EPA-450/4-74-007
(OAQPS NO. 1.2-025)
GUIDELINES FOR AIR QUALITY
MAINTENANCE PLANNING AND ANALYSIS
VOLUME 6 :
OVERVIEW OF AIR QUALITY
MAINTENANCE AREA ANALYSIS
ENVIRONMENTAL PROTECTION AGENCY
Office of Air and Waste Management
Office of Air Quality Planning and Standards
Research Triangle Park, North Carolina 27711
September 1974
-------
OAQPS GUIDELINE SERIES
The guideline series of reports is being issued by the Office of Air Quality
Planning and Standards (OAQPS) to provide information to state and local
air pollution control agencies; for example, to provide guidance on the
acquisition and processing of air quality data and on the planning and
analysis requisite for the maintenance of air quality. Reports published in
this series will be available - as supplies permit - from the Air Pollution
Technical Information Center, Research Triangle Park, North Carolina
27711; or, for a nominal fee, from the National Technical Information
Service, 5285 Port Royal Road, Springfield, Virginia 221 51.
Publication No. EPA-450/4-74-007
(OAQPS Guideline No. 1.2-025)
-------
FOREWORD
This document is the sixth in a series comprising Guidelines for Air
Quality Maintenance Planning and Analysis. The intent of the series is to
provide State and local agencies with information and guidance for the prepa-
ration of Air Quality Maintenance Plans required under 40 CFR 51. The volumes
in this series are:
Volume 1: Designation of Air Quality Maintenance Areas
Volume 2: Plan Preparation
Volume 3j_ Control Strategies
Volume 4: Land Use and Transportation Consideration
Volume 5: Case Studies in Plan Development
Volume 6: Overview of Air Quality Maintenance Area Analysis
Volume 7j_ Projecting County Emissions
Volume 8: Computer-Assisted Area Source Emissions Cridding
Procedure
Volume 9_^ Evaluating Indirect Sources
Volume 10: Reviewing New Stationary Sources
Volume II: Air Quality Monitoring and Data Analysis
Volume 12: Applying Atmospheric Simulation Models to Aij^ Quality
Maintenance Areas
Additional volumes may be issued.
All references to 40 CFR Part 51 in this document are to the regulations
as amended through July I974.
NOTE
This guideline is being released in its present form in order to
allow its immediate use by State and local agencies. This guideline
may be reissued in the near future in order to incorporate comments
and suggested improvements offered by the EPA Regional Offices and by
State and local agencies and other concerned groups.
-------
PREFACE
The purpose of Volume 6 1s to present an overview of the role
of analysis in designating air quality maintenance areas, in determining
whether a maintenance plan is needed and in evaluating whether a proposed
plan is sufficient to avoid exceeding ambient air quality standards.
Volume 6 presents a brief discussion of the several elements needed
for a complete analysis. Elements of analysis required to estimate
present air quality include the present emission inventory, air quality
data base and meteorological data base which may be used in atmospheric
simulation models of varying complexity. Model estimates are used to
supplement the observations obtained by monitoring to assess present
air quality. Additional elements needed to estimate future air quality
levels are the projection and allocation of changes in emissions which
may result from implementation of various air quality maintenance plans
or by default. Projected future levels of emissions must be related
to future air quality using atmospheric simulation models.
Next, the roles played by the various elements of analysis in
assessing the need for an air quality maintenance plan and in the
formulation of such plans are discussed more specifically. Finally, a
synopsis of the Volumes in the Guidelines for Air Quality Maintenance
Planning and Analysis which deal specifically with analysis (i.e.
Volumes 7-13) is presented. Interrelationships among the contents of
each Volume are described in terms of previously discussed relation-
ships among the elements of analysis.
iv
-------
TABLE OF CONTENTS
Preface i v
List of Figures vi
List of Tables vi
1.0 Introduction 1
2.0 Elements of Analysis 3
2.1 The Emission Inventory 4
2.2 The National Emission Data System 5
2.3 Inventory Projection 6
2.4 Air Quality Measurement 8
2.5 The SAROAD Air Quality Data System 10
2.6 Meteorological Measurement 10
2.7 Atmospheric Simulation Models 12
3.0 Air Quality Analysis as Applied to Air Quality Maintenance- • • .14
3.1 Update of Emission Inventory, Air Quality and
Meteorological Data Bases 15
3.2 Estimates of Current Air Quality 16
3.3 Air Quality Projection 18
3.4 Analysis in AQM Plan Formation 19
3.4.1 Air Quality Maintenance Strategies 21
3.4.2 Air Quality Maintenance Tactics 25
3.5 Analysis of Proposed New Sources 28
4.0 Framework for Air Quality Maintenance Analysis Guidelines ... .30
5.0 References 34
Bibliographic Data Sheet 37
-------
LIST OF FIGURES
1. Flexible Air Quality Maintenance Process
LIST OF TABLES
1. Role of Analysis Guidelines in Assessing the Need for
Maintenance Plans and in Plan Formulation
v1
-------
1.0 INTRODUCTION
Air quality analysis is concerned with estimating quantitative
answers to the following questions:
1. What is the present distribution of emissions?
2. What is the present distribution of pollutant concentrations?
3. What is the relationship between pollutant emissions and
concentrations?
4. How are total emissions within the area of interest likely to
change due to alternative rates of growth and development, alternative
emission control regulations and incentives, and alternative land use
and transportation plans?
5. How are emission patterns likely to change?
6. What pollutant concentrations are associated with those
alternative emission distribution scenarios?
7. How might emissions be limited in order to stay within the
set of compatible emission distributions?
8. How shall the impact of proposed new stationary point or
indirect sources be related to air quality goals?
These questions lead to three general analytical activities:
(a) the development, maintenance, and improvement of the data base
and techniques for relating the data to present air quality (questions
1-3);
(b) the description of probable air quality given an emission
scenario (questions 4-6); and
(c) the prescription of a set of emission distributions given an
air quality goal (questions 7 and 8).
1
-------
Activity (a) includes measurement of pollutant concentrations,
emissions, regional meteorology and development of models relating
concentrations to emissions for various meteorological conditions.
