United States
Environmental
Protection
Agency
Office of Air Quality
Planning and Standards
Research Triangle Park, NC 27711
EPA-450/4-91-013
July 1991
AIR
GUIDELINE FOR REGULATORY
APPLICATION OF THE
URBAN AIRSHED MODEL
-------
-------
GUIDELINE FOR REGULATORY APPLICATION OF THE
URBAN AIRSHED MODEL
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Air Quality Planning and Standards
Technical Support Division
Source Receptor Analysis Branch
Research Triangle Park, NC 27711
July 1991
-------
-------
; DISCLAIMER
I '
| This report has been reviewed by the Office of Air Quality
Planning and Standards, U.S. Environmental Protection Agency, and
has been approved for publication. Any mention of trade names or
commercial products is not intended to constitute endorsement or
recommendation for use.
-------
-------
CONTENTS
Page
DISCLAIMER
TABLES v
|
ABBREVIATIONS . ............ vii
ACKNOWLEDGEMENTS . . . !. ix
1 INTRODUCTION i
2 MODELING PROTOCOL 5
I
2.1 Protocol Development Process .... 6
2.2 Contents of Protocol Document .... 8
3 DOMAIN AND DATA BASE ISSUES i........ 11
i
3.1 Episode Selection ! 11
3.2 Size of the Modeling Domain ;. 15
3.3 Horizontal Grid Cell Size 15
, 3.4 Number of Vertical Layers . . 17
3.5 Meteorological Data 19
j 3.5.1 Wind fields 19
3.5.2 Data needs for wind field i
development 23
3.5.3 Mixing heights ; 24
3.5.4 Clear-sky assumption for photolysis rate
calculations 25
3.6 Air Quality .'....' 26
3.6.1 Initial and boundary conditions 26
3.6.2 Performance Evaluation Data . 31
3.7 Emissions ............ 32
3.7.1 VOC speciation 35
| 3.7.2 Spatial gridding of area sources 35
| 3.7.3 Mobile sources . 36
3.7.4 Episode-specific adjustments ....... 37
; 3.7.5 Biogenic emissions 38
3.7.6 Point-source and plume-rise cutoff levels . 39
! 3.7.7 Consistency with national inventories ... 40
111
-------
CONTENTS f Continued)
Page
4 DATA QUALITY ASSURANCE AND MODEL DIAGNOSTIC ANALYSES . . 41
4.1 Step 1: Quality Assurance Testing of Component
Fields 43
4.2 Step 2: Diagnostic Testing of the Base Case
Meteorological episodes 44
4.3 Step 3: Additional Base Meteorological Episode
Sensitivity Testing 46
5 MODEL PERFORMANCE EVALUATION ........ 49
5.1 Performance Measures ..... 49
5.1.1 Graphical performance procedures ..... 51
5.1.2 Statistical performance measures ..... 53
5.2 Assessing Model Performance Results ........ 55
6 ATTAINMENT DEMONSTRATION ..... 59
6.1 Developing Attainment-Year Model Inputs ...... 59
6.2 Construction of Attainment Year Emission Control
Strategies ..... 60
6.3 Performing Attainment-Year Simulations to Assess
Various Control Strategies ..... 62
6.4 Using Modeling Results in the Attainment
Demonstration ..... 63
6.5 Exceptions to Guidance Document .......... 64
REFERENCES .
APPENDIX A RECOMMENDED MODELING PROTOCOL CONTENTS ,
APPENDIX B IDENTIFICATION OF METEOROLOGICAL REGIMES
CORRESPONDING WITH HIGH OBSERVED OZONE
65
69
77
APPENDIX C PERFORMANCE MEASURE FORMULATIONS . . 81
IV
-------
TABLES
Number
Example Table of Contents for Protocol
Document
2 Default Boundary Condition Concentrations for
Carbon Bond-IV Species .......... 28
-------
VI
-------
ABBREVIATIONS
AIRS Aerometric Information Retrieval System
BEIS Biogenic Emissions Inventory System
CAAA Clean Air Act Amendments
GARB California Air Resources Board
i
CMSA Consolidated Metropolitan Statistical Area
CO Carbon Monoxide
CSC Computer Sciences Corporation
DWM Diagnostic Wind Model (UAM preprocessor program)
EKMA Empirical Kinetic Modeling Approach !
