Fine Particulate Matter (PM2.5) NAAQS

State Implementation Plan

For

The Tennessee Portion of the

Chattanooga, Tennessee-Georgia PM2.5 Nonattainment Area

Prepared by:

Chattanooga-Hamilton County Air Pollution Control Bureau

6125 Preservation Drive

Chattanooga, Tennessee  37416-3638

September 28, 2009

TABLE OF CONTENTS

List of Tables		iii

List of Figures		iv

List of Appendices	v

Acknowledgements	vi

Executive Summary	1

1.0	Introduction	2

	1.1	Fine Particulate Matter PM2.5	2

	1.2	PM2.5 National Ambient Air Quality Standard	3

	1.3 	Chattanooga, Tennessee-Georgia PM2.5 Nonattainment Area	3

	1.4	State Implementation Plan Requirements for PM2.5 Nonattainment
Areas	9

2.0	Attainment Modeling	10

	2.1	Emissions Inventory	10

		2.1.1  2002 Emissions Inventory Development	11

		2.1.2  Controls Applied	15

		2.1.3  2009 Emissions Inventory Development	17

	2.2	Model Selection	22

		2.2.1  Selection of Photochemical Grid Model	23

		2.2.2  Selection of Meteorological Model	24

		2.2.3  Selection of Emissions Processing System	25

		2.2.4  Selection of Modeling Year	27

	2.3	Modeling Domains	28

	2.4	Boundary Conditions	32

	2.5	Initial Conditions	32

	2.6	Model Performance Evaluation	32

3.0	Attainment Demonstration	34

	3.1	Speciated Modeled Attainment Test	34

	3.2	Attainment Test Results	35

	3.3	Supplemental Analyses/Weight of Evidence	36

		3.3.1  Emissions Reductions Not Modeled	37

		3.3.2  Status of Modeled Emissions Reductions	43

		3.3.3  Status of Attainment for Surrounding Counties	43

4.0	PM2.5 RACT/RACM	44

	4.1	Permitted Stationary Sources	45

		4.1.1  Local Regulations	46

		4.1.2  Sensitivity Modeling	47

	4.2	Gasoline Dispensing Facilities	49

	4.3	On-Road Mobile Sources	49

	4.4	Non-Road and Stationary Internal Combustion Engines	49

	4.5	Open Burning	50

	4.6	Home Heating with Wood	50

5.0	Motor Vehicle Emissions Budgets	51

	5.1	Transportation Conformity	51

	5.2	Transportation Budget for Hamilton County	51

	5.3	Pollutants to Be Considered	51

		5.3.1  Sulfur Dioxide (SO2)	51

		5.3.2  Nitrogen Oxides (NOX)	51

		5.3.3  Volatile Organic Compounds (VOCs)	52

		5.3.4  Ammonia	52

		5.3.5  Road and Construction Dust	52

6.0	Contingency Measures	53

7.0	Reasonable Further Progress	54

LIST OF TABLES

Table 1-1	Annual PM2.5 Design Values	4

Table 1-2	Components of Speciated PM2.5 Monitored at the Siskin
Drive/UTC Site	6

Table 2-1	Summary Emissions Inventory for 2002 for Hamilton County,
Tennessee	12

Table 2-2	Summary Emissions Inventory for 2002 for Catoosa County,
Georgia	12

Table 2-3	Summary Emissions Inventory for 2002 for Walker County,
Georgia	13

Table 2-4	Summary Emissions Inventory for 2002 for Jackson County,
Alabama	13

Table 2-5	PM2.5 Emissions for 2002 by Emission Source Category	14

Table 2-6	Sulfur Dioxide (SO2) Emissions for 2002 by Emission Source
Category	14

Table 2-7	Nitrogen Oxides (NOX) Emissions for 2002 by Emission Source
Category	14

Table 2-8	Ammonia Emissions for 2002 by Emission Source Category	15

Table 2-9	Summary Emissions Inventory for 2009 for Hamilton County,
Tennessee	19

Table 2-10	Summary Emissions Inventory for 2009 for Catoosa County,
Georgia	19

Table 2-11	Summary Emissions Inventory for 2009 for Walker County,
Georgia	20

Table 2-12	Summary Emissions Inventory for 2009 for Jackson County,
Alabama	20

Table 2-13	PM2.5 Emissions for 2009 by Emission Source Category	21

Table 2-14	Sulfur Dioxide (SO2) Emissions for 2009 by Emission Source
Category	21

Table 2-15	Nitrogen Oxides (NOX) Emissions for 2009 by Emission Source
Category	21

Table 2-16	Ammonia Emissions for 2009 by Emission Source Category	22

Table 2-17	Vertical Layer Definition for MM5 and CMAQ	31

Table 2-18	MFE and MFB for the Siskin Drive/UTC FRM (470654002)	33

Table 3-1	2009 Quarterly Mean and Annual PM2.5 Design Values	36

Table 3-2	2009 Modeled Species and PM2.5 Annual Design Values	36

Table 3-3	Surrounding Area 2009 Modeled Annual PM2.5 Design Values	43

Table 4-1	Maximum PM2.5 Concentration Reductions from Sensitivity
Modeling	48

LIST OF FIGURES

Figure 1-1	Locations of Federal Reference Monitors in the Chattanooga,
Tennessee-Georgia PM2.5 Nonattainment Area	5

Figure 1-2	Continuous PM2.5 Monitor Frontal Passage Trace	8

Figure 2-1	MM5 Horizontal Domain (Blue) with Nested CMAQ 36-km Domain	29

Figure 2-2	ASIP/VISTAS Region 12-km Grid	30

Figure 3-1	Hamilton County Emissions of PM2.5	39

Figure 3-2	Hamilton County Emissions of PM10	39

Figure 3-3	Hamilton County Emissions of Sulfur Dioxide (SO2)	40

Figure 3-4	Hamilton County Emissions of Nitrogen Oxides (NOX)	40

Figure 3-5	Hamilton County Emissions of Toluene	41

Figure 3-6	Hamilton County Emissions of Xylenes	41

Figure 3-7	Hamilton County Emissions of Ammonia	42

LIST OF APPENDICES

Appendix 1	PM2.5 Conceptual Description

Appendix 2	Draft Documentation of the Base G2 and Best & Final 2002 Base
Year, 2009 and 2018 Emission Inventories for VISTAS

Appendix 3	Methodology for Augmenting State/Local/Tribal 1999 Point
Source Emissions Inventories with PM10 and PM2.5 Emissions for the 1999
National Emissions Inventory

Appendix 4	VISTAS/ASIP Technical Support Documents

Appendix 5	Modeling Protocol for Association for Southeastern Integrated
Planning (ASIP)

Appendix 6	Application of GEOS CHEM for 2002 Boundary Conditions for
CMAQ Modeling of the VISTAS Region

Appendix 7	Local Area Modeling Performance

Appendix 8	Hamilton County SMAT Process	

Appendix 9	Letter from Bowater Newsprint – Calhoun Operations

Appendix 10	Chattanooga-Hamilton County Air Pollution Control Industrial
Point Source Emissions

Appendix 11	EPA NEI PM2.5 Document

Appendix 12	RACT Industrial Point Source Review

Appendix 13	House Heating Fuel – 2000 U.S. Census

Appendix 14	Planning Assumptions for Development of 2009 CHCNGA TPO
TransCAD Model 2009 Emissions Results

Appendix 15	NOX Insignificance Demonstration

ACKNOWLEDGEMENTS:  VISTAS CONTRACTORS

VISTAS Technical Coordinator, Pat Brewer, NOX Insignificance Evaluation

Air Resource Specialists:  Monitoring Data Analysis

Alpine Geophysics:  Emissions AND Air Quality Modeling, Technical
Advisor for Emissions Inventory

Atmospheric Research and Analysis, Inc.:  Operation of continuous
monitors - Millbrook, NC

Baron Applied Meteorology:  Meteorological Modeling

Desert Research Institute:  Carbon Source Attribution including sample
filter preparation, sample analysis using Gas Chromatography and Mass
Spectrometry, source attribution using Chemical Mass Balance analyses
and Positive Matrix Factorization

Earth Tech, Inc:  BART Modeling, CALPUFF Training

ENVIRON:  Emissions and Air Quality Modeling

Georgia Institute of Technology:  Emissions and Air Quality modeling
sensitivities

Harvard University:  GEOS-Chem global model

ICF Consulting:  Integrated Planning Model for future electric utility
generation

MACTEC:  2002, 2009, and 2018 Emissions Inventory and Projections,
CALPUFF training

E.H. Pechan and Associates:  2002 Mobile Inventory

Research Triangle Institute:  Chemical Analysis of Monitoring Samples

System Applications International:  Meteorological Characterization 

Tennessee Valley Authority:  Operation of continuous monitors at Great
Smoky Mountain National Park and preparation of final project report and
draft manuscript 

TRC, Inc:  BART Modeling

University of California, Riverside:  Emissions and Air Quality Modeling

Woods Hole:  Analysis of Carbon 14 isotope in carbon samples as part of
carbon source attribution project

EXECUTIVE SUMMARY

eded the 15.0 μg/m3 National Ambient Air Quality Standard (NAAQS).  The
Chattanooga-Hamilton County Air Pollution Control Bureau, the Tennessee
Division of Air Pollution Control, the Georgia Air Protection Branch,
and the Alabama Department of Environmental Management have collaborated
to develop a PM2.5 NAAQS State Implementation Plan (SIP) for each
jurisdiction.  This document contains the SIP for the Hamilton County,
Tennessee portion of the nonattainment area.

Air quality modeling conducted for Hamilton County by the Association
for Southeastern Integrated Planning (ASIP) indicates that the
nonattainment area will attain the annual NAAQS in 2009.  The model
results are based upon projected 2002-2009 emissions reductions from
sources within the area and are supported by downward trends in actual
emissions.  Hamilton County expects to attain the PM2.5 NAAQS based upon
implementation of Federal and regional measures.  The 2009 modeled PM2.5
design values are provided below.

2009 Modeled Annual PM2.5 Design Values

Federal Reference Monitor	PM2.5 Design Value (µg/m3)

Siskin Drive/UTC (470654002)	13.6

Maxwell Road/East Ridge (470650031)	14.4

Soddy-Daisy High School (470651011)	12.3

Walker County, Georgia (132950002)	13.9



As required by the EPA Clean Air Fine Particle Implementation Rule,
sources of PM2.5 and its precursors within Hamilton County were
evaluated to verify that they were meeting at least reasonably available
control measures (RACM) levels of emission controls.  This evaluation
and the results of sensitivity modeling that was performed indicate that
no additional reductions are available from local sources that would
result in advancement of the attainment date to earlier than 2009.

1.0  INTRODUCTION

1.1  Fine Particulate Matter (PM2.5)

Fine particulate matter, also known as fine particles and PM2.5, refers
to airborne particles less than or equal to 2.5 micrometers (μm) in
diameter.  Fine particles are treated as though they are a single
pollutant, but they come from many different sources and are composed of
many different compounds.  PM2.5 exposure adversely affects human
health, especially respiratory and cardiovascular systems.  Individuals
particularly sensitive to PM2.5 exposure include children, people with
heart and lung disease, and older adults.

PM2.5 can be liquid, solid, or can have a solid core surrounded by
liquid.  PM2.5 can include material produced by combustion,
photochemical reactions, and can contain salt from sea spray and soil
like particles.  Particles are distinguished based on the method of
formation.  Primary particles are particles directly emitted into the
atmosphere and retain the same chemical composition as when they were
released.  Secondary particles are those formed through chemical
reactions involving atmospheric oxygen, water vapor, hydroxyl radical,
nitrates, sulfur dioxide (SO2), nitrogen oxides (NOX), and organic gases
from natural and anthropogenic sources.  PM2.5 can therefore be composed
of varying amount of different species, including:

Sulfates

Nitrates (usually found in the form of ammonium nitrate)

Ammonium

Hydrogen ion

Particle bound water

Elemental carbon

Organic compounds

Primary organic species (from cooking and combustion)

Secondary organic compounds

Crustal material (includes calcium, aluminum, silicon, magnesium, and
iron)

Sea salt (generally only found at coastal monitoring sites)

Transitional metals

Potassium (generally from wood burning or cooking)

The most significant sources of PM2.5 and its precursors are coal-fired
power plants, industrial boilers, and other combustion sources.  These
emissions are often transported over large distances.  Other sources of
PM2.5 emissions include mobile sources, area sources, biogenic sources,
fires, wind blown dust, and oceans.

A variety of meteorological and geographic factors influences the
concentration levels of fine particles, including both the regional and
local distribution of urbanized areas, primary and precursor emissions
sources, and natural features such as oceans and forests.  Since PM2.5
concentrations can exceed the National Ambient Air Quality Standards
(NAAQS) at any time of the year, the United States Environmental
Protection Agency (EPA) mandates the year round monitoring of the
pollutant throughout the country (40 CFR 58, App. D).

1.2  PM2.5 National Ambient Air Quality Standard

μg/m3) or less.  The annual standard is met when the 3-year average of
a monitoring site's annual mean concentration is 15.0μg/m3 or less. 
The 3-year averages at each monitoring site (of annual means or 98th
percentiles) are called design values.  The design value for an entire
area is the single highest design value among the monitoring sites
located in the area.

Since the 1977 amendments to the Clean Air Act (CAA), areas of the
country that violated the ambient standard for a particular pollutant
were formally designated as nonattainment for that pollutant.  This
formal designation concept was retained in the 1990 Amendments (CAAA). 
With the implementation of the PM2.5 standard, areas designated
nonattainment under Section 172 of the CAAA (Subpart 1) have five years
from designation to attain the standard.

1.3  Chattanooga, Tennessee-Georgia PM2.5 Nonattainment Area

The EPA does not plan to designate geographic areas as attainment or
nonattainment for the 2006 24-hour PM2.5 standard until December 2008 or
later.  Therefore, this State Implementation Plan (SIP) does not address
attainment and/or nonattainment issues with the 24-hour standard.

