Energy Answers Arecibo, LLC

Imagine the result
Energy Answers Arecibo, LLC
PSD Air Quality Modeling Analysis
Amendment for Startup Periods
For the proposed
Arecibo Renewable Energy Project
Arecibo, Puerto Rico
Barrio Cambalache, Arecibo, Puerto Rico
Submitted February 2012
Energy Answers Arecibo
Renewable Energy Project
Arecibo, Puerto Rico
PSD Air Quality Modeling
Amendment for Startup
Prepared for:
Energy Answers Arecibo, LLC
Prepared by:
ARCADIS
801 Corporate Center Drive
Suite 300
Raleigh, North Carolina 27607
Tel 919.854.1282
Fax 919.854.5448
Our Ref.:
NCENRGY1.0005
Date:
February 2012
Table of Contents
1.0 Introduction
1
2.0 Project and Site Description
2
3.0 Regulatory Applicability
2
4.0 Source Description and Operating Scenarios
4
4.1 Boiler Operating Load Scenarios
5
4.2 Boiler Startup and Shutdown
6
4.3 Other Sources
7
4.4 Pollutants Evaluated
8
5.0 Modeling Methodology
8
5.1 Model Selection
8
5.2 Meteorological Data
9
5.3 Surface Characteristics
9
5.4 Dispersion Coefficients
10
5.5 Receptor Arrays
10
5.6 Good Engineering Practice Stack Height and Building Downwash
11
5.7 Source Input Data
12
5.7.1 Model Setup – Source and Source Group Naming Convention
15
6.0 Model Results for Evaluating Significance
16
6.1 Identifying the Significant Impact Area (SIA)
18
6.2 Full (Cumulative) Impact Analysis
19
6.2.1 Background Air Quality
19
6.2.2 Off-Site Source Inventory
20
6.2.3 AERSCREEN Concentration Gradient Evaluation for Sources to the
South
21
6.3 Evaluating 1-hour NO2 Cumulative Impacts
22
7.0 Environmental Justice
23
8.0 References
24
i
Table of Contents
Figures
2-1
Project Location Map
2-2
Site Location Map
2-3
Site Layout With Emission Points
5-1
Site and Surface Observation Stations Location Map
A
Emission Rate Calculations
B
Air Modeling Files on DVD
Appendix
ii
Energy Answers
PSD Air Quality
Modeling Analysis
Revised for Startup
1.0 Introduction
Energy Answers International, Inc. (Energy Answers) is proposing to construct and
operate a 77 megawatt (MW) renewable energy facility at the former site of the Global
Fibers paper mill in Barrio Cambalache, Arecibo, Puerto Rico, referred herein as the
Arecibo Renewable Energy Project (AREP). The proposed facility will consist of two
spreader-stoker boilers, each with a heat input rating of 500 million British Thermal
Units (MMBTU) per hour. The primary fuel for the plant will be processed refuse fuel
(PRF), which is derived from municipal solid waste (MSW) by first shredding the
material and then removing much of the metal for recycling. Energy Answers was the
developer, owner and co-operator of a similar PRF-fired facility in Rochester,
Massachusetts, called the SEMASS Recovery Facility. Energy Answers plans to
install state-of-the-art air pollution controls on its boilers for controlling its potential
emissions of regulated air pollutants.
The proposed power plant falls under one of the 28 named source categories under
the New Source Review permitting program with a major source threshold of 100 tons
per year. Potential emission estimates for the proposed plant exceed the major source
threshold of 100 tons per year. And since the site is located in an attainment area for
criteria air pollutants, the proposed plant is subject to the Prevention of Significant
Deterioration (PSD) permitting process.
Accordingly, Energy Answers prepared an application for PSD permit to construct,
including an ambient air impact analysis using air dispersion modeling methods. The
PSD application with a dispersion modeling analysis was submitted to EPA Region 2 in
February 2011. Following EPA’s release of an updated version of the AERMOD
dispersion model, a revised air modeling analysis was submitted in July 2011 per the
request of EPA Region 2. A revised modeling analysis was submitted in October 2011
to address a change in potential emissions of condensable particulate matter and also
to address comment to the July submittal. Subsequent changes to the potential
emission rate of Nitrogen Oxides (NOx) and Carbon Monoxide (CO) during startup
periods warranted further analysis, which is the focus of this submittal.
The modeling analysis was completed in accordance with the modeling protocols
submitted in May 2011 and September 2011 (PM10/PM2.5 Addendum) and approved
July 5, 2011 and October 11, 2011.
1
Energy Answers
PSD Air Quality
Modeling Analysis
Revised for Startup
2.0 Project and Site Description
Energy Answers is proposing to construct a 77-MW renewable energy facility to be
fueled primarily by PRF to produce steam and electricity. The boilers will fire
approximately 2,106 tons per day of PRF on average. MSW will be processed to
produce PRF, and the PRF will then be combusted to produce steam used to generate
electricity. The facility proposes to supplement the PRF with processed urban wood
waste, tire derived fuel, automotive shredder residue in amounts limited to 50%, 20%
and 20% by weight of the total fuel feed, respectively.
The facility will be located in Barrio Cambalache, Municipality of Arecibo, Puerto Rico.
Figure 2-1 shows the location of the site on the island, and Figure 2-2 provides the
location of the site on a United States Geological Survey (USGS) topographic map.
The approximate UTM coordinates for the facility are 742.688 km E and 2,042.698 km
N (UTM Zone 19) with the design plant grade at approximately 20 feet (3.2 meters)
above mean sea level (MSL). The facility will be built such that the waste receiving,
waste processing, and energy recovery operations are conducted within the
boundaries of the site.
The topography in the immediate vicinity of the site is generally flat. The shoreline is
approximately 1 mile to the north. To the south, the terrain becomes hilly and
eventually mountainous (complex). A review of USGS 7.5-minute quadrangle map
indicates that most of the surrounding terrain within 5 kilometers (km) of the site is
below the proposed stack height. A scaled design site layout is provided in Figure 2-3.
The nearest Class I area to the proposed plant site is the Virgin Island National Park on
the Island of St. John, located approximately 125 miles to the east.
3.0 Regulatory Applicability
Energy Answers is required to obtain a Permit-to-Construct from the EPA and the
Puerto Rico Environmental Quality Board (PREQB) prior to beginning construction.
Based on the design processing rates, this facility is subject to the requirements under
the PSD regulations contained in 40 CFR Part 52.21 since the site location is currently
designated in attainment of the National Ambient Air Quality Standards (NAAQS)
except for lead. Specific regulatory applicability requirements and emission limitations
are discussed in the PSD air permit application for the proposed AREP submitted to
EPA Region 2 on February 7, 2011 and in subsequent responses to comments.
2
Energy Answers
PSD Air Quality
Modeling Analysis
Revised for Startup
Pursuant to 40 CFR 52.21(k) and (l), facilities that require review under the PSD
regulations must conduct an air quality impact analysis using air dispersion modeling
methods for each pollutant emitted in major (significant) quantities. The purpose of the
analysis is to demonstrate whether the proposed installation will meet applicable
NAAQS and PSD allowable increments during startup conditions.
This air quality analysis begins with a preliminary analysis of the significant increase in
potential emissions from a proposed new source. The results of the preliminary
analysis are compared with accepted significant impact levels (SIL) for each pollutant
to determine whether a full impact analysis is necessary and, if so, to define the area
where the analysis must be completed. If the preliminary analysis indicates that
predicted ambient air impacts are below the SIL, it is deemed insignificant or de
minimis, and no further analysis is required. Should potential air quality impacts
exceed the SIL, a full impact analysis must be conducted with respect to the NAAQS
and PSD allowable increments, including off-site emission sources.
The applicable SILs, NAAQS and allowable PSD increments are defined in 40 CFR
Part 50 and 51, or provided in New Source Review Workshop Manual (USEPA 1990)
and guidance memorandums from EPA (USEPA 2010a-g). Table 3-1 below lists these
thresholds used for this air quality analysis. Further details on the air quality analysis
procedures are given in Section 5 and in the modeling protocol documents that were
submitted to EPA under separate cover.
3
Energy Answers
PSD Air Quality
Modeling Analysis
Revised for Startup
Table 3-1: Ambient Air Quality Standards, PSD Increments, and Significant
Impact Levels
Pollutant
SO2
PM10
PM2.5
CO
NO2
Pb
Averaging
Period
National Ambient Air
Quality Standard
3
(µg/m )
1-hour
196
PSD
Increment
Class II
3
(µg/m )
---
3-hour
1,300
512
25
91
5
20
1
24-hour
365
Annual
80
(b)
(b)
SIL
3
(µg/m )
7.8
(a)
24-hour
150
30
5
Annual
Revoked
17
1
24-hour
35
9
1.2
Annual
15
4
0.3
1-hour
40,000
---
2,000
8-hour
10,000
---
1-hour
188
---
7.5
Annual
3-month
(rolling)
100
25
1
0.15
---
---
500
(c)
(a) EPA recommended a non-binding interim SIL of 3 ppb for the 1-hour SO2 NAAQS in August 2010
(USEPA, 2010d). Assuming a conversion based on the 3-hour secondary standards for SO2, the
SIL would be 7.8 µg/m3.
(b) The EPA is revoking the two existing primary standards of 140 ppb evaluated over 24 hours, and
30 ppb evaluated over an entire year because they will not add additional public health protection
given a 1-hour standard at 75 ppb. Nevertheless, this analysis addresses these time averaging
periods for reference.
(c) EPA provided a non-binding interim SIL of 4 ppb for the 1-hour NO2 NAAQS in June 2010
(USEPA, 2010c). Converting to a mass-based value and rounding to a whole number results in a
1-hour NO2 SIL of 7.5 µg/m3.
4.0 Source Description and Operating Scenarios
The proposed AREP will have the following air emission sources:
•
Two (2) spreader-stoker boilers with a maximum heat input rating of 500
MMBTU/hr each, equipped with three (3) 167 MMBTU/hr No 2 Fuel Oil-fired
burners each;
•
One (1) cooling tower, with 4-cells (air-cooled condenser type);
•
Fly and bottom ash transfer, processing and storage operations;
4
Energy Answers
PSD Air Quality
Modeling Analysis
Revised for Startup
•
Three (3) Storage Silos (lime, pulverized activated carbon, flyash);
•
One (1) diesel fuel-fired emergency generator; and
•
One (1) diesel fuel-fired emergency firewater pump
Energy Answers proposes to install advanced air quality control systems that qualify as
the Best Available Control Technology (BACT) for its operations. Independently
operating air quality control systems will be proposed for each boiler, consisting of the
following technologies:

An activated carbon injection system to remove heavy metals, including
mercury and dioxins/furans;

A Turbosorp Dry Circulating Fluid Bed Scrubber system to remove acid
gases using lime injection

A fabric filter (baghouse) to control particulate emissions (including metals);
and,

A regenerative selective catalytic reduction (RSCR) system for reducing
emissions of NOx and CO.
4.1 Boiler Operating Load Scenarios
Under normal operating conditions, the boilers are expected to operate at an average
heat input rating of 500 MMBTU/hr each. For the purposes of this air quality impact
analysis, 500 MMBTU/hr is defined as the 100% load scenario. This analysis includes
multiple scenarios where one boiler is undergoing startup while the second is operating
at 80%, 100%, 110% load corresponding to 400 MMBTU/hr, 500 MMBTU/hr,
550MMBTU/hr, respectively, or is inactive. It is noted that the proposed emission rates
at these load conditions are based on vendor guarantees and do not depend upon the
amount or type of supplemental fuel (ASR, TDF, UWW) used at a given time.
In October 2011, Energy Answers submitted the PSD Air Quality Modeling Analysis
(Revised PM10/PM2.5 Analysis) to USEPA Region 2 which addresses each of these
operating scenarios under normal operating conditions. There are no changes to
normal operating conditions from what was modeled in October 2011. Details for
potential air quality impacts during normal operations are given in that report. This
analysis represents a revision to startup conditions only.
5
Energy Answers
PSD Air Quality
Modeling Analysis
Revised for Startup
4.2 Boiler Startup and Shutdown
The proposed AREP will use No. 2 fuel oil for startup and shutdown, and intermittently
during short-term plant upsets in order to maintain boiler temperatures. Each boiler
unit will be started up using auxiliary burners firing No. 2 fuel oil (ultra low sulfur
content) to preheat the flue gas until the temperature can be maintained at or above
1800°F. At that point, PRF will be introduced into the boiler. Energy Answers
estimates that a cold start will take approximately 7 hours. Previously, emission
estimations indicated that emissions during startup periods could be expected to be
lower than emissions during normal operations while firing PRF. However, a closer
examination of the emission factors used for NOx and CO indicated that emissions
were likely under-estimated. Furthermore, although Energy Answers initially thought
that the RSCR could be brought on line prior to firing the fuel oil burners, the vendor
has indicated that the RSCR will not begin to effectively control both NOx and CO
during the startup period. This is because the temperature of the boiler flue will not be
sufficient to enable proper atomization of ammonia for NOx reduction. There also
concern that the flue during startup will have a cooling effect on the catalysts for some
portion of the startup period so they will not support the chemical reactions necessary
for controlling NOx and CO. Emission calculations for startup periods have been
adjusted to account for these uncertainties, conservatively assuming that no control by
the RSCR is achieved during startup. The revised emission calculations for startup are
given in Appendix A.
For the purposes of this modeling demonstration, the startup sequence has been
broken down into three phases to represent the gradual ramp-up period of preheating
the boiler before reaching the point when PRF can be introduced (80% load). Initially,
the boiler will be fired at approximately 35 MMBTU/hr for an estimated 4.5 hours. At
that point, the boiler firing rate will be increased to approximately 250 MMBTU/hr. After
about an hour, the boiler firing rate will be increased to 400 MMBTU/hr until the proper
temperature is reached and PRF is introduced into the combustor. This third phase is
expected to require about 1.5 hours. Emission rate calculations for each of the startup
periods are provided in Appendix A.
Shutdown is expected to take an estimated 6 hours or less to complete. During
shutdown events, the general procedure will be to stop feeding PRF and fire fuel oil
until the burnout of remaining PRF is completed and the grates are clear. Fuel oil
burners will begin firing when PRF feed has stopped. Energy Answers will take
measures to minimize emissions during shutdown by keeping the air quality control
system functioning until the grates are cleared of PRF and PRF burnout has been
6
Energy Answers
PSD Air Quality
Modeling Analysis
Revised for Startup
completed. Emissions during the shutdown process are not subject to change at this
time. No additional limits from the proposed BACT for normal operations are
requested for shutdown. Therefore, no additional modeling has been completed for
the shutdown periods.
Emissions estimated to occur during startup were modeled to demonstrate compliance
with the short-term averaged standards (1-hour and 8-hour CO, 3-hour and 24-hour
SO2; 24-hour PM10 and PM2.5). Due to the increase in potential emissions of NOx from
previous estimates, potential impacts of NO2 on a 1-hour average are also provided.
Energy Answers expects that each boiler will undergo approximately sixteen (16)
startup and shutdown events per year.
Modeling was completed for four potential startup scenarios: one each for the
emissions while one boiler is undergoing startup and the second boiler is inactive, or
operating at 80%, 100%, and 110% load. Energy Answers proposes to accept a timeof-day (TOD) restriction for initiating startup of either boiler. Startup will begin between
7:00 AM and 12:00 PM only. Also, simultaneous startup of the boilers will not occur. It
is understood that this proposed TOD limit is consistent with the recent EPA guidance
memorandum (USEPA 2011) issued for the purposes of conducting the air quality
impact analysis for the 1-hour NO2 and SO2 standards pursuant to the PSD permitting
requirements. The three stages of startup are represented in each run, as are the four
possible operating conditions for the second boiler (0%, 80%, 100%, and 110%).
With respect to the need to address the new 1-hour standard for SO2 during startup
and shutdown periods, we reference the approved modeling protocol which requests
an exception from evaluating potential impacts of SO2 on a 1-hour averaging period
during startup and shutdown.
4.3 Other Sources
Emissions from the following sources have not changed from the October analysis:

Cooling Tower

Ash Processing Operations

Storage Silos

Firewater Pump

Emergency Diesel Generator

Fugitive Emissions.
7
Energy Answers
PSD Air Quality
Modeling Analysis
Revised for Startup
Therefore, please reference the PSD Air Quality Modeling Analysis (Revised
PM10/PM2.5 Analysis) submitted in October 2011 for further details on emission
calculations, assumptions and model input information for these listed sources.
4.4 Pollutants Evaluated
The Facility will have a potential to emit CO, NO2, SO2, PM, PM10, PM2.5,VOC, Lead,
Beryllium, Fluoride, Mercury, Sulfuric Acid Mist, MWC Organics, MWC Acid Gases and
GHG. With the exception of lead, each of these is projected to exceed the applicable
PSD significant emission rate (SER) threshold. Potential emissions from the facility are
below the applicable PSD SER levels for all other PSD regulated pollutants listed in 40
CFR Subpart 52. Accordingly, the facility is subject to the PSD air quality impact
analysis requirements for CO, NO2, SO2, PM10 and PM2.5. There are no applicable
ambient air standards for the other constituents and, therefore, no air quality modeling
impact analysis is required. This analysis focuses on emissions of CO, NO2, SO2,
PM10 and PM2.5 during startup periods.
5.0 Modeling Methodology
The modeling analysis was completed in accordance with the modeling protocols
submitted in May 2011 and September 2011 (PM10/PM2.5 Addendum) and approved
July 5, 2011 and October 11, 2011. This section provides a summary of the model
selection, land use classification, receptor grid specifications, meteorological data set,
receptor grid arrays, Good Engineering Practice (GEP) stack height analysis, building
downwash parameters, emission source input data, and the background ambient air
concentrations to be used for this analysis.
5.1 Model Selection
Energy Answers used the most current version of EPA’s AERMOD (11353) dispersion
model to predict ambient concentrations in simple, complex and intermediate terrain.
The AERMOD Modeling System includes preprocessor programs (AERMET (11059),
AERSURFACE (updated January 2008), and AERMAP (11103)) to create the required
input files for meteorology and receptor terrain elevations. AERMOD is the
recommended model in USEPA’s Guideline on Air Quality Models (40 CFR Part 51,
Appendix W) (USEPA 2005). The regulatory default option was used. Specifically, the
regulatory default option directs AERMOD to use:

The elevated terrain algorithms requiring input of terrain height data for
receptors and emission sources;
8
Energy Answers
PSD Air Quality
Modeling Analysis
Revised for Startup

Stack tip downwash (building downwash automatically overrides);

The calms processing routines;

Buoyancy-induced dispersion; and

The missing meteorological data processing routines.
5.2 Meteorological Data
Careful consideration must be given to selecting a location from which to obtain
meteorological data that were representative of conditions at the proposed project site.
Per the recommendation of USEPA and the approved modeling protocol, one year of
meteorological data (August 1992 to August 1993) obtained from the Puerto Rico
Energy Power Authority (PREPA) facility in Cambalache Barrio (located within one mile
of the proposed AREP site) was used. Figure 5-1 shows the proximity of the PREPA
station to the proposed Energy Answers site.
The Cambalache data include wind direction, wind speed, temperature, solar radiation,
sigma theta, sigma phi, and temperature difference between levels. Additional
meteorological parameters required for executing AERMOD including cloud cover,
ceiling height, pressure, and relative humidity were extracted from the 1992-1993 San
Juan surface station Hourly US Weather Observation data. Additionally, substitutions
for missing data (winds) were extracted from the 1992-1993 San Juan surface station
data. The onsite parameters, as well as the National Weather Service (NWS) surface
and upper air input files for AERMOD were prepared using the AERMET utility.
Further details regarding the meteorological data can be found in the May 2011
protocol.
5.3 Surface Characteristics
The inputs to AERMET for surface characteristics (surface roughness, Albedo and
Bowen ratio) were determined based on land use in the area surrounding the
Cambalache meteorological site. Surface characteristics surrounding the San Juan
International Airport are also incorporated as part of the AERMET data substitution
technique that is available when processing onsite data. These parameters remain
unchanged from what was described in the May 2011 approved protocol. Per EPA
direction, the AERSURFACE utility was not used for this project. Rather, surface
roughness numbers were calculated per the Alaska Department of Environmental
Conservation (ADEC) Guidance for AERMET Geometric Means, (ADEC 2009)
developed by the State of Alaska. Further details on the derivation of the albedo,
9
Energy Answers
PSD Air Quality
Modeling Analysis
Revised for Startup
Bowen ratio, and surface roughness coefficients used in this analysis can be
found in the May 2011 modeling protocol.
5.4 Dispersion Coefficients
In addition to surface characteristics for AERMET, it is necessary to select appropriate
dispersion coefficients when executing AERMOD (and other dispersion models).
Generally, the dispersion coefficients are determined using the USEPA-preferred land
use classification technique in 40 CFR 51, Appendix W (also known as the “Auer”
technique). Based on a review of land use in the vicinity of the site, approximately 20
percent of the area within three (3) kilometers is urban while rural land use constitutes
approximately 80 percent. Given this land use for the area, the “rural” dispersion
parameter is selected for this demonstration.
5.5 Receptor Arrays
Coarse and fine grid receptors grids are used to evaluate potential impacts. The
dense grid is a Cartesian system that covers of 8 km by 8 km in area centered at the
proposed project location. Receptors begin at the project boundary. Receptor spacing
from the project boundary is specified as follows:

Inner grid = 25 m spacing out to a distance of 200 m;

Second grid = 50 m spacing out to a distance of 400 m;

Third grid = 100 m spacing to 0.5 km;

Fourth grid = 500 m spacing out to a distance of 4 km;

Outer grid = 1,000 m spacing out to a distance of 8 km.
The coarse grid also includes a polar coordinate grid extending out to 24 km from the
center of the project location. Grid radials are spaced every ten degrees and rings are
placed at 1-km intervals beginning 2 km from the project location center. To ensure
that the receptor grid captured the maximum predicted 24-hour PM2.5 impact, the
above described grid was revised to extend the 100-meter spaced grid out to
approximately 2 kilometers.
Receptor elevations are assigned using the EPA’s AERMAP software tool (version
11103), which is designed to extract elevations from USGS National Elevation Dataset
data at 1 degree (approximately 90 m) resolution in GeoTIFF format (USGS 2002).
While 7.5-minute DEM data would be preferable for better resolution, these data are
10
Energy Answers
PSD Air Quality
Modeling Analysis
Revised for Startup
not available for Puerto Rico. The one degree datum is acceptable internationally and
adequately captures changes in elevation such as the mountainous region southwest
of the subject site.
5.6 Good Engineering Practice Stack Height and Building Downwash
Section 123 of the Clean Air Act, as amended, required the EPA to promulgate
regulations to assure that the degree of emission limitation required for the control of
any air pollutant under an applicable SIP is not affected by that portion of any stack
height which exceeds GEP or by any other dispersion technique. These regulations
have been promulgated under 40 CFR 51, dated July 8, 1985. A GEP stack height
analysis is required for new and existing air pollution sources subject to a modeling
analysis in order to determine if wake effect and downwash conditions need to be
accounted for in the dispersion modeling analysis. Building wake effects may cause
the predicted concentrations near a point source to be higher.
The formula for GEP stack height is given as:
HGEP = HB + 1.5LB where:



HGEP = formula GEP stack height;
HB = the building’s height above stack base; and
LB = the lesser of the building’s height or maximum projected width.
A second definition of GEP stack height is “regulatory” GEP stack height. Regulatory
GEP stack height is either 65 meters (m) or formula GEP stack height, whichever is
greater. Sources are not allowed to take credit for ambient air concentrations that
result from stacks that are higher than regulatory GEP stack height.
The EPA Building Profile Input Program (BPIP) (USEPA 1995) was used to evaluate
GEP stack height for each of the proposed stacks and to produce the model input
parameters necessary to account for building wake effects, based on the dimensions
of buildings in the vicinity of the stacks when the stack height is determined to be below
GEP. The “PRIME” version of BPIP (BPIPPRM) (Schulman et al. 1997) is used for
models such as AERMOD for calculating potential air quality impacts with the building
“cavity” region. BPIPPRM requires a digitized blueprint of the facility’s buildings and
stacks as well as other nearby structures. The position and height of buildings relative
to the stack positions must be evaluated in the GEP analysis. Coordinates for each
11
Energy Answers
PSD Air Quality
Modeling Analysis
Revised for Startup
building tier corner were identified using a digitized geo-referenced AutoCAD survey.
BPIPPRM input and output files are provided on the attached DVD.
The stack heights for all sources at the proposed facility are determined to be below
the GEP stack height. Table 5-1 provides a summary of the GEP analysis for the
sources included within this study. Therefore, building downwash effects are taken into
account in this dispersion modeling analysis.
Table 5-1: GEP Stack Height Values
Stack Height
Base Elevation
GEP Stack Height
Stack ID
(m)
Differences (m)
Value (m)
BOILER1
95.52
1
101.73
BOILER2
95.52
1
101.73
GEN
10
1
65
FIREPUMP
10
0.38
65
COOL1
10.7
1
65
COOL2
10.7
1
65
COOL3
10.7
1
65
COOL4
10.7
1
65
20
1
65
TRANS1
16.5
1
65
TRANS2
16.5
1
65
SILO1
13.1
1
65
SILO2
30.5
1
65
SILO4
38.1
1
65
ASH
5.7 Source Input Data
The air dispersion model program AERMOD requires the input of certain site-specific
data to produce results that are representative of the actual site conditions. These
data include stack coordinates, height, diameter, emission rates exit temperature and
exit flow rate. The primary sources of emissions at the new facility are the boiler units.
12
Energy Answers
PSD Air Quality
Modeling Analysis
Revised for Startup
The boiler emissions will be exhausted from a tall stack which contains two identical
flues (one for each of the two identical boilers). The two identical flues will be adjacent
to each other within an outer concrete shell. Table 5-2 provides a list of these data for
the maximum, (110% firing rate), average (100% firing rate), and minimum (80% firing
rate) operating scenarios. Note that the emission rates represent the worst case
emissions regardless of the fuel mix including the proposed supplemental fuels.
Emission rates from normal operating conditions remain unchanged from the October
2011 analysis PSD Air Quality Modeling Analysis (Revised PM10/PM2.5 Analysis).
Figure 2-3 shows the approximate location of each modeled emission point.
Whereas the flues were merged for the modeling analysis submitted in October 2011
for normal operations, each flue is modeled separately for the purposes of this
demonstration for startup periods. This was necessary since it is no longer appropriate
to merge flues when the exit velocities differ appreciably as they will during startup.
The following four startup conditions were identified and modeled:
1. One boiler in startup; The second boiler inactive (0% load)
2. One boiler in startup; The second boiler at minimum (80%) load.
3. Once boiler in startup; The second boiler at average (100%) load.
4. One boiler in startup; The second boiler at maximum (110%) load.
Emission rate data and stack flow and temperature data for the boiler startup phases
are provided in Table 5-3 and in Appendix A.
13
Energy Answers Arecibo
PSD Air Quality
Modeling Analysis
Table 5-2 Source Input Parameters – Normal Operations
Source ID
Vent #
Load
Stack
Height
(m)
Stack
Diameter
(m)
110%
Boilers 1 &
2
P-5
(each unit -
P-6
100%
95.52
2.13
80%
normal ops)
Revised for Startup
Exit
Velocity
(m/s)
Temperature
(K)
32.17
434.82
1.54 (a)
1.93(b)
1.413(a)
29.09
429.82
1.405 (a)
1.76(b)
22.35
424.82
1.13 (a)
1.41(b)
PM10
(g/s)
PM2.5
(g/s)
NOx
(g/s)
SO2
(g/s)
CO
(g/s)
1.93(b)
5.53
4.11
5.61
1.288(a)
1.76(b)
5.04
3.74
5.11
1.034(a)
1.41(b)
4.05
3.01
4.11
Gen
P-16
(c)
10
0.152
99.4
779
0.014(c)
0.014(c)
0.032(c)
1.85E-4(c)
0.243(c)
Firepump
P-17
(c)
10
0.152
49.2
708
0.007(c)
0.007(c)
0.016(c)
1.0E-4(c)
0.122(c)
Cool1
P-11
--
10.7
9.14
7.62
310.93
0.0412
1.44E-4
--
--
--
Cool2
P-12
--
10.7
9.14
7.62
310.93
0.0412
1.44E-4
--
--
--
Cool3
P-13
--
10.7
9.14
7.62
310.93
0.0412
1.44E-4
--
--
--
Cool4
P-14
--
10.7
9.14
7.62
310.93
0.0412
1.44E-4
--
--
--
Ash
P-15
--
20
1.52
15.52
310.93
4.79E-4
4.79E-4
--
--
--
Trans1
P-3
--
16.5
0.83
17.47
310.93
1.61E-4
1.61E-4
--
--
--
Trans2
P-4
--
16.5
0.83
17.47
310.93
1.61E-4
1.61E-4
--
--
--
Silo1
P-9
--
13.1
0.18
18.59
310.93
8.04E-6
8.04E-6
--
--
--
Silo2
P-7
--
30.5
0.18
18.59
310.93
8.04E-6
8.04E-6
--
--
--
Silo4
P-9
--
38.1
0.18
18.59
310.93
8.04E-6
8.04E-6
--
--
--
(a) Estimated emissions based on 24 mg/dscm for PM10 and 22 mg/dscm for PM2.5.
(b) Estimated emissions based on 30 mg/dscm.
(c) A 50% operating factor is applied to the emergency generator and fire water pump to reflect a 30 minute duration of routine equipment testing. For NO2 annual average emissions, an
operating factor of 500 hours per year is applied.
14
Energy Answers
PSD Air Quality
Modeling Analysis
Revised for Startup
Table 5-3 Source Input Parameters – Boiler Startup
Startup
Phase
Stack
Height
(m)
Stack
Diameter
(m)
Exit
Velocity
(m/s)
Temp
(K)
PM10
(g/s)
PM2.5
(g/s)
NOx
(g/s)
SO2
(g/s)
CO
(g/s)
1
95.52
2.13
1.72
366
0.0725
0.0489
3.09
0.00671
1.96
2
95.52
2.13
13.74
408
0.518
0.349
22.05
0.0479
14.02
3
95.52
2.13
21.93
416
0.828
0.558
35.28
0.0767
22.43
5.7.1 Model Setup – Source and Source Group Naming Convention
To represent conditions during the three phases of startup, three sources are specified
in the model called “B1SU1”, “B1SU2”, and “B1SU3”. Emissions from each phase of
startup are governed by the use of the EMISFACT feature, which provides the control
for when each phase begins and ends for the startup sequence described in Section
4.2.
For the preliminary SIL analysis, the source naming convention for each startup phase
includes a letter A-F to correspond to the six possible start times between 7:00 AM and
12:00 PM. In addition, the active boiler is identified as “B280”, “B2100”, and “B2110” to
represent the three loading conditions if the second boiler is active.
Twenty-four source groups are specified for each SIL model run, to represent the six
possible start times and three possible operating conditions of the active boiler plus a
fourth “inactive” condition of the second boiler. The naming convention for the source
groups follows the following pattern:
“SU” + letter A-F (corresponding to start time of 7AM-12PM) + 0/80/100/110 (corresponding to the
firing rate percentage of second boiler)
For example, the scenario representing the condition when one boiler is in startup
beginning at 7:00 AM while the second boiler is at 80 percent load has the source
group name “SUA80”. The same scenario but with a start time of 8:00 AM is called
“SUB80”, and so on through “SUF80” representing a start time of 12:00 PM.
15
Energy Answers
PSD Air Quality
Modeling Analysis
Revised for Startup
For the NAAQS analysis, separate model runs are provided for each time startup is
initiated. Separate runs are provided due to the addition of the offsite sources to the
model run and the convenience of specifying the “ALL” source group when specifying
the MAXDCONT and other output files. Therefore, six model runs are completed for
