, 24-hr
12.58
NA
16.3
28.9
35
PM2 5, annual
1.3
NA
7.6
8.9
15
Notes:
- "USAF" values are for receptors within USAF Plant 42; "other" is for receptors elsewhere; USAF Plant 42 receptors
are not ambient air with respect to its own emissions.
- Background concentrations for USAF receptors were added hour-by-hour to modeled concentrations before computing
98th percentile total impact, rather than a single background value being added to the modeled impact as for
the other cases.
Sources:
N02 USAF: Supplemental Information p3-2. pdf.22
N02 other: Updated Analyses Memo Table 7, p. 11 pdf. 11, "Normal Operations - No PHPP Fire Water Pump"
PMi0: PSD Application Table 6-7, p.6-8 pdf.53
PM2.5: Updated Analyses Memo Table 9, p. 15 pdf. 15
8.4.3.5 Startup and shutdown analyses
Combustion turbine CO and NOx emissions during startup and shutdown (SU/SD) are estimated
to be substantially higher than during normal operations, and thus the applicant also modeled for
shutdown, the condition having the highest emissions. Modeled stack parameters such as exit
temperature and exhaust velocity were consistent with a 20% operating load; the ambient
temperature used represented worst-case meteorological conditions, emission into a cool morning
stable layer. Since shutdown duration may not exceed half an hour, worst case hourly emissions
consist of a half-hour of normal operations followed by a shutdown event. For CO, this is 1/2 of
15.16 lb/hr, plus 337 lb, for a combined rate of 344.6 lb/hr per turbine (PSD Application p.6-9
pdf.54). For NOx, this is 1/2 of 16.6 lb/hr, plus 57 lb, for a combined rate of 65.3 lb/hr per
turbine (Updated Analyses Memo Table 7, p. 11 pdf. 11). Emergency generator testing was not
included in the NOx modeling, since it would not be undergoing testing during source shutdown.
This 1-hour N02 analysis continues to use the conservative assumptions discussed above for the
analysis of normal operations. The model results are shown in Table 8-7 for the preliminary or
Project-only analysis, and in Table 8-8 for the cumulative impacts analysis. The results
demonstrate that emissions from PHPP will comply with the 1-hour N02 NAAQS and both the 1-
hour and 8-hour CO NAAQS under shutdown conditions (and therefore for startup conditions,
for which emissions are lower). We note that the applicant was not required to, and did not,
perform a cumulative impact analysis for CO, as its emissions are below the SILs; however, for
informational purposes, Project impacts were added to background concentrations of CO for a
rough comparison to the NAAQS.
61
Palmdale (SE 09-01)
Fact Sheet/Ambient Air Quality Impact Report
August 2011
-------
Ta
)le 8-7: PHPP Significant
Impacts, Startup/Shutdown
NAAQS pollutant
& averaging lime
Projeel-onlv \ 1 odeled
1 in pad
Significant Impact
Level (SI 1.). fig/m'
Project significant
impact?
CO, 1-hr
674.6
2000
No
CO, 8-hr
489.1
500
No
N02, 1-hr
136.4
7.5 (4 ppb)
Yes
Sources:
CO: PSD Application Table 6-9, p.6-9 pdf.54
N02 1-hr: Supplemental Information p3-3. pdf.23
Table 8-8: PHPP Compliance with NAAQS, Startup/Shutdown
NAAQS
Project-
All
pollutant &
onlv
Sources
Cumulative
averaging
Modeled
Modeled
Background
impact w/
time
impact
Impact
Concentration
background
NAAQS
CO, 1-hr
674.6
NA
3,680
4,354.6
40,000 (35 ppm)
CO, 8-hr
489.1
NA
1,840
2,329.1
10,000 (9 ppm)
N02, 1-hr;
USAF
(not
modeled)
136.4
(hourly)
180.3
188 (100 ppb)
N02, 1-hr;
other
(not
modeled)
109.7
77.1
186.9
188 (100 ppb)
Notes:
- There are no PSD increments defined for CO or for 1 -hour N02.
