Designations Methodology

	

3.1	An Overview of EPA’s Nine-Factor Analysis

In June 2007, the Administrator issued guidance to the EPA Regional
offices regarding the factors that could be considered as the basis for
nonattainment area boundaries.  In that guidance, EPA identified certain
factors that were deemed appropriate to consider in making nonattainment
area boundary recommendations and final boundary determinations. EPA
noted they would consider these same factors, along with any other
relevant information, in evaluating modifications to the boundary
recommendations from states and tribes. EPA recommended that states and
tribes consider the following nine factors in assessing whether to
include an area in the designated nonattainment area boundary:

Emission data 

Air quality data 

Population density and degree of urbanization (including commercial
development)

Traffic and commuting patterns

Growth rates and patterns 

Meteorology (weather/transport patterns) 

Geography/topography (mountain ranges or other air basin boundaries) 

Jurisdictional boundaries (e.g., counties, air districts, Reservations,
metropolitan planning organizations (MPOs))  

Level of control of emission sources

EPA pointed out that this list of recommended factors was not intended
to be exhaustive, and states and tribes were free to submit additional
information on factors they believed were relevant for EPA to consider.
In general, EPA said that a state’s or tribe’s demonstration
supporting the boundary recommendation for an area should show that: 1)
violations are not occurring in the excluded portions of the recommended
area, and 2) the excluded portions do not contain emission sources that
contribute to the observed violations. EPA indicated that a state or
tribal submittal that addressed only whether monitored violations are
occurring in an area would not suffice as the sole justification for
designating the boundaries of a nonattainment area.  Appendix A contains
the EPA Guidance Memorandum to Regional Administrators.  The following
sections discuss each of the nine factors in more detail.

3.1.1	Factor 1: Emissions Data 

The emissions analysis is an examination of emissions that contribute to
ambient PM2.5, including carbonaceous particles (carbon), inorganic
particles (crustal), SO2, NOx, VOCs, and ammonia. Emissions data are
derived from the 2005 National Emissions Inventory (NEI), version 1, and
are given in tons per year. Emissions data indicate the potential for a
county to contribute to observed violations, making it useful in
assessing boundaries of nonattainment areas.  

Emission data from the 2005 NEI Version 1 were used for technical
analyses for purposes of area designations for the 2006 24-hour PM2.5
NAAQS.

The 2005 NEI Version 1 is based on the 2002 NEI Version 3, with the
following additions: 

1)	2005 emissions for electric generating units (EGUs) baed on CEM/heat
input reported to EPA by EGUs;

2)	2005 version 1 National Mobile Inventory Model (NMIM) run for onroad
and nonroad emissions; 

3)	Facility closures from state/local and tribal agencies; and,

4)	2005 wildfire/managed burn data.

For more information about the NEI, go to: 
http://www.epa.gov/ttn/chief/net/2005inventory.html.

For technical analyses for 24-hr PM2.5 designations, PM2.5 emissions
data were derived from the 2005 NEI, version 1, and are given in tons
per year.  Emissions that contribute to ambient PM2.5, including
carbonaceous particles (carbon), inorganic particles (crustal), SO2,
NOx, VOCs, and ammonia, are included in the following sets of data. 
Detailed monthly and annual emissions data for each State, by county,
are provided in Microsoft Excel spreadsheet format.

Appendix B includes 2005 NEI data.  

See also item 2.B. at
http://www.epa.gov/ttn/naaqs/pm/pm25_2006_techinfo.html

3.1.2 	Factor 2: Air Quality Data

The air quality data factor involves consideration of data from the
national network of Federal Reference Method monitors operated to
measure total fine particle mass for determining compliance with the
PM2.5 NAAQS.  In the designations process, EPA also considered fine
particle chemical composition data from the Chemical Speciation Network,
the IMPROVE monitoring network in national parks and wilderness areas,
and for a limited number of sites, a supplemental analysis to evaluate
chemical composition for areas having only FRM monitoring (measuring
PM2.5 mass only) and no chemical speciation monitor.  