EPA U.S. Environmental Protection Agency
EPS Emissions Preprocessor System for the UAM
GMISS Gridded Model Information Support System
MSA Metropolitan Statistical Area
NAAQS National Ambient Air Quality Standard(s)
NO Nitric Oxide
N6X Nitrogen oxides
NO2 Nitrogen Dioxide ;
NSR New source review j
NTIS National Technical Information Service
I -
NWS National Weather Service !
OAQPS EPA Office of Air Quality Planning and Standards
OMS EPA Office of Mobile Sources f
ORD EPA Office of Research and Development
PWD Predominant wind direction
RACT Reasonably available control technology
RFP Reasonable further progress (a type of tracking
required under Section 182 of the CAAA)
ROM EPA Regional Oxidant Model
SAI Systems Applications International
VI1
-------
ABBREVIATIONS f Continued ~>
SCC Source Category Code
SCRAM BBS EPA Support Center for Regulatory Air Models
Bulletin Board System
SIP State Implementation Plan
UAM Urban Airshed Model
UV Ultraviolet
VOC Volatile Organic Compound
VMT Vehicle Miles Traveled
Vlll
-------
ACKNOWLEDGEMENTS
Mr. Dennis C. Doll is the principal author of this document.
Dr. Richard D. Scheffe, Dr. Edwin L. Meyer, and Mr. Shao-Hang Chu
also contributed significantly to the guidance document's content.
Mrs. Cynthia J. Baines was responsible for typing and assembling
the manuscript.
A draft version of this guidance document Has been subject to
p'ublic review and comment through inclusion in pocket A-88-04 for
'
public comment on the Fifth Conference on Air Qusility Modeling held
f
in March 1991. In addition, the authors would like to acknowledge
the efforts of a working group set up to review and advise them
regarding the issues addressed in the guidance document. While all
the recommendations do not necessarily reflect j the views of each
working group member nor their organizations, members' free
exchange of ideas and the generoxis amount of time they spent in
reviewing various drafts added materially to the guidance
document's content. Working group members (in alphabetical order)
ijncluded Dr. C.S. Burton (Systems Applications International
Incorporated (SAI, Inc.)), Mr. C. Durrenberger (Texas Air Control
Board), Mr. G. Gipson (U.S. Environmental Protection Agency (EPA),
Office of Research and Development (ORD)), Mr. T. Helms (EPA,
Office of Air Quality Planning and Standards), Dr. H. Jeffries
('university of North Carolina), Mr. M. Koerber (Lake Michigan Air
i
Directors Consortium), Dr. C. Liu (South Coast Air Quality
Management District), Mr. T. McGuire (California Air Resources
Board), Mr. R. Morris (SAI, Inc.), Dr. S. T. Rao (New York
Department of Environmental Conservation), Dr.
P. Roth (private
consultant), Mr. K. Schere (EPA, ORD), Dr. T. Tesche (private
consultant), Mr. J. Vimont (U.S. National Park Service), Mr. D.
IX
-------
Wackter (Connecticut Department of Environmental Protection), and
Dr. S. Ziman (Chevron USA, Inc.).
Finally, the authors would like to thank Ms. Jeanne R.
Eichinger (CSC) for her thorough review and many helpful editorial
comments.
-------
CHAPTER 1
INTRODUCTION |
The Clean Air Act Amendments (CAAA) of 1990 require ozone
nonattainment areas designated as extreme, severe, serious, or
multi-State moderate to demonstrate attainment of the ozone
National Ambient Air Quality Standard (NAAQS) through photochemical
grid modeling or any other analytical method determined by the
Administrator to be at least as effective.* The Environmental
Protection Agency (EPA) has adopted the Urban Airshed Model (UAM)
as the guideline model for photochemical grid modeling applications
involving entire urban areas. Procedures described in this
guidance document are intended to satisfy the CAAA attainment
demonstration requirements, foster technical credibility, and
promote consistency among UAM regulatory applications.
This guidance document provides recommendations and general
procedural guidance for exercising the UAM (described in
References 1-5) in regulatory applications. However, methodologies
and procedures discussed in this guidance document generally apply
fjor other urban-scale photochemical grid jmodels as well.
J The CAAA does not specify the method for demonstrating
attainment for within-State moderate areas. Thus, the EPA has
determined that the use of the Empirical Kinetic Modeling
Approach (EKMA) may be sufficient for demonstrating attainment
fpr these areas. However, the use of a photochemical grid model
is preferred.