- PHPP emissions are not significant for CO, so no cumulative analysis is required; "cumulative impact" here is PHPP-
only plus background.
- "USAF" values are for receptors within USAF Plant 42; "other" is for receptors elsewhere; USAF Plant 42 receptors
are not ambient air with respect to its own emissions. Project-only impacts were not modeled for 1 -hour N02
startup/shutdown, rather only the full cumulative impact was modeled.
- Background concentrations for USAF receptors were added hour-by-hour to modeled concentrations before computing
98th percentile total impact, rather than a single background value being added to the modeled impact as for the
other cases."Project-only" and "all sources" are the same except for 1-hr N02 "other" receptors.
Sources:
CO: PSD Application Table 6-9, p.6-9 pdf.54; Project-only plus background
N02 USAF: Supplemental Information p3-3. pdf.23
N02 other: Updated Analyses Memo Table 7, p. 11 pdf. 11, "Startup/Shutdown - No PHPP Emergency generator"
8.5 Class I Area Analysis
The Class I area analysis was performed using CALPUFF Version 5.8 for long range transport,
which required additional detailed meteorological data as explained in the applicant's Class I
Modeling Protocol. Additionally, the applicant used CALPUFF to assess PSD Class I increment
consumption, regional haze, and acid deposition. The Class I modeling protocol was provided to
the Federal Land Managers (FLMs) for the two relevant Class I areas, the Cucamonga and the
San Gabriel Wilderness Areas. The FLMs raised no objections to the protocol or the modeling
62
Palmdale (SE 09-01)
Fact Sheet/Ambient Air Quality Impact Report
August 2011
-------
itself.
8.5.1 Class I Increment Consumption Analysis
The results of the PHPP Class I increment analysis are shown in Table 8-9; for the PSD pollutants
for which there are applicable increments, PHPP impacts are less than the Class I Significant
Impact Levels (SILs), and therefore the applicant has demonstrated that the Project will not cause
or contribute to any Class I PSD increment violation.
Table 8-9: P]
IPP Class I Increment Impacts
Significant
(lass 1 PSI)
Pollutant and
Project Impact.
1111 pad Level.
Increment.
Class 1 Area
averaging lime
iig/m''
iig/nr'
iig/nr'
Cucamonga
Wilderness Area
N02, annual
0.0010
0.1
2.5
PM10, 24-hr
0.059
0.3
8
PM10, annual
0.003
0.2
4
San Gabriel
Wilderness Area
N02, annual
0.0017
0.1
2.5
PM10, 24-hr
0.122
0.3
8
PM10, annual
0.004
0.2
4
Source: PSD Application, Table 6-10, p.6-11 pdf.56
8.5.2 Visibility and Deposition in Class I areas
The PSD regulations at 40 C.F.R. section 52.21 require that PSD permit applicants address
potential impairment to visibility (e.g., regional haze, plume blight) for Class I areas. The
deposition of nitrogen is another potential concern due to potential effects on soils, vegetation,
and other biological resources.
For Cucamonga Wilderness Area (WA), which is located greater than 50 km from the Project, a
Class I regional haze analysis was conducted. The modeling considered the two CTGs' emissions
of H2S04, NOx, PM10, PM2.5, and S02. The applicant used CALPUFF to predict visibility
impacts at Class I areas. Visibility impacts are assessed using the extinction coefficient (bext),
which represents the scattering of light by air pollutants, which appears as haze that reduces
visibility. The results of the CALPUFF modeling for the three meteorology years (2001-2003)
are shown in Table 8-10 and indicate that changes in light extinction (bext), averaged over a 24-
hour period, at Cucamonga WA is predicted to be below the 5% change threshold46.
46 "Federal Land Managers' Air Quality Related Values Workgroup (FLAG) Phase I Report" (December 2000), U.S.