Models are then tested against measured conditions to see how well
they work (validation), adjusted in order to maximize their accuracy
(calibration), and improved when new or better data indicate that
such improvement is desirable and possible.
Activity (b) involves the application of the model to alternative
emission patterns so as to estimate their air quality implications.
Elements of this activity usually Include some emission projection
and manipulation prior to the application of the model. Typically,
the present emission inventory may be altered by estimating the
effects of emission control and new source development. Appropriate,
available models may then be applied to estimate the air Quality (or
change in air quality) associated with the new emission inventory
(or change in the emission inventory). In other words, Activity (b)
involves the use of models to make a series of "if—then" statements
which estimate the air quality associated with emission scenarios.
In performing Activity (c) models are applied in reverse to
derive a set of emission patterns which are compatible with a given
air quality goal or standard. This compatible set defines limits
within which the emission pattern must He if the air quality goal or
standard is to be achieved and/or maintained. Some of the emission
patterns in the compatible set will be unrealistic, so a thorough
-------
presceiptive analysis would proceed to narrow the possibilities by
excluding those patterns that are incompatible with social, economic,
and technological realities even though they would produce acceptable
air quality. Realistic patterns can only be derived after thorough
coordination with personnel and agencies responsible for other aspects
of the area's economic and social welfare. This requirement is needed
because emission limitation and allocation impacts on both the rate
and form of land use and transportation development, and these, in turn,
affect property values, economic development, and future expectations.
Furthermore, air quality is not the only environmental consideration
which constrains the form and/or intensity of development. Other forms
of pollution (water, noise, solid waste) and a host of resource
conservation issues are involved in the planning and management of the
future. While air duality constraints and constraints imposed by other
environmental requirements may be supplied piecemeal, the technical,
legal and managerial task of learning to live within these constraints
requires unprecedented coordination among the many groups in the public
and private sectors which influence development trends.
2.0 ELEMENTS OF ANALYSIS
Elements of analysis include development of a current emission
inventory, projection of future emissions and hypothetical emission
patterns, measurement and interpretation of air quality data, measure-
ment and interpretation of meteorological data, use of an inventory
and measured data to construct and calibrate appropriate atmospheric
simulation models, and use of models to project future air quality
3
-------
associated with projected emissions.
2.1 The Emission Inventory
The emission inventory is central to any air quality control
program, for the only practical means for altering air quality is
by the limitation, reduction, and redistribution of emissions. The
inventory is used to identify the sources of emission, to formulate
control strategies, as a base from which to construct future emission
scenarios and as a principal input to atmospheric simulation models.
An emission inventory is constructed from a source inventory,
emission factors and supplemental source inquiry. The source
inventory consists of a list of the number, size, location and (in
the case of large sources) emission release characteristics affecting
the height of a source's effluent. Only "point sources" are listed
individually. A point source, for purposes of diffusion modelling,
is defined as one which emits over 100 tons/year of any pollutant.
The numerous small sources are listed collectively as area sources.
They may be aggregated on a county basis or disaggregated to a network
of grid squares within counties using such schemes as the Computer
2
Assisted Area Source Emissions Gridding Procedure (CAASE).
Emission factors are used to estimate the emissions from a
source or aggregation of sources given the source type, size, process,
number, or other relevant indicator of activity level. Since it is
impractical to measure the emission of each and every emission source,
emission factors are frequently used to estimate emissions from a
4
-------
source, given certain of the source's design and operating characteristics.
These factors are developed by measuring emissions from representative
sources of a given type and using the assumption that similar sources
will emit at similar rates. Emission factors are compiled, published,
o
and updated in "Compilation of Air Pollution Emission Factors," AP-42.
Nationwide emission factors are, of necessity, typical or
average values. Direct measurement of emissions, inquiry of sources as
to their emissions, process, process rate, emission control equipment,
etc. are desirable, especially for large sources, to supplement the
use of emission factors. The use of specific known emission data for a
source is, of course, preferable to use of generalized emission factors.
Growth, attrition, modification and emission control con-
tinuously alter the emission pattern and require frequent, periodic
updating of the emission inventory. A systematic process for inventory
update not only maintains a relevant inventory, it also develops,
over a period of time, emission and development trend data on which
short range emission projections may be based.
2.2 The National Emission Data System
The wealth of detail inherent in an emission inventory and
the diverse forms in which data are retrieved from the inventory suggest
automatic data processing. A format for data collection, storage,
manipulation and retrieval called the National Emission Data System
(NEDS) has been developed by EPA. The NEDS system serves as: (1) a
central storage center; (2) a service to state and local agencies;
(3) an organizing system for the collection, modification, and updating
5
-------
of emission Inventories; (4) a base for National and sub-National
emission projection; and (5) the basic data set for formulating
National air pollution control strategies and for approving and
monitoring State Implementation Plans. It should, then, be obvious
why EPA Insists that all emission Inventories be kept 1n common
format and why a National Emission Data System has been developed.
2.3 Inventory Projection
Longer range forecasting of emissions Involves the
estimation of the number, type, and location of new sources, the
attrition and modification of existing sources, and the emission
factors applicable to both existing and new sources. These estimates
are then used to alter the existing or base emission inventory.
The estimation of the direction, rate, and form of future
changes must account for Inertia, design, and chance. Inertia is
the tendency of past trends to continue Into the future. The
strength of this factor depends on the persistence and uniformity of
the historical record. Design 1s the deliberate prescription and/or
limitation of development patterns, rates and practices through
regulation and incentive. The design factor limits, enhances, or
redirects the trajectory of past trends. Chance Includes all those
unknown and unexpected factors which influence the future. The
probable effects of some chance factors may be able to be described
statistically by noting the variance 1n past trends; others can be
limited by design. But forecasting in the social sciences is beset
6
-------
with a large residue of chance factors which are essentially
unpredictable.
The variance 1n the projected trajectory of past trends
and the possibility of altering the probabilities of various
trajectories through design lead one to speak of the future 1n terms
of scenarios—chains of possible events leading to alternative
futures. These scenarios are more like recipes than predictions.
They say, in effect, "Here 1s a set of possible futures; if you
like (don't like) one or more of them, then insure that the chain of
events leading there will (will not) occur."