On January 5, 2005, EPA published the annual PM2.5 nonattainment
designation for the Chattanooga, Tennessee-Georgia area.  This area
consists of Hamilton County in Tennessee, Catoosa and Walker Counties in
Georgia, and the portion of Jackson County in Alabama that is described
as U.S. Census 2000 block group identifier 01–071-9503–1.  Although
Jackson County is not contiguous to the rest of the nonattainment area,
a coal-fired power plant located in this census tract is a major
contributor to the PM2.5 problem.  The effective date of the
nonattainment designation was April 5, 2005, and was based on air
quality monitoring for the years 2001, 2002, and 2003.

Four PM2.5 federal reference monitors are located in the nonattainment
area – three in Hamilton County and one in Walker County.  Federal
reference monitors (FRM) collect samples that are used to determine if
an area is in attainment with the PM2.5 NAAQS.  A list of the monitors
and associated design values is provided below in Table 1-1.  The FRM
located at Soddy-Daisy High School is at the most rural site of the
three Hamilton County FRMs and has always shown attainment.  The other
three monitors show exceedances of the annual standard for the 2001-2003
averaging period, which is the period the EPA used in the designation of
the nonattainment area.  The locations of the four FRMs in the
nonattainment area are shown in Figure 1-1, which follows.

Table 1-1.  Annual PM2.5 Design Values

μg/m3)

	2001-2003*	2002-2004	2003-2005	2004-2006

Siskin Drive/UTC (470654002)	15.2	14.7	15.0	15.1

Maxwell Road/East Ridge (470650031)	16.1	15.7	16.1	15.5

Soddy-Daisy High School (470651011)	14.1	13.8	13.8	13.5

Walker County, Georgia (132950002)	15.5	15.2	15.8	15.2

*Averaging period used by the EPA in the nonattainment area designation

Figure 1-1.  Locations of Federal Reference Monitors in the

Chattanooga, Tennessee-Georgia PM2.5 Nonattainment Area

As mentioned in Section 1.1 of this document, PM2.5 is composed of many
species from varying sources.  In order to fully understand the nature
of the problem in Hamilton County and to ascertain an appropriate
response, the PM2.5 components must be determined.  Fine particulate
speciation monitors are collocated with the FRMs at the Siskin Drive/UTC
and Walker County, Georgia, sites.  The speciation monitor located at
the Siskin Drive/UTC site is part of the EPA Speciation Trends Network
(STN).  Speciation monitors provide concentrations of total PM2.5,
ammonium, potassium, sodium, nitrates, sulfates, elemental carbon,
organic carbon, and forty-seven metals.  Major components of PM2.5 from
the Siskin Drive/UTC speciation monitor are shown below in Table 1-2.

Table 1-2.  Components of Speciated PM2.5 Monitored at the Siskin
Drive/UTC Site

Year	Concentration (µg/m3)

	Total PM2.5	Organic Carbon	Sulfates	Nitrates	Ammonium	Elemental Carbon

2002	16.90	5.34	4.99	0.95	1.65	0.66

2003	17.55	5.46	5.11	1.26	1.85	0.75

2004	15.26	4.40	4.85	1.18	1.63	0.65

2005	17.84	4.94	5.39	1.10	1.88	0.68



Air emissions from both regional and local sources contribute to the
nonattainment problem in Hamilton County.  The two major components of
the speciated total PM2.5 are organic carbon and sulfates.  Organic
carbon makes up approximately thirty percent of the speciated PM2.5. 
The organic carbon is composed of both primary PM2.5, such as direct
emissions from engines, and secondary PM2.5 that results from
atmospheric reactions of precursors.  The organic carbon from diesel
engines includes unburned fuel and engine lubrication oil.  Many diesel
engines vent crankcase emissions to the atmosphere.  Two cycle engines
have lubricating oil mixed with their fuel resulting in emissions of
unburned lubricating oil.  Aromatic compounds, biogenic compounds, and
organic products of incomplete combustion are examples of secondary
PM2.5 precursors.  Aromatic compounds include toluene, xylenes, and
trimethylbenzenes that are emitted from industrial sources and form
secondary organic aerosols.  Biogenic compounds include terpenes and
sesquiterpenes that are emitted by trees and other vegetation.  As shown
by the EPA’s Biogenic Emission Inventory System (BEIS3.09) model,
organic carbon from biogenic precursors accounts for a significant
portion of the organic carbon in the nonattainment area between March
and December.  Forest fires, prescribed burning, and home heating with
wood are all sources of organic products of incomplete combustion. 
Prescribed burns in Hamilton County are normally conducted in the spring
and fall.  Significant home heating in this area is performed for five
months of the year.

Sulfates are formed from atmospheric reactions of SO2 and also make up
approximately thirty percent of the total PM2.5 at the monitor.  Some of
these reactions also involve ammonia.  The primary sources of SO2 in
Hamilton County include industrial combustion equipment and mobile
sources.  Most industrial sources in Hamilton County use natural gas as
the primary fuel.  Since natural gas contains a negligible amount of
sulfur, minimal SO2 emissions result from its combustion.  When natural
gas is curtailed, some industrial sources use fuel oil as a backup. 
Number 2 fuel oil typically contains no more than 0.2% sulfur, but
Number 6 fuel oil could contain up to two percent sulfur.  There is also
some industrial use of coal as a fuel in Hamilton County.  The
combustion of Number 2 and Number 6 fuel oils and coal results in
increasingly higher levels of SO2 emissions than those from natural gas.
 However, the SO2 emissions from combined industrial and mobile sources
within Hamilton County are significantly less than those from regional
coal-fired electric utilities, none of which are located in the county. 
In 2002, the combined SO2 emissions from all sources within Hamilton
County were less than ten percent of the SO2 emissions from the TVA
Widows Creek Fossil Plant in Jackson County, Alabama, which is the
closest of four regional coal-fired power plants located within 75 miles
of the county.

Nitrates, which make up less than ten percent of the speciated PM2.5,
are formed from atmospheric reactions of NOX with ammonia.  Combustion
processes are the primary source of NOX emissions both inside and
outside the county.  These processes include on-road mobile sources;
non-road mobile sources; space heating; industrial processes; forest
fires; and burning of vegetation.  Regional coal-fired electric
generation plants located outside the county also contribute NOX
emissions.

The ammonia that reacts with NOX and SO2 to form ammonium nitrate and
ammonium sulfate, respectively, is produced primarily from automobile
catalysts, sewage treatment, and agriculture sources both locally and
regionally.  Elemental carbon, or soot, makes up a small portion of the
total PM2.5 and is usually formed by incomplete combustion of fuels such
as gasoline or diesel fuel.

Once PM2.5 is emitted to or formed in the atmosphere, it may remain
suspended for weeks, but is eventually removed by either dry or wet
deposition.  The concentration in the air is dependent on wind and
weather conditions.  If there is no wind, the PM2.5 remains in the
vicinity, and the concentration continues to rise until reaching a state
of equilibrium in which the rate of deposition is equal to the rate of
development of fine particulate.  When weather fronts pass through the
Hamilton County area, monitored increases in PM2.5 are observed followed
by rapid decreases with the frontal passage.  One example of this
phenomenon is charted in Figure 1-2, which follows.  The profile is for
a day during which the National Oceanic and Atmospheric Administration
weather records for February 13, 2007, indicate the passage of a front
between 10:53 AM and 11:53 AM.  The characteristic rapid drop in PM2.5
is shown beginning at that time by the green line labeled RIV, which is
the output of a continuous PM2.5 monitor located at the Siskin Drive/UTC
site.

Figure 1-2.  Continuous PM2.5 Monitor Frontal Passage Trace

A study performed for the Visibility Improvement State and Tribal
Association of the Southeast (VISTAS), Characterization of Meteorology
and its Relationship to Fine Particulate Mass and Visibility in the
VISTAS Region, was completed on December 8, 2006.  The analysis was
conducted over the VISTAS sites and included specific information for
Hamilton County.  The study showed that high PM2.5 concentrations in the
area are associated with the following:

Moderate temperatures

Low surface or aloft wind speeds

South surface winds

East aloft winds

Stable atmospheric conditions

Very high PM2.5 concentrations on the previous day

A local surface wind study using Chattanooga/Lovell Field data conducted
by the Southern Regional Climate Center showed calm winds 26.27% of the
time with a mean wind of 6.28 knots.  The most common direction for
surface winds was from the south.  This study used wind data from 1984
through 1992.

Appendix 1 provides further conceptual information on ambient PM2.5
concentrations in Hamilton County.

1.4  State Implementation Plan Requirements for PM2.5 Nonattainment
Areas

The April 5, 2007, EPA Clean Air Fine Particle Implementation Rule
(PM2.5 Implementation Rule) requires that a State Implementation Plan
(SIP) be developed and submitted to the EPA no later than three years
from the date of nonattainment designation.  A SIP documents how and
when an area will be brought into attainment with a national air quality
standard.  There are several required elements in the PM2.5 SIP.  These
include an emissions inventory, a modeled attainment demonstration, an
analysis of reasonably available control technologies (RACT) and
reasonably available control measures (RACM), and the adoption of the
RACT/RACM levels of emission controls if the addition of those controls
could move up the attainment date by a year.

The PM2.5 Implementation Rule establishes a hierarchy of precursor
pollutants for PM2.5: SO2 is always considered a precursor, NOX is
presumptively a precursor, and VOCs and ammonia are presumed not to be
precursors.  The rule allows States to exempt NOX by demonstrating that
NOX emissions are not a significant contributor to an area’s PM2.5
nonattainment problem and that the area is not a source of a PM2.5
transport problem.  

The rule also requires that contingency measures be included in the SIP
for areas designated nonattainment for PM2.5.  The contingency measures
are to be implemented in the event the nonattainment area does not meet
attainment by the date specified in the SIP.

2.0  ATTAINMENT MODELING

In 2001, the southeastern states formed the Visibility Improvement State
and Tribal Association of the Southeast (VISTAS) to address compliance
with the Regional Haze Rule that was published in 1999 (64 FR 35714). 
The member states are Alabama, Florida, Georgia, Kentucky, Mississippi,
North Carolina, South Carolina, Tennessee, Virginia, and West Virginia. 
VISTAS found that the modeling requirements for the Regional Haze Rule
were similar to the requirements for the modeled attainment
demonstration for PM2.5 nonattainment areas.  The member states formed a
new organization, the Association for Southeastern Integrated Planning
(ASIP), in order to conduct modeled attainment demonstrations for the
PM2.5 nonattainment areas.  VISTAS and ASIP used the same emissions
inventory and model.

As required by the EPA’s April 2007 Guidance on the Use of Models and
Other Analyses for Demonstrating Attainment Goals for Ozone, PM2.5, and
Regional haze (Attainment Modeling Guidance), modeling data files are
archived and will be maintained by Tennessee Division of Air Pollution
Control at 615-532-0554.

2.1  Emissions Inventory

An emissions inventory is a comprehensive quantification of all
emissions in a region that cause or contribute to a nonattainment area
failing to meet an air standard.  In modeled attainment demonstrations,
the emission inventory is developed for a past year so that emissions
can be quantified as accurately as possible.  For the annual PM2.5
standard, the EPA provided guidance for selecting the past (base) year. 
The base year was recommended to be 2002 as per a November 18, 2002,
letter titled “2002 Base Year Emission Inventory SIP Planning: 8-hr
Ozone, PM2.5, and Regional Haze Programs” signed by Lydia N. Wegman. 
Both National Emission Inventory (NEI) data and monitored PM2.5 values
were available for the 2002 base year.

VISTAS developed a regional emissions inventory, which included input
from each member state air agency as well as input from the local
agencies in Davidson County, TN, Hamilton County, TN, Shelby County, TN,
Knox County, TN, Jefferson County, AL, Jefferson County, KY, Buncombe
County, NC, Forsyth County, NC and Mecklenburg County, NC.  A detailed
discussion of the inventory development is provided in the Documentation
of the Base G 2002 Base Year, 2009 and 2018 Emission Inventories for
VISTAS located in Appendix 2.

There are five different emission inventory source classifications:
stationary point sources, area sources (including stationary area
sources), off-road and on-road mobile sources, and biogenic sources.  In
Hamilton County, stationary point sources have a fixed location and
require permits to operate unless they have been defined as
“insignificant activities” in the local air pollution control
regulations.  Industrial sources make up the bulk of stationary point
sources in the county, and their emissions are inventoried annually. 
Electric generating units (EGUs) are the single largest contributor to
PM2.5 from point source emissions in the region.  No EGUs are located in
Hamilton County.

Area sources in Hamilton County include stationary area sources, fires,
animal husbandry, and paved and unpaved roads.  Stationary area sources
are those sources whose individual emissions are relatively small, but
due to the large number of these sources, the collective emissions from
the source category could be significant.  For example, dry cleaners and
service stations are stationary area sources found in Hamilton County.

Non-road (or off-road) mobile sources are equipment that move but do not
use the roadways, such as construction equipment, aircraft, railroad
locomotives, lawn and garden equipment, etc.  On-road mobile sources are
automobiles, buses, trucks, and motorcycles that use the roadway system.
 Emissions from on-road sources are estimated by vehicle type and road
type.  Biogenic sources are natural sources like trees, crops, grasses,
and the natural decay of plants.  Emissions from non-road, on-road, and
biogenic sources are estimated county wide.

2.1.1  2002 Emissions Inventory Development

The starting point of the 2002 emissions inventory was EPA’s 1999 NEI.
 Several agencies provided more recent inventories.  Selected large
ammonia sources were updated, 2002 continuous emissions monitoring based
SO2 and NOX emissions were added for electric utilities, emissions of
PM2.5 and PM10 (particulate matter less than or equal to 10 μm in
diameter) were added if not in the submittal, and quality
assurance/quality control checks were performed.

The “Base G” emissions inventory was released in August 2006 and,
where appropriate, incorporated 2002 NEI submittals (including Hamilton
County’s NEI submittal).  However it was still necessary, in
finalizing the inventory, to augment stationary point source emissions
in accordance with Appendix C of EPA’s Documentation of the Final 1999
National Emissions Inventory (Version 3) for Criteria Pollutants and
Ammonia – Point Sources, which can be found in Appendix 3.  For EGUs,
normalization factors were developed for producing a 2002 typical year
inventory.  The normalization factor is derived by averaging the 2000 to
2004 heat input and then dividing the result by the actual 2002 heat
input.  If the unit did not operate all five years, then the average
heat input was only calculated for the years it operated.  The final
version of the emissions inventory is Base G4, which was used for the
modeled attainment demonstration.