Forest Service, National Park Service, U.S. Fish And Wildlife Service, http://www2.nature.nps.gov/air/Permits/flag/
63
Palmdale (SE 09-01)
Fact Sheet/Ambient Air Quality Impact Report
August 2011
-------
Table 8.10: Class I Area Regional Haze CALPUFF Modeling Results
Class 1 Area
Maximiiin Predicted
% Change in hlAI
Significance
Threshold
(%)
2001
2002
2003
Cucamonga WA
1.77
2.14
1.92
5
Applicants are not required to perform a cumulative effects analysis of new source growth if the
visibility impact of their proposed source is less than 5%. Based on the Class I regional haze
results, emissions from the facility are not expected to have an adverse impact on visibility in the
Cucamonga WA.
For San Gabriel WA, which is within 50 km of the Project, the impact of the Project on visibility
impairment, also known as plume blight, was assessed. The EPA VISCREEN screening model
was used to estimate visibility impairment to the San Gabriel WA from the CTG emissions.
Effects of plume blight are assessed as changes in plume perceptibility (AE) and plume contrast
(Cp) for sky and terrain backgrounds. A Level 1 analysis, using default meteorological data and
no site-specific conditions, was conducted. Because the Level 1 results of AE and Cp were above
the screening thresholds, a Level 2 analysis was conducted. A detailed discussion of the
VISCREEN plume blight impact analysis is presented in Section 6.2.4 of the applicant's PSD
permit application.
The results of the VISCREEN modeling runs are presented in Tables 8-11 and 8-12. The
VISCREEN results are presented for the two default worst-case theta angles - theta equal to 10
degrees representing the sun being in front of an observer, and theta equal to 140 degrees
representing the sun being behind the observer. A negative plume contrast means the plume has a
darker contrast than the background sky.
Table 8-1 la: Class I VISCREEN Modeling Results of
Changes in Plume Perceptibility (AE)
Background
Distance
Plume Perceptibility (\K)
Theta 10
Theta 140
Criteria
Sky
47.4
0.135
0.261
2.00
Terrain
34.6
0.806
0.072
2.00
Palmdale (SE 09-01)
Fact Sheet/Ambient Air Quality Impact Report
August 2011
64
-------
Table 8-1 lb: Class IVISCREEN Modeling Results of
Changes in Plume Contrast (Cp)
Background
Distance
P
nine Contrast i
C,.)
Tlicta 10
Tlicta 140
Criteria
Sky
47.4
0.001
-u.uuy
u.u5
Terrain
34.6
0.005
0.001
0.05
The results from the VISCREEN model show that changes in plume perceptibility and plume
contrast for sky and terrain backgrounds are below the criteria thresholds. Therefore, the plume
would not be perceptible against a sky or terrain background.
For Cucamonga WA and San Gabriel WA, a deposition analysis was conducted for nitrogen
compounds which considered Project emissions of NOx and conversion of NOxto nitrate and
nitric acid. The results from the deposition analysis are presented in Table 8-12.
Table 8-12: Class I Nitrogen Deposition CALPUFF Modeling Results
( lass 1 Area
.Maximum Predicted Nitrogen
Deposition - Annual average (g/ha/vr)
Deposition
Analysis
Threshold
(g/ha/yr)
2001
2002
2003
Cucamonga WA
0.496
0.521
0.458
5
San Gabriel WA
0.718
0.396
0.607
5
The Deposition Analysis Threshold was established by the Federal Land Managers, and represents
a level below which deposition is deemed to have no adverse effect, and does not require further
analysis.47 The maximum deposition rates modeled for PHPP are below the Class I Area Nitrogen
Deposition Analysis Threshold of 0.005 kilograms per hectares per year, or below 5 grams per
hectare per year (g/ha/yr), and therefore no further deposition analysis is necessary.