An Important set of scenarios results from the projection
of past trends and current designs under the assumptions of no new
design and no major disruptions. These scenarios represent the range
of "expected" future conditions. They are useful 1n deciding whether
intervention (new design) Into expected trends is desirable (i.e.
whether an air quality maintenance plan 1s needed), and as a starting
place for Intervention strategy formulation. The user of such projections
should, however, be continuously aware of two caveats:
1. There is not one "expected" future. The trend data
upon which the projection is based has some variance. The further into
the future the projection goes, the greater 1s the spread or range of
the projection. It 1s the fringes of this range that should be
critically examined for acceptability, not the mean.
2. The "expected" future and Its variants comprise only
one subset of the possible futures. If the "expected" future is
7
-------
treated like a prediction to be accommodated rather than a scenario to
be evaluated, the result may be a "self-fulfilling prophecy."
The foregoing discussion of forecasting may be summarized
thus: the analytical tools for predicting the future are useful but
dull; they can narrow the range of possibilities but cannot pinpoint
what will, 1n fact, happen. The emission Inventory can be projected
a few years Into the future with some confidence by noting growth
trends, compliance schedules, and approved building permits. Beyond
that, the type, number, and especially the location of new sources
becomes increasingly difficult to predict.
2.4 A1r Quality Measurement
Both the amount of emission control needed and the progress
of a control program are indicated by actual measurement of pollutant
concentrations in the ambient air. The word "indicated" is used
advisedly, for air quality cannot be directly measured at all places
and times, and the confounding factor attributable to changing
meteorology frequently complicates the Interpretation of air quality
data. The use of monitoring data to estimate the spatial-temporal
distribution of pollutant concentrations must nearly always be
supplemented by the use of atmospheric simulation models.
The air quality monitoring and data system serves several
functions:
. Initially, it 1s the sole Indicator of air quality.
. It is used 1n conjunction with the emission Inventory
and meteorological data to select, and calibrate regional atmospheric
simulation models.
8
-------
. Once models are available, the air quality data serve
as a check on the accuracy and validity of the models and provide
a means for improving them.
. Where the legal acceptability of model estimates is
questioned, the measured air quality data provide supporting evidence.
. The monitor readings and the model estimates together
allow the estimation of the component of concentration due to natural
and extra-regional sources, and allow one to draw inferences about
air quality between monitoring sites.
Not only is it impractical to measure air quality every
place, but it is often inconvenient to measure it all the time. The
air quality monitoring network may be supplemented by a temporal
distribution model as well as by diffusion modeling. It has been
observed that pollutant concentrations generally are distributed log-
A
normally in time. This observation allows estimation of the frequency
of occurrence of various levels of concentration from sample measurements.
In other words, the log-normal model can be used to fill in the temporal
blanks in the same way that diffusion models fill in the spatial
blanks between monitors, provided there are at least some representative
monitoring sites operated continuously which can be used to estimate
the standard geometric deviation of air quality data in the area.
It will frequently be desirable to include mobile as well
as fixed monitors within a monitoring system. Mobile monitors might
be applied to the following situations:
9
-------
. To determine concentrations around the site of a
proposed large point or complex source as part of an environmental
impact statement or pre-permit investigation.
. To expand the areal coverage obtainable from a limited
number of monitors.
. To spot check on model estimates which appear too high
or low, or are legally disputed.
. To check local concentrations around existing large
point sources.
. To check CO concentrations near existing points of high
ireu i IL. ueiib i iy .
2.5 The SAROAD Air Quality Data System
The large quantity of data taken by a monitoring network
and the several statistical analyses and summaries routinely performed
on the data suggest automatic data processing and storage. The
Storage And Retrieval Of Aerometrlc Data (SAROAD) system has been
developed and should be used for this purpose. Detailed description
of the selection, placement and operation of monitors, and of the
SAROAD system has been documented in several EPA publications.
2.6 Meteorological Measurement
The capacity of the ambient air to assimilate wastes
without adverse effect depends on the atmospheric dilutive capacity.
This capacity can vary over a wide range with time and location;
therefore, it is highly desirable that meteorological conditions be
10
-------
specified when estimating the relationship between emissions and
the resulting concentrations.
Principal meteorological variables affecting air quality include:
0) Wind speed. The more air that moves by a source of
emissions, per unit of time, the less the concentration (mass/volume) will
be.
(2) Wind direction is obviously important in estimating the
location of concentration relative to the location of emission sources.
(3) Atmospheric stability is a composite variable describing
the turbulence, and therefore the dispersion rate, of the atmosphere.
The lapse rate (the rate of temperature change with height) is an
important indicator of stability.
(4) Mixing height is defined as the height above the surface
through which relatively vigorous vertical mixing occurs. It is
commonly identified as a layer of the atmosphere with a temperature
lapse rate conducive to strong mixing, "capped" by a temperature
inversion (temperature increases with height) which restricts such
mixing. A low mixing height restricts the volume of air throughout
which emissions within the layer may be dispersed. It is often
associated with high ground-level concentrations.
Several other meteorological variables are of interest in
special circumstances. Temperature affects air quality indirectly by
influencing the rate of emissions resulting from the demand for space
heating and air conditioning. Humidity and solar radiation intensity
are important to the rates of certain atmospheric reactions (formation
of oxidants, sulfates, aerosols).
11
-------
Variations in meteorological conditions for an area can be
described with several levels of precision. The level of precision
depends on the amount of data available. In general,
descriptions of meteorological variations involves four levels of
detail. At the least detailed level, meteorology variation is ignored
by assuming that worst case conditions will be duplicated annually.
The next level of precision assumes that measured conditions at a
nearby airport or U.S. Weather Service station are similar to
conditions over the entire region. The third level is attained
when conditions in a particular area are treasured in detail. This
is especially important in areas where interactions of specified
sources and their immediate surroundings are of interest. Finally,
the three dimensional wind field over the region may be estimated.
This requires numerous meteorological monitoring sites and the use
of various theoretical concepts such as mass balance, conservation
of momentum, etc.
Since the siting of instruments and the interpretation of
data frequently are dominated by local influences on the important
meteorological parameters it is not feasible to give simple, universally
applicable guidance on this important subject. It is suggested that
the services of A competent meteorologist be obtained to assist in
this portion of the analysis.