-------
Acceptance criteria for alternative models is beyond the scope of
this guidance document. Use of alternative urban-scale
photochemical grid models as well as regional-scale photochemical
grid models other than the EPA Regional Oxidant Model (ROM) must be
addressed on a case-by-case basis through the EPA Regional Offices.
The UAM source code is maintained and distributed by the
Source Receptor Analysis Branch, Technical Support Division of the
EPA Office of Air Quality Planning and Standards (OAQPS). Users
will be informed of modifications or enhancements to the UAM
through the Support Center for Regulatory Air Models Bulletin Board
System (SCRAM BBS). Additionally, the UAM source code, user's
guide, and test case data base are available from the National
Technical Information Service (NTIS)(703-737-4600). The NTIS
document numbers are noted in the reference list.
This guidance document provides recommendations and procedures
for conducting an ozone analysis with the UAM for ozone attainment
demonstrations. Some of the recommendations and procedures
described were adopted from the California Air Resources Board
(GARB) Technical Guidance Document; Photochemical Modeling6, and
will be referenced as such throughout the text.
Steps needed to conduct an urban-scale photochemical modeling
study using the UAM typically consist of the following:
1. Establish a protocol for the modeling study.
2. Compile air quality, meteorological, and emissions data
to develop UAM input files for each meteorological
episode to be used in the attainment demonstration model
simulations.
introduction
-------
3. Execute the UAM for each meteorological episode.
• - . • • I
4. Conduct diagnostic ancilyses on each meteorological
episode simulation. The principal purpose of diagnostic
analyses is to ensure that the model properly
characterizes physical and chemical phenomena (e.g., wind
fields, spatial and temporal emission patterns)
instrumental in leading to observed ozone concentrations.
The visible product is enhanced model performance (i.e.,
better spatial and temporal agreement with observed
data). Diagnostic model simulations uncover potential
model input data gaps that, when corrected, may lead to
improved model performance.
Exercise the UAM for each meteorological episode and
conduct a series of performance measures to determine
overall model performance in replicating observed ozone
concentrations and patterns. Model performance evalua-
tion should also be done for ozone precursors (e.g., NO,
NO2) if suitable monitoring data are available.
•
•
For each meteorological episode, estimate emissions and
air quality for the projected attainment year required
under the CAAA. Perform model simulations for each
episode to determine whether the ozone NAAQS can be met
in the attainment year.
If the model simulations for the attainment year do not
show attainment for each modeled episode, develop
emission control measures on selected source categories
(for example, volatile organic compound (VOC) and/or NO
controls on selected source categories, alternative fuel
scenarios, etc.).
-------
r
8. Perform model simulations for the emission control
measures to demonstrate attainment of the ozone NAAQS for
each meteorological episode. If the control measures do
not show attainment, repeat steps 7 and 8 as an iterative
process until attainment is shown for each modeled
episode.
These steps are addressed in subsequent chapters of this
guidance document as follows::
Chapter 2: Modeling Protocol
Chapter 3:
Domain and Data Base Issues
•Episode selection
•Domain selection/grid spatial allocation
•Meteorological/air quality data
•Emission inventories
Chapter 4: Data Quality Assurance and Model Diagnostic
Analyses
Chapter 5: Model Performance Evaluation
Chapter 6: Attainment Demonstration
.Introduction '•"'"
-------
CHAPTER 2
MODELING PROTOCOL
Regulatory application of the UAM potentially affects a broad
spectrum of society. The UAM modeling domains may encompass
multiple geopolitical boundaries (counties, cities, and States)
Therefore, the
with a potentially large regulated community.
development of a Modeling Protocol is required, j This Protocol is
necessary to (l) promote technical credibility, (2) encourage the
participation of all interested parties, (3) provide for consensus
building among all interested parties concerning modeling issues,
and (4) provide documentation for technical decisions made in
applying the model as well as the procedures followed in reaching
these decisions. '
The Protocol should detail and formalize procedures for
conducting all phases of the modeling study, such as (1) describing
the background and objectives for the study, (2) creating a
schedule and organizational structure for the study, (3) developing
the input data, (4) conducting diagnostic and model performance
evaluations, (5) interpreting modeling results, (6) describing
procedures for using the model to demonstrate whether proposed
strategies are sufficient to attain the ozone NAAQS, and (7)
producing documentation and data analyses that must be submitted
for EPA Regional Office review and approval.