9. Additional Impact Analysis
In addition to assessing the ambient air quality impacts expected from a proposed new
source, the PSD regulations require that EPA evaluate other potential impacts on 1) soils
and vegetation; 2) growth; and 3) visibility impairment. 40 C.F.R. § 52.2l(o). The depth
of the analysis generally depends on existing air quality, the quantity of emissions, and the
47 "Guidance on Nitrogen and Sulfur Deposition Analysis Thresholds", Attachment to Letter from Christine L. Shaver,
National Park Service and Sandra V. Silva, U.S. Fish and Wildlife Service to S. William Becker, STAPPA/ALAPCO,
January 3, 2002 (files DatNotifyLetter.pdf, nsDATGuidance.pdf) http://www.nature.nps. gov/air/Permits/flag/
65
Palmdale (SE 09-01)
Fact Sheet/Ambient Air Quality Impact Report
August 2011
-------
sensitivity of local soils, vegetation, and visibility in the source's impact area.
9.1 Soils and Vegetation
For the soils and vegetation analysis, the applicant considered as part of the impact area the 400
meter significant impact area considered in the initial PSD application for the Project. In the
applicant's July 2010 supplement (Section 5.0), the applicant provided additional information on
the vegetation and soils inventory in the project area, a discussion of the potential impacts to
those soils and vegetation types with respect to the five Class II areas (within 50 km of the
project) discussed in Section 9.2, Visibility Impairment, and a discussion of nitrogen deposition.
Also, the applicant noted there are no federal habitat areas of concern within 20 miles of the
PHPP.
For most types of soils and vegetation, ambient concentrations of criteria pollutants below the
secondary NAAQS will not result in harmful effects because the secondary NAAQS are set to
protect public welfare, including vegetation, crops, and animals. No harmful effects are expected
from this project because the total estimated maximum ambient concentrations presented in Table
9-1 are below the primary NAAQS (listed in Table 8-1 of Section 8) and secondary NAAQS for
N02 (100 |ig/m3) and PM2..j(35 |ig/m3 for 24-hour periods; and 15.0 |ig/m3 over an annual
period). There are no secondary NAAQS for CO.
The initial application (dated March 2009) used EPA's "Screening Procedure for the Impacts of
Air Pollution Sources on Plants, Soils and Animals" (1980)48 to determine if maximum modeled
ground-level concentrations of N02 and CO could have an impact on plants, soils, and animals.
The modeled impacts of N02 and CO emissions from the facility, individually, and in addition to
the background concentrations of N02 and CO, are below the minimum impact level for sensitive
plants. The following table summarizes information in this regard from the PSD application (Table
6-17, Soils and Vegetation Analysis).
Table 9. 1
Project Maximum Concentrations and EPA Guidance Levels
Criteria Pollutant
and Guidance
Averaging l ime
KPA Screenin«
Concent rat ion
( ii g/m'')
Modeled Maximum
Concent rat ions
(jig/m')
Modeling
Averaging
time
N02 4-Hours
3,760
419.7
1 hour
N02 8-Hours
3,760
419.7
1 hour
N02 1-Month
564
419.7
1 hour
N02 Annual
94
29.2
Annual
CO Weekly
1,800,000
1,806.4
8 hour
48 Screening Procedure for the Impacts of Air Pollution Sources on Plants, Soils, and Animals," EPA 450/2-81-078,
December 1980.
66
Palmdale (SE 09-01)
Fact Sheet/Ambient Air Quality Impact Report
August 2011
-------
As part of the July 2010 supplement regarding additional impacts to vegetation, the applicant also
reviewed a document developed by the U.S. Department of Agriculture entitled "A Screening
Procedure to Evaluate Air Pollution Effects in Region 1 Wilderness Areas" (1991). As a
complement to the EPA 1980 screening procedure document, the applicant determined that for
the NOx "sensitive" species of alfalfa, which is found nearby the project, the modeled air
concentrations (Table 9-1) demonstrate that the impacts are below the significance criteria.
The applicant also considered soil acidification and eutrophication as part of the July 2010
supplement regarding additional impacts on soil. Nitrogen deposition in soil can have beneficial
effects to vegetation if they are lacking these elements; however, gaseous emissions impacts on
soils at levels greater than vegetation requirements can cause acidic conditions to develop. Soil
acidification and eutrophication can occur as a result of atmospheric deposition of nitrogen.