PM2.5 Design Values

Areas are designated nonattainment for the 24-hr PM2.5 NAAQS because
they have at least one ambient monitoring location which violates the
standard of 35 µg/m3.  A determination of NAAQS compliance is made by
considering the “design value” for each site.  The design value for
a site is the 3-year average of three annual 98th percentile
concentration values. The specific methodology for calculating the PM2.5
design values, including computational formulas and data completeness
requirements, is described in 40 CFR Part 50, Appendix N.  Only PM2.5
measurements produced with the Federal Reference Method (FRM) or a
Federal Equivalent Method (FEM) can be used for NAAQS comparisons. 

FRM measurement data residing in EPA’s Air Quality System (AQS) are
used to calculate the 24-hr PM2.5 design values.  Individual
measurements which are judged to be “exceptional” in accordance with
the Exceptional Events Rule (such as days with poor air quality caused
by wildfire or dust events) are not included in these calculations. 
State, Local, and Tribal monitoring agencies are required to certify
data submitted to AQS on an annual basis, specifically by June 30th of
the subsequent year.  EPA typically extracts ambient data from AQS for
regulatory purposes shortly after that certification due date.  EPA then
calculates NAAQS design values and posts design value results on a
public website.  The State and Tribal recommendations sent to EPA in
December 2007 were based on data from 2004-2006.  The 2005-2007 PM2.5
design values were initially posted to www.epa.gov/airtrends in August
of 2008 (utilizing a data file extracted from AQS on July 8, 2008).
These data were referenced by EPA in its August letters responding to
State and Tribal designation recommendations.  

Although annual data certification is required by June 30th of the
following year, additional time is allowed for EPA to review and
finalize its concurrence /non-concurrence of State requests to exclude
data from regulatory calculations, as appropriate, due to “exceptional
events.”  In order to provide the most up-to-date design values to
support 24-hr designations, EPA re-extracted all 24-hr PM2.5 FRM
measurement data from AQS on December 18, 2008 and re-computed final
PM2.5 design values. 

Appendix C provides a summary listing of the final PM2.5 design values
for the periods 2004-2006 and 2005-2007 for each area to be designated
nonattainment for the 24-hour standard.  It also includes more detailed
site specific monitoring information for these periods.  This file is
also posted on www.epa.gov/airtrends.  

PM2.5 Chemical Composition Monitoring Data

Fine particle chemical composition data (also known as “speciation
data”) is important in order to determine the composition of measured
PM2.5 and potential contributing emission sources.  The speciation data
used to support the various designation analyses include data from the
routine urban and rural speciation monitoring networks -- the Chemical
Speciation Network (CSN) and Interagency Monitoring of Protected Visual
Environments (IMPROVE), as well as limited measurements from Federal
Reference Method (FRM) filters.  For these 24-hr NAAQS analyses, data
for the highest PM2.5 days were selected to represent the 24-hr design
values as measured by the FRM.  These speciation measurement data were
adjusted using the SANDWICH procedure to represent the chemical
constituents of FRM mass.  Depending on availability, completeness and
proximity to the PM2.5 design value monitor location, different years of
data may have been used for specific analyses.  In all cases, however,
the data are judged to be representative of the most recent 3-years of
PM2.5 levels.

Appendix D is   HYPERLINK
"http://www.epa.gov/ttn/naaqs/pm/presents/pm2.5_chemical_composition.pdf
"  The Chemical Composition of PM2.5 to support PM Implementation (PDF)
, a presentation on the general role of speciation data to support the
fine particle designations process and related analyses.  It was
presented at the PM2.5 Implementation and Designations Workshop for
State and Tribal representatives, held June 20-21, 2007 in Chicago, IL. 
The presentation illustrates how EPA derives fine particle composition
associated with PM2.5 mass measurements, how the typical high day and
average composition varies spatially and temporally, and how these data
relate to potential emission sources.  

Appendix E includes a summary of the major chemical components of PM2.5
for a number of areas being designated as nonattainment in December
2008.  These data are derived from the Chemical Speciation Network and
are presented for the 2005-2007 period.  

Appendix F includes a report entitled “Limited 2004-06 Speciation Data
Derived from FRM filters for 20 Areas.”  The report describes an
analysis coordinated by EPA for selected areas without speciation
monitoring to evaluate the chemical composition of filters obtained from
FRM mass-based monitors.  See also:    HYPERLINK
"http://www.epa.gov/ttn/naaqs/pm/docs/available_new_speciation_data_pm2.
5_naa.pdf" 
http://www.epa.gov/ttn/naaqs/pm/docs/available_new_speciation_data_pm2.5
_naa.pdf .