All issues concerning the modeling study must be thoroughly
addressed during the Protocol development. Modifications to the
Protocol as the study progresses should not be needed unless
-------
significant, unforeseen procedural and/or technical issues are
encountered. All parties involved in the study should agree to
Protocol modifications through the Modeling Policy Oversight
Committee (see below). It is especially important that the State/
local agencies and EPA Regional Office(s) overseeing the study
concur on Protocol modifications.
2.1 Protocol Development Process
Ordinarily, the State agency responsible for developing the
ozone State Implementation Plan (SIP) is also the lead agency
responsible for developing the Modeling Protocol. For domains
encompassing parts of more than one State, the responsible State
agencies need to develop the Modeling Protocol jointly. The
Protocol should describe the modeling policy and technical
objectives of the study. This will require input from various EPA
and State/local personnel dealing with regulatory policy issues and
from others with modeling expertise. It is likely that Modeling
Policy Oversight and Technical Committees will be needed for
addressing these issues. The composition and responsibilities of
the Committees should be defined in the Modeling Protocol.
Responsibilities of the Modeling Policy Oversight Committee
may be, at a minimum, to set the objectives of the study, set the
schedule, determine resource needs, and implement any modifications
to the Protocol as the modeling study proceeds. The Committee
should include representatives from the appropriate EPA Regional
Office(s), State/local agencies, the regulated community, and
public interest groups. It is important that appropriate policy-
oriented personnel be identified for membership on the Committee.
Responsibilities of the Technical Committee may be, at a
minimum, to develop the Protocol's technical specifications
Protocol
-------
concerning emission inventories, meteorological data, air quality
data, data quality assurance, development of emission control
strategies, model diagnostic analyses, model performance evaluation
procedures, and interpretation of model results. The Technical
Committee should include appropriate technically-oriented members
from the EPA Regional Office(s), State/local agencies, the
regulated community, and public interest groups.
For some areas, regional modeling is being planned to
establish initial and boundary conditions for urban-area modeling
attainment demonstrations. The urban-area
development should be coordinated with the
Modeling Protoco1
regional Modeling
Protocol. Some members of the urban-area Policy Oversight and
Technical Committees would probably also be members of the regional
Policy Oversight and Technical Committees.
i The Modeling Protocol must be submitted to the appropriate EPA
Regional Modeling Contact for review and approval. The EPA
Regional Modeling Contact should be a member of the Policy
Oversight and/or Technical Committees so that rapid review and
approval of the Protocol is assured.
Recommendations
A Protocol Document is required for each UAM application used
for an ozone attainment demonstration. This -Protocol should
describe the methods and procedures to be used for conducting
the photochemical modeling study.
Additionally, it is recommended that both a Policy Oversight
Committee and a Technical Committee be established to develop
A regional Modeling Protocol is being prepared by the EPA
OAQPS and will be available by late 1991. This regional Modeling
Protocol will facilitate coordinating regional model applications
to support the nonattainment area SIP UAM applications.
-------
r
the Modeling Protocol. The composition and responsibilities
of the Committees should be defined in the Protocol.
The Modeling Protocol and any modifications to it should be
agreed upon -by all parties involved in the study, through the
Policy Oversight Committee. It is especially important that
the State/local agency participants and EPA Regional Office(s)
overseeing the modeling study concur on any Protocol
modifications. Protocol modifications should be documented
for subsequent public review.
For some nonattainment areas, regional modeling is being
planned to provide initial and boundary conditions as well as
other inputs for the urban-area modeling attainment
demonstrations. Procedures for coordinating the development
of the urban-area Modeling Protocol with the regional Modeling
Protocol should be clearly described.
It is especially important that close technical coordination
be maintained during the Protocol development among nearby
urban-area domains within a regional modeling domain.
Procedures should be established for coordinating the Modeling
Protocols among these areas, and these coordination procedures
should be clearly specified in each nonattainment area's
Modeling Protocol.
The Modeling Protocol must be submitted to the appropriate EPA
Regional Modeling Contact for review and approval.
2 . 2 Contents of Protocol
The recommended contents of the Protocol Document (Table 1)
are patterned after those described in a GARB Technical Guidance
Document . 6
Recommendations
It is recommended that, at a minimum, the components listed in
Table 1 be included in the Protocol Document for each
attainment demonstration modeling study. A description of
each component is presented in Appendix A.