Stationary area source emissions, which were submitted to comply with
the Consolidated Emissions Reporting Rule (CERR), were obtained from the
EPA.  The emissions were evaluated and missing data added.  A
“typical” fire year inventory was developed covering wildfires,
prescribed burning, agricultural fires, and land clearing fires. 
Ammonia emissions from livestock and fertilizers were calculated using
Version 3.6 of the Carnegie Mellon University ammonia model.  Paved and
unpaved road emissions were estimated using EPA approved methodology.

Most non-road emissions were estimated using the EPA’s NONROAD2005c
model.  Aircraft engines, railroad locomotives, and commercial marine
emissions were not included in the NONROAD model and were estimated
separately.  These “other non-road” emissions were based on
emissions estimates developed for the EPA’s 1999 NEI Version 2. 
Mobile6.2 was used to estimate emissions from on-road mobile sources. 
Biogenic emissions were estimated using the BEIS3.09 model.  Tables 2-1
through 2-8 summarize the emissions inventory used in the 2002 model. 
This data reflects best data available with some data from modeling,
some calculated and some CEM data. 

Table 2-1.  Summary Emissions Inventory for 2002 for Hamilton County,
Tennessee

Table 2-2.  Summary Emissions Inventory for 2002 for Catoosa County,
Georgia

Table 2-3.  Summary Emissions Inventory for 2002 for Walker County,
Georgia

Table 2-4.  Summary Emissions Inventory for 2002 for Jackson County,
Alabama

Table 2-5.  PM2.5 Emissions for 2002 by Emission Source Category

County	PM2.5 Emissions (tons)

	Point	Area	Non-road	On-road	Total

Hamilton, TN	567	1,000	292	183	2,042

Catoosa, GA	0	675	37	37	750

Walker, GA	2	920	19	29	970

Jackson, AL	963	1,209	88	41	2,301

Total	1,532	3,804	437	291	6,064



Table 2-6.  Sulfur Dioxide (SO2) Emissions for 2002 by Emission Source
Category

County	Sulfur Dioxide (SO2) Emissions (tons)

	Point	Area	Non-road	On-road	Total

Hamilton, TN	1,721	507	539	461	3,228

Catoosa, GA	0	272	51	92	415

Walker, GA	203	761	21	71	1,056

Jackson, AL	45,357	576	167	99	46,199

Total	47,281	2,115	778	723	50,898



Table 2-7.  Nitrogen Oxides (NOX) Emissions for 2002 by Emission Source
Category

County	Nitrogen Oxides (NOX) Emissions (tons)

	Point	Area	Non-road	On-road	Total

Hamilton, TN	2,856	639	6,428	11,610	21,532

Catoosa, GA	0	162	671	2,377	3,211

Walker, GA	46	384	269	1,699	2,398

Jackson, AL	26,862	310	1,418	2,195	30,785

Total	29,763	1,495	8,787	17,881	57,926

Table 2-8.  Ammonia Emissions for 2002 by Emission Source Category

County	Ammonia Emissions (tons)

	Point	Area	Non-road	On-road	Total

Hamilton, TN	17	172	2.1	357	549

Catoosa, GA	0	468	0.2	74	542

Walker, GA	0	763	0.3	56	819

Jackson, AL	2	1,280	0.6	77	1,361

Total	20	2,684	3.2	563	3,270



2.1.2  Controls Applied

Several control measures already in place or being implemented over the
next few years will reduce stationary point, highway mobile, and
non-road mobile sources emissions.  The Federal control measures that
have impacts on air quality in Hamilton County were incorporated into
the Base G4 emissions inventory and are discussed in the following
sections.

Tier 2 Vehicle Standards

Federal Tier 2 vehicle standards will require all passenger vehicles in
a manufacturer’s fleet, including light-duty trucks and sport utility
vehicles (SUVs), to meet an average standard of 0.07 grams of NOX per
mile.  Implementation began in 2004, and was to be completely phased in
by 2007.  The Tier 2 standards will also cover passenger vehicles over
8,500 pounds gross vehicle weight rating (the larger pickup trucks and
SUVs), which are not covered by the current Tier 1 regulations.  For
these vehicles, the standards will be phased in beginning in 2008, with
full compliance in 2009.  The new standards require vehicles to be
seventy-seven percent to ninety-five percent cleaner than those on the
road today.  The Tier 2 rule also reduced the sulfur content of gasoline
to thirty parts per million (ppm) starting in January of 2006.  Most
gasoline sold in Hamilton County prior to January 2006 had a sulfur
content of about 300 ppm.  Sulfur occurs naturally in gasoline, but
interferes with the operation of catalytic converters on vehicles
resulting in higher NOX emissions.  Lower-sulfur gasoline is necessary
to achieve the Tier 2 vehicle emission standards.

Heavy-Duty Diesel Highway Vehicles Standards

New EPA standards designed to reduce NOX and volatile organic compound
(VOC) emissions from heavy-duty gasoline and diesel highway vehicles
began to take effect in 2004.  A second phase of standards and testing
procedures, which began in 2007, will reduce particulate matter
emissions from heavy-duty highway engines and will also reduce highway
diesel fuel sulfur content to fifteen ppm since the sulfur damages the
particulate filters that are required to meet the new standard.  The
total program is expected to achieve a ninety percent reduction in
particulate matter emissions and a ninety-five percent reduction in NOX
emissions for these new engines using low-sulfur diesel fuel, compared
to existing engines using higher-sulfur content diesel fuel.

Large Non-road Diesel Engines Rule

In May 2004, the EPA promulgated new rules for large non-road diesel
engines, such as those used in construction, agricultural, and
industrial equipment, to be phased in between 2008 and 2014.  The
non-road diesel rules also reduce the allowable sulfur in non-road
diesel fuel by over ninety-nine percent.  Non-road diesel fuel currently
averages 3,400 ppm sulfur.  The rule limits non-road diesel sulfur
content to 500 ppm in 2006 and 15 ppm in 2010.  The combined engine and
fuel rules would reduce NOX and particulate matter emissions from large
non-road diesel engines by over ninety percent, compared to current
engines using higher-sulfur content diesel fuel.

Non-road Spark-Ignition Engines and Recreational Engines Standard

The new standard, effective in July 2003, regulates emissions of NOX,
hydrocarbons, and carbon monoxide for groups of previously unregulated
non-road engines.  The new standard will apply to all new engines sold
in the United States and imported after these standards began and will
apply to large spark-ignition engines (forklifts and airport ground
service equipment), recreational vehicles (off-highway motorcycles and
all-terrain-vehicles), and recreational marine diesel engines.  The
regulation varies based upon the type of engine or vehicle.

The large spark-ignition engines contribute to ozone formation and
ambient carbon monoxide and particulate matter levels in urban areas. 
Tier 1 of this standard was implemented in 2004 and Tier 2 started in
2007.  Like the large spark-ignition, recreational vehicles contribute
to ozone formation and ambient carbon monoxide and particulate matter
levels.  For the model year 2006 off-highway motorcycles and
all-terrain-vehicles, the new exhaust emissions standard was phased-in
by fifty percent and for model years 2007 and late, by one hundred
percent.  Recreational marine diesel engines over thirty-seven kilowatts
are used in yachts, cruisers, and other types of pleasure craft. 
Recreational marine engines contribute to ozone formation and
particulate matter levels, especially in marinas.  Depending on the size
of the engine, the standard began phasing-in in 2006.  When all of the
non-road spark-ignition engines and recreational engines standards are
fully implemented, an overall seventy-two percent reduction in emissions
of hydrocarbons, eighty percent reduction in NOX emissions, and
fifty-six percent reduction in carbon monoxide emissions are expected by
2020.  These controls will help reduce ambient concentrations of ozone,
carbon monoxide, and PM2.5.

NOX SIP Call

In October 1998, the EPA made a finding of significant contribution of
NOX emissions from certain states and published a rule that set ozone
season NOX budgets for the purpose of reducing regional transport of
ozone (63 FR 57356).  This rule, referred to as the “NOX SIP Call,”
called for ozone season controls to be put on utility and industrial
boilers, as well as internal combustion engines in twenty-two states in
the eastern United States.  A NOX emissions budget was set for each
state, and the states were required to develop rules that would allow
the state to meet their budget.  The emission budgets were to be met by
the beginning of 2004.  The NOX SIP Call resulted in regional reductions
pf PM2.5 and PM2.5 precursors that reduced PM2.5 in this nonattainment
area.  

Clean Air Interstate Rule

On May 12, 2005, the EPA promulgated the “Rule to Reduce Interstate
Transport of Fine Particulate Matter and Ozone (Clean Air Interstate
Rule); Revisions to Acid Rain Program; Revisions to the NOX SIP Call,”
referred to as CAIR.  This rule established the requirement for States
to adopt rules limiting NOX and SO2 emissions and a model rule for the
states to use in developing their rules.  The purpose of the CAIR is to
reduce interstate transport of precursors to fine particulate and ozone.
 The CAIR applies to certain large fossil-fuel fired boilers and
turbines that serve electricity generators.

This rule provides annual state caps for NOX and SO2 in two phases, with
the Phase I caps for NOX and SO2 starting in 2009 and 2010,
respectively.  Phase II caps become effective in 2015.  The EPA is
allowing the caps to be met through a cap and trade program if a state
chooses to participate in the program.  When fully implemented, the CAIR
will reduce SO2 emissions in the eastern United States by over seventy
percent and NOX emissions by over sixty percent from 2003 levels.

Though the rule was remanded to EPA by the U.S. Court of Appeals for the
District of Columbia, TVA is continuing to operate their controls set
forth in the CAIR rule pending a new rule being developed.  These
controls have resulted in regional reductions of both nitrates and
sulfates in the area.

2.1.3  2009 Emissions Inventory Development

A regional emissions inventory was developed for the 2009 modeling. 
Projections for EGUs, non-EGU point sources, area sources, non-road, and
on-road mobile sources were used to produce a 2009 emission inventory. 
Future EGU emissions were estimated using the EPA Integrated Planning
Model, which is used to examine air pollution control policies for
various pollutants for the entire electric power system throughout the
contiguous United States.  The general approach for future year non-EGU
point source emissions projections was to use growth and control data
consistent with the data used in EPA’s Clean Air Interstate Rule
analysis, and supplement these data with stakeholder input.  The
following activities were performed for non-EGU point sources.

Obtained, reviewed, and applied the most current growth factors
developed by EPA

Obtained, reviewed and applied any state-specific, or sector specific
growth factors submitted by stakeholders

Obtained and incorporated information on sources that shutdown after
2002 and set their emissions to zero in the projection inventories

Obtained, reviewed, and applied control assumptions for programs that
are presently adopted and controls that have future compliance dates. 
Since the modeling dealt with controls that are presently required to be
operated and controls that will be operated in a future year as the
future year modeling was conducted for 2009 and 2018.

For area source categories, except fires, VISTAS/ASIP generated future
emissions were based on the 2002 inventory year.  Growth factors
supplied by the states or the CAIR emission projections were applied to
project the controlled emissions to the appropriate year.  In some
cases, the EPA’s Economic Growth and Analysis System Version 5 growth
factors were used if no growth factor was available from either the
states or the CAIR growth factor files.  For fires, the “typical”
year inventory developed in 2002 was used.

For source categories with emissions estimated using the EPA’s NONROAD
model, the model growth assumptions were used to create the 2009 future
year inventory.  The NONROAD model takes into consideration regulations
affecting emissions from these categories.  For the commercial marine,
railroad locomotives, and airport emissions, VISTAS calculated the
future growth in emissions using detailed inventory data (both before
and after controls) for 1996 and 2010, obtained from the CAIR Technical
Support Document.  When available, state-specific growth factors were
used.

MOBILE6.2, which takes into consideration regulations that affect
emissions from this source category, was used to estimate emissions for
on-road mobile source emissions in the 2009 inventory.  The vehicle
miles traveled (VMT) input from the MOBILE6.2 2002 base year inventory
provided the basis for the initial 2009 VMT input.  The 2009 VMT
estimate assumed a linear growth rate for each state, county, and
vehicle type as derived from the VMT data assembled by the EPA for their
most recent heavy-duty diesel rulemaking.  In the final 2009 VMT input
estimate for MOBILE6.2, several states provided independent forecast
data to either replace or augment the forecast data based on the
heavy-duty diesel rule.  Tables 2-9 through 2-16 summarize the emissions
inventory that was used for the 2009 model.

Table 2-9.  Summary Emissions Inventory for 2009 for Hamilton County,
Tennessee

Table 2-10.  Summary Emissions Inventory for 2009 for Catoosa County,
Georgia



Table 2-11.  Summary Emissions Inventory for 2009 for Walker County,
Georgia

Table 2-12.  Summary Emissions Inventory for 2009 for Jackson County,
Alabama



Table 2-13.  PM2.5 Emissions for 2009 by Emission Source Category

County	PM2.5 Emissions (tons)

	Point	Area	Non-road	On-road	Total

Hamilton, TN	504	1,073	243	131	1,951

Catoosa, GA	0	731	33	27	792

Walker, GA	1	988	15	21	1,025

Jackson, AL	1,156	1,252	70	30	2,507

Total	1,661	4,044	360	209	6,274



Table 2-14.  Sulfur Dioxide (SO2) Emissions for 2009 by Emission Source
Category

County	Sulfur Dioxide (SO2) Emissions (tons)

	Point	Area	Non-road	On-road	Total

Hamilton, TN	1,700	541	223	54	2,517

Catoosa, GA	0	273	9	11	292

Walker, GA	195	762	4	8	969

Jackson, AL	46,578	532	79	12	47,200

Total	48,473	2,107	314	85	50,978



Table 2-15.  Nitrogen Oxides (NOX) Emissions for 2009 by Emission Source
Category

County	Nitrogen Oxides (NOX) Emissions (tons)

	Point	Area	Non-road	On-road	Total

Hamilton, TN	2,949	645	5,602	7,514	16,710

Catoosa, GA	0	169	584	1,621	2,374

Walker, GA	46	398	200	1,194	1,838

Jackson, AL	5,663	317	1,287	1,437	8,704

Total	29,764	1,495	8,786	17,881	57,926





Table 2-16.  Ammonia Emissions for 2009 by Emission Source Category

County	Ammonia Emissions (tons)

	Point	Area	Non-road	On-road	Total

Hamilton, TN	17	173	2.4	415	608

Catoosa, GA	0	521	0.3	87	608

Walker, GA	0	844	0.3	66	910

Jackson, AL	8	1,408	0.6	88	1,504

Total	25	2,944	3.6	656	3,629



2.2  Model Selection

To ensure that a modeling study is defensible, care must be taken in the
selection of the models to be used.  The models selected must be
“scientifically appropriate” for the intended application and be
“freely accessible” to all stakeholders.  Scientifically appropriate
means that the models address important physical and chemical phenomena
in sufficient detail, using peer-reviewed methods.  Freely accessible
means that model formulations and coding are freely available for review
and that the models are available to stakeholders, and their
consultants, for execution and verification at no or low cost.