The applicant determined that project-specific modeling for nitrogen deposition was not
warranted because the estimated nitrogen deposition rates were negligible as a plant growth
influence and because the effects of deposition on eutrophication were insignificant, as described
below.
When considering soil acidification, the applicant referred to the CALPUFF modeling conducted
for the PHPP's Class I analysis. The applicant also referred to the nitrogen deposition modeling
analysis (using CALPUFF) performed for a similar project, the Victorville 2 (VV2) Hybrid Power
Project.49 CALPUFF incorporates the atmospheric chemistry and chemical transformations to
determine nitrogen deposition and provides results in units of kilograms per hectare per year,
which can be converted to pounds per unit area. For the VV2 project, the modeled maximum
annual deposition rate was considered to be very low.
The PHPP is nearly identical to the VV2 hybrid solar-gas plant, with the exception of a larger
natural gas-fired auxiliary boiler; the PHPP boiler is 110 MMBtu/hr, while the VV2 boiler is 40
MMBtu/hr. Additionally, the predominant wind direction for PHPP is the northeast of the power
block, which is similar to the predominant wind direction for VV2. (There have not been
pertinent upgrades to the CALPUFF model since the VV2 2008 analysis.). Because of the
similarities between the PHPP and VV2, and VV2's fence line deposition of 1.2 ounces of
nitrogen per acre, the applicant determined that the nitrogen deposition rates for PHPP also
would be considered negligible as a plant growth influence, and therefore no additional nitrogen
deposition analysis was performed.
In sum, based on our consideration of the information and analysis provided by the applicant, we
do not believe that emissions associated with the Project will result in adverse impacts on soils or
vegetation.
49 EPA Region 9 issued the initial PSD permit to the Victorville 2 Hybrid Power Project in 2010. EPA proposed the
PSD permit in 2008, with Docket I.D. number EPA-R09-OAR-2008-0406.
fhttp://www.regulations.gov/#!docketDetaikD=EPA-R09-OAR-2008-04Q6'). The initial PSD permit was issued in 2010
with Docket I.D. number EPA-R09-OAR-2008-0765 fhttp ://www.regulations.gov/#!docketDetaikD=EPA-R09-QAR-
2008-0765 1
67
Palmdale (SE 09-01)
Fact Sheet/Ambient Air Quality Impact Report
August 2011
-------
9.2 Visibility Impairment
Using procedures in EPA's Workbook for Plume Visual Impact Screening and Analysis50, the
applicant evaluated visibility impairment for one Class I area and five Class II areas. The five
Class II areas included three state parks, one woodland, and one wilderness area.
In the initial PSD application, the applicant presented visibility impairment (e.g., plume blight) for
the Class I area of San Gabriel Wilderness Area (see Section 8.5.2 of the application), which is
located within 50 km of the proposed PHPP. The applicant provided supplemental application
information for visibility impairment in July 2010 for five Class II areas identified as potentially
sensitive state or federal parks, forests, monuments, or recreation areas within 50 km of the
project. These five areas with their approximate closest distances to PHPP were:
• Antelope Valley Indian Museum State Park (23 km)
• Saddleback Butte State Park (26 km)
• Antelope Valley California Poppy State Reserve (26 km),
• Arthur B. Ripley Desert Woodland (37 km), and
• Sheep Mountain WA (43 km)
The applicant performed a Level 1 and Level 2 VISCREEN analysis for all five areas. The results
of this analysis were below the significance criteria for three of the five areas. A further refinement
in VISCREEN of plume perceptibility for the two exceptions - Saddleback Butte State Park and
Antelope Valley Indian Museum State Park - was performed for the worst-case daytime
meteorological conditions; the result is that the plume would not be perceptible at either site
during daylight hours, based on low plume perceptibility and contrast predicted by VISCREEN.