startup scenario; one for each startup beginning at 7:00 AM through 12:00 PM. The
four possible operational conditions for the second boiler are modeled, giving a total of
twenty-four model runs.
6.0 Model Results for Evaluating Significance
Following USEPA guidance (USEPA, 1990), a preliminary analysis was conducted to
determine if the emissions from the proposed facility during startup resulted in a
significant impact on ambient air quality. For each of the criteria pollutants subject to
PSD review (NO2, SO2, CO, PM10 and PM2.5), the proposed facility’s emissions during
startup were modeled using AERMOD. Modeling was completed for twenty-four
potential startup scenarios as described above. A time of day restriction is requested
for initiating the 7 hour startup, beginning between 7:00 AM and 12:00 PM. Table 6-1
provides maximum results for startup under the multiple scenarios. Results in Tables
6-1 are limited to the 1-hour and 8-hour CO, 1-hour NO2, 3-hour SO2, 24-hour SO2, 24hour PM10, and 24-hour PM2.5 due to the relatively short period that the boilers undergo
startup. Per the approved protocol, Energy Answers did not model 1-hour SO2 impacts
during startup and shutdown periods due to the statistical form of the standard and the
intermittency of startup conditions. As mentioned previously, this analysis, however,
includes a demonstration with respect to the 1-hour NO2 standard due to the change in
potential emissions during startup from the previous estimations.
16
Energy Answers
PSD Air Quality
Modeling Analysis
Revised for Startup
Table 6-1: Model Results - Significant Impact Levels Evaluation – Boiler Startup
Parameter
2nd Boiler
Operating
Level
Averaging
Period
Class II SIL
(µg/m3)
Maximum
Concentration
(µg/m3)
UTM
Northing
(meters)
UTM Easting
(meters)
Distance from
Stack
(meters)
110%
1
2000
126.5
742577.13
2042376.00
158
110%
8
500
28.0
742539.82
2042384.01
167
100%
1
2000
126.1
742577.13
2042376.00
158
100%
8
500
28.1
742539.82
2042384.01
167
80%
1
2000
124.7
742577.13
2042376.00
158
80%
8
500
28.3
742539.82
2042384.01
167
0%
1
2000
106.7
742577.13
2042376.00
158
0%
8
500
23.8
CO
110%
100%
PM10
80%
0%
110%
100%
PM2.5
80%
(a)
(b)
(c)
(d)
24
24
24
24
24
24
5
5
5
2042384.01
167
742402.13
2042601.0
239
(b, c)
742402.13
2042601.0
239
(b, c)
742402.13
2042601.0
239
2.65
2.65
2.65
(b, c)
5
2.65
742402.13
2042601.0
239
1.2
(b)
(c)
742658.29
2042987.81
463
(b)
(c)
742658.29
2042987.81
463
(b)
(c)
742596.19
2042949.99
426
(b)
(c)
1.2
1.2
1.14 (1.54 )
1.12 (1.51 )
1.04 (1.41 )
0%
24
1.2
0.47 (0.47 )
742477.13
2042501.00
154
110%
3
25
19.9
742602.13
2043051.00
526
110%
24
5
3.23
742658.29
2042987.81
463
100%
3
25
19.6
742602.13
2043051.00
526
100%
24
5
3.16
742658.29
2042987.81
463
80%
3
25
18.9
742596.19
2042949.99
426
80%
24
5
2.96
742596.19
2042949.99
426
0%
3
25
0.16
742552.13
2042401.00
147
0%
24
5
0.025
742539.82
2042384.01
167
110%
1
7.5
146
742577.13
2042376.00
158
100%
1
7.5
146
742577.13
2042376.00
158
80%
1
7.5
145
742577.13
2042376.00
158
0%
1
7.5
130.4
742577.13
Includes a 0.8 default ambient ratio per March 01, 2011 Modeling Guidance Memo..
Predicted impacts using estimated PM10/PM2.5 emissions based on 24/22 mg/dscm.
Predicted impacts using estimated PM10/PM2.5 emissions based on 30 mg/dscm.
Impacts for 1-hour SO2 are not required per approved modeling protocol
2042376.00
463
SO2 (d)
NO2 a
24
742539.82
(b, c)
17
Energy Answers
PSD Air Quality
Modeling Analysis
Revised for Startup
Since the maximum impacts for CO, SO2, PM10 and PM2.5 - 22 mg/dscm are predicted
to be below significance for all averaging times under all scenarios, no further analysis
was necessary. These emissions are not considered to cause or contribute to an
exceedance of an ambient air quality standard or PSD increment. A full, cumulative,
multisource analysis is not required for these pollutants and averaging times. Since
maximum impacts of NO2 on a 1-hour basis were found to exceed the SIL, an
additional full impact multi-source analysis is required. The full impact analysis for NO2
on a 1-hour average is discussed in the following sections. Note that the annual
averaging periods are not relevant when modeling startup conditions and, therefore,
are not evaluated as part of this demonstration. Furthermore, maximum impacts of
PM2.5 during startup are shown herein to be below those predicted during normal
3
operations with both units active (equal to 1.95 µg/m at 30 mg/dscm – see PSD Air
Quality Modeling Analysis (Revised PM10/PM2.5 Analysis) submitted October 2011).
Since startup conditions do not represent the worst-case for potential PM2.5 impacts
and considering that startup occurs intermittently, requiring less than 24 hours to
complete, and results in lower impacts than both units fully operational, a multisource
analysis is not included here for the startup emissions. The full multisource modeling
demonstration for PM2.5 is provided in the PSD Air Quality Modeling Analysis (Revised
PM10/PM2.5 Analysis) submitted in October 2011.
6.1 Identifying the Significant Impact Area (SIA)
Considering the probabilistic form of the standard and commentary provided in the
March 1, 2011 USEPA memo regarding intermittent emissions and the overly
conservative representation of intermittent emissions when modeling them as if they
occur every day over the 1-year period (as in this case), the average of the maximum
SIA distances determined for each startup scenario is offered as a practical yet
conservative measure for determining the SIA distance. As discussed in the
referenced memo, the over-estimation is due to the improbable circumstance that the
maximum emissions during the startup process occur on the worst-case
meteorological hour when in fact, the facility is restricted to only 32 startups per year for
both boilers combined. The calculated average SiA distance for this analysis is
approximately 10.5 km based on the distances for the maximum, or highest first-high,
1-hour impacts among the various load scenarios as determined from the preliminary
impact analysis. Note that the Ambient Ratio Method is applied for the SIA evaluation.
This maximum SIA distances the various startup scenarios were found to range
between 7 and 15 km.
18
Energy Answers
PSD Air Quality
Modeling Analysis
Revised for Startup
6.2 Full (Cumulative) Impact Analysis
A cumulative air modeling analysis was completed in accordance with EPA’s Guideline
on Air Quality Models (40 CFR 51 Appendix W) to demonstrate compliance with the 1hour NAAQS for NO2. This 1-hour cumulative modeling analysis is required following
the SIL evaluation described above in which potential concentrations of NO2 were
found to exceed the respective interim SIL on the 1-hour averaging period as shown in
Table 6-1. In the cumulative modeling analysis, emissions from existing off-site
sources and representative background concentrations are included to assess the
ambient impact at the receptor location within the SIA. The 8th highest daily 1-hour
maximum concentration at each receptor (98th percentile) was used for comparing the
impacts to the 1-hour NO2 NAAQS.
If the full impact analysis indicates a potential modeled exceedance, the determination
as to whether the proposed facility may potentially cause or contribute to this modeled
exceedance may be based on both spatial (at locations where the SIL is exceeded)
and temporal (at the time of a potential modeled exceedances in terms of year, month,
day, and hour) conditions. This is demonstrated (where necessary) by using the
MAXDCONT report generated by AERMOD.
6.2.1 Background Air Quality
Background air monitoring data must also be evaluated for the purposes of conducting
a cumulative (full) impact analysis for demonstrating that potential emissions do not
result in an exceedance of the NAAQS. Per USEPA recommendation and the
approved modeling protocol, the most recent three years of background data is
referenced for the 1-hour NO2 impacts. For the purposes of this analysis, a tiered
approach was followed in accordance with the recommendations made in the March 1,