The following models were selected by ASIP for use in PM2.5 modeling.

Air Quality Model:  The EPA’s Models-3/Community Multiscale Air
Quality (CMAQ) modeling system is a “one-atmosphere” photochemical
grid model capable of addressing ozone, particulate matter, visibility,
and acid deposition at regional scale for periods up to one year.

Meteorological Model:  The Pennsylvania State University/National Center
for Atmospheric Research Mesoscale Meteorological Model (MM5) is a
nonhydrostatic, prognostic meteorological model routinely used for
urban-scale and regional-scale photochemical, fine particulate matter,
and regional haze regulatory modeling studies.

Emissions Model:  The Sparse Matrix Operator Kernel Emissions (SMOKE)
modeling system is an emissions modeling system that generates hourly
gridded speciated emission inputs of mobile, non-road mobile, area,
point, fire and biogenic emission sources for photochemical grid models.

The following sections outline the criteria for selecting a modeling
system that is both defensible and capable of meeting the study's goals.
 These criteria were used in selecting the modeling system used for this
modeling attainment demonstration.

2.2.1  Selection of Photochemical Grid Model

Criteria

For a photochemical grid model to qualify as a candidate for use in a
SIP, a State needs to show that it meets the following general criteria
for a NAAQS attainment demonstration.

The model has received a scientific peer review.

The model can be demonstrated applicable to the problem on a theoretical
basis.

Data bases needed to perform the analysis are available and adequate.

Available past appropriate performance evaluations have shown the model
is not biased toward underestimates or overestimates.

A protocol on methods and procedures to be followed has been
established.

The developer of the model must be willing to make the source code
available to users for free or for a reasonable cost, and the model
cannot otherwise be proprietary.

Overview of CMAQ

The photochemical model selected for this study was CMAQ Version 4.5. 
During the course of the modeling, VISTAS updated the secondary organic
aerosol module in CMAQ to include important processes missing in the
standard CMAQ model.  Secondary organic aerosol module improvements are
detailed in Section 1.3.3.3.1 of the Technical Support Document for
VISTAS Emissions and Air Quality Modeling to Support Fine Particulate
State Implementation Plans (Technical Support Document for VISTAS),
which can be found in Appendix 4.  For more than a decade, the EPA has
been developing the Models-3 CMAQ modeling system with the overarching
aim of producing a “One-Atmosphere” air quality modeling system
capable of addressing ozone, fine particulate matter, visibility and
acid deposition within a common platform.  The original justification
for the Models-3 development emerged from the challenges posed by the
1990 CAAA and the EPA’s desire to develop an advanced modeling
framework for “holistic” environmental modeling utilizing
state-of-science representations of atmospheric processes in a high
performance computing environment.  The EPA completed the initial stage
of development with Models-3 and released the CMAQ model in mid-1999 as
the initial operating science model under the Models-3 framework.  The
most recent rendition is CMAQ version 4.4, which was released in October
2004.

Another reason for choosing CMAQ as the atmospheric model is the ability
to do one-atmospheric modeling.  Since the same modeling exercise is
required for the ozone and PM2.5 attainment demonstrations SIPs, as well
as the regional haze SIP, having a model that can handle both ozone and
particulate matter is essential.  A number of features in the
theoretical formulation and technical implementation of CMAQ make the
model well suited for annual particulate matter modeling.

The configuration used for this modeling demonstration, as well as a
more detailed description of the CMAQ model, can be found in the VISTAS
Modeling Protocol (Appendix 5).

2.2.2  Selection of Meteorological Model

Criteria

Meteorological models, either through objective, diagnostic, or
prognostic analysis, extend available information about the state of the
atmosphere to the grid upon which photochemical grid modeling is to be
carried out.  The criteria for selecting a meteorological model are
based on both the models ability to accurately replicate important
meteorological phenomena in the region of study, and the model's ability
to interface with the rest of the modeling systems, particularly the air
quality model.  With these issues in mind, the following criteria were
established for the meteorological model to be used in this study.

Non-hydrostatic formulation

Reasonably current, peer reviewed formulation

Simulates cloud physics

Publicly available at no or low cost

Output available in I/O API format

Supports four dimensional data assimilation

Enhanced treatment of planetary boundary layer heights for air quality
modeling

Overview of MM5

The non-hydrostatic MM5 model is a three-dimensional, limited-area,
primitive equation, prognostic model that has been used widely in
regional air quality model applications.  The basic model has been under
continuous development, improvement, testing and open peer-review for
more than 20 years and has been used worldwide by hundreds of scientists
for a variety of mesoscale studies.

MM5 uses a terrain-following non-dimensionalized pressure, or
“sigma,” vertical coordinate similar to that used in many
operational and research models.  In the non-hydrostatic MM5, the sigma
levels are defined according to the initial hydrostatically-balanced
reference state so that the sigma levels are also time-invariant.  The
gridded meteorological fields produced by MM5 are directly compatible
with the input requirements of “one atmosphere” air-quality models
using this coordinate.  MM5 fields can be easily used in other regional
air quality models with different coordinate systems by performing a
vertical interpolation, followed by a mass-conservation re-adjustment. 
Distinct planetary boundary-layer parameterizations are available for
air-quality applications, both of which represent sub-grid-scale
turbulent fluxes of heat, moisture, and momentum.  One scheme uses a
first-order eddy diffusivity formulation for stable and neutral
environments and a modified first-order scheme for unstable regimes. 
The other scheme uses a prognostic equation for the second-order
turbulent kinetic energy, while diagnosing the other key boundary layer
terms.

Initial and lateral boundary conditions are specified for real-data
cases from mesoscale three-dimensional analyses performed at 12-hour
intervals on the outermost grid mesh selected by the user.  Surface
fields are analyzed at three-hour intervals.  A Cressman-based technique
is used to analyze standard surface and radiosonde observations, using
the National Meteorological Center’s spectral analysis, as a first
guess.  The lateral boundary data are introduced using a relaxation
technique applied in the outermost five rows and columns of the coarsest
grid domain.  The configuration used for this modeling demonstration, as
well as a more detailed description of the MM5 model, can be found in
the modeling protocol (Appendix 5).

2.2.3  Selection of Emissions Processing System

Criteria

The principal criterion for an emissions processing system is that it
accurately prepares emissions files in a format suitable for the
photochemical grid model being used.  The following list includes
clarification of this criterion and additional desirable criteria for
effective use of the system.

File system compatibility with the I/O API

File portability

Ability to grid emissions on a Lambert conformal projection

Report capability

Graphical analysis capability

MOBILE6 mobile source emissions

Biogenic Emissions Inventory System (BEIS3) model

Ability to process emissions for the proposed domain in a reasonable
amount of time

Ability to process control strategies

No or low cost for acquisition and maintenance

Expandable to support other species and mechanisms

Overview of SMOKE

The Sparse Matrix Operator Kernel Emissions (SMOKE) Processing System
Prototype was originally developed at the Micro-computing Center of
North Carolina.  As with most emissions models, SMOKE is principally an
emission processing system and not a true emissions modeling system in
which emissions estimates are simulated from “first principles.” 
This means that, with the exception of mobile and biogenic sources, its
purpose is to provide an efficient, modern tool for converting emissions
inventory data into the formatted emission files required by an air
quality simulation model.  For mobile sources, SMOKE actually simulates
emissions rates based on input mobile-source activity data, emission
factors, and outputs from transportation travel-demand models.

SMOKE was originally designed to allow emissions data processing methods
to utilize emergent high-performance-computing as applied to
sparse-matrix algorithms.  Indeed, SMOKE is the fastest emissions
processing tool currently available to the air quality modeling
community.  The sparse matrix approach utilized throughout SMOKE permits
both rapid and flexible processing of emissions data.  The processing is
rapid because SMOKE utilizes a series of matrix calculations instead of
less efficient algorithms used in previous systems.  The processing is
flexible because the processing steps of temporal projection, controls,
chemical speciation, temporal allocation, and spatial allocation have
been separated into independent operations wherever possible.  The
results from these steps are merged together at a final stage of
processing.

SMOKE contains a number of major features that make it an attractive
component of the modeling system.  The model supports a variety of input
formats from other emissions processing systems and models.  It supports
both gridded and county total land use scheme for biogenic emissions
modeling.  SMOKE can accommodate emissions files from up to 10 countries
and any pollutant can be processed by the system.  For additional
information about the SMOKE model please refer to the modeling protocol
(Appendix 5).

2.2.4  Selection of Modeling Year

A crucial step to SIP modeling is the selection of the period of time to
model to represent current air quality conditions and to project changes
in air quality in response to changes in emissions.  The year 2002 was
selected as the base year for several reasons.

The Attainment Modeling Guidance identifies specific goals to consider
when selecting one or more episodes for use in demonstrating reasonable
progress in attaining the regional haze air quality goals.  The EPA
recommends that episode selection derive from the following four
principal criteria.

Simulate a variety of meteorological conditions

Model time periods in which observed concentrations are close to the
appropriate baseline design value or visibility impairment

Model periods for which extensive air quality/meteorological data bases
exist

Model a sufficient number of days so that the modeled attainment test
applied at each monitor violating the NAAQS is based on multiple days

VISTAS adopted a logical, stepwise approach in implementing the
Attainment Modeling Guidance in order to identify the most preferable,
representative modeling year.  This approach included the following
steps.

Representativeness of Meteorological Conditions:  The VISTAS identified
important meteorological characteristics and data sets in the VISTAS
region directly relevant to the evaluation of candidate annual modeling
episodes.

Initial Episode Typing:  At the time of selection in 2003,
meteorological and air quality data were available for 2002 for model
inputs and model performance evaluation.  VISTAS used Classification and
Regression Tree (CART) analyses to evaluate the candidate modeling
years.  The year 2002 was found to be representative of conditions in
the other years.  Subsequently, these analyses were repeated with the
meteorological and air quality monitoring data for 2000 to 2004 to
evaluate how well the 2002 modeling year represented the full 2000-2004
baseline period.  This analysis confirmed that visibility and PM2.5
concentrations in 2002 were representative of the five-year baseline
period.  This analysis is discussed in more detail in Appendix 5.

Data Availability:  In parallel with the CART analyses, episode
characterization analyses, collaborative investigations by VISTAS states
(e.g., North Carolina Department of Air Quality, Georgia Department of
Natural Resources, and Florida Department of Environmental Protection)
intensively studied the availability of PM2.5, meteorological and
emissions data, and representativeness of alternative baseline modeling
periods from a regulatory standpoint.  Additionally, 2002 was the year
that the EPA was requiring states to provide emissions inventory data
for the Comprehensive Emissions Reporting Rule, it was appropriate to
use 2002 as the modeling year to take advantage of the 2002 inventory.

Years to be used by other Regional Planning Organizations:  VISTAS also
considered what years other regional planning organizations would be
modeling, and several had already chosen calendar year 2002 as the
modeling year.

After a lengthy process of integrated studies, the episode selection
process culminated in the selection of calendar year 2002 (January 1
through December 31) as the most current, representative, and pragmatic
choice for modeling.  All of the EPA criteria for model year selection
were directly considered in this process together with many other
considerations (e.g., timing of new emissions or aerometric data
deliveries by the EPA or the states to the modeling teams).

2.3  Modeling Domains

Horizontal Modeling Domain

The CMAQ model was run in the one-way nested grid mode.  This allowed
the larger outer domains to feed concentration data to the inner nested
domain.  One-way nesting is believed to be appropriate for the generally
stagnant or near stagnant conditions experienced in the region.  Two-way
nesting was not considered due to numerical and computational
uncertainty associated with the technique.

The horizontal coarse grid modeling domain boundaries were determined
through a national effort to develop a common grid projection and
boundary.  Since this national modeling domain was used in the VISTAS
regional haze modeling, it was used for the attainment demonstration as
well.  A smaller 12-km grid, modeling domain was selected in an attempt
to balance location of areas of interest, such as ozone and fine
particulate matter nonattainment areas, as well as Class 1 and
wilderness areas for regional haze.  Processing time was also a factor
in choosing a smaller 12-km grid, modeling domain.

The coarse 36-km horizontal grid domain covers the continental United
States.  This domain was used as the outer grid domain for MM5 modeling
with the CMAQ domain nested within the MM5 domain.  Figure 2-1 shows the
MM5 horizontal domain as the outer most, blue grid with the CMAQ 36-km
domain nested in the MM5 domain.

Figure 2-1.  MM5 Horizontal Domain (Blue) with Nested CMAQ 36-km Domain

To achieve finer spatial resolution in the VISTAS states, a one-way
nested high resolution (12-km grid resolution) was used.  Figure 2-2
shows the 12-km grid, modeling domain for the VISTAS region.  This is
the modeling domain for which the attainment test results are based. 
VISTAS conducted a study to determine if using a finer grid resolution
provided different modeling results.  Since the EPA attainment test uses
the modeling results to determine the relative reductions in PM2.5, it
was determined that effectively the same attainment test results are
obtained from 12-km grid modeling or 4-km grid modeling.  Since 4-km
grid modeling takes significantly more time and resources to run, the
ASIP decided to use the VISTAS 12-km grid modeling results for this
attainment demonstration.  This is in accordance with the Attainment
Modeling Guidance.  According to EPA, “areas without strong gradients
in primary particulate matter will likely have little benefit from fine
scale resolution.”  Primary particulate matter is not a significant
contributor to this area’s nonattainment problem as shown in the
monitored data.