Based on the VISCREEN results, w believe that the Project would not contribute to visibility
impairment.
9.3 Growth
The growth component of the additional impact analysis considers an analysis of general
commercial, residential, industrial and other growth associated with the PHPP. 40 C.F.R. §
52.21(o). The PHPP is expected to employ 36 employees, with an ample work force in the
Southern California area to accommodate the PHPP estimated peak of 767 construction workers;
impacts to the local population and housing needs are therefore expected to be minimal.
Therefore, we do not expect this project to result in any significant growth.
The applicant provided growth-related information in its initial PSD application and in
supplemental application materials submitted to EPA in July 2010 and July 2011. The July 2011
supplement includes Attachment A, which is an updated version of the socioeconomics analysis
PHPP prepared for its July 2008 California Energy Commission (CEC) Application for
50 "Workbook for Plume Visual Impact Screening and Analysis (Revised)", EPA, EPA-454/R-92-023, 1992.
68
Palmdale (SE 09-01)
Fact Sheet/Ambient Air Quality Impact Report
August 2011
-------
Certification (AFC). The applicant's original July 2008 CEC AFC socioeconomics analysis was
based on 2000 Census data; Attachment A of the July 2011 supplement includes updated
information based on the available 2010 Census data regarding population and population growth
projections.
The applicant's initial PSD application growth analysis (Section 6.3.2) stated that .. no long-
term growth is expected during project operations." A Project labor force of 36 employees was
estimated. The July 2010 supplement further discussed the Project's potential growth-inducing
activities. Additional details in this supplement included a summary of growth-inducing impacts
associated with employment. The information submitted indicates that for the construction and
operating phases of the Project, impacts to the population and housing needs are expected to be
minimal, and are expected not to induce substantial population growth.
With regards to the question of whether the Project's power generation would induce growth, the
applicant anticipates that the Project would likely displace the older once-through cooling
facilities in the Southern California region that are expected to be retired in the future. Therefore,
rather than induce growth, PHPP would supply energy to accommodate the existing demand and
projected growth in the Southern California region.
In sum, based on our consideration of the information and analysis provided by the applicant, we
do not expect the Project to result in any significant growth.
10. Endangered Species
Pursuant to section 7 of the Endangered Species Act (ESA), 16 U.S.C. 1536, and its
implementing regulations at 50 C.F.R. Part 402, EPA is required to ensure that any action
authorized, funded, or carried out by EPA is not likely to jeopardize the continued
existence of any endangered or threatened species or result in the destruction or adverse
modification of such species' designated critical habitat. EPA has determined that this
PSD permitting action is subject to ESA section 7 requirements.
The applicant and EPA identified two federally-listed species,the desert tortoise Gopherus
agassizii) and the arroyo toad (Bufo californica), that might be affected by the proposed
PSD permitting action for the Project. In March 2009, a Draft Biological Assessment
(BA) was submitted by the applicant to EPA and the U.S. Fish and Wildlife Service
(FWS). Based on discussions between the applicant and FWS, in August 2009, the
applicant submitted to EPA and FWS an Addendum to the BA. The BA Addendum
further detailed that the PHPP "... may affect but is not likely to adversely affect the
desert tortoise and will have no effect on the arroyo toad." In July 2011, the applicant
submitted a second Addendum to the BA to EPA and FWS, outlining updates to the
Project scope and a further analysis supporting the conclusion that the PHPP may affect,
but is not likely to adversely affect, the federally-listed desert tortoise and will have no
effect on the federally-listed arroyo toad.
69
Palmdale (SE 09-01)
Fact Sheet/Ambient Air Quality Impact Report
August 2011
-------
In a letter dated August 5, 2011, EPA requested FWS's written concurrence with EPA's
determination under ESA section 7 that the proposed PSD permit for the PHPP is not
likely to adversely affect the desert tortoise or arroyo toad.
EPA will proceed with its final PSD permit decision after making a determination that
issuance of the permit will be consistent with ESA requirements. In making this
determination, EPA will consider actions taken, or to be taken, by the applicant to ensure
ESA compliance.