2.7 Atmospheric Simulation Models
The fundamental question in air quality management is: How
can air pollution be reduced/limited? The intuitively obvious answers
12
-------
are: (1) reduce/limit emissions, (2) raise the height of emission,
(3) move emission sources farther away (spread them out), and (4)
curtail/prevent emissions when meteorological conditions are especially
conducive to high concentrations. These qualitative prescriptions
Indicate the direction of measures for Improving and maintaining
acceptable air quality, but are insufficient for the development of
strategies for a specific reduction 1n concentration, maintenance
of concentrations below specified levels, or estimating the effects of
particular emission sources and measures to control them. In order
to estimate the effect of proposed measures and to justify the
expense of implementing those measures, the quantitative Impact on
air quality that will be achieved by a specified change in emissions
and/or conditions of emission release must be estimated. A mathematical
relationship connecting cause and effect in this manner is called
a model. Models may be used 1n four ways to support air quality and
management programs:
1. Atmospheric simulation models may be used 1n conjunction
with measured air quality to estimate the current air quality distribution.
2. Some model must be used to estimate the change in present
air quality which would be effected by a projected or hypothetical
change in the quantity and distribution of emissions.
3. Some model must be used to estimate the change in the
air quality distribution which would be effected by the addition of
a proposed new source to the emission inventory.
4. Available models and their data requirements should be
13
-------
used to rationalize the selection of data Improvement projects.
Indeed, Improvement of the accuracy and detail of the modeling capability
1s a major purpose of the data upgrade program.
Generally, models which are able to provide air quality
estimates having the most detailed temporal and spatial resolution
also require the most detailed emission and meteorological data bases.
The availability of models which can be used to reliably estimate
1n detail spatial and temporal variations in pollutant concentrations
greatly increases the number of air quality management schemes which
can be examined explicitly. Thus, 1t would be possible to distinguish
between two proposals resulting in the same overall amounts of emissions
but having different source configurations. The ability to estimate
air quality with a fine degree of temporal and spatial detail has
Important Implications on the responsibility to maintain adequate
air quality, because 1t enables the analyst to examine a great many
more cost-effective and less socially disruptive control strategies
explicitly. The benefits accruing with these Increased analytical
capabilities provide a strong Incentive for maintaining and improving
emission Inventory, meteorological and air quality data bases in
order to improve modeling capabilities.
3.0 AIR QUALITY ANALYSIS AS APPLIED TO AIR QUALITY MAINTENANCE
The air quality analyst has five general tasks associated with
air quality maintenance (AQM). The first is to develop and Implement
a procedure whereby the emission inventory, meteorological and air
quality data bases can be routinely improved and updated. Such a
14
-------
procedure provides a means for determining whether previously
projected growth and development is, in fact, occurring, and an
increasingly better capability for estimating present air quality
as well as that likely to occur in the near future. The second
task is the collection and periodic evaluation of the current air
quality data to see how well air quality is, in fact, being maintained.
The third task is a projection, performed at five year intervals
(at least), to estimate whether the combination of new emissions due
to general growth and development, and expected emission reductions
of existing sources, could lead to unacceptable air quality during a
10-year projection period. The fourth task is assistance in formu-
lating an emission limitation and allocation process which will
insure that air quality standards will be maintained. This task is
not required unless the projection indicates a reasonable probability
that a National air quality standard could be exceeded in the
projection period. The final task is the analysis of proposed new
point and indirect sources to see if any air quality standard would
be exceeded if their emissions were added to the emission inventory
or as the result of inadequate design and operating procedures at
the proposed source.
3.1 Update of Emission Inventory, Air Quality and Meteorological
Data Bases
Efforts should be made to initiate and implement a
routine reporting procedure whereby emission estimates from existing
15
-------
sources can be checked and updated as necessary. Similarly, emission
reporting requirements from new sources should be linked as closely
as feasible with existing requirements for construction and operating
permits so that the emission inventory will reflect the current
state of affairs as well as possible. To facilitate inclusion of
new data in automated data processing procedures and to aid in the
periodic revaluation of emission projections, it is highly desirable
that the new information obtained from the reporting procedures
described above be recorded in a standardized fashion (i.e. NEDS format).
Inclusion of the latest available emission information will greatly
assist in estimating current air quality accurately.
Initial application of models to estimate air quality
should identify the most urgent needs 1br air quality and meteorological
data in the area. Such an analysis should serve as the rationale
for improving these data bases in a systematic fashion within the
constraints imposed by the resources available for monitoring.
3.2 Estimates of Current Air Quality
The success of existing air pollution control programs and
the ability to accommodate new emission sources in the region are
measured by the historical record and present state of the air quality
distribution. Furthermore, the detail and accuracy with which the other
AQM tasks are performed depend on the detail and accuracy of periodic air
quality distribution estimates. So the performance of this task and
the improvement of the tools needed to carry it out should receive
serious and regular attention.
16
-------
The analyst should be 1n a position to answer five questions
about the air quality distribution: (1) what is it now? (2) how has
it changed in the past several years? (3) how has it responded to
changes in the emission inventory? (4) what is the present relationship
between the emission inventory and the air quality distribution? and
(5) how would it be changed by a change in the emission inventory?
Questions 1-3 can be answered with the records of emissions and
concentrations from the emission inventory and monitoring network.
The answers to questions 4 and 5 require the use of models.
The next step, then, in AQM analysis, after provision for
routine updating and improvement of emission air quality and meteorological
data, 1s the assembly of the emission, inventory, monitoring data and
meteorological data, and the selection and exercise of the best models
that the quality of that data will support, to determine the present
air quality distribution, in space and time, for the baseline year.
In many areas, the state of the emission inventory, monitoring
network, meteorological data, technical expertise and computing
facilities will not allow the use of other than very simple models
for the first (1975) AQM iteration. It should be recognized that
simple models limit the accuracy and detail of the analysis, and also
limit the strategy options that can be considered if a SIP revision
for AQM is necessary. For example, the effects of alternative
emission patterns resulting from land use and transportation options
cannot be evaluated unless a model is available that will accept
emission location as an input. It is particularly important that
17
-------
improvement of atmospheric simulation modeling capability, and the
data quality improvement that 1t Implies, be major objectives of the
air pollution control agency.