Figure 2-2.  ASIP/VISTAS Region 12-km Grid

Vertical Modeling Domain

The CMAQ vertical structure is primarily defined by the vertical grid
used in the MM5 modeling.  The MM5 model employed a terrain following
coordinate system defined by pressure, using thirty-four layers that
extend from the surface to 100 mb.  Table 2-17, which follows, lists the
layer definitions for both MM5 and for CMAQ.  A layer-averaging scheme
is adopted for CMAQ to reduce the computational cost of the CMAQ
simulations.  The effects of layer averaging were evaluated in
conjunction with the VISTAS modeling effort and were found to have a
relatively minor effect on the model performance metrics when both the
thirty-four layer and nineteen layer CMAQ models were compared to
ambient monitoring data.

Table 2-17.  Vertical Layer Definition for MM5 and CMAQ

MM5 Layers	CMAQ Layers

Layer	Sigma	Pressure (mb)	Height (m)	Depth (m)	Layer	Sigma	Pressure (mb)
Height (m)	Depth (m)

34	0.000	100	14,662	1,841	19	0.000	100	14,662	6,536

33	0.050	145	12,822	1,466	–	0.050	145	–	–

32	0.100	190	11,356	1,228	–	0.100	190	–	–

31	0.150	235	10,127	1,062	–	0.150	235	–	–

30	0.200	280	9,066	939	–	0.200	280	–	–

29	0.250	325	8,127	843	18	0.250	325	8,127	2,966

28	0.300	370	7,284	767	–	0.300	370	–	–

27	0.350	415	6,517	704	–	0.350	415	–	–

26	0.400	460	5,812	652	–	0.400	460	–	–

25	0.450	505	5,160	607	17	0.450	505	5,160	1,712

24	0.500	550	4,553	569	–	0.500	550	–	–

23	0.550	595	3,984	536	–	0.550	595	–	–

22	0.600	640	3,448	506	16	0.600	640	3,448	986

21	0.650	685	2,942	480	–	0.650	685	–	–

20	0.700	730	2,462	367	15	0.700	730	2,462	633

19	0.740	766	2,095	266	–	0.740	766	–	–

18	0.770	793	1,828	259	14	0.770	793	1,828	428

17	0.800	820	1,569	169	–	0.800	820	–	–

16	0.820	838	1,400	166	13	0.820	838	1,400	329

15	0.840	856	1,235	163	–	0.840	856	–	–

14	0.860	874	1,071	160	12	0.860	874	1,071	160

13	0.880	892	911	158	11	0.880	892	911	158

12	0.900	910	753	78	10	0.900	910	753	155

11	0.910	919	675	77	–	0.910	919	–	–

10	0.920	928	598	77	9	0.920	928	598	153

9	0.930	937	521	76	–	0.930	937	–	–

8	0.940	946	445	76	8	0.940	946	445	76

7	0.950	955	369	75	7	0.950	955	369	75

6	0.960	964	294	74	6	0.960	964	294	74

5	0.970	973	220	74	5	0.970	973	220	74

4	0.980	982	146	37	4	0.980	982	146	37

3	0.985	986.5	109	37	3	0.985	986.5	109	37

2	0.990	991	73	36	2	0.990	991	73	36

1	0.995	995.5	36	36	1	0.995	995.5	36	36

0	1.000	1,000	0	0	0	1.000	1,000	0	0

2.4  Boundary Conditions

The GEOS-CHEM global chemical transport model, which is managed and
supported by the Atmospheric Chemistry Modeling Group at Harvard
University, was used to develop the boundary conditions of the modeling
domain.  Appendix 6 contains a report titled Application of GEOS CHEM
for 2002 Boundary Conditions for CMAQ Modeling of the VISTAS Region by
Ivar Tombach dated October 5, 2007, which details the process.

The following three full-year 2002 simulations were run.

A baseline simulation with best estimates of 2002 emissions

A background simulation modified from the baseline by shutting off U.S.
anthropogenic emissions which included fuel, industrial, and
agricultural sources but not biomass burning

A natural simulation modified by shutting off anthropogenic emissions
worldwide.

Three dimensional concentration fields with three-hour temporal
resolution were archived from each simulation to serve as boundary
conditions for CMAQ.  A model performance evaluation was conducted
through comparisons of the baseline simulation to Interagency Monitoring
of Protected Visual Environments (IMPROVE) and Clean Air Status and
Trends Network (CASTNET) observations in the United States.

2.5  Initial Conditions

CMAQ default initial concentrations were used with a spin up period of
approximately fifteen days to eliminate any significant influence of the
initial conditions.

2.6  Model Performance Evaluation

A model performance evaluation was accomplished for VISTAS as part of
the Regional Haze SIP process.  A detailed evaluation of the model
performance is included in Section 7.0 of the modeling protocol
(Appendix 5), in Chapter 3 and Appendix B of the Technical Support
Document for ASIP Emissions and Air Quality Modeling to Support Fine
Particulate State Implementation Plans (Appendix 4) and in the Technical
Support Document for VISTAS (Appendix 4).  ASIP, as discussed
previously, utilized the modeling for regional haze to also demonstrate
that the area will attain the PM2.5 standard by the end of 2009 as
required.

Though there are no EPA designated metrics, there are metrics that have
been used in several studies.  Appendix B of the Attainment Modeling
Guidance provides examples of the metrics and the goals and criteria for
their use in other modeling studies.  The following metrics and
performance measures are presented to demonstrate adequate model
performance.

Mean fractional bias (MFB) and mean fractional error (MFE) are the
performance metrics used for the evaluation.  The goal for MFB is within
thirty percent.  The goal for the MFE is less than or equal to fifty
percent.  The criterion for the MFB is within ± sixty percent.

As stated in Appendix B of the Attainment Modeling Guidance, 

“. . .proposed to use asymptotically approaching and criteria when
data are greater than 2.5 µm, approaching +200% MFE and ±200% MFB for
extreme small model and observed data (formula of logarithmic MFB and
MFE are proposed).  Based on combined modeling studies … for more
abundant conditions MFE and MFB are typical in the range of:

Sulfates:	MFE = 30% – 77%	MFB =	–45% – +51%  (> 2 µg/m3)

Nitrates:	MFE = 55% – 125%	MFB =	+3% – +82%  (> 1 µg/m3)

Organic carbon:	MFE = 35% – 95%	MFB =	–70% – +35%  (> 1.5 µg/m3)

Elemental carbon:	MFE = 50% – 95%	MFB =	–45% – +50%  (> 0.5
µg/m3)

PM2.5:	MFE = 50% – 85%	MFB =	–55% – +60%  (> 5 µg/m3)”

Bugle plots and time series plots for each of the components of
particulate matter show MFE and MFB.  The bugle plots are shaped as
shown because the goal and criteria lines are adjusted based on the
average concentration of the observed species.  The bugle plots and time
series plots are located in Appendix 7 entitled Local Area Model
Performance.

Modeling some components resulted in better model performance than
others.  Elemental carbon, sulfates, and ammonium all met the criteria
for both MFB and MFE.  Nitrates did not meet fractional bias criteria at
low concentrations.  Organic carbon met the criteria but did not meet
the goal.  Though not all values are in the desired range, the
performance at the Siskin Drive/UTC FRM (470654002) in Hamilton County,
which is critical for the modeled attainment demonstration, is
satisfactory.  Table 2-18 shows the MFEs and MFBs for the Siskin Drive
FRM.

Table 2-18.  MFE and MFB for the Siskin Drive/UTC FRM (470654002)

Pollutant	Mean Fractional Error (MFE)	Mean Fractional Bias (MFB)

Sulfates	37.55%	–17.00%

Nitrates	118.07%	–86.17%

Organic Carbon	59.81%	–55.52%

Elemental Carbon	40.15%	–23.46%

PM2.5	40.63%	–31.56%

3.0  ATTAINMENT DEMONSTRATION

An attainment demonstration consists of analyses that estimate whether
selected emissions reductions will result in ambient concentrations that
meet the NAAQS and an identified set of control measures which will
result in the required emissions reductions.  The necessary emission
reductions for both of these attainment demonstration components may be
determined by relying on results obtained with air quality models.

Section 3.0 of the Attainment Modeling Guidance recommends applying both
a modeled attainment test and a subsequent screening test (or
unmonitored area analysis) to the air quality modeling results to
determine if the annual PM2.5 NAAQS will be met.  Additional technical
or corroboratory analyses may also be used as part of a “supplemental
analysis” or a more stringent “weight of evidence” determination
to supplement the modeled attainment test and to further support a
demonstration of attainment of the NAAQS.

3.1  Speciated Modeled Attainment Test

The purpose of a modeling assessment is to determine if control
strategies currently being implemented and proposed control strategies
will lead to attainment of the annual PM2.5 NAAQS by the attainment year
of 2009.  The modeling is applied in a relative sense, similar to the
8-hour ozone attainment test.  However, the PM2.5 attainment test is
more complicated and reflects the fact that PM2.5 is a mixture.  In the
test, ambient PM2.5 is divided into major components, with a separate
relative response factor (RRF) and future design value (DVF) calculated
for each of the PM2.5 components.  Since the attainment test is
calculated on a per species basis, the attainment test for PM2.5 is
referred to as the Speciated Modeled Attainment Test (SMAT).  In its
entirety, SMAT consists of four basic steps.

First, the observed quarterly mean PM2.5 and quarterly mean composition
for each monitor is calculated.  This is achieved by multiplying the
monitored quarterly mean concentration of PM2.5 from FRMs by the
monitored fractional composition of PM2.5 species for each quarter
[e.g., (20% sulfate) × (15.0 µg/m3 PM2.5 mass) = 3.0 µg/m3 sulfate
mass)].  The monitored quarterly mean concentration of PM2.5 from FRMs
are the five-year baseline design values that are the result of
averaging the three current design values that straddle the modeling
base year.  The fractional composition of PM2.5 species is derived from
speciation monitoring site data that has been processed by the
“sulfate, adjusted nitrate, derived water, inferred carbonaceous
material balance approach,” or SANDWICH method, so that speciation and
FRM masses are equivalent.  The mean composition derived from the
SANDWICH method includes the percent of PM2.5 that can be attributed to
sulfates, nitrates, organic carbon, elemental carbon, other primary
inorganic particulate matter (or crustal material), ammonium, and
particle bound water.  Data from the STN monitor at the Siskin Drive/UTC
site was used in conjunction with all four FRMs in the nonattainment
area.

The second step is to use model results to derive component specific
RRFs for each monitor for each quarter.  The RRF is the ratio of the
model’s future projections to the baseline current projections for
each component.

For the third step, the component specific RRFs are applied to the
observed air quality concentrations to project quarterly species
estimates.  For each quarter, the current quarterly mean component
concentrations from the first step are multiplied by the
component-specific RRFs obtained in the second step.  This leads to an
estimated future quarterly mean concentration for each component.

The fourth step sums the quarterly components to get a quarterly mean
PM2.5 value.  These quarterly mean values are then averaged to produce a
future year annual average PM2.5 estimate, or DVF, for each FRM site. 
This final value is then compared to the NAAQS (15.0 µg/m3) to
determine if attainment is reached.  Documentation of the SMAT
methodology and results are detailed in Sections 8.2 through 8.2.2 of
the Modeling Protocol, which can be found in Appendix 5, and in the
Hamilton County SMAT Process, which can be found in Appendix 8.

3.2  Attainment Test Results

The 2008 annual average data and the three-year average data is as
follows for all of the PM2.5 FRM monitors in this nonattainment area.  

The area monitors show that the annual average concentrations currently
are at a level that demonstrates attainment.  The 2008 annual average
concentrations are 12.7 micrograms per cubic meter at the Siskin
Drive/UTC site (470654002), 12.9 micrograms per cubic meter at the
Soddy-Daisy site (470651011), 13.9 micrograms per cubic meter at the
East Ridge site (470650031), and 12.5 micrograms per cubic meter at the
Walker County, Georgia site (1329540002).

Three-year annual averages (Design Value) for 2006-2008 are as follows:

Federal Reference Monitor	2008

Annual	2006-2008

Average

Siskin Drive/UTC (470654002)	12.7	14.2

East Ridge (470650031)	12.3	13.8

Soddy-Daisy (470651011)	11.4	12.9

Walker County, Georgia (132950002)	12.5	*

		*Incomplete Data for 2007

  The 2009 future design values for the four FRM sites in the
nonattainment area are provided in Tables 3-1 and 3-2, which follow. 
Since the 2009 design value at each site is less than the annual PM2.5
NAAQS of 15.0 µg/m3, the area has passed the attainment test portion of
the attainment demonstration.



Table 3-1.  2009 Quarterly Mean and Annual PM2.5 Design Values

μg/m3)

	Quarter	Annual

	1st	2nd	3rd	4th

	Siskin Drive/UTC (470654002)	12.2	13.7	15.3	13.0	13.6

Maxwell Road/East Ridge (470650031)	13.8	14.8	15.6	13.4	14.4

Soddy-Daisy High School (470651011)	12.2	13.2	14.7	9.3	12.3

Walker County, Georgia (132950002)	12.6	14.9	16.0	12.2	13.9



Table 3-2.  2009 Modeled Species and Annual PM2.5 Design Values

Federal Reference Monitor	Modeled Concentrations (μg/m3)*

	Average Crustal Material	Average Elemental Carbon	Average Organic
Carbon	Average Sulfates	Average Nitrates	Average Ammonium	Average
Particle Bound Water	Blank Filter Mass	PM2.5 Design Value

Siskin Drive/UTC (470654002)	0.970	0.498	5.230	3.845	0.178	1.207	1.135
0.5	13.6

Maxwell Road/East Ridge (470650031)	1.023	0.522	5.614	4.068	0.190	1.277
1.201	0.5	14.4

Soddy-Daisy High School (470651011)	0.902	0.448	4.795	3.458	0.146	1.080
1.020	0.5	12.3

Walker County, Georgia (132950002)	0.978	0.509	5.446	3.914	0.188	1.230
1.153	0.5	13.9

*Passively collected mass, or the mass of the blank filter, has a
constant value of 0.5 μg/m3 for each FRM.