11. Environmental Justice Analysis
Executive Order 12898, entitled "Federal Actions To Address Environmental Justice in
Minority Populations and Low-Income Populations," states in relevant part that "each
Federal agency shall make achieving environmental justice part of its mission by
identifying and addressing, as appropriate, disproportionately high and adverse human
health or environmental effects of its programs, policies, and activities on minority
populations and low-income populations." Section 1-101 of Exec. Order 12898, 59 Fed.
Reg. 7629 (Feb. 16, 1994).
EPA determined that there may be minority or low-income populations potentially
affected by its proposed action on the PHPP PSD permit application, and determined that
it would be appropriate to prepare an Environmental Justice Analysis for this action. EPA
therefore prepared an Environmental Justice Analysis, which is included in the
administrative record for EPA's proposed PSD permit for the Project. EPA's analysis
concludes that the Project will not cause or contribute to air quality levels in excess of
health standards for the pollutants regulated under EPA's proposed PSD permit for the
Project, and that therefore the Project will not result in disproportionately high and
adverse human health or environmental effects with respect to these air pollutants on
minority or low-income populations residing near the proposed Project, or on the
community as a whole.
12. Clean Air Act Title IV (Acid Rain Permit) and Title V
(Operating Permit)
The applicant must apply for and obtain an acid rain permit and a Title V operating permit.
The applicant will apply for these permits after the facility is constructed, as these permits
are not required prior to construction. The District has jurisdiction to issue the Acid Rain
Permit and the Operating Permit for the facility.
13. Comment Period, Hearing, Public Information Meeting,
Procedures for Final Decision, and EPA Contact
70
Palmdale (SE 09-01)
Fact Sheet/Ambient Air Quality Impact Report
August 2011
-------
The comment period for EPA's proposed PSD permit for the Project begins on August
11, 2011. Any interested person may submit written comments on EPA's proposed PSD
permit for the Project. All written comments on EPA's proposed action must be received
by EPA via email by September 14, 2011, or postmarked by September 14, 2011.
Comments must be sent or delivered in writing to Lisa Beckham at one of the following
addresses:
E-mail: R9airpermits@epa. gov
U.S. Mail: Lisa Beckham (AIR-3)
U.S. EPA Region 9
75 Hawthorne Street
San Francisco, CA 94105-3901
Phone: (415) 972-3811
Comments should address the proposed PSD permit and facility, including such matters
as:
1. The Best Available Control Technology (BACT) determinations;
2. The effects, if any, on Class I areas;
3. The effect of the proposed facility on ambient air quality; and
4. The attainment and maintenance of the NAAQS.
Alternatively, written or oral comments may be submitted to EPA at the Public Hearing
for this matter that EPA will hold on September 14, 2011, pursuant to 40 C.F.R. §
124.12, to provide the public with further opportunity to comment on the proposed PSD
permit for the Project. At this Public Hearing, any interested person may provide written
or oral comments, in English or Spanish, and data pertaining to the proposed permit.
Prior to the Public Hearing, EPA will also hold a Public Information Meeting for the
purpose of providing interested parties with additional information and an opportunity for
informal discussion of the proposed Project.
The date, time and location of the Public Information Meeting and the Public Hearing are
as follows:
Date: September 14, 2011
Time: 4:00 p.m. - 6:00 p.m. (Public Information Meeting)
7:00 p.m. - 10:00 p.m. (Public Hearing)
Location: Larry Chimbole Cultural Center
Manzanita Ballroom, 2nd Floor
38350 Sierra Highway
Palmdale, California 93550-4611
71
Palmdale (SE 09-01)
Fact Sheet/Ambient Air Quality Impact Report
August 2011
-------
English-Spanish translation services will be provided at both the Public Information
Meeting and the Public Hearing.
If you require a reasonable accommodation, by August 31, 2011 please contact Terisa
Williams, EPA Region 9 Reasonable Accommodations Coordinator, at (415) 972-3829, or
Williams. T erisa@epa. gov.