Appendix G includes the data file for the report “Limited 2004-06
Speciation Data Derived from FRM filters for 20 Areas.”  See also: 
http://www.epa.gov/ttn/naaqs/pm/docs/archive_filters_chemical_analysis_d
ata_2004-2006.xls

Appendix H contains the report “Derivation of the Contributing
Emissions Score,” and includes speciation data summaries (2004-2005)
for CSN and IMPROVE sites that were used in developing the Contributing
Emissions Score.  For CSN raw data, see   HYPERLINK
"http://www.epa.gov/cgi-bin/htmSQL/mxplorer/query_spe.hsql" 
http://www.epa.gov/cgi-bin/htmSQL/mxplorer/query_spe.hsql .  For raw
IMPROVE data, see http://vista.cira.colostate.edu/views/.

3.1.3 	Factor 3: Population Density and Degree of Urbanization 

This analysis is an examination of the population for each urban area,
as well as the population density for each county in that area.
Population data indicate the likelihood of population-based emissions
that might contribute to violations.

The information used to derive the Factor 3 values for the 2006 PM2.5
NAAQS was taken from the U.S. Census Bureau estimates for 2000 and 2005
(  HYPERLINK "http://www.census.gov/popest/datasets.html" 
http://www.census.gov/popest/datasets.html ).  The population density
was presented as the number of people per square mile, and the
statistics were rounded to two significant figures.

3.1.4 	Factor 4: Traffic and Commuting Patterns 

The traffic and commuting analysis includes an examination of the number
of commuters in each county who drive to another county within an urban
area, the percent of total commuters in each county who commute to other
counties within the metropolitan area, and the total Vehicle Miles
Traveled (VMT) for each county in millions of miles.

Vehicle Miles Traveled

Vehicle miles traveled (VMT) data developed for the National Emissions
Inventory 2005 version 2 were used in technical analyses of areas for
purposes of area designations for the 2006 24-hour PM2.5 NAAQS.  The
2005 VMT data used for the 9-factor analysis have been derived using

methodology such as that described in "Documentation for the  2005
Mobile National Emissions Inventory, Version 2," December 2008, prepared
for the Emission Inventory Group, U.S. EPA.  A copy of this document can
be found in the docket for this rulemaking. 

For 2005, a full VMT database at the county, roadway type, and vehicle
type level of detail was developed from Federal Highway Administration
(FHWA) information.  For States and local areas that submitted VMT data
that were incorporated in the 2002 NEI, the 2002 NEI VMT data were
extrapolated to 2005 using growth factors developed from the FHWA data,
and these new VMT data replaced the baseline FHWA-based VMT data.  The
resulting VMT database prepared for 2005 include data for all 50 States,
the District of Columbia, Puerto Rico, and the Virgin Islands for each
of the 12 Highway Performance Modeling System (HPMS) functional roadway
types and the 28 MOBILE6 vehicle classes, for a total of 336 records per
year per county.  The data were prepared in the NMIM National County
Database Base Year VMT table format. At this point, States had the
opportunity to submit 2005 estimates that replaced EPA’s estimates for
2005.  

Appendix I includes a description of the methodology for preparing VMT
estimates:  “Methodology for Preparing VMT Estimates for the National
Emission Inventory:  2003, 2004, and 2005.”

Appendix J includes VMT totals for 2005 for each county in the United
States, in millions of miles:  “VMT Data Used for Technical Analyses
for Area Designations for the 2006 24-hour PM2.5 NAAQS.”  

Commuting Data

Information from the U.S. Census 2000 County-to-County Worker Flow Files
(downloaded from   HYPERLINK
"http://www.census.gov/population/www/cen2000/commuting.html on
6/4/2003"  http://www.census.gov/population/www/cen2000/commuting.html
on 6/4/2003 ) was processed to develop summaries of the number of
workers 16 years old and over who are commuting between counties in and
around areas under consideration for PM2.5 designations.  To manage the
exceptionally large files that match every pair of counties in the
nation, only county pairs with greater than 10 commuters were included
in the tables.  The four statistics considered in the designations
process from these data were:  

1) “Number commuting into any violating counties” indicates the
number of residents going from this county into any violating county
(including residents who commute within the same county)

2) “Percent commuting into any violating counties” indicates the
number of residents going from this county into any violating county
(including residents who commute within the same county) divided by the
total number of commuters residing in that county.