Protocol
-------
TABLE 1
j
EXAMPLE TABLE OF CONTENTS FOR PROTOCOL DOCUMENT
•
UAM Modeling Study Design '
Background and Objectives
Schedules
Deliverables
Management Structure/Technical Committees
Participating Organizations
Relationship to Regional Modeling Protocols
Relationship to Other Urban-Area Modeling Protocols
Relationship to Planning/Strategy Groups
Domain and Data Base Issues
Applicable Preprocessor Programs (e.g., ROM-UAM
Interface System)
Data Bases:
•Air quality
•Meteorology
Base Meteorological Episode Selection
Modeling Domain
Horizontal Grid Resolution
Number of Vertical Layers
Emission Inventory
Specification of Initial and Boundary Conditions
Wind Field Specification
Mixing Depths
Sources of Other Input Data [
Diagnostic Analyses :
Quality Assurance Tests of Input Components
Diagnostic Tests of Base Case Simulation
Test Results/Input Modifications
Model Performance Evaluation
Performance Evaluation Tests
f
Attainment Demonstrations
Identification of Attainment-Year Mandated Control
Measures
Methodologies for Generating ! Control Strategy
Emission Inventories i
Procedures for Attainment Demonstration
Submittal Procedures
Data Analyses Review
Documentation Review and Approval
-------
Protocol
10
-------
CHAPTER 3
DOMAIN AND DATA BASE ISSUES
| Described in this chapter are the following topics: episode
i I
selection, domain selection, meteorological data, air quality data,
and emissions inventories. Choices made in each topic area are
often interrelated. Accordingly, decisions concerning a particular
topic area probably will be based on consideration of several
areas. In several topic areas, recommendations are made concerning
minimum requirements for data availability and modeling resolution.
To reduce uncertainties in modeling inputs and outputs, users are
encouraged to exceed these minimum recommendations whenever
possible. i •
3.1 Episode Selection
A major component of the Modeling Protocol is selection of
meteorological episodes. In genercil, episode selection involves a
review (described below) of several multiday periods during which
high ozone was monitored. At least 1 day is chosen as the day of"
primary interest for each selected episode. Model simulations
typically begin at least one day prior to the day of primary
interest. This minimizes the effects of assumed initial conditions
on predicted concentrations for the critical dayL The length of a
modeled episode is generally a minimum of 48 hours, and the last
day in this period-—the day of primary interest—ris referred to as
the "primary day." j
Data Issues
11
-------
Episodes that have a high probability of covering different
sets of meteorological conditions corresponding with high ozone
concentrations should be modeled. Clearly, a trained meteorologist
familiar with local and regional weather patterns should be
consulted in the selection process. Conditions resulting in
distinctly different source-receptor configurations should be the
prime consideration in distinguishing different meteorological
regimes. Generally, conditions reflecting both poorly defined wind
flow (stagnation) and better defined flow (transport) will need to
be included. It is important to coordinate episode selection with
those responsible for Modeling Protocols in nearby domains,
particularly when observed exceedances may result from
"overwhelming transport."7
The following,approach is recommended for selecting episode
days for use in modeling:
1. Identify the meteorological regimes associated with high-
ozone episodes. The procedure recommended for
identifying meteorological regimes is described in
Appendix B.
2. Select candidate episode days for modeling from the
period from 1987 to the present time. Place each
candidate episode day in the appropriate meteorological
regime (see Appendix B).
3. Rank each candidate episode day within each regime
according to the magnitude of the peak observed ozone
concentration (ranked highest are days with the highest
observed daily maximum ozone from among all sites in or
near a nonattainment Consolidated Metropolitan Statis-
tical Area/Metropolitan Statistical Area [CMSA/MSA]).