3.3  Supplemental Analyses/Weight of Evidence

The Attainment Modeling Guidance asserts that all attainment
demonstrations should be accompanied by supplemental analysis that
further supports the modeling conclusions.  This supplemental analysis
can include additional analyses of air quality, emissions, and
meteorological data and may consider modeling outputs other than the
results of the attainment test.  If the attainment test results fall
short of the standard, the results of corroboratory analyses may be used
in a weight of evidence determination to show that attainment is likely
despite modeled results, which may be inconclusive.

The Attainment Modeling Guidance defines the guideline for supplemental
analysis/weight of evidence for the annual PM2.5 standard as follows:

Sites with a DVF less than 14.5 µg/m3 should submit basic supplemental
analysis to confirm the outcome of the model attainment test.

Sites with a DVF between 14.5 µg/m3 and 15.5 µg/m3 should submit a
weight of evidence demonstration to aggregate supplemental analysis to
support the model attainment demonstration.

Sites with a DVF greater than or equal to 15.5 µg/m3 should consider
additional control measures to ensure attainment, as more qualitative
analysis is unlikely to support a conclusion differing from the outcome
of the modeled attainment test.

All monitoring sites in the nonattainment area have DVFs lower than 14.5
µg/m3.  Therefore, the following section is a supplemental analysis to
corroborate modeling results, rather than a weight of evidence analysis
to show attainment.

3.3.1  Emissions Reductions Not Modeled

Electric Utilities

Georgia Rules for Air Quality Control Chapter 391-3-1 Rule (sss)
“Multipollutant Control of Electric Steam Generating Units” requires
additional SO2 reductions that were not included in the 2009 Emissions
Inventory for the attainment model.  By June 1, 2009, the Bowen Plant
Unit 2 must operate selective catalytic reduction and flue gas
desulfurization.  The changes will result in 19,328 tons reduction in
SO2 emissions between June and December 2009.

Bowater Newsprint – Calhoun Operations

Bowater Newsprint – Calhoun Operations (Bowater) is a paper
manufacturing plant located thirty miles from the Maxwell Road/East
Ridge monitoring site.  This source is not in the nonattainment area and
the permits for this source are not within the jurisdiction of the
Chattanooga-Hamilton County Air Pollution Control Bureau.    The
facility provided the State of Tennessee with updated emissions that
contained weight of evidence information showing emissions reductions
near this nonattainment area that were less than the modeled emissions. 
In an October 31, 2007, letter (Appendix 9) from Bowater’s Director of
Environmental Affairs, Mr. J.W. O’Grady, the company reports the SO2
emissions were overestimated in the 2009 emissions inventory used in the
ASIP model.  The error was due to overestimates of coal sulfur content
by approximately thirty-four percent and future coal firing by
approximately sixteen percent.  The ASIP modeling showed that the
emissions of SO2 were 5,379 tons in 2009.  In their letter, Bowater
indicates that using the correct coal sulfur content and future coal
firing would lower the SO2 emissions used as model input by forty
percent.

In addition to the correction of SO2 emissions input, Bowater provided
clarification of the company’s seasonal coal-firing rates.  The ASIP
model inventories assumed that SO2 emissions were equally distributed
throughout the year.  The company’s boilers are used primarily to heat
river water.  During the summer months when the river is warmer, monthly
coal usage is almost half that of the winter months.  According to the
Georgia Institute of Technology sensitivity modeling, which will be
detailed in Section 4.1.2 of this document, the conversion of non-EGU
sources’ SO2 emissions to sulfates is ten times higher in the summer
than the winter.  Thus Bowater’s actual fuel use profile, with less
coal burning in the summer results in decreased sulfates.

3.3.2  Status of Modeled Emissions Reductions

Fuel Changes

On-road and non-road diesel fuel and gasoline will contain far less
sulfur in 2009 than in 2002.  Despite expected growth of on-road and
non-road vehicle uses, federal fuel sulfur reductions result in SO2
emission reductions of 317 tons per year from non-road vehicles and 402
tons per year from on-road mobile sources.  As of November 21, 2007,
three of the four fuel terminals for diesel fuel in Hamilton County
stock only ultra-low sulfur diesel fuel.  The local sulfur levels in
gasoline are about 30 ppm at this time. 

One hundred eighty-five diesel school buses are operated by a contractor
in the Chattanooga urban area.  One hundred eighty-two of these were
built in 2007, and the remaining three were built in 2008.  All of these
buses are equipped with diesel particulate filters.  The contractor
reported, in November 2007, that they had not experienced any
difficulties with the performance of the school buses.  The City of
Chattanooga also operates diesel-powered garbage trucks that were all
built in 2007.  They too report no problems with the diesel particulate
filters.

Industrial Sources

Hamilton County sources make up a small portion of the sources in the
modeling domain.  Emissions of PM2.5 and precursors have continued to
decrease.  Traditional heavy manufacturing industries continue their
long term decrease in Hamilton County.  In this jurisdiction, there were
243 industrial point sources in 1999.  Of these, twenty-two were Title V
major sources and eighty-eight were synthetic minor sources.  Today
there are 207 industrial point sources, which include nineteen Title V
sources and seventy-five synthetic minor sources.  Industrial source
emissions for the period have trended downward as well.  Figures 3-1
through 3-7, which follow, chart these emissions trends as reported in
the Chattanooga-Hamilton County Air Pollution Control Bureau emissions
inventories between 1999 and 2007.  Individual industrial point source
emissions from the inventories are provided in Appendix 10.  Emissions
of PM2.5 were not inventoried prior to 2002.  With greater availability
of emission factors, the PM2.5 emissions inventory has improved.

Figure 3-1.  Hamilton County Emissions of PM2.5

Figure 3-2.  Hamilton County Emissions of PM10

Figure 3-3.  Hamilton County Emissions of Sulfur Dioxide (SO2)

Figure 3-4.  Hamilton County Emissions of Nitrogen Oxides (NOX)

Figure 3-5.  Hamilton County Emissions of Toluene

Figure 3-6.  Hamilton County Emissions of Xylenes

Figure 3-7.  Hamilton County Emissions of Ammonia

Electric Utilities

The CAIR is in place, and electric utilities are complying with the
rule.  Georgia Rules for Air Quality Control Chapter 391-3-1 Rule (sss)
“Multipollutant Control of Electric Steam Generating Units” has been
adopted.  Under this rule, the Hammond Plant near Rome, Georgia is
required to have desulfurization equipment installed and operating by
December 31, 2008.  The Bowen Plant in Bartow County is required to have
operational desulfurization equipment installed and operating on Units 3
and 4 by December 31, 2008.  The combined SO2 emissions from the two
plants are estimated to be reduced by 92,104 tons per year by the end of
2008.  A wind from 184° carries the emissions from the Hammond Plant
directly to the Maxwell Road/East Ridge monitor (470650031).  A wind
from 161° carries emissions from the Bowen plant to the Maxwell
Road/East Ridge Monitor (470650031).

Miscellaneous Sources

In 2005, a seasonal open burning ban was put in place in Hamilton
County.  One small city has closed its air curtain destructor, and the
City of Chattanooga no longer burns brush in their air curtain
destructor.  The air curtain destructor is maintained for use only for
major storm brush cleanup as allowed by Federal rules.  Citizen open
burning has also decreased.  For example, in 1994, there were 7,131 open
burning permits issued in Hamilton County.  For the 2006-2007 burning
season, 5,510 permits were issued.  Decreased open burning is known to
reduce elemental carbon, organic carbon, and direct PM2.5 emissions.

Chattanooga has continued to encourage bicycle use.  Both the City and
major employers have installed bicycle racks in the downtown area.  All
Chattanooga Regional Transportation Authority mainline buses are
equipped with bike racks.  Approximately three thousand riders use the
bike racks on the buses.

3.3.3  Status of Attainment for Surrounding Counties

In addition to the modeling results for the nonattainment area, all
monitors in the counties surrounding this nonattainment area also show
attainment in 2009.  Table 3-3 lists the 2009 DVFs for FRMs located in
counties that surround Hamilton County.

Table 3-3.  Surrounding Area 2009 Modeled Annual PM2.5 Design Values

Federal Reference Monitor	PM2.5 Design Value (µg/m3)

DeKalb County, Alabama (010491003)	13.3

Madison County, Alabama (010890014)	12.8

Floyd County, Georgia (131150005)	14.0

McMinn County, Tennessee (471071002)	13.2

4.0  PM2.5 RACT/RACM

The PM2.5 Implementation Rule (40 CFR 51.1010) requires that “For each
PM2.5 nonattainment area, the state shall submit with the attainment
demonstration a SIP revision demonstrating that it has adopted all
reasonably available control measures (including RACT for stationary
sources) necessary to demonstrate attainment as expeditiously as
practical and to meet any reasonable further progress requirements.  The
SIP revision shall contain the list of potential measures considered by
the state and information and analysis sufficient to support the
state’s judgment that it has adopted all RACM including RACT.”  RACM
can apply to industrial sources, mobile sources, open burning, heating,
etc.  RACT is a subset of RACM that applies to industrial sources, and
it is defined as “the lowest emissions limitation that a particular
source is capable of meeting by the application of control technology
that is reasonably available considering technological and economic
feasibility.”  The PM2.5 Implementation Rule further states, “In
determining whether a particular emission reduction measure or set of
measures must be adopted as RACM under Section 172(c)(1) of the Act, the
state must consider the cumulative impact of implementing the available
measures.  Potential measures that are reasonably available considering
technical and economic feasibility must be adopted as RACM if,
considered collectively, they would advance the attainment date by one
year or more.”

Speciated data from the STN monitor at the Siskin Drive/UTC site was
used help identify principle sources of emissions of PM2.5 and its
precursors to be considered for emission controls or reductions. 
According to this monitoring data, other primary particulate matter and
elemental carbon each make up less than five percent of the total PM2.5
mass.  Existing local regulations already address the primary local
sources of elemental carbon – on-road mobile sources, industrial
boilers, process heaters, and open burning.  Additional regulation of
local sources of direct PM2.5 emissions is unlikely to result in a
measurable reduction of PM2.5 due to their small contribution to the
total, the emission and opacity limitations that are in place for all
permitted stationary sources of particulate matter emissions in Hamilton
County, and the uncertainty of some of the PM2.5 emission factors.  The
EPA has determined that their emission factors for PM2.5 from natural
gas combustion are overestimated.  As a result, the EPA reduced such
emissions in the 2002 NEI by about ninety-five percent.  Documentation
to support this reduction is found in Appendix 11.

Monitoring data reveals that sulfates are one of the largest components
of PM2.5 in this area.  Depending on the specific year, sulfates account
for twenty-five percent to thirty-two percent of the total PM2.5 mass. 
As stated previously, sulfates are pollutants caused by SO2 emissions,
primarily from coal-fired power plants outside of Hamilton County.  In
2002, combined SO2 emissions from all sources within Hamilton County
were 3,222 tons per year compared to 44,206 tons per year from the TVA
Widows Creek Fossil Plant in Jackson County, Alabama, which is the
closest of four regional coal-fired power plants located within 75 miles
of the county.  Primary local sources of SO2 emissions are motor
vehicles (both on-road and non-road) and industrial boilers that use a
fuel other than natural gas.  Ultra-low sulfur transportation fuels are
being phased in for both on-road and non-road vehicles.  The primary
fuel that is used by most industrial combustion sources in Hamilton
County is natural gas, which contains a negligible amount of sulfur.

Organic carbon is also one of the largest components of PM2.5 in this
area.  The portion of PM2.5 that is organic carbon ranges from
twenty-eight percent to thirty-two percent.  Organic carbon is produced
from condensation of high molecular-weight VOCs; reactions in the
atmosphere of toluene, xylenes, and trimethylbenzenes that form
secondary organic aerosols; atmospheric reactions of biogenic compounds;
open burning; home heating with wood; on-road and non-road vehicles; and
stationary internal combustion engines.

Approximately 24 tons of high molecular-weight VOCs are emitted annually
in the county, all of which are generated by one Title V industrial
source.  Scrubbers, demisters, and electrostatic precipitators are used
to control these emissions to RACT levels in accordance with Section
4-41, Rule 26 of the Chattanooga-Hamilton County Air Pollution Control
Ordinance (the Ordinance).  There are currently 34 tons per year of
toluene and 41 tons per year of xylenes emitted annually from permitted
industrial sources in Hamilton County.  These emissions are regulated
under Section 4-41, Rule 25 of the Ordinance, which requires BACT for
VOCs.  No emissions of trimethylbenzenes are reported in the local
emissions inventory.  Reactive biogenic emissions from the nonattainment
area and the surrounding counties were estimated by the EPA BEIS3.09
model to be 94,358 tons per year in 2002.  Reactive biogenic VOCs
produced in the nonattainment area alone in 2002 totaled 13,141 tons per
year.

Nitrates are generated in the atmosphere from NOX, which is a product of
combustion.  Nitrates constitute between six percent and eight percent
of the monitored PM2.5.  Seventy-three industrial sources emit NOX
within Hamilton County.  Only two of these have NOX emissions that are
greater than one hundred tons per year.  An additional twenty sources
have NOX emissions greater than twenty tons per year but less than one
hundred tons per year.  All other sources have NOX emissions of less
than twenty tons per year.  The largest industrial NOX source is a
cement kiln that is subject to and meeting the NOX RACT requirements of
the NOX SIP call.  These requirements are in effect for the entire year
rather than just applying during the ozone season.

Ammonia emissions are the source of the ammonium component of the PM2.5
total mass.  On-road mobile sources are the largest contributor to
ammonia emissions in Hamilton County.  Current emissions of ammonia from
permitted stationary sources in the county are inventoried at 17 tons
per year.

4.1.  Permitted Stationary Sources

Section 4-8 of the Ordinance requires that all stationary sources that
emit regulated air pollutants above insignificant levels be permitted. 
There are 207 permitted stationary sources in Hamilton County.  Nineteen
of these are Title V sources, seventy-five are synthetic minor sources,
and 113 area minor sources.  Title V sources have potential emissions of
greater than one hundred tons per year of particulate matter, SO2, NOX,
VOCs, or carbon monoxide; twenty-five tons per year of all hazardous air
pollutants (HAPs) combined; or ten tons per year of any individual HAP. 
Synthetic minor sources have accepted federally enforceable permit
conditions that limit their emissions to below the Title V applicability
thresholds.  Semi-annual compliance reports and annual compliance
certifications are required to be submitted by all Title V sources.  In
addition, most synthetic minor sources and several minor sources are
required to submit periodic reports showing compliance with permit
requirements.