3) “Number commuting into statistical area” indicates the number of
commuters residing in this county and traveling into this particular
violating combined statistical area (CSA) or core based statistical area
( CBSA) (including residents who commute within the same county).  In
cases where no CBSA was identified, then the commuters traveling into
the particular violating county were considered.

4) “Percent commuting into statistical area” indicates the number of
commuters residing in this county and traveling into this particular
violating CSA or CBSA (including residents who commute within the same
county) divided by the total number of commuters residing in that
county.  In cases where no CBSA was identified, then the commuters
traveling into the particular violating county were considered.

For items 3 and 4, if the violating monitor(s) in the area were located
in a county within a CSA, the CSA was considered to be the statistical
area.  If the violating monitors were not located in a CSA but in a
CBSA, then the CBSA was considered to be the statistical area.  If the
violating monitor was not located in a CSA or CBSA, then the violating
county was considered to be the statistical area.

Appendix K includes county to county worker flow data for relevant areas
(from 2000 Census data).   

3.1.5	Factor 5: Growth Rates and Patterns 

The growth analysis is an evaluation of actual and/or projected percent
population growth for counties in an area over a period of time (e.g.,
ten years).  

This factor looks at the population and VMT trends for each area from
2000 to 2005, as well as patterns of population and VMT growth. A county
with rapid population or VMT growth is generally an integral part of an
urban area and could be an appropriate county for implementing
mobile-source and other emission-control strategies, thus  warranting
inclusion in the nonattainment area.

3.1.6	Factor 6: Meteorology 

The analysis of meteorological data considers the contribution of
meteorological conditions such as wind speed and wind direction to high
PM2.5 concentrations.  EPA evaluated wind trajectories and pollution
roses in evaluating this factor.

Wind Trajectories

Meteorology plays a major role in the formation and transport of fine
particulate matter over large areas. To better take into account the
transport of PM2.5 precursors and primary emissions to the violating
monitor, the HYSPLIT trajectory model developed by NOAA was used to
calculate wind trajectories 48 hours backward in time from the violating
monitor to show the path the air mass took on its way to the site.
Trajectories were run for those days with PM2.5 concentrations greater
than the 98th percentile for a particular year. Depending upon when an
area first violated the 24-hour standard, either the three year period
from 2004 through 2006 or 2005 or 2007 was used. A trajectory was
started from eight equally spaced times spread across the span of each
day (i.e., at three hour intervals). The initial start height for each
backward trajectory was set to the mixing height at the start time. The
mixing height at the start time was calculated by the HYSPLIT model. Two
additional evenly spaced starting heights relative to the starting
mixing height and a fourth height at 10 meters were used to calculate
trajectories in the same manner described above. The trajectories were
plotted using Google Earth and animations were created to show the path
the air masses took on their way to the violating site. The trajectory
paths were plotted as white lines with the locations of the air mass
along the path represented as colored dots to show the time the air mass
passed over a location. 

The county trajectory weight corresponding to the values used for the
Contributing Emissions Score (CES) were also plotted to show the
meteorological adjustment used for the CES. The trajectory weights
provide an indication of the likelihood of a county being upwind of a
violating monitor on days with high PM2.5 concentrations. The locations
of electric generating units were also displayed as yellow push pins on
the map. The violating site was marked by a yellow star. Where
appropriate, an area's trajectories were divided into two seasons based
on the time of year when the day above the 98th percentile occurred. The
warm season was designated as the months of May through September, while
the cold season was all other months of the year.  A more detailed
description of the CES and its components is provided in given in
appendix H, Derivation of the Contributing Emissions Score.  

Appendix L, Back Trajectory Information, includes a number of static
images showing the trajectories for days greater than the annual 98th
percentiles during either 2004 through 2006 or 2005 through 2007 for the
following areas:  Madison WI, Paducah KY, Clarksville TN, Parkersburg
WV, Knoxville TN, Louisville KY, Huntington WV, and Cleveland OH.  The
last two pages of the document contain two examples of trajectory
animations for the Paducah, KY area from September 10, 2005 and July 26,
2007. The concentration at the violating site for each day is noted next
to the site which is marked by a yellow star. The trajectory paths are
represented by white lines with the locations of the air mass at a
particular time being represented by colored dots. The clock at the top
of the animation shows the time of day.