1
2011 guidance memorandum (USEPA 2011). The following tiers were used for
developing a conservative representation of background concentrations for conducting
the cumulative 1-hour assessments (as described in the modeling protocol approved
by EPA):
Tier 1:
Maximum 1-hour value in recent 3 years;
1
The modeling protocol included an additional tier, but based on comments in the EPA approval letter of July
5,2011, only three tiers are included.
19
Energy Answers
PSD Air Quality
Modeling Analysis
Revised for Startup
Tier 2:
3 year average of the maximum 1-hour values in each year of the
most recent 3 years;
Tier 3:
3 year average of the 98 percentile of the daily maximum 1-hour
concentrations of NO2.
th
The tiered approach provides a mechanism for progressively evaluating ambient
concentrations using a simple conservative assumption (Tier 1) to a more data
intensive statistical computation (Tier 3). For this analysis, a background value of 65.2
3
µg/m is used for NO2 calculated from the most recent 3 year period (2005-07) from the
monitor in Catano (Monitor ID 72-033-0008) according to the Tier 2 approach. This
value is unchanged from the value used for the October 2011 analysis.
6.2.2 Off-Site Source Inventory
Per the EPA’s Draft New Source Review Workshop Manual (October 1990), the scope
of the off-site sources that must include in a cumulative impact analysis, starts by
defining the SIA. This was done in the process of completing the SIL evaluation
described above. Initial air dispersion modeling in the February 2011 PSD application
indicates that the predicted maximum impacts for NO2 that are equal to and greater
than the interim 1-hour SIL occurred out to a distance of approximately 11 km from the
site. As a result, major and minor facilities within this distance from the site were
identified and incorporated in the full impact analysis, and the major sources that are
located within an additional 50 km past the pollutant-specific SIA distance must be
evaluated.
The process of identifying potential off-site sources included in this analysis started by
consulting the PREQB Air Quality Division and USEPA Region 2. Energy Answers
reviewed permit files, including copies of the air permits and permit applications.
Energy Answers also coordinated with PREQB on obtaining necessary modeling input
data directly from some of the sources via data requests made by PREQB. In addition
to these efforts, the EPA’s Air Facility System and National Emissions Inventory
databases were searched for major sources in the modeling inventory area. The offsite source inventory is unchanged from the inventory used for the October 2011
analysis.
20
Energy Answers
PSD Air Quality
Modeling Analysis
Revised for Startup
6.2.3 AERSCREEN Concentration Gradient Evaluation for Sources to the South
The USEPA AERSCREEN model was executed for each of the major sources listed by
PREQB located on the south side of the central mountain range on the island. This
was done per the recommendation of EPA to provide further evidence supporting the
conclusion that the major sources located to the south of the central mountain range do
not have the potential to produce plumes with significant concentration gradients within
the SIA and, therefore, do not need to be included in the cumulative modeling analysis.
AERSCREEN uses a conservative set of meteorological conditions, actual stack
parameters and geographical location, and actual terrain elevation data surrounding
the source to approximate the plume characteristics. Stack data used for
AERSCREEN were collected as part of the off-site inventory data collection efforts.
Based on historical average temperature records for Puerto Rico, the minimum and
maximum temperatures used for AERSCREEN are 69 °F (294 K) and 88 °F (304 K).
AERSCREEN input and output files are included with the October 2011 modeling
analysis.
AERSURFACE was used to estimate the surface roughness coefficients, albedo, and
Bowen ratio around each source for input to AERSCREEN based on available NED
data for the island. (Although AERSURFACE was not used for the AERMOD
demonstration for the PSD ambient impact analysis due to the age of the available
surface data, it is sufficiently accurate for the purposes of this screening analysis.)
AERSCREEN was used to estimate the distance out from each of the sources that the
maximum air impact occurs and give a conservative indication to the general trend of
plume dispersion with distance. Unit emission rates were used at each source;
therefore, the resultant concentrations reported are relative values rather than absolute
values.
Table 6-2 below lists the distances of the maximum impact concentrations obtained
from AERSCREEN used in the October 2011 modeling analysis. This data indicates
that the facilities to the south do not have the potential to produce a plume with a
significant concentration gradient affecting the SIA of the proposed AREP. For each of
these four facilities evaluated, the maximum concentrations are estimated to
essentially “level out” before reaching the project study area. Thus, it is reasonable to
conclude that any measureable impact associated with these facilities is captured
within the background monitoring data or is insignificant. Please refer to the October
2011 modeling analysis for further details regarding the AERSCREEN evaluation of
these sources.
21
Energy Answers
PSD Air Quality
Modeling Analysis
Revised for Startup
Table 6-2: AERSCREEN Model Results for Sources Located to the South of the
Central Mountain Range
Source
Location
Distance to
Maximum
Concentration
(m)
Approximate
Distance to
Project Area
(m)
Cemex de Puerto
Rico, Inc.
Ponce
477
49,000
Destilleria Serralles
Ponce
1,376
51,200
Ecoelectrica LP
Penuelas
6,550
53,600
PREPA Costa Sur
Guayanilla
3,780
51,200
6.3 Evaluating 1-hour NO2 Cumulative Impacts
Multisource modeling was completed for all receptors used in the preliminary analysis.
The MAXDCONT utility is relied upon for determining whether the proposed AREP is a
3
significant contributor (i.e. contributing 7.5 µg/m or more) to the cumulative impact at
th
the times and locations of predicted exceedances. The 8 highest value is taken,
adjusted by a factor of 0.8 per the Tier 2 Ambient Ratio Method (ARM), and then added
to the background concentration. In executing the model, the adjustment per the ARM
3
3
was made by specifying a threshold value of [(188 μg/m – 65.2 μg/m ) ÷ 0.8 = ] 153.5
for the MAXDCONT report. As discussed in Section 6.2.1, the background value is
taken as the 3-year average of the maximum 1-hour values measured between 20052007 at the monitor in Catano, PR. A review of the MAXDCONT table indicates that
there are no modeled exceedances of the standard at the receptors and times when
the potential AREP impacts are significant. When exceedances are predicted to occur,
the proposed AREP is shown to have an insignificant contribution.
All model input and output files are provided on DVD in Appendix B. It should be
noted that the results reported in the MAXDCONT tables show exceedances at
different levels and locations than the October 2011 analysis because the receptor
field in this analysis was not limited to only those where Energy Answers is
significant. All receptors as described in Section 5.5 were included.
22
Energy Answers
PSD Air Quality
Modeling Analysis
Revised for Startup
7.0 Environmental Justice
Energy Answers prepared an Environmental Justice Evaluation for the proposed
AREP, which consolidates several analyses and public outreach efforts made in and
around the Arecibo area. This evaluation is supplied to EPA under separate cover at
the time this report is submitted. The Environmental Justice study was performed
following the EPA guidelines and definitions. The EPA defines the concept of
environmental justice as the fair treatment and meaningful involvement of all people
regardless of race, color, national origin, or income with respect to the development,