3.3 Air Quality Projection
The constraints on emissions, existing emissions, and capacity
of the atmosphere to assimilate new emissions should be known to developers
and land use/transportation planners in all urban areas so that develop-
ment patterns leading to unacceptable air quality are not proposed or accepted.
The assessment of future air quality potential is done in
two steps. The first Is the application of quick-look procedures, as
given in "Guidelines for Designation of Air Quality Maintenance Areas,"
OAQPS No. 1.2-016. The second step 1s a more thorough analysis of the
air quality potential for the purpose of assuring that a SIP revision
for maintenance is really necessary and for describing the nature and
extent of potential air quality problems.
The more thorough analysis to determine whether a maintenance
plan is needed is, in turn, divided into 4 steps:
(1) Estimation of current air quality, as described in
the previous section;
Q
(2) Projection of county-wide emission growth potential;
(3) Allocation of projected emissions on a subcounty
basis;10'11
(4) Application of projected emission data in atmospheric
12
simulation models to estimate future air quality levels.
18
-------
Depending on the sophistication and spatial resolution of
available models, it may be possible to by-pass step (3). For example,
the disaggregation of projected county non-methane hydrocarbon (NMHC)
and NO emissions may not be necessary for the 1975 analysis if
^
models capable of estimating the effect of small emission pattern
changes on oxidant and NCL concentration distribution are not generally
available. This is not to say that the location of these emissions
makes no difference, but rather that the difference cannot presently
be quantified. Models capable of accepting NMHC and NO emission
/\
patterns should be available by 1980 or sooner, so the preparation
of a gridded emission inventory for non-methane hydrocarbons and MO
/\
should be a goal of the emission inventory upgrade program.
If the analysis uses a model which accepts source location
as an input, then urban planners should be consulted to select a
probable worst case distribution of projected emissions. Remember
that the AQM analysis is to establish only whether a reasonable
probability of NAAQS exceedance exists through 1985 without further
constraints on land use or transportation or more stringent emission
regulations. Therefore, the projection is concerned with what could
happen, not what will happen.
It is prudent to repeat at this point that a projection is
not a prediction. The probable worst case projections asked for in
the analysis phase of the AQM process are scenarios which need not occur.
3.4 Analysis in AQM Plan Formation
The SIP should contain a set of legally enforceable measures
19
-------
which restrict the rate and distribution of emissions from mobile and
stationary sources so that the air quality distribution remains within
acceptable limits. Under the Clean Air Act "acceptable limits" are
presently defined by Primary and Secondary NAAOS. In the case where
Secondary NAAOS cannot be attained in the near-term future due to the
lack of appropriate fuel or control technology for existing sources,
attainment of Secondary NAAPS may be programmed for a more distant
date. Regulations relating to "significant deterioration" may, in
the future, prescribe acceptable limits more stringent than secondary
standards for TSP and S02.
In any case, the SIP must show how the pollutant concentration
limits are to be related to emission limits, and how the emissions are
to be maintained within those emission limits.
The analytical tools for developing the SIP Revisions
will have been explicated during the AOM analysis. These will be:
an updated emission inventory, air Quality and meteorological data, models
based on current and historical data, projections of future development,
emission estimates for existing and expected sources, and projected air
quality. The results of this analysis will have shown that the potential
amount and distribution of emissions of one or more pollutants are
not consistent with secondary air quality standards through 1985
(otherwise a SIP revision (AOMP) would not be needed). The analytical tasks
associated with SIP revision are: (1) to use the available models to
20
-------
explore the emission distributions that would be compatible with
acceptable air quality objectives, (2) to suggest feasible emission
limitation measures for maintaining emission distributions within the
compatible set, (3) to test suggested measures against the air quality
objectives using available models, (4) to assist land use and trans-
portation planners 1n defining the air quality constraints on those
activities, (5) to develop a periodic air quality assessment program
for monitoring the effectiveness of the SIP, and (6) to maintain a
program for updating and upgrading the data and models needed for
the foregoing tasks.
Each of the above tasks has been discussed earlier and is
discussed in detail in Volumes 7-13 in the Guidelines for Air Quality
Maintenance and Analysis series. The remaining discussion will be
devoted to the role of analysis 1n selecting certain strategic and
tactical options which may be features of an air quality maintenance
plan.
3.4.1 Air Quality Maintenance Strategies
Technically, there 1s only one AQM strategy, and that is
to keep the emission pattern within the acceptable set. This requires
a knowledge of what the acceptable set of emission patterns 1s (through
the use of valid atmospheric simulation models), the authority to deny
building or operating permits which would produce an emission pattern
outside the acceptable set, and the will to use that authority.
While numerous emission patterns may be compatible with
AQM, their effects on land use potential--who gets to emit how much
21
-------
and where—may differ substantially. The principal difference between
alternative AOM strategies is the way they address distribution
problems—such questions as the land use, property rights, development
pattern and development rate.
Numerous approaches to AOM are conceivable, but three
will serve for illustrative purposes. These may be called accommodation,
specification, and flexible. The accommodation strategy temporarily
avoids the distributional AOM problems by pushinq them into the future.
This is done by requiring sufficient reduction of existing emissions
(through technological innovation, attrition, etc.) to compensate for
expected new emissions for a qiven time period (e.g., 5 years). No
consideration is given to the spatial and temooral distribution of
emissions.
The specification strategy approaches the distributional
aspects of AOM directly. A specific emission distribution pattern
is selected from the acceptable set and imposed directly on the land
through emission density zoning and vehicle density limitations. The
pattern selected is not the prerogative of the analyst, but of the
public and planners.
The flexible strategy represents the broad middle ground
between the previous, extreme strategies. The emissions from proposed
new construction and land development projects are added to the existing
emission inventory and the result is tested for compatibility with
AOM using emission-air Quality models. Both expected reduction in
22
-------
existing emissions and land use/transportation plans may also be
considered during the new project impact analysis. A necessary
condition for construction permit approval is the estimation that
the emission pattern would remain within the acceptable set.
Sufficient conditions for permit approval should include conformity
with other land use/transportation plans and restrictions.