Pollution Roses

Pollution rose graphics were produced for PM2.5 federal reference method
(FRM) ambient monitoring sites located in or near one of the 58
identified “potential” nonattainment areas.  A rose plot was
generated for each ambient monitoring site.  Each rose figure combines
24-hour FRM data (2005 to 2007) with available same-day meteorological
24-hour resultant wind speed and wind direction information.  Each rose
provides a visual indication of the predominant direction and associated
speed in which the wind was blowing on each PM2.5 sample day. 
Meteorological information (i.e., wind speed and wind direction) from
the National Weather Service (NWS) reporting station nearest each
ambient monitoring site were aggregated to a 24-hour basis and paired
with the site’s PM2.5 24-hour average concentrations.  Hourly NWS wind
speed and wind direction data were aggregated to a daily (24-hour)
“resultant” basis by calculating vector averages.  

On the rose graphics, the center of the plot (i.e., the crosshair
intersection of a north-south line and an east-west line) represents the
ambient monitor location.  Colored symbols (triangles and dots),
depicting the reported 24-hour average PM2.5 concentrations, are plotted
around the monitor with their relative position (to the monitor center)
denoting the 24-hr average resultant wind speed and direction.  More
specifically:

The color and size of the symbols reflects the concentration of PM2.5 on
a given day. Small blue symbols indicate a 24-hour average concentration
of 30 µg/m3 or less; medium-sized yellow symbols indicate 24-hour
average concentrations between 30 and 35 (29.9 to 34.4) µg/m3; large
red symbols identify daily concentrations between 35 and 40 (35.5 to
40.4) µg/m3; and large black symbols identify concentrations greater
than 40 (> 40.5).  Thus, the red and black symbols show exceedances of
the 24-hour PM2.5 NAAQS.

The symbol shape indicates that “season” corresponding to the
reported observation.  Triangle symbols are used for readings collected
in the “cool” season (October through April, inclusive) and dot
symbols are used to mark observations that were collected in the
“warm” season (May through September).

The location of the plotted symbol in relation to the center of the
diagram indicates the direction from which the wind was predominantly
traveling that day (i.e., the 24-hour resultant wind direction). Thus, a
symbol in the top left quadrant of the pollution rose demonstrates that
the wind was emanating from the northwest direction on the date of that
particular monitor reading.  High concentration markers, especially when
grouped together, identify the typical directional location or
“source” of high particulate (and/or particulate precursor)
emissions.

The symbol’s distance from the center of the plot represents the
“resultant wind speed” for the day.  Resultant wind speeds are
indicated by the distance of the symbol from the center of the plot. 
The center of the plot indicates a wind speed of zero; concentric
reference rings are drawn at 2 miles per hour increments up to 12 miles
per hour (i.e., at 2, 4, 6, 8, 10, and 12 mph).  Resultant speeds
greater than 12 miles per hour are capped at 12 mile per hour, and
plotted on the outer reference ring.  Thus, symbols located close to the
center of the diagram indicate a slower resultant wind speed (and
perhaps stagnation conditions); symbols located further away from the
center indicate higher wind speeds and ostensibly, more possibility of
pollutant transport over a longer distance.

	

	There are several situations in which reported PM2.5 observations are
not plotted for a particular site: 

Exceptional event-flagged days that have been approved by EPA (i.e., AQS
concurrence code =’Y’) are not shown.  Note that a July, 2008 AQS
extraction was the source of the PM2.5 data used to generate the
pollution roses but that subsequent exceptional event approvals were
made and considered in the designation process.  Thus, in some
situations, exceedances are plotted (and included in associated counts)
that would not have been if a newer dataset were utilized.  In these
situations, the identified 98th percentiles and 24-hour design values
are probably overstated.  

Days in which there is insufficient meteorological data available for
the nearest NWS site are not plotted.

If wind data are available but indicate extreme variability (as
identified by a ratio of resultant speed / scalar speed of less than
0.4) then the observation also is not shown.  