implementation, and enforcement of environmental laws, regulations, and policies.
The main purpose the analysis is to evaluate whether the community that the proposed
project will be located is an environmental justice community given its race and/or
origin or rather that the proposed community is considered economically
disadvantaged when compared to other areas.
Energy Answers has taken extensive measures related to Public Outreach, which are
described in the Environmental Justice Evaluation. Additionally, Energy Answers
prepared an environmental justice study as part of the Environmental Impact
Statement (EIS) for the development of the proposed AREP. These studies were
performed in compliance with the Environmental Quality Board, “Regulation for
presentation, evaluation, and procedures of environmental documents,” Regulation No.
6510.
The proposed AREP is located in Cambalache and the predicted maximum impacts
from the proposed AREP during startup are located in the immediate vicinity of the
facility (within 550 meters of the boiler stack – see Table 6-1). The findings of the
Environmental Justice Evaluation submitted to USEPA Region 2 in October 2011
indicate no disproportionate impacts are predicted to occur in the low-income barrios
around Arecibo. The findings of this evaluation are consistent with the conclusions
drawn from the October 2011 analysis. As further evidence, an additional model run
(representing a typical condition of 7:00 AM start time with the second boiler operating
at 100% load) was completed using the same receptor field used for the October 2011
analysis, and is included on the DVD in Appendix B.
23
Energy Answers
PSD Air Quality
Modeling Analysis
Revised for Startup
8.0 References
Auer, August H. Jr. 1978: “Correlation of Land Use and Cover with Meteorological
Anomalies.” Journal of Applied Meteorology, pp 636-643. 1978.
Alaska Department of Environmental Conservation (ADEC). 2009. “ADEC Guidance
re AERMET Geometric Means.” Revised April 7, 2009.
PREQB, 1993. “Source Specific Acidic Deposition. Impacts for Permit Applications,”
L. Sedefian. March 4.
PREQB. 1997. Policy DAR-1: Guidelines for the Control of Toxic Ambient Air
Contaminants. November.
PREQB. 2006. PREQB DAR-10: Guidelines on Dispersion Modeling Procedures for
Air Quality Impact Analyses. May.
Schulman, et al. 1997. “The PRIME Plume Rise and Building Downwash Model,”
Addendum to ISC3 User’s Guide. November.
United States Environmental Protection Agency (USEPA). 1980. “A Screening
Procedure for the Impacts of Air Pollution Sources on Plants, Soils and
Animals.” EPA 450/2-81-078. December 12.
USEPA. 1987. Ambient Monitoring Guidelines for Prevention of Significant
Deterioration, EPA-450/4-87-007. Revised May 1987. Research Triangle
Park, NC.
USEPA. 1990. Draft EPA NSR Workshop Manual: PSD and NonAttainment Area
Permitting Manual. October.
USEPA. 1995. User's Guide To The Building Profile Input Program. EPA-454/R-93038. Revised February 8, 1995.
USEPA. 1996. PCRAMMET User’s Guide. EPA-454/B-96-001. OAQPS, Research
Triangle Park, NC.
USEPA. 2000. Meteorological Monitoring Guidance for Regulatory Modeling
Applications. EPA-454/R-99-005. OAQPS, Research Triangle Park, NC.
24
Energy Answers
PSD Air Quality
Modeling Analysis
Revised for Startup
USEPA. 2004a. User's Guide for the AMS/EPA Regulatory Model – AERMOD. EPA454/B-03-001. September.
USEPA. 2004b. User's Guide For The AERMOD Terrain Preprocessor (AERMAP).
EPA-454/B-03-003. October.
USEPA. 2005. Guideline on Air Quality Models. November.
USEPA. 2008. AERSURFACE User’s Guide. EPA-454/B-08-001. OAQPS,
Research Triangle Park, NC.
USEPA. 2010a. Notice Regarding Modeling for New Hourly NO2 NAAQS. Office of
Air Quality Planning and Standards (OAQPS), Air Quality Modeling Group
(AQMG). February 25.
USEPA. 2010b. Modeling Procedures for Demonstrating Compliance with PM2.5
NAAQS. Office of Air Quality Planning and Standards (OAQPS).
Memorandum from Stephen D. Page to Regional Air Division Directors dated
March 23, 2010.
USEPA. 2010c. General Guidance for Implementing the 1-hour NO2 national Ambient
Air Quality Standard in Prevention of Significant Deterioration Permits,
Including an Interim 1-hour NO2 Significant Impact Level. Office of Air Quality
Planning and Standards (OAQPS). Memorandum from Anna Marie Wood to
Regional Air Division Directors dated June 28, 2010.
USEPA. 2010d. Applicability of Appendix W Modeling guidance for the 1-hour NO2
National Ambient Air Quality Standard. Office of Air Quality Planning and
Standards (OAQPS). Memorandum from Tyler Fox to Regional Air Division
Directors dated June 28, 2010.
USEPA. 2010e. Guidance Concerning Implementation of the 1-hour NO2 NAAQS for
the Prevention of Significant Deterioration Program. Office of Air Quality
Planning and Standards (OAQPS). Memorandum from Stephen D. Page to
Regional Air Division Directors dated June 29, 2010.
25
Energy Answers
PSD Air Quality
Modeling Analysis
Revised for Startup
USEPA. 2010f. Guidance Concerning the Implementation of the 1-hour SO2 NAAQS
for the Prevention of Significant Deterioration Program. Office of Air Quality
Planning and Standards (OAQPS). Memorandum from Stephen D. Page to
Regional Air Division Directors dated August 23, 2010.
USEPA. 2010g. General Guidance for Implementing the 1-hour SO2 National
Ambient Air Quality Standard in Prevention of Significant Deterioration
Permits, Including an Interim 1-hour SO2 Significant Impact Level. Office of
Air Quality Planning and Standards (OAQPS). Memorandum from Anna Marie
Wood to Regional Air Division Directors.
USEPA. 2010h. Applicability of Appendix W Modeling Guidance for the 1-hour SO2
National Ambient Air Quality Standard. Office of Air Quality Planning and
Standards (OAQPS). Memorandum from Tyler Fox to Regional Air Division
Directors dated August 23, 2010.
USEPA. 2011. Additional Clarification Regarding Application of Appendix W Modeling
Guidance for the 1-hour NO2 National Ambient Air Quality Standard. Office of
Air Quality Planning and Standards (OAQPS). Memorandum from Tyler Fox
to Regional Air Division Directors dated March 1, 2011.
United States Geological Survey (USGS). 2002. The National Map – Elevation, Fact
Sheet 106-02. http://egsc.usgs.gov/isb/pubs/factsheets/fs10602.html, U.S.
Department of the Interior. November.
26
Figures
I
AT L A N T I C O C E A N
SITE LOCATION
Dorado
Isabela
Aguadilla
Camuy
Hatillo
Loíza
#
*
Vega Baja
Manatí
Toa Baja
Arecibo
San Juan
Fajardo
Carolina
Moca
Luquillo
Toa Alta
Aguada
Rincón
San Sebastián
Morovis
Lares
Añasco
Río Grande
P
u
Ciales
Utuado
e
r
t
Las Marías
Jayuya
Corozal
o
R
i
Comerío
Orocovis
o
Juncos
Ceiba
Naguabo
Cidra
Adjuntas
San Germán
Humacao
San Lorenzo
Aibonito
Villalba
Cayey
Coamo
Yauco
Yabucoa
Peñuelas
Cabo Rojo
c
Caguas
Mayagüez
Maricao
Gurabo
Naranjito
Ponce
Patillas
Juana Díaz
Salinas
Lajas
Maunabo
Guayama
Santa Isabel
0
10
20
Miles
PROJECT NUMBER:
CITY:NOVI DIV/GROUP:ENV
DB:
PIC:
PM:
TM:
TR:
Arroyo
Guánica
ENERGY ANSWERS INTERNATIONAL, INC.
ARECIBO, PUERTO RICO
PROJECT LOCATION MAP
FIGURE
2-1
I
TM:
TR:
SITE LOCATION
0.4
0.8
PM:
0
PROJECT NUMBER:
CITY:NOVI DIV/GROUP:ENV
DB:
PIC:
Miles
SITE LOCATION
ENERGY ANSWERS INTERNATIONAL, INC.
ARECIBO, PUERTO RICO
#
*
SITE LOCATION MAP
P
P uu ee rr tt oo
R
R ii cc oo
FIGURE
2-2
I
Oceano
Oceano Atlantico
Atlantico
La
La Boca
Boca
CAMBALACHE, ARECIBO (18.471553; -66.701673)
Puerto
Puerto Arecibo
Arecibo
Arecibo
Corcovado
Cano
Cano Tiburones
Tiburones
Canal
Canal Perdomo
Perdomo
!
Tierras Nuevas Poniente
Garrochales
SITE (18.460300; -66.701606)
Arecibo
Barceloneta
Laguna
Laguna Tortuguero
Tortuguero
Barceloneta
La Luisa
Animas
Tiburones Imbéry
Coto Norte
Rio
Rio Cibuco
Cibuco
Ceiba
Monserrate
Vega Alta
Búfalo
Rio
Rio Tanama
Tanama
Rio
Rio Grande
Grande de
de Arecibo
Arecibo
San Antonio
Rio
Rio Indio
Indio
Rio
Rio Grande
Grande de
de Manati
Manati
Bajadero
Dorado
Sabana
Vega Baja
Vega Baja
Manatí
Sabana Hoyos
Rafael Capó
Breñas
Puerto
Puerto de
de Tortuguero
Tortuguero
!
San José Candelaria
Río Lajas
Mucarabones
Toa Alta
La Alianza
Miranda
Pajonal
San Juan NHS
Ensenada
Ensenada de