The type of strategy selected depends, on the public
acceptance of the constraints necessary to carry them out and on the
analytical capabilities which exist in the air quality maintenance area.
Generally, the accommodation approach requires the l^east analytical
sophistication, while the specification strategy requires not only a
sophisticated approach, but a preat deal of confidence in the approach
as well.
The accommodation strategy was incorporated in the initial
SIPs. Needed emission reduction was identified and required in the SIPs
to reduce existing (1970) concentrations plus the increases in those
concentrations expected due to growth of emissions in the 1970-1975 period
to levels which were consistent with NAAOS. AOM could be approached by
a series of such 5 year plans. One problem with this approach is the
"moving target" presented to large, existing sources in order to accommodate
future growth. The installation of new, more efficient control equipment
at frequent intervals would be both disruptive and expensive. Another
problem involves long-range major land use and transportation planning
projects. A five, or even ten, year time horizon is too short for major
land use and transportation planning. A third problem with this strategy
is that the model on which it is based (proportional rollback) ignores
location of emissions.
23
-------
The specification approach occupies the opposite end of the
spectrum of strategies. In order to choose one of the many emission
patterns from the compatible set identified analytically, qreat confidence
must exist in the accuracy of predicting air quality from emissions
and in predicting future social, economic, and technological desires
and capabilities. Such confidence is presently lacking. Nevertheless,
there is a powerful reason for attempting to develop the analytical
capabilities to make such a strategy feasible. In order to avoid the
obvious equity problems involved in a first-come-first-served
allocation of emissions, emission rights could be assigned to the land
as part of the zoning restrictions on that land, and land use.
Transportation routes, and transportation models could be designed
to be consistent with each other, land use requirements and with AQM.
A Utopian goal? Certainly the analytical techniques needed to define
the "optimum" combination of planning elements, including AOM, is
Utopian. However, if total emissions in a region must be limited,
then some rational and eauitable method of allocating those limited
emissions to potential "users" must be selected or else the allocation
will probably be irrational and inequitable. The allocation of emission
ceilings to the land does involve difficult decisions as to which land
receives which ceilings, but so does any other method of distributing a
limited quantity of emissions,
The difficulties of predicting future events, conditions
and capabilities lead to the desirability of a flexible process for
AOM rather than a static plan. Such a process is also suggested by
24
-------
the multiplicity of emission patterns compatible with AOM. If
models are available which can test the compatibility of a large
proposed new source, or a major development or rezoning request for MM
compatibility, then the process approach to AQM can he implemented. A
diagram of the "flexible process" approach and the role of analysis in
such an approach is shown in Figure 1.
The key feature of the flexible AQM process is the
building permit, One condition for permit issuance would be compatibility
of the proposed construction with ADM considerations. Those con-
siderations would include, at a minimum, the calculation of the combined
air auality impact of the proposed and existing emissions and the
assurance that air quality limits would not be exceeded. A second
consideration, closely related to land use/transportation planning,
would be that the proposed source was the type desired in that location
and that its emissions would not preclude other development likely or
desired in the vicinity. In order to avoid saying "no" to valuable
development due to prior commitments, some advanced planning is clearly
necessary.
3.4.2 Air Quality Maintenance Tactics
Tactics may be thought of as the means by which maintenance
strategies are implemented. Six types of emission control tactics
include: (1) source category emission standards, (2) fuel quality
standards, (3) emission disincentives (charges, taxes), (4) land-based
emission limits, (5) supplementary controls (stack height, time variable
emission limits), and (6) indirect tactics (land-use pattern, transportation
25
-------
Hin ^UHLII
BACKGROUN
METEOROLO
MEASURED
AIR QUALITY
FROM MONITORING
NETWORK
EMISSION
INVENTORY
GY AND TOPOGRAPHY
i
MODEL 4_
REVISION
GROWTH TRAS?P°£T
FORECAST -* PLANNI
EXPECTED EMISSION
FACTOR REDUCTION
AND SOURCE
ATTRITION
PROPOSED
SOURCE EM
REVISE
INVENTORY
* NO *
A-^DISAPPROVE) t
REVISE ^N VF<; m^
ALLOCATION "APPROVE)
f
L
,,'
EMISSK
ALLOCATIOI\
AND PRIOR
NFW
SSIONS
YES
CAN/SHOULD (DI$AP
\LLOCATION BE
REVISED TO «
ACCOMMODATE?
CALCULATED
AIR QUALITY 4—
DISTRIBUTION
AND
ATION A <•<— -•
NG
)N
PLANS »
ITIES
1 ' 1
' i
r
EMISSION AND
AIR QUALITY
MODELS
i
r
ALLOWABLE
EMISSION
DISTRIBUTION
ALTERNATIVES
4
EMISSIONS
AVAILABLE FOR
DISTRIBUTION
I
EMISSION
ACCOUNTING
AND PERMIT
APPROVAL
i
r
+ ARE THE AVAILABLE EMISSIONS
SUFFICIENT TO COVER THE
W 4— PROPOSED SOURCE? (MANDATORY)
PROVE)
IS THIS ALLOCATION
N0 COMPATIBLE WITH PLANS AND
PRIORITIES? (DESIRABLE)
Figure 1. Flexible air quality maintenance process.
26
-------
routes and modes, qrowth policy, etc.). These tactics can be combined
in endless ways to implement one or a mixture of AQM control strategies.
The first three tactics are designed to limit emissions from individual
as well as groups of sources. Atmospheric simulation models relating
emissions to air quality are needed to assess the effect of such
tactics in improving air quality. Depending on the type of source
and the pollutants of interest, use of point, area and/or line source
models may be required. The fourth tactic has the same objective as
the first three, but it is more suitable for analysis using area source
models. The fifth tactic is to take fuller advantage of atmospheric
dilutive capacity and/or short term variations in that capacity. For
such a tactic to be successful, highly reliable analytical techniques
for meteorological prediction and source contribution estimation,
as well as clear placement of responsibility for difficult control
decisions and enforcement, are necessary. For these reasons, EPA
policy has been to confine such tactics to isolated point sources.