The permitting process that each source goes through requires an
analysis of the pollutant emissions and emission controls.  In addition,
all permitted sources are evaluated annually for compliance with all
applicable regulations.  This evaluation consists of an on-site
inspection and review of required records and reports.  Sources
inspections are also conducted in conjunction with complaint
investigations and routine patrols.  A total of seven staff members
maintain certification to perform visible emission evaluations.

4.1.1  Local Regulations

The Ordinance includes rules that regulate emissions from both new and
existing stationary sources.  The following Federal rules and standards
have been incorporated by reference in the Ordinance.

Federal New Source Performance Standards are adopted by reference into
the Ordinance.  They provide for emissions control based on the
installation date of specific components.

National Emissions Standards for Hazardous Air Pollutants were
promulgated as required by the 1990 Clean Air Act Amendments and are
also adopted by reference into the Ordinance.  These standards have
resulted in reductions of both particulate matter emissions and
emissions of certain aromatic compounds, such as toluene and xylenes,
which react in the atmosphere to produce secondary organic aerosols. 
Some standards use particulate matter as a surrogate for certain HAPs.

Hamilton County was a nonattainment area for total suspended particulate
matter (TSP) and for the one-hour ozone standard.  The community came
into attainment for TSP in 1979 and for the one-hour ozone standard in
1989.  The area has successfully completed the requirements of an Early
Action Compact for the eight-hour ozone standard.  As a result, the area
is being designated as an attainment area for this standard.  In
addition, the largest source of NOX emissions in the nonattainment was
subject to the NOX SIP call.

These earlier nonattainment experiences and the Early Action Compact
resulted in the adoption of rules that have significantly reduced direct
PM2.5 emissions, emissions of VOCs that result in secondary organic
aerosols, and emissions that result from fuel combustion.  Various local
rules and regulations were adopted to bring the area into attainment
with the TSP, one-hour ozone, and eight-hour ozone standards.  Local
rules and regulations of the Ordinance that are in effect include the
following.

Section 4-8(e)(2) requires best available control technology (BACT) for
sources of particulate emissions that impact the former nonattainment
area and that were constructed or modified after September 16, 1980.

Section 4-41, Rule 2 regulates emissions of NOX.

Section 4-41, Rule 3 regulates the opacity of visible emissions.

Section 4-41, Rule 4 limits fuel sulfur content.

Section 4-41, Rule 6 regulates and restricts open burning.  This
includes a county-wide ban on open burning during ozone season.

Section 4-41, Rule 8 regulates particulate matter emissions from fuel
burning equipment.

Section 4-41, Rule 10 regulates particulate matter emissions from
process equipment.

Section 4-41, Rule 11 regulates handling, processing, and storing of
material in the open air and requires that measures be taken to prevent
particulate matter from becoming airborne.  This includes particulate
matter emissions from roads, parking areas, open fields, etc.

Section 4-41, Rule 13 regulates emissions of SO2.

Section 4-41, Rule 25 regulates emissions of VOCs such as toluene and
xylenes that react in the atmosphere to form secondary organic aerosols.
 Rule 25.3 requires BACT or lowest achievable emission rate (LAER)
levels of emissions control, depending on potential emission rates, for
sources of VOC emissions that were constructed or modified after
December 26, 1979.  Rule 25.10 requires gasoline dispensing facilities
to utilize Stage 1 vapor recovery.  180 gasoline dispensing facilities
in Hamilton County are subject to this requirement.  The use of Stage 1
vapor recovery minimizes fugitive emissions of gasoline.  Although
toluene and xylenes combined comprise only about 1.7 percent by weight
of gasoline emissions, the large quantity of gasoline dispensed contains
a considerable amount of these compounds.

Section 4-41, Rule 26 requires RACT for sources of particulate matter
emissions that were in existence before 1978.  Three coal-fired boilers
at one industrial source in Hamilton County comply with this rule by the
use of baghouses to control particulate matter emissions and the use of
continuous opacity monitors.

Section 4-41, Rule 27 requires BACT or “reasonable and proper
emissions limitations,” depending on potential emission rates, for
sources of particulate matter emissions that were constructed or
modified after August 29, 1995.

4.1.2  Sensitivity Modeling

The Georgia Institute of Technology conducted sensitivity modeling for
each of the ASIP states.  Seasonal sensitivities were determined from
modeled PM2.5 concentration responses to uniform emission reductions
throughout Tennessee.  Sensitivities were determined for emissions of
NOX from point sources, SO2 from non-EGU sources, and PM2.5 from point
sources for both summer and winter seasons.  Since these sensitivities
are for statewide reductions in emissions and are not county specific,
the theoretical maximum reductions in PM2.5 concentrations that could be
realized in Hamilton County are arrived at by applying the sensitivities
to the statewide emissions of each pollutant and assuming that all of
these emissions are concentrated in Hamilton County.  The resulting
absolute maximum reductions that result from completely eliminating all
emissions of each specified pollutant, in turn, from Hamilton County are
small, ranging from PM2.5 concentration reductions of 0.0554 μg/m3 for
NOX point source emissions to 0.249 μg/m3 for SO2 non-EGU source
emissions.  These calculated PM2.5 concentration reductions for Hamilton
County are given below in Table 4-1.

Table 4-1.  Maximum PM2.5 Concentration Reductions from Sensitivity
Modeling

Pollutant	Emissions (tons/day)	Season	Sensitivity (ng/m3)/ (ton/day)
PM2.5 Reduction from Total Elimination of Each Pollutant

	Hamilton County	Tennessee (Statewide)



	NOX from point sources	7.501	275.6	Summer	-0.174	0.0479 µg/m3



577.8	Winter	-0.0958	0.0554 µg/m3

SO2 from non-EGU sources	8.786	230.5	Summer	-1.08	0.249 µg/m3



226.7	Winter	-0.0958	0.0217 µg/m3

PM2.5 from point sources	1.899	29.8	Summer	-5.25	0.157 µg/m3



31.7	Winter	-3.51	0.111 µg/m3



Even though potential PM2.5 reductions that are available from local
sources are insufficient to bring the area into attainment before 2009,
a further review was conducted as described in the PM2.5 Implementation
Rule (page 20612): “A state must consider RACT and RACM for all its
nonattainment areas.  However EPA believes that if the state projects
that an area will attain the standard within 5 years of designation as a
result of existing measures (i.e. projected to have a design value of
14.5 or lower), then the state may conduct a limited RACT and RACM
analysis that does not involve additional air quality modeling.  A
limited analysis of this type would involve reasonably available
measures, the estimate of potential emissions reductions, and the time
needed to implement these measures.”

Seventeen permitted sources in Hamilton County emit greater than 10 tons
per year of TSP, as reported in the local emissions inventory for 2007. 
Based on available emission factors, two sources emit greater than 10
tons per year of direct PM2.5.  Nineteen sources emit greater than 10
tons per year of NOX.  Ten sources emit greater than 10 tons per year of
SO2.  Thirteen sources emit more than 40 tons per year of TSP, NOX, and
SO2 combined.  The permits of each of these thirteen sources were
examined in detail and were all found to require at least RACT levels of
emissions control.  The results of this review are given in Appendix 12.
 This review and the results of the sensitivity modeling indicate that
no additional reductions are available from local permitted stationary
sources that would result in attainment in 2008 rather than 2009.

4.2  Gasoline Dispensing Facilities

180 gasoline dispensing facilities are permitted in Hamilton County. 
Stage 1 vapor recovery became a requirement in the county in 2005.  Each
dispensing facility receives an annual inspection.  Stage 1 vapor
recovery results in a reduction of fugitive aromatic compound emissions
of toluene and xylenes of approximately 20 tons per year.  The use of
Stage 1 vapor recover is a RACT level of emissions control.  No
additional reductions of aromatic compound emissions from gasoline
dispensing facilities are practical.

4.3  On-Road Mobile Sources

Automobiles and light trucks registered in Hamilton County are subject
to annual inspection and maintenance requirements of Tennessee State
Rule 1200-3-29.  This rule requires a successful vehicle inspection
prior to registration renewal for 1975 model year and newer vehicles. 
inspection and maintenance of on-road mobile sources results in
reductions in fugitive emissions of aromatic compounds, emissions of
VOCs from incomplete combustion, and NOX emissions.  The resulting
reduction in emissions of aromatic compounds, such as toluene and
xylenes, is approximately 20 tons per year.  The existing inspection and
maintenance program constitutes RACM for on-road mobile sources.  No
additional emissions reductions from on-road mobile sources are planned.

Highway vehicles in Hamilton County emit 352 tons of ammonia per year. 
Ammonia is produced by vehicle catalytic converters, which are regulated
at the federal level.  The preamble of the PM2.5 Implementation Rule
states, “The NARSTO Fine Particulate Assessment indicates that
reducing ammonia emissions where sulfates concentrations are high may
reduce PM2.5 mass concentrations, but may also increase the acidity of
particles and precipitation.”  Based on the high sulfates in this area
and previous experience with acid rain, reductions of ammonia emissions
are not practical.

4.4  Non-Road and Stationary Internal Combustion Engines

Non-road and stationary internal combustion engines result in organic
carbon and other primary particulate matter.  This category of engines
includes a wide variety of non-road diesel engines without controls,
railroad and barge diesel engines, two cycle engines where lubricating
oil is mixed with gasoline, and numerous small gasoline engines.  All of
these engines are regulated at the federal level.  However, state and
local governments could institute voluntary programs to retrofit after
market controls on existing non-road sources.

A previous retrofit program had resulted in the installation of 105
diesel oxidation catalysts on local school buses.  This program was
completed in approximately three years.  About one year was required to
contract and install the catalysts.  It is unlikely that another
retrofit program generating sufficient reductions could be instituted in
time to result in early attainment.

Section 4-41, Rule 2 of the Ordinance limits industrial process NOX
emissions to 300 ppm.  Because many stationary internal combustion
engines cannot meet this requirement, their use is limited.  The
ordinance does allow for the use of stationary internal combustion
engines to power emergency generators.  No emissions reductions beyond
federal requirements for non-road internal combustions engines are
planned.

4.5  Open Burning

Open burning is regulated by Section 4-41, Rule 6 of the Ordinance. 
This rule requires open burning permits, bans open burning from May 1
through September 30, and prohibits the burning of brush cleared for
road building and trash in Hamilton County.  In addition, open burning
is only allowed within the City of Chattanooga if no alternative is
available.  The City of Chattanooga provides a brush pickup service, and
the collected brush is chipped.  Existing restrictions constitute RACM
for open burning and will remain in effect.

4.6  Home Heating with Wood

Information from the 2000 US Census (Appendix 13) showed that 2.2% of
the households in twenty counties in and around the nonattainment area
used wood as the primary heating fuel.  Within Hamilton County, only 712
households, or 0.6% of the total number, heat primarily with wood.

Heating requirements are typically designated in degree days.  According
to the American Gas Association, a degree day accrues for every degree
the average outside temperature is below 65°F during a 24-hour period. 
June, July, August, and September require no heating in this area.  In
Chattanooga, 3,384 degree days comprise the design heating requirements.
 This compares with design heating requirements of 6,282 degree days in
Chicago and 7,966 degree days in Minneapolis.

The Federal New Source Performance Standards for Residential Wood
Heaters (40 CFR Part 60 Subpart AAA) applies to residential wood heaters
manufactured after July 1, 1988, and sold after July 1, 1990.  In the
last seventeen years, it would be expected that a considerable portion
of wood burning stoves have been replaced and that new stoves have been
installed in new construction.  Replacing remaining older stoves would
result in some particulate matter and VOC emissions reductions. 
Particulate matter emission factors are 83 pounds per ton for
conventional stoves and from 26.6 to 28 pounds per ton for stoves
manufactured in accordance with Subpart AAA.  VOC emission factors are
28 pounds per ton and 17.2 pounds per ton for conventional stoves and
Subpart AAA compliant stoves, respectively.

The small portion of households using wood heating, the mild local
climate, and the normal purchases of Subpart AAA compliant wood burning
stoves in the nonattainment area indicate that accelerated replacement
of the remaining older stoves will not be able to advance the attainment
date.  No short term efforts to upgrade or further regulate home heating
are planned.

5.0  MOTOR VEHICLE EMISSIONS BUDGETS

5.1  Transportation Conformity

The purpose of transportation conformity is to ensure that Federal
Transportation actions occurring in nonattainment and maintenance areas
do not hinder the area from attaining and maintaining the NAAQS.  This
means that the level of emissions estimated for the Transportation
Improvement Program and Long Range Transportation Plan must not exceed
the motor vehicle emission budgets (MVEB) defined in this attainment
demonstration.  The MVEB for Hamilton County was developed through
consultation with the Interagency Consultation Partners and the
Chattanooga Hamilton County North Georgia Transportation Planning
Organization.  Georgia and Alabama are addressing transportation budgets
for their portions of the nonattainment area.

5.2  Transportation Budget for Hamilton County

The budget was developed following the 2005 EPA Guidance for Creating
On-Road Mobile Source Emission Inventories for PM2.5.  Mobile source
inventories were developed using the latest planning assumptions, the
most recent recalibrated travel demand model, and EPA’s latest motor
vehicle model, Mobile 6.2.03.  Average annual activity and meteorology
was used in modeling.  Planning Assumptions for Development of 2009
CHCNGA TPO TransCAD Model 2009 Emissions Results are contained in
Appendix 14.  The transportation budget for PM2.5 is 125.56 tons per
year for Hamilton County.

5.3  Pollutants to Be Considered

5.3.1  Sulfur Dioxide (SO2)

The PM2.5 Implementation Rule provides that SO2 must be addressed as a
significant PM2.5 precursor, and sources of SO2 emissions must be
evaluated for control measures.  The rule further states that sulfur
oxides are mostly emitted from fossil fuel combustion with a smaller
portion of emissions from other industrial processes.  According to the
EPA, SO2 emissions from motor vehicles would not require a budget due to
the current de minimis levels of those emissions from on-road vehicles
and further considering the projected decline in SO2 emissions from
vehicles due to low-sulfur gasoline and low-sulfur diesel fuel. 
Hamilton County agrees with the EPA, and therefore no motor vehicle
budget was developed for SO2 on-road mobile sources.