Data Issues
12
-------
I
Select the episode days for modeling from among the three
highest ranked episode days from each meteorological
. j
regime. In choosing from among the top-ranked episode
days, consider the availability and quality of air
quality and meteorological data bases; the availability
of supporting regional modeling analyses, the number of
monitors recording daily maximum ozone concentrations
greater than 0.12 ppm (i.e., pervasiveness), number of
hours for which ozone in excess of 0.12 ppm is observed,,
frequency with which the observed meteorological condi-
tions correspond with observed exceedances, and model
performance (discussed in Chapter 5). For example, the
top-ranked episode day within a meteorological regime may
have only routine air quality and meteorological data
bases available for use in the modeling. The third-
highest day, however, may have occurred during an
intensive field study, so that a more comprehensive data
base is available. Thus, the third-highest day may be
more desirable for modeling than the top-ranked day. As
another example, the three highest-ranked episode days
may have air quality and meteorological data bases of
similar quality and quantity, and the number of
i
monitoring sites recording daily maximum ozone greater
than 0.12 ppm may also be similar. If|model performance
on the initially chosen day is questionable, the
Technical Committee may wish to consider a second- or
third-ranked day from the three highest-ranked days for
a regime. The day with the overall best model
performance may be selected as the primary day for
modeling in the attainment demonstration. Note that a
more comprehensive model performance evaluation may be
needed for the selected day, as described in Chapter 5.
"Data
13
-------
5. At least 1 day should be modeled from each identified
meteorological regime. Further, a minimum of 3 days from
among all meteorological regimes should be modeled for
the attainment demonstration (e.g., three meteorological
regimes each containing 1 primary episode day, or two
meteorological regimes with at least 2 primary days from
one of those regimes). Using the model results in the
attainment demonstration is described in Section 6.4-.
States may want to consider a technique other than the one
outlined in steps 1-5 for selecting modeling episodes. Any such
techniques should be described in the Modeling Protocol and
approved by the appropriate EPA Regional Office.
Consideration of several meteorological regimes that
correspond to observed daily maximum ozone levels above 0.12 ppm is
important, because certain emission control strategies that are
effective in reducing peak ozone under some meteorological
conditions may be less so under others. The goal is to develop
strategies that are robust with respect to effectiveness over most
scenarios.
Recommendations
It is recommended that episodes for modeling be selected from
the period from 1987 to the present time. Selected episodes
should represent different meteorological regimes observed to
correspond with ozone >0.12 ppm (as described above). When
selecting episodes, both stagnation and transport conditions
should be examined. A minimum of 3 primary episode days
should be simulated.
Primary episode days falling within each meteorological regime
are ranked according to the highest observed daily maximum
ozone concentration measured within or near the nonattainment
CMSA/MSA. Episodes may be chosen to include any of the three
top-ranked days in each regime. In addition to considering
the magnitude of the highest observed daily maximum ozone
Data Issues
14
-------
concentration in making this choice, data availability and
quality, model performance, availability of regional modeling
analyses, pervasiveness, frequency with which observed
meteorological conditions coincide with exceedances, and
duration of observations >0.12 ppm may be considered.
I ' .
Other techniques for selecting episodes should be described in
the Protocol Document and approved by the appropriate EPA
Regional Office. i
3-2 Size of the Modeling Domain
The size and location of the modeling domain define the data
requirements for the modeling. In selecting a modeling domain,
consideration should be given to (l) typical wind patterns, (2) the
location of major area and point emission sources, (3) the location
of air quality monitors and important receptor locations, and
(4) the need to mitigate effects of uncertainty in upwind boundary
conditions. Generally, the domain should be set as large as
feasible in order to reduce the dependence of predictions on
uncertain boundary concentrations and to provide flexibility in
simulating different meteorological episodes. It is generally much
easier to subsequently reduce the size of a modeled area than it is
to subsequently increase it. !
Once UAM input data for a sufficiently large domain have been
assimilated and processed, the size of the modeling.domain can be
reduced for modeling purposes by specifying domain boundary values
in the UAM. Procedures for reducing the size jof the domain are
described in Reference 2. This could save resources in simulating
modeling episodes in which light or poorly defined wind fields
.
result in a smaller domain being adequate. In contrast, expanding
domain dimensions would require reconstructing most of the UAM
input files.
Data Issues
15
-------
Recommendations
It is recommended that the domain's downwind boundaries be
sufficiently far from the CMSA/MSA that is the principal focus
of the modeling study to ensure that emissions from the
CMSA/MSA occurring on the primary day for each selected
episode remain within the domain until 8:00 p.m. on that day.
The extent of the upwind boundaries will depend on the
proximity of large upwind source areas and the adequacy of
techniques7 used to characterize incoming precursor
concentrations. Large upwind emission source areas should be
included in the modeling domain to the extent practicable.