Nevertheless, urban sources should utilize stacks with heights as needed
in accordance with "good engineering practice." The sixth tactic
includes those measures which influence the rate or pattern of emissions
by affecting the type, pattern density and/or intensity of activity in the
area. Land use and transportation planning clearly fit this category.
Growth policy is also an indirect air quality control tactic. These
tactics are indirect because they do not limit the amount or location
of emission specifically. Industrially zoned land may be developed
for any number of source types whose emission factors may differ
27
-------
by orders of magnitude. The emissions from a transportation corridor
depend on the mix of transportation modes employed. The emissions
associated with a given population depend on the mix and pattern of
the urban activities.
Land use and transportation patterns are nevertheless
highly influential over the types of air quality control measures
that can be practically employed. Conversely, the state of emission
control technology limits the extent and intensity of certain land
uses. For such tactics to be effective 1n maintaining air quality
within defined limits, 1t 1s essential that analysis by the air
pollution control agency be closely coordinated with the work of
land use/transportation planners. The analysis performed by the
control agency using models with locatlonal inputs can be used to
define a number of combinations of emission density limitations
associated with numerous patterns of areas with differing land uses.
Such a procedure should aid the planners in selecting sets of patterns
which are consistent with maintenance requirements and are at the
same time realistic and responsive to the area's social and economic
needs.
3.5 Analysis of Proposed New Sources
All point and Indirect sources of certain sizes and types
should receive an air quality impact analysis prior to approval
regardless of where they are located. The analysis for specific
proposed new sources is necessary, because the reliability with
which the assessment of the source's impact on air quality can be
28
-------
made is greatly enhanced once the exact location, design characteristics
(e.g. stack height, volume throughput for point sources, or peak
expected demand-capacity ratios for indirect sources) and size of
the source are known. Within an urban area, smaller and more types of
proposed sources should receive more scrutiny than is necessary in
undeveloped areas. If an area is operating under an AQM plan or process,
all new emission sources should be subject to some kind of limitations
on emissions, and/or location, but it is clearly impractical to subject
each one to an air quality impact analysis. The restraints on small
source development should be approached through design standards, emission
density zoning, urban design, transportation planning or other collective means.
Analysis for larger point and indirect sources should include
the following estimates: (1) the probable emissions created or induced
by the proposed source, (2) the probable concentrations that would
result from the proposed source's emissions, (3) the present and expected
concentrations due to existing and expected new emissions in the
area significantly affected by the proposed new source, (4) the sum of
(2) and (3) for each relevant pollutant and averaging time and comparison
with applicable air quality standards, and (5) (possibly) an estimate
of the new emissions that could be permitted in the proposed location,
or suggested locations where the proposed source could locate.
Analytical techniques which might be used to estimate the impact of
individual new sources are suggested in Volumes 9 and 10 in Guidelines
14 ic
for Air Quality Monitoring Planning and Analysis. *
29
-------
4.0 FRAMEWORK FOR AIR OUALITY MAINTENANCE ANALYSIS GUIDELINES
Since there are several Volumes 1n the Guidelines for Air
Quality Maintenance Planning and Analysis series which describe the
various elements of analysis, the review and use of these documents
is likely to be facilitated if their relationship to the analytical
procedure described in this volume is enuniciated. There are seven
Volumes (Volumes 7-13) which discuss the elements of analysis in
greater detail. Volume 7, "Projecting County Emissions," contains
sugg^.:t^d -.cthcdclcc;-:;::; fzr pr:jecting future el—:::4?": ?* •"»*+?<***
pollutants for the county as a whole, starting with an updated,
current emission inventory in an appropriate National Emission Data
System (NEDS) format. Several levels of effort for making county-wide
projections, with their attendant advantages and disadvantages, are
discussed. Volume P, "Computer-Assisted Area Source Emissions
Gridding Procedure," presents one method which may be useful in
allocating present and projected county-wide area source emissions
within the county. The methodology described in Volume 8 may be
particularly reliable for estimating distributions of pollutants
which are closely related to living patterns, into the near (e.g.
less than 5 years) future. Volume 13, "Sub-County Emission Allocation,"
is a more general treatment of rationales which might be used as
bases for allocating projected county-wide area source emissions and,
to the extent possible, projected point source emissions within the
30
-------
county. Volume 11, "Air Ouallty Monltorinq and Data Analysis,"
contains guidance concernlg monitoring network design and instrument
siting, acceptable instrumentation, air Quality data evaluation
procedures, interpretation of air quality data as it relates to
NAAQS, establishment of baseline air quality levels and air quality
trends evaluation. The information in Volume 11 serves as useful
guidance for updating and Improving air cuality data bases. In
addition, interpretation of air quality data in accordance with
the guidance in Volume 11 will aid in the efforts to validate
and/or calibrate reliable atmospheric simulation models. Volume 12,
"Applying Atmospheric Simulation Models to Air Quality Maintenance
Areas," discusses the data requirements needed by several categories
of models. Such information can be of use in identifying the
level of effort needed for monitoring meteorological variables in
an area and in identifying key emission information which should
have a high priority. Volume 12 also describes the spatial and temporal
resolution available with various types of models, as well as the
pollutants for which each type of model is appropriate. Examples
of each type of model are cited and described briefly. Volume 12
can be used as a means for weighing the data requirements (and
their attendant costs) of each modeling approach against the analytical
capability afforded with each approach. An essential part of an
air quality maintenance plan is the review of major new proposed
sources. This requirement results from (1) the need to determine
whether a proposed source is consistent with a prescribed maintenance
31
-------
plan, and (2) the increased reliability of modeling techniques
when specific locational, size and design Information about a
source is available. Volume 10, "Reviewing New Stationary Sources,"
and Volume 9, "Evaluating Indirect Sources," address the need for
impact analysis of specific sources. Table 1 summarizes the roles
of Volumes 7-13 in the analysis of air quality maintenance areas
and in the formulation of maintenance procedures.
32
-------
TABLE 1—ROLE OF ANALYSIS GUIDELINES IN ASSESSING THE NEED FOR
MAINTENANCE PLANS AND IN PLAN FORMULATION
Volume Number Functions
7 Project county-wide emissions to future years (out
to 10 years). Projections can be made to simulate
the "most probable" occurrence, or the "worst" of
a set of county-wide projections.