A note on the plot shows the number of exceedance days not plotted for
one of the last two reasons.  

The distance between the ambient monitoring site and the NWS site is
provided on each plot.  For about 90% of the air quality monitoring
sites, the distance to the NWS site is within 30 miles.  For cases where
the distance between the air quality and meteorological site is great
and/or where there are significant topographic differences separating
the sites, the meteorological data may not be fully representative of
conditions at the air monitoring site.  This issue should be taken into
consideration when evaluating such data.  

Pollution roses for the 58 areas included in the designations process
are provided in Appendix M.

An example pollution rose and detailed description is shown below.

Pollution Rose Key

The top line of the title shows the “potential” nonattainment area
in which the monitoring site is located (or nearby).  The
“potential” areas are typically based on (in this priority order)
existing PM2.5  nonattainment area boundaries (as prescribed in the 2006
FR notice), a combined statistical area (CSA) if the site is part of
one, a core-based statistical area (CBSA) if the site is located in one
of those that does not map to a CSA; or else just a county-state if the
county is not part of CBSA.  The area name in the top title is followed
by the county and state names in parenthesis.   A subtitle (the third
printed line) identifies the AQS site code represented by the plot.  The
site in the example plot has an AQS site ID of 42095002, and is located
in Northampton County, PA, which is identified as part of the potential
nonattainment area of Allentown, PA.

A legend in the right top corner shows the symbol colors used to
identify the daily concentration range and the symbol markers used to
show season.

The rose plot body shows the individual 24-hour average concentrations
for the site plotted around the site (center) directionally relevant to
the compass bearing the wind was blowing from and distance-relevant
according to the resultant wind speed that day.   The example plot shows
that there are sample days occurring in every wind quadrant but that
most of the days appear to be plotted to the left (or west), more
specifically between NNW and SSW.  Most of the high concentration days
for the example site appear to occur when the wind blows from the SW at
speeds of about 4-8 miles per hour. 

2005-2007 PM2.5  summary information for the plotted site is shown in a
table at the bottom left of the figure.  The table shows the annual 98th
percentiles, annual counts of (non-concurred) NAAQS exceedances, and
also the 3-year 24-hour standard design value (shown in red if violating
the NAAQS and in blue if not).  A note below the table tells how many
non-concurred exceedances were not plotted because of missing met data
or variable wind conditions.  In the example plot, the site had 23 total
exceedances over the 3-year period 2005-2007: 8 in 2005, 7 in 2006, and
8 in 2007.   The annual 98th percentiles for the site are 36.2 µg/m3
for 2005; 38.3 µg/m3 for 2006; and 37.9 µg/m3 in 2007.  The 3-year
design value for the site is 37 µg/m3.  According to the note below the
table, 2 exceedances (out of the 23 total) were not plotted because of
missing met data or variable winds.

A two-line green-font text entry at the bottom right corner of the
figure identifies the specific NWS site from which meteorological data
were used to generate the plot (bottom line), and the distance between
that NWS site and the ambient monitoring location and (top line).  In
the example plot, met data from the Allentown Lehigh Valley
International Airport (NWS ID = 1437) were used to generate the rose
plot; that NWS site is approximately 5.9 miles from the PM2.5 monitoring
site.   A key for the plotted wind speeds (i.e., the concentric circles)
is shown just below the plot body (above the “5” in the example). 

3.1.7	Factor 7: Geography/Topography 

The geography/topography factor involves an examination of physical
features of the land that might have an effect on the airshed and,
therefore, on the distribution of particulate matter over an area. For
example, an area located in a valley bordered by mountains could
experience very different effects to the airshed from an area with
generally uninterrupted flat terrain.  Topography was an issue primarily
in western nonattainment areas.  Topographical maps are included in a
number of  area-specific technical analyses.

3.1.8	Factor 8: Jurisdictional Boundaries 

The jurisdictional boundaries factor involves consideration of existing
boundaries that may facilitate or affect implementation of programs to
improve air quality.  The planning and organizational structure of an
area was considered to provide insights into how air quality planning
and enforcement is carried out in a potential nonattainment area. 
Examples of jurisdictional boundaries include counties, air districts,
Tribal Reservations, metropolitan planning organizations (MPOs), and
existing nonattainment areas. 