5.3.2  Nitrogen Oxides (NOX)

The PM2.5 Implementation Rule provides that NOX be addressed as a
significant PM2.5 precursor unless demonstrated otherwise by the State. 
The insignificance demonstration for NOX as a PM2.5 precursor is found
in Appendix 15.  Modeling for seventy percent and one hundred percent
reductions of NOX shows minimal changes in PM2.5.

5.3.3 Volatile Organic Compounds (VOCs)

The PM2.5 Implementation Rule explains that the State is not required to
address VOCs as a PM2.5 attainment plan precursor unless the State or
EPA provides an appropriate technical demonstration for a specific area
showing that VOC emissions from sources in the State significantly
contribute to PM2.5 concentrations in a given nonattainment area.  No
demonstration is being submitted.

5.3.4  Ammonia

The EPA’s policy on ammonia presumes that ammonia is not a PM2.5
attainment plan precursor.  The State is not required to address ammonia
unless the State or EPA provides an appropriate technical demonstration
showing that ammonia emissions from sources in the State significantly
contribute to PM2.5 concentrations in the nonattainment area.  No
demonstration is being submitted.

5.3.5  Road and Construction Dust

The PM2.5 Implementation Rule states that crustal material is only a
minor part of PM2.5 annual average concentrations, and that construction
dust will not be significant in all areas.  Construction dust is not a
significant concern in Hamilton County.  The temperate climate and few
unpaved roads in this primarily urban county leave little opportunity
for road or construction dust to become significant contributors to
PM2.5.  In addition, regular rainfall removes dust from paved roads and
helps keep unpaved roads from becoming a source of significant PM2.5
crustal material.  No demonstration is being submitted.

6.0  CONTINGENCY MEASURES

The PM2.5 Implementation Rule requires contingency measures that would
take effect in the event a nonattainment area fails to meet the PM2.5
NAAQS.  These contingency measures must be fully adopted rules or
control measures that are ready to be implemented quickly upon failure
of an area to meet the standard by the attainment date.

There are no available contingency measures in Hamilton County that can
result in significant PM2.5 reductions from industrial sources as
indicated by Table 4-1 in Section 4.1.2 of this document.  The table
shows that elimination of all of the industrial emissions in the county
will have little effect on the PM2.5 concentration.

-

9

;

j

l

‚

ƒ

„

´

µ

¶

â

ä



<

m

…

·

å

 

Q



N

P

Q

V

X

Z

z

{

}

~

®

°

è

é

ê

ë

ï

ñ

ô

ÿ

OQ

~

±

ë

h

h

h×

h×

h

h×

%

hß

②朁ꍤ‗氀჆

最ꍤ‗氀჆

②朁ꍤ‗氀჆

最ꍤ‗氀჆

②朁ꍤ‗氀჆

最ꍤ‗氀჆

②朁ꍤ‗氀჆

最ꍤ‗氀჆

②朁ꍤ‗氀჆

最ꍤ‗氀჆

h

②朁ꍤ‗氀჆

最ꍤ‗氀჆

②朁ꍤ‗氀჆

最ꍤ‗氀჆

②朁ꍤ‗氀჆

②朁ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

②朁ꍤ‗氀჆

最ꍤ‗氀჆

②朁ꍤ‗氀჆

最ꍤ‗氀჆

②朁ꍤ‗氀჆

jø

②朁ꍤ‗氀჆

 h¯

 h¯

  h¯

②朁ꍤ‗氀჆

②朁ꍤ‗氀჆

②朁ꍤ‗氀჆

②朁ꍤ‗氀჆

②朁ꍤ‗氀჆

②朁ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

②朂ꍤ‗氀჆

最ꍤ‗氀჆

②朁ꍤ‗氀჆

û

②朂ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

②朂ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

②朂ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

②朂ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

②朂ꍤ‗氀჆

②朁ꍤ‗氀჆

最ꍤ‗氀჆

②朁ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

②朂ꍤ‗氀჆

最ꍤ‗氀჆

②朁ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

②朂ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

②朂ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

②朂ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

②朂ꍤ‗氀჆

②朁ꍤ‗氀჆

最ꍤ‗氀჆

②朁ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

②朂ꍤ‗氀჆

最ꍤ‗氀჆

②朁ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

②朂ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

②朂ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

②朂ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

②朂ꍤ‗氀჆

②朁ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

②朂ꍤ‗氀჆

最ꍤ‗氀჆

②朁ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

②朁ꍤ‗氀჆

②朂ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

②朁ꍤ‗氀჆

②朂ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

②朁ꍤ‗氀჆

②朂ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

②朁ꍤ‗氀჆

②朂ꍤ‗氀჆

②朁ꍤ‗氀჆

最ꍤ‗氀჆

②朁ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

②朂ꍤ‗氀჆

最ꍤ‗氀჆

②朁ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

②朂ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

②朂ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

②朂ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

②朂ꍤ‗氀჆

②朁ꍤ‗氀჆

最ꍤ‗氀჆

②朁ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

②朂ꍤ‗氀჆

最ꍤ‗氀჆

②朁ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

②朂ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

②朂ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

②朂ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

②朂ꍤ‗氀჆

②朁ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

②朂ꍤ‗氀჆

最ꍤ‗氀჆

②朁ꍤ‗氀჆



②朂ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

②朂ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

②朂ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

②朂ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

②朂ꍤ‗氀჆

②朁ꍤ‗氀჆

最ꍤ‗氀჆

②朁ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

②朂ꍤ‗氀჆

最ꍤ‗氀჆

②朁ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

②朁ꍤ‗氀჆

②朂ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

②朁ꍤ‗氀჆

②朂ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

②朁ꍤ‗氀჆

②朂ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

②朁ꍤ‗氀჆

②朂ꍤ‗氀჆

②朁ꍤ‗氀჆

②朁ꍤ‗氀჆

gꍤ‗氀჆

②朂ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

②朁ꍤ‗氀჆

gꍤ‗氀჆

②朁ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

②朂ꍤ‗氀჆

gꍤ‗氀჆

②朂ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

②朁ꍤ‗氀჆

gꍤ‗氀჆

②朂ꍤ‗氀჆

最ꍤ‗氀჆

②朁ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

②朂ꍤ‗氀჆

最ꍤ‗氀჆

②朁ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

②朂ꍤ‗氀჆

最ꍤ‗氀჆

②朁ꍤ‗氀჆

gꍤ‗氀჆

②朁ꍤ‗氀჆

最ꍤ‗氀჆

②朂ꍤ‗氀჆

最ꍤ‗氀჆

gꍤ‗氀჆

最ꍤ‗氀჆

②朁ꍤ‗氀჆

最ꍤ‗氀჆

②朂ꍤ‗氀჆

②朂ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

②朁ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

②朂ꍤ‗氀჆

最ꍤ‗氀჆

②朁ꍤ‗氀჆

最ꍤ‗氀჆

gꍤ‗氀჆

②朂ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

②朁ꍤ‗氀჆

gꍤ‗氀჆

②朂ꍤ‗氀჆

最ꍤ‗氀჆

②朁ꍤ‗氀჆

最ꍤ‗氀჆

gꍤ‗氀჆

②朁ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

②朂ꍤ‗氀჆

gꍤ‗氀჆

②朂ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

②朁ꍤ‗氀჆

gꍤ‗氀჆

②朂ꍤ‗氀჆

最ꍤ‗氀჆

②朁ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

②朂ꍤ‗氀჆

最ꍤ‗氀჆

②朁ꍤ‗氀჆

最ꍤ‗氀჆

gꍤ‗氀჆

最ꍤ‗氀჆

②朂ꍤ‗氀჆

最ꍤ‗氀჆

②朁ꍤ‗氀჆

②朁ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

②朂ꍤ‗氀჆

最ꍤ‗氀჆

gꍤ‗氀჆

最ꍤ‗氀჆

②朁ꍤ‗氀჆

最ꍤ‗氀჆

②朂ꍤ‗氀჆

②朂ꍤ‗氀჆

最ꍤ‗氀჆

②朁ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

②朂ꍤ‗氀჆

最ꍤ‗氀჆

②朁ꍤ‗氀჆

最ꍤ‗氀჆

!

"

H

I

o

p

š

›

¿

À

è

é

	

#



ƒ

 

¥

¦

Ã

Ä

Ó

Û

Ü

gꍤ‗氀჆

最ꍤ‗氀჆

②朂ꍤ‗氀჆

最ꍤ‗氀჆

②朁ꍤ‗氀჆

gꍤ‗氀჆

②朂ꍤ‗氀჆

最ꍤ‗氀჆

②朁ꍤ‗氀჆

最ꍤ‗氀჆

gꍤ‗氀჆

②朁ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

②朂ꍤ‗氀჆

gꍤ‗氀჆

②朂ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

②朁ꍤ‗氀჆

gꍤ‗氀჆

②朂ꍤ‗氀჆

最ꍤ‗氀჆

②朁ꍤ‗氀჆

最ꍤ‗氀჆

	

-

!

"

$

最ꍤ‗氀჆

②朂ꍤ‗氀჆

最ꍤ‗氀჆

②朁ꍤ‗氀჆

最ꍤ‗氀჆

ఀ$

*

.

2

5

7

=

A

E

H

I

K

Q

最ꍤ‗氀჆

最ꍤ‗氀჆

②朂ꍤ‗氀჆

最ꍤ‗氀჆

②朁ꍤ‗氀჆

②朁ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

②朂ꍤ‗氀჆

最ꍤ‗氀჆

ఀ~

‚

…

‡

“

—

š

›

£

§

ª

最ꍤ‗氀჆

②朁ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

②朂ꍤ‗氀჆

②朂ꍤ‗氀჆

最ꍤ‗氀჆

②朁ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

ఀÔ

Ö

Ü

â

å

è

é

ë

ñ

÷

ù

û

ý

最ꍤ‗氀჆

②朂ꍤ‗氀჆

最ꍤ‗氀჆

②朁ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

②朂ꍤ‗氀჆

最ꍤ‗氀჆

②朁ꍤ‗氀჆

h«V

hà

h

hƒ

hƒ

h«V

最ꍤ‗氀჆

	È

	€

②朁ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

②朁ꍤ‗氀჆

	È

	€

②朁ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

②朁ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

②朁ꍤ‗氀჆

最ꍤ‗氀჆

②朁ꍤ‗氀჆

②朁ꍤ‗氀჆

最ꍤ‗氀჆

②朁ꍤ‗氀჆

最ꍤ‗氀჆

②朁ꍤ‗氀჆

②朂ꍤ‗氀჆

②朁ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

②朂ꍤ‗氀჆

②朁ꍤ‗氀჆

最ꍤ‗氀჆

②朂ꍤ‗氀჆

②朁ꍤ‗氀჆

最ꍤ‗氀჆

②朂ꍤ‗氀჆

②朁ꍤ‗氀჆

̀Ĥ␷㠀$⑈愀Ĥ摧⪁

②朁ꍤ‗氀჆

最ꍤ‗氀჆

*

②朁ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

②朁ꍤ‗氀჆

最ꍤ‗氀჆

②朁ꍤ‗氀჆

②朁ꍤ‗氀჆

̀Ĥꐓx␷㠀$⑈愀Ĥ摧狙Y̀Ĥ␷㠀$⑈愀Ĥ摧⪁

最ꍤ‗氀჆

最ꍤ‗氀჆

②朁ꍤ‗氀჆

②朁ꍤ‗氀჆

最ꍤ‗氀჆

②朁ꍤ‗氀჆

最ꍤ‗氀჆

②朁ꍤ‗氀჆

最ꍤ‗氀჆

 h

②朁ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

②朁ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

②朁ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

②朁ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

②朁ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

②朁ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

最ꍤ‗氀჆

②朁ꍤ‗氀჆

最ꍤ‗氀჆

*Georgia Rules for Air Quality Control Chapter 391-3-1 Rule (sss)
“Multipollutant Control of Electric Steam Generating Units” requires
additional power plant SO2 emission reductions after the end of 2008. 
Bowen Plant Unit 1 is required to be equipped with flue gas
desulfurization by June 1, 2010.  The SO2 emissions reductions expected
from this unit are 33,135 tons per year or 19,328 tons reduction in the
last seven months of 2010.  These reductions were not modeled for 2009. 
The reductions are not required to demonstrate attainment of the annual
PM2.5 standard, but constitute contingency measures.

7.0  REASONABLE FURTHER PROGRESS

Clean Air Act §172(c)(2) requires plans for nonattainment areas to
require “reasonable further progress,’’ (RFP) as defined in CAA
§171(1).  The Act requires nonattainment areas to make “such annual
incremental reductions in emissions of the relevant air pollutant as are
required by this part or may reasonably be required by the Administrator
for the purpose of ensuring attainment of the applicable national
ambient air quality standard by the applicable date.”  In general
terms, the goal of RFP is to allow areas to achieve generally linear
progress toward attainment.  The RFP requirements were included in the
CAA to assure steady progress toward attaining air quality standards, as
opposed to deferring implementation of all measures until the end date
by which the standard is to be attained.

40 CFR §51.1009 states that, consistent with CAA §172(c)(2), SIPs for
PM2.5 nonattainment areas must demonstrate reasonable further progress
as provided in §51.1009(b) through (h).  Paragraph (b) of that section
states that “if the State submits to EPA an attainment demonstration
and State implementation plan for an area which demonstrates that it
will attain the PM NAAQS within five years of the date of designation,
the State is not required to submit a separate RFP plan.  Compliance
with the emission reduction measures in the attainment demonstration and
State implementation plan will meet the requirements for achieving
reasonable further progress for the area.”  Consistent with the
provisions of §51.1009(b), Hamilton County has submitted a SIP that
demonstrates that the Hamilton County portion of the nonattainment area
will attain the annual PM2.5 NAAQS within five years of designation. 
Therefore, no additional provisions for RFP are required.  Compliance
with the emission reduction measures in this attainment demonstration
and SIP will meet the requirements for achieving reasonable further
progress for the Hamilton County, Tennessee area.

	Page   PAGE  vi 

		

		

Page 1 of 2		

		

C-HCAPCB	-  PAGE  14 -	PM 2.5 SIP