Also, if large uncertainty is anticipated for domain boundary
conditions, the upwind boundaries should be located at a
distance sufficient to minimize boundary effects on the model
predictions in the center of the domain. Sensitivity analyses
described in Section 4.3 assist in determining the effects of
boundary conditions on predicted values.
3.3 Horizontal Grid Cell Size
The horizontal dimension of each model grid square is based
upon (1) the sensitivity of predicted concentrations to horizontal
grid size, (2) the resolution of observed meteorological and air
quality data and/or estimated emissions data, and (3) limitations
imposed by other considerations such as a required minimum domain
size. Generally, large grid square dimensions result in smoothing
of the emission gradients, wind fields, and spatially varying
mixing heights, which in turn leads to a smoothing of the predicted
concentration field. Also, larger grid cell dimensions reduce both
computer storage space and computational time.
The following should be considered when selecting the
horizontal grid cell size:
1.
The grid cells should be small enough to reflect emission
gradients and densities in urban areas, particularly
those resulting from large point sources and major
terrain or water features that may affect air flow
Data issues
16
-------
2. Sensitivity studies conducted by the EPA suggest that
peak ozone predictions may increase as grid size
decreases j
1
3. Practical limitations on the grid cell size are the
resolution of the emission inventories eind the density of
meteorological and air quality monitoring networks
i
Previous modeling studies have used horizontal grid cell sizes
generally in the range of 2 x 2 km to 8 x 8 km. A grid size of
5 ;x 5 km has generally been compatible with computer resource
requirements and emission inventory development.
1 Recommendations j
It is recommended that the size of the horizontal grid cells
i should not be greater than 5x5 km. Grid cell sizes coarser
than this should be justified and should, at a minimum,
1 address items 1-3 above. Smaller grid cell sizes are
i encouraged because they allow more accurate gridding of area
: and mobile sources. Additionally, emissions from major point
sources are better characterized by smaller grid cell sizes.
However, grid cell sizes smaller than 2 x 2 km are not
! recommended because of potential model formulation
inconsistencies for those grid sizes. J
3.4
of Vertical Layers
! In specifying the number of vertical layers] issues analogous
to those raised for horizontal grid cell dimensions must be
addressed. Again, a compromise is generally needed between the
number of vertical layers and the adequacy of available data bases
and computer resources. It is important that sources with tall
stacks or sources having plumes with high buoyancy be assigned to
an appropriate altitude. Pollutants in elevated, buoyant, point-
source plumes often have effective release heights in layers well
above the surface. Increased vertical resolution allows more
Data
17
-------
accurate representation of the vertical layer at which these
emissions interact with emissions occurring closer to the ground.
Also, increased vertical resolution minimizes dilution that may
result from placing emissions into artificially large vertical
layers. Finally, increased vertical resolution improves the
simulation of when and where plumes are mixed to ground level.
Simulation of the chemistry between individual plumes and the
environment can be greatly affected by how well the model simulates
mixing of these plumes with the ambient air.
Previous applications of the UAM have generally used four or
five vertical layers, with two layers between the surface and the
morning mixing height (diffusion break in the UAM) and three layers
between the mixing height and the top of the modeling domain.
Sensitivity studies suggest that using fewer than three layers
above the mixing height may artificially dilute elevated point-
source plumes, which may cause the model to underpredict near-
surface ozone and precursor concentrations.
Users of the UAM should consider specifying greater detail for
the horizontal and vertical grid cell size than the minimum
recommended in this guidance document. This is encouraged
particularly in modeling domains containing complex terrain or
land/water interfaces. Wind field models can typically produce
wind fields for many more vertical layers than the minimum number
given here.3 The number of vertical layers considered in the UAM
is more likely to be constrained by the time-consuming calculations
needed to simulate atmospheric chemistry.
Recommendations
Based on previous model applications, it is recommended that
a minimum of five vertical layers be used in the modeling
study, with at least three layers above the morning mixing
height (diffusion break in the UAM). Additionally, it is
Data Issues
18
-------
recommended that the top of the modeling domain (region top in
the UAM) be specified above the mixing height by at least the
depth of one upper-layer cell. This can be done by setting
the region top value equal to the maximum mixing depth plus
the minimum depth of the upper-layer cells.
Previous applications have typically used 50 m for the minimum
depth of the vertical layers below the diffusion break and 100
or 150 m for the vertical layers above the diffusion break.
It is recommended that 50 m be used as the minimum thickness
for layers below the diffusion break and 100 m as the minimum
thickness for