3.1.9	Factor 9: Level of Control of Emission Sources 

The level of control of emission sources factor involves the
consideration of emission control programs and associated emission
reductions that would be currently in place at the time the designations
are finalized.  In considering this factor, EPA recognized that the most
recent national emission inventory data available was from the year
2005.  EPA requested that the States and Tribes provide any additional
information regarding emission controls put in place through 2008.  EPA
also used information maintained by EPA’s Clean Air Markets Division
on power plant emissions and future planned emission controls.  Data on
future planned emission controls for electric utilities were obtained
from the National Electric Energy Data System (NEEDS) database and
unit-level annual emissions data was obtained from EPA’s national
electric utility emissions database.  The NEEDS / emissions database
(May 2008 version) used in the designations process is included as
Appendix N. 

3.2	Contributing Emissions Score

The Contributing Emissions Score (CES) is an EPA-developed method to
help identify counties that are potentially contributing to violations
of the 2006 24-hour national ambient air quality standards (NAAQS) for
fine particles (PM2.5). The CES is a tool EPA used in conjunction with
the factors recommended by EPA guidance to assess the contribution, if
any, of a given area to the ambient air quality of a nearby area not
meeting the 2006 24-hour PM2.5 NAAQS.  The CES results help to determine
an appropriate potential nonattainment area boundary for an area that
should be designated nonattainment for the 24- hour NAAQS. It expands on
the methodology previously used for calculating the Weighted Emissions
Score (WES), which EPA developed to help inform decisions about nearby
contributing areas for the PM2.5 NAAQS during the nonattainment area
designation process completed in 2004. That process addressed areas
violating the 1997 annual PM2.5 NAAQS. The CES is only one piece of
information considered by EPA in the analysis for evaluating PM2.5
nonattainment areas for the 2006 24-hour PM2.5 NAAQS. 

For the 24-hour PM2.5 designation process, the CES incorporates
additional information that was not included in the WES calculation,
reflecting differences between an annual versus a 24-hour standard. It
specifically considers the influence of emissions and meteorology on
days with high 24-hour fine particle concentrations, and incorporates
the seasonal composition of fine particle mass and source receptor
distance relationships.  The latter information is relevant for a
24-hour standard, for which individual days affect attainment status,
unlike an annual standard. The CES estimates the effect of pollutant
transport with a method similar to the Weighted Emissions Potential
(WEP) and the Emission Impact Potential (EIP). These daily PM2.5
assessment techniques, developed by non-EPA authors, use wind
trajectories and distance to better characterize the potential impact of
nearby areas. It is important to note that the CES represents a very
generalized approach and has been implemented as such across the country
using only county level data for consistency purposes. For this reason,
EPA recognizes that if more refined or updated data exist for a specific
area (e.g., higher resolved meteorological data or more recent emissions
data), state and local agencies have the opportunity to provide that
information to EPA for the revision of the Contributing Emissions
Scores. 

 “Vector average” means that the daily 24-hour wind speed and wind
direction have been calculated by considering 24 individual observations
of wind speed and direction, and summing (and then averaging) these as
vector quantities.  If the wind direction is constant during the 24
hours, the 24-hour speed will simply be the average of the hourly wind
speeds.  If the wind direction changed from hour to hour, the 24-hour
speed will be less than the simple average of the hourly wind speed and
the direction will be a type of average of the hourly directions. 
Together, the vector resolved wind direction and speed represent the net
forward path of an air mass during the 24 hours, accounting for any
meandering due to shifts in wind direction.

 For example, if the wind was from the east for 12 hours at one speed
and then from the west for 12 hours at the same speed, the vector
resolved wind speed and direction would both be zero.  The ratio of
scalar to resultant wind speeds would be zero.  Such a day would not be
plotted on the wind rose because doing so could give the impression that
there was little transport potential when in reality there was
opportunity for the air measured at the site to have been appreciably
influenced by emissions at some distance.  

 These concepts were originally presented to an audience of
representatives from

USEPA, state and local environmental agencies at the PM2.5
Implementation Program

and the Area Designation Process for the 2006 PM2.5 Standards workshop
in Chicago, IL )

 PAGE   

3- PAGE   4 