8, 13 Distribution of emissions projected using Volume
7 within the county. Distribution can be made to
simulate "most probable" distribution or the "worst"
of a set of realistic distributions. Methods can
also be used to distribute present area source
emissions within a county.
11 Used as guidance for developing and maintaining a
technically adequate air quality monitoring network
and in interpreting data from this system properly.
Such data can be used to validate and/or calibrate
atmospheric simulation models.
12 Used to relate present and future emissions to
present and future air quality under various
meteorological conditions. Used to determine the
set of emission development patterns which will
meet NAAQS, and to prescribe emission limitations
which should be met to ensure conformity with NAAQS.
9, 10 To evaluate the impact of a proposed source on air
quality in its immediate vicinity. Used to
determine whether a proposed source is consistent
with maintenance requirements and as possible bases
for granting or refusing to grant construction or
operating permits.
33
-------
5.0 REFERENCES
(1) U. S. EPA, OA, Applied Technology Division; "Guide for
Compiling a Comprehensive Emission Inventory," Publication No.
APTD-1135; A1r Pollution Technical Information Center, U. S.
EPA, Research Triangle Park, N. C. 27711 (June 1972).
(2) Research Triangle Institute; "Computer Assisted Area Source
Emissions Gridding Procedure (CAASE)"; Users Manual; Prepared for
EPA, OAQPS Under Contract No. 68-02-1014; Research Triangle Park,
N. C. 27711; (January 1974).
(3) U. S. EPA, OAWP, OAQPS; "Compilation of Air Pollutant
Emission Factors (second edition)"; Publication NO. AP-42; Air
Pollution Technical Information Center, U. S. EPA, Research Triangle
Park, N. C. 27711; (April 1973).
(4) Larsen, R. I.; "A Mathematical Model for Relating Air Quality
Measurements to Air Quality Standards"; Publication No. AP-89; A1r
Pollution Technical Information Center, U.S. EPA, Research Triangle
Park, N. C. 27711; (November 1971).
(5) U. S. EPA, OAQPS; "Air Quality Monitoring and Data Analysis";
Guidelines for Air Quality Maintenance Planning and Analysis; Volume 11.
OAQPS No. 1.2-030; (September 1974).
(6) U. S. EPA, OAP; "SAROAD Users Manual," OAP Publication No.
APTD-0663; (November 1971); Air Pollution Technical Information Center,
U. S. EPA, Research Triangle Park, N. C. 27711.
34
-------
(7) Fair, D. H.; SAROAD Station Coding Manual for Aerometric Sampling
Networks," OAP Publication No. APTD-0907; (February 1972); Air Pollution
Technical Information Center; U. S. EPA; Research Triangle Park, N. C.
27711.
(8) U. S. EPA, OAQPS; "Guidelines for Designation of Air Quality
Maintenance Areas"; Guidelines for Air Quality Maintenance Planning
and Analysis; Volume 1; OAQPS 1.2-016; (January 1974).
(9) U. S. EPA, OAQPS; "Projecting County Emissions"; Guidelines for
Air Quality Planning and Analysis; Volume 7; OAQPS No. 1.2-026;
(September 1974).
(10) U. S. EPA, OAQPS; "Computer-Assisted Area Source Emissions
Gridding Procedure"; Guidelines for Air Quality Maintenance Planning
and Analysis; Volume 8; OAQPS 1.2-027; (September 1974).
(11) U. S. EPA, OAQPS; "Subcounty Emission Allocation"; Guidelines
for Air Quality Maintenance Planning and Analysis; Volume 13; (September
1974).
(12) U. S. EPA, OAQPS; "Applying Atmospheric Simulation Models to Air
Quality Maintenance Areas"; Guidelines for Air Quality Maintenance
Planning and Analysis; Volume 12; OAQPS No. 1.2-031; (September 1974).
(13) 40 CFR 51.13(h); "Use of Supplementary Control Systems and
Implementation of Secondary Standards"; Federal Register; p. 25,700;
(September 14, 1973).
(14) U. S. EPA, OAQPS; "Evaluating Indirect Sources"; Guidelines for
Air Quality Maintenance Planning and Analysis; Volume 9; OAQPS No. 1.2-028;
(September 1974).
35
-------
(15) U. S. EPA, OAQPS; "Reviewing New Stationary Sources,"
Guidelines for A1r Quality Maintenance Planning and Analysis;
Volume 10; OAQPS No. 1.2-029; (September 1974).
36
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1 REPORT NO.
EPA 450/4-74-007
2.
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
Guidelines for Air Quality Maintenance Planning and
Qofufne §: Overview of Air Quality Maintenance Area
5. REPORT DATE
September 1974
6. PERFORMING ORGANIZATION CODE
7. AUTROR(S)
8. PERFORMING ORGANIZATION REPORT NO.
OAQPS Guideline No. 1.2-025
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Source Receptor Analysis Branch
Monitoring and Data Analysis Division, OAQPS, EPA
Research Triangle Park, North Carolina 27711
10. PROGRAM ELEMENT NO.
2AC129
11. CONTRACT/GRANT NO.
12. SPONSORING AGENCY NAME AND ADDRESS
13. TYPE OF REPORT AND PERIOD COVERED
14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
16. ABSTRACT
A description of all the major requirements in analyzing whether an area should be
designated as one requiring an air quality maintenance plan and whether a plan is
sufficient to meet designated air quality goals is presented. Elements of air quality
maintenance area analysis include design and update of emission, meteorological and
air quality data bases, use of dispersion models and air quality data to estimate
present air quality, projection of future emissions and emission distribution pat-
terns, use of models to estimate future air quality, and assessment of whether control
strategies are sufficient to meet air quality requirements in the future.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS
c. COSATI Field/Group
Air Pollution
Airborne Wastes
Atmosphere Contamination Control
Atmospheric Models
Meteorology
Air Quality Maintenance
Emission Inventory
Emission Projection
Air Quality Projection
13/02
13. DISTRIBUTION STATEMENT
Release Unlimited
19. SECURITY CLASS (ThisReport/
None
21. NO. OF PAGES
44
20. SECURITY CLASS (This page)
None
22. PRICE
EPA Form 2220-1 (9-73)
37
------- |