de Boca
Boca Vieja
Vieja
Cano
Cano de
de San
San Antonio
Antonio
Rio
Calle
Rio de
de Bayamon
Bayamon
Calle Hortensia
Hortensia
Canal
Levittown
Laguna
Canal Levittown
Laguna del
del Condado
Condado
Laguna
Laguna la
la Torrecilla
Torrecilla
Canal
Canal Bahia
Canal Hondo
HondoCanal
Bahia de
de San
San Juan
Juan
Ingenio
Cataño
Levittown
Laguna
Luis Munoz
Munoz Marin
Marin Intl
Intl
Laguna Los
Los Corozas
CorozasLuis
Toa Baja
Laguna
Cano
Laguna de
de Pinones
Pinones
!
Cano Campanero
Campanero
SAN JUAN INTERNATIONAL AIRPORT
Canal
Canal Suarez
Suarez
Sabana Seca
(18.434270; -66.001576)
Campanilla
Canal
Canal Blasina
Blasina
Candelaria Arenas
Canal
Laguna
Canal Puerto
Puerto Nuevo
Nuevo
Laguna San
San Jose
Jose
Rio
Rio Cocal
Cocal
Carolina
San Juan
Santa Bárbara
San Juan
Bayamón
Bayamon
Rio
Rio Grande
Grande de
de Loisa
Loisa
Rio
Rio Guaynabo
Guaynabo Guaynabo
Galateo
Florida
Barahona
H. Rivera Colón
Corozal Corozal
Fránquez
Trujillo Alto
Pájaros
Presa
Presa
Rio
Rio de
de la
la Plata
Plata
PROJECT NUMBER: NCENRGY.0003.0006
CITY:NOVI DIV/GROUP:ENV DB: PIC: PM: TM: TR:
G:\GIS\Project Files\EnergyAnswers\ProjectLocations.mxd
Ciales
Lago
Lago Dos
Dos Bocas
Bocas
Morovis
Lago
Lago Loisa
Loisa
Rio
Rio Limon
Limon
Rio
Rio Caonillas
Caonillas
0
Naranjito
Cayuco
6
12
SCALE IN KILOMETERS
Utuado
Rio Vivi
Vivi
Utuado Rio
Rio
Rio Gurabo
Gurabo
Celada
Aguas Buenas ENERGY ANSWERS INTERNATIONAL, INC.
Gurabo
Bairoa
Lago
Lago Caonillas
Caonillas
ARECIBO, PUERTO RICO
Caguas
Palomas
Orocovis
Orocovis
SITE AND SURFACE
OBSERVATION
Caguas
STATIONS
LOCATION
MAP
Rio
Rio Turabo
Turabo
Comerio Comerío
Jayuya
Juncos
Santa Clara
FIGURE
Rio
Rio Pellejas
Pellejas
Rio
Rio Cidra
Cidra
Lago
Lago de
de Matrullas
Matrullas
Barranquitas
Lago
Lago de
de Cidra
Cidra
Cidra
San Lorenzo
5-1
Appendix A
Emission Rate Calculations
APPENDIX A
ENERGY ANSWERS ARECIBO
Potential Emissions Calculations
During Startup - Phase 1
Firing No. 2 Fuel Oil
Boiler startup procedures require a gradual ramp up in heat levels at approximately the following rate:
Time Elapsed
0 to 4.5 hours:
4.5 - 5.5 hours:
5.5 - 7 hours:
Average Heat Input
35 MMBTU/hr
250 MMBTU/hr
400 MMBTU/hr
% Load
7
50
80
Flow (ACFM)
13,034
104,119
166,126
(DSCFM)
10,430
59,350
99,610
Temp (F)
200
275
290
Startup is completed by the end of the 7th hour.
Startup Phase 1:
No. 2 Fuel Oil Heating Value:
Fuel Use Rate - 7% load:
Pollutant
PM
PM10
PM10
PM10
PM2.5
PM2.5
PM2.5
SO2
NOx
VOC
CO
Filterable
Filterable
Condensable
Total
Filterable
Condensable
Total
Ammonia Slip - 10 ppmv @ 7%O2 - 0- 4.5 hr
140000 BTU/gal
250 Gal/hour
Emission
Factor
lb/1000 gal
lb/hr
2.0
1.0
1.3
2.3
0.25
1.3
1.55
0.213
98
0.2
62.3
---
0.5
0.25
0.325
0.575
0.0625
0.325
0.388
0.053
24.5
0.05
15.6
---
6.30E-02
3.15E-02
4.10E-02
7.25E-02
7.88E-03
4.10E-02
4.88E-02
6.71E-03
3.09E+00
6.30E-03
1.96E+00
0.03
5.60E-04
2.75E-03
4.20E-04
4.20E-04
4.20E-04
8.17E-04
3.73E-02
4.80E-02
1.26E-03
8.40E-04
4.20E-04
2.36E-04
3.33E-04
4.20E-04
3.30E-03
2.10E-03
7.97E-02
1.40E-03
1.40E-04
6.88E-04
1.05E-04
1.05E-04
1.05E-04
2.04E-04
9.33E-03
1.20E-02
3.15E-04
2.10E-04
1.05E-04
5.90E-05
8.33E-05
1.05E-04
8.25E-04
5.25E-04
1.99E-02
3.50E-04
1.76E-05
8.66E-05
1.32E-05
1.32E-05
1.32E-05
2.57E-05
1.17E-03
1.51E-03
3.97E-05
2.65E-05
1.32E-05
7.43E-06
1.05E-05
1.32E-05
1.04E-04
6.62E-05
2.51E-03
4.41E-05
Emission Rate
g/s
mg/dscm
0.3624
0.1812
0.236
0.417
0.04530
0.236
0.281
---------
ppmvd
--------------0.48
305
0.65
319
HAP
Arsenic
Benzene
Beryllium
Cadmium
Chromium
Ethylbenzene
Fluoride
Formaldehyde
Lead
Manganese
Mercury
Methyl Chloroform
Naphthalene
Nickel
POM
Selenium Compounds
Toluene
Xylenes
Notes:
1) Emission factors taken from AP-42 "Compilation of Air Pollutant Emission Factors", 5th edition, Tables 1.3-1 and 1.3-2.
2) Sulfur content = 15 ppmw
3) Emission factor for NOx and CO based on vendor data equivalent to 0.7 lb/MMBTU for NOx and 0.445 lb/MMBTU for CO.
APPENDIX A
ENERGY ANSWERS ARECIBO
Potential Emissions Calculations
During Startup - Phase 2
Firing No. 2 Fuel Oil
Boiler startup procedures require a gradual ramp up in heat levels at approximately the following rate:
Time Elapsed
0 to 4.5 hours:
4.5 - 5.5 hours:
5.5 - 7 hours:
Average Heat Input
35 MMBTU/hr
250 MMBTU/hr
400 MMBTU/hr
% Load
7
50
80
Flow (ACFM)
13,034
104,119
166,126
(DSCFM)
10,430
59,350
99,610
Temp (F)
200
275
290
Startup is completed by the end of the 7th hour.
Startup Phase 2:
No. 2 Fuel Oil Heating Value:
Fuel Use Rate - 50% load:
Pollutant
PM
PM10
PM10
PM10
PM2.5
PM2.5
PM2.5
SO2
NOx
VOC
CO
Filterable
Filterable
Condensable
Total
Filterable
Condensable
Total
Ammonia Slip - 10 ppmv @ 7%O2 - 4.5- 5.5 hr
140000 BTU/gal
1786 Gal/hour
Emission
Factor
lb/1000 gal
lb/hr
2.0
1.0
1.3
2.3
0.25
1.3
1.55
0.213
98
0.2
62.3
---
3.57
1.79
2.32
4.11
0.45
2.32
2.77
0.38
175
0.36
111
---
4.50E-01
2.25E-01
2.93E-01
5.18E-01
5.63E-02
2.93E-01
3.49E-01
4.79E-02
2.21E+01
4.50E-02
1.40E+01
0.196
5.60E-04
2.75E-03
4.20E-04
4.20E-04
4.20E-04
8.17E-04
3.73E-02
4.80E-02
1.26E-03
8.40E-04
4.20E-04
2.36E-04
3.33E-04
4.20E-04
3.30E-03
2.10E-03
7.97E-02
1.40E-03
1.00E-03
4.91E-03
7.50E-04
7.50E-04
7.50E-04
1.46E-03
6.66E-02
8.57E-02
2.25E-03
1.50E-03
7.50E-04
4.21E-04
5.95E-04
7.50E-04
5.89E-03
3.75E-03
1.42E-01
2.50E-03
1.26E-04
6.19E-04
9.45E-05
9.45E-05
9.45E-05
1.84E-04
8.39E-03
1.08E-02
2.84E-04
1.89E-04
9.45E-05
5.31E-05
7.49E-05
9.45E-05
7.43E-04
4.73E-04
1.79E-02
3.15E-04
Emission Rate
g/s
mg/dscm
0.4549
0.2275
0.296
0.523
0.05687
0.296
0.353
---------
ppmvd
--------------0.60
383
0.8
400
HAP
Arsenic
Benzene
Beryllium
Cadmium
Chromium
Ethylbenzene
Fluoride
Formaldehyde
Lead
Manganese
Mercury
Methyl Chloroform
Naphthalene
Nickel
POM
Selenium Compounds
Toluene
Xylenes
Notes:
1) Emission factors taken from AP-42 "Compilation of Air Pollutant Emission Factors", 5th edition, Tables 1.3-1 and 1.3-2.
2) Sulfur content = 15 ppmw
3) Emission factor for NOx and CO based on vendor data equivalent to 0.7 lb/MMBTU for NOx and 0.445 lb/MMBTU for CO.
APPENDIX A
ENERGY ANSWERS ARECIBO
Potential Emissions Calculations
During Startup - Phase 3
Firing No. 2 Fuel Oil
Boiler startup procedures require a gradual ramp up in heat levels at approximately the following rate:
Time Elapsed
0 to 4.5 hours:
4.5 - 5.5 hours:
5.5 - 7 hours:
Average Heat Input
35 MMBTU/hr
250 MMBTU/hr
400 MMBTU/hr
% Load
7
50
80
Flow (ACFM)
13,034
104,119
166,126
(DSCFM)
10,430
59,350
99,610
Temp (F)
200
275
290
Startup is completed by the end of the 7th hour.
Startup Phase 3:
No. 2 Fuel Oil Heating Value:
Fuel Use Rate - 80% load:
Pollutant
140000 BTU/gal
2857 Gal/hour
Emission
Factor
lb/1000 gal
lb/hr
Emission Rate
g/s
mg/dscm
ppmvd
PM
Filterable
2.0
5.71
7.20E-01
0.4337
---
PM10
PM10
PM10
PM2.5
PM2.5
PM2.5
SO2
NOx
VOC
Filterable
1.0
1.3
2.3
0.25
1.3
1.55
0.213
98
0.2
2.86
3.71
6.57
0.71
3.71
4.43
0.61
280
0.57
3.60E-01
4.68E-01
8.28E-01
9.00E-02
4.68E-01
5.58E-01
7.67E-02
3.53E+01
7.20E-02
0.2168
0.282
0.499
0.05421
0.282
0.336
-------
------------0.6
365
0.78
62.3
---
178
---
2.24E+01
0.329
---
382
5.60E-04
2.75E-03
4.20E-04
4.20E-04
4.20E-04
8.17E-04
3.73E-02
4.80E-02
1.26E-03
8.40E-04
4.20E-04
2.36E-04
3.33E-04
4.20E-04
3.30E-03
2.10E-03
7.97E-02
1.40E-03
1.60E-03
7.86E-03
1.20E-03
1.20E-03
1.20E-03
2.34E-03
1.07E-01
1.37E-01
3.60E-03
2.40E-03
1.20E-03
6.74E-04
9.51E-04
1.20E-03
9.43E-03
6.00E-03
2.28E-01
4.00E-03
2.02E-04
9.90E-04
1.51E-04
1.51E-04
1.51E-04
2.94E-04
1.34E-02
1.73E-02
4.54E-04
3.02E-04
1.51E-04
8.50E-05
1.20E-04
1.51E-04
1.19E-03
7.56E-04
2.87E-02
5.04E-04
Condensable
Total
Filterable
Condensable
Total
CO
Ammonia Slip - 10 ppmv @ 7%O2 - 5.5- 7.0 hr
HAP
Arsenic
Benzene
Beryllium
Cadmium
Chromium
Ethylbenzene
Fluoride
Formaldehyde
Lead
Manganese
Mercury
Methyl Chloroform
Naphthalene
Nickel
POM
Selenium Compounds
Toluene
Xylenes
Notes:
1) Emission factors taken from AP-42 "Compilation of Air Pollutant Emission Factors", 5th edition, Tables 1.3-1 and 1.3-2.
2) Sulfur content = 15 ppmw
3) Emission factor for NOx and CO based on vendor data equivalent to 0.7 lb/MMBTU for NOx and 0.445 lb/MMBTU for CO.
Appendix B
Air Modeling Files on DVD