All information submitted by the applicant is available as part of the administrative record.
The proposed air permit, fact sheet/ambient air quality impact report, permit application
and other supporting information are available on the EPA Region 9 website at
http://www.epa.gOv/region09/air/permit/r9-permits-issued.html#pubcomment. The
administrative record may also be viewed in person, Monday through Friday (excluding
Federal holidays) from 9:00 AM to 4:00 PM, at the EPA Region 9 address above. Due to
building security procedures, please call Lisa Beckham at (415) 972-3811 at least 24 hours
in advance to arrange a visit. Hard copies of the administrative record can be mailed to
individuals upon request in accordance with Freedom of Information Act requirements as
described on the EPA Region 9 website at http://www.epa.gov/region9/foia/.
Additional information concerning the proposed PSD permit may be obtained between the
hours of 9:00 a.m. and 4:00 p.m., Monday through Friday, excluding holidays, from:
E-mail: R9airpermits@epa. gov
U.S. Mail: Lisa Beckham (AIR-3)
U.S. EPA Region 9
75 Hawthorne Street
San Francisco, CA 94105-3901
Phone: (415) 972-3811
EPA's proposed PSD permit for the Project and the accompanying fact sheet/ambient air
quality impact report are also available for review at the following locations: Antelope
Valley Air Quality Management District, 43301 Division Street, Suite 206, Lancaster, CA
93535, (661) 723-8070; Palmdale City Library, 700 East Palmdale Boulevard, Palmdale,
CA 93550-4742, (661) 267-5600; Lancaster Regional Library, 601 W. Lancaster
Boulevard, Lancaster, CA 93534-3398, (661) 948-5029; Lake Los Angeles Library,
16921 East Avenue O, Palmdale, CA 93591-3045, (661) 264-0593; and Quartz Hill
Library, 42018 N. 50th Street West, Quartz Hill, CA 93536-3590, (661) 943-2454.
All comments that are received will be included in the public docket without change and
will be available to the public, including any personal information provided, unless the
comment includes Confidential Business Information (CBI) or other information whose
disclosure is restricted by statute. Information that is considered to be CBI or otherwise
protected should be clearly identified as such and should not be submitted through e-mail.
If a commenter sends e-mail directly to the EPA, the e-mail address will be automatically
captured and included as part of the public comment. Please note that an e-mail or postal
72
Palmdale (SE 09-01)
Fact Sheet/Ambient Air Quality Impact Report
August 2011
-------
address must be provided with comments if the commenter wishes to receive direct
notification of EPA's final decision regarding the permit.
EPA will consider all written and oral comments submitted during the public comment
period before taking final action on the PSD permit application and will send notice of the
final decision to each person who submitted comments and contact information during the
public comment period or requested notice of the final permit decision. EPA will respond
to all substantive comments in a document accompanying EPA's final permit decision and
will make the Public Hearing proceedings available to the public.
EPA's final permit decision will become effective 30 days after the service of notice of the
decision unless:
1. A later effective date is specified in the decision; or
2. The decision is appealed to EPA's Environmental Appeals Board pursuant to 40 CFR
124.19; or
3. There are no comments requesting a change to the proposed permit decision, in which
case the final decision shall become effective immediately upon issuance.
14. Conclusion and Proposed Action
EPA is proposing to issue a PSD permit for the PHPP. We believe that the proposed
Project will comply with PSD requirements, including the installation and operation of
BACT, and will not cause or contribute to a violation of the applicable NAAQS or
applicable PSD increments. We have made this determination based on the information
supplied by the applicant and our review of the analyses contained in the permit
application and other relevant information contained in our administrative record. EPA
will make this proposed permit and this Fact Sheet/AAQIR available to the public for
review, and make a final decision after considering any public comments on our proposal.
Palmdale (SE 09-01)
Fact Sheet/Ambient Air Quality Impact Report
August 2011
73
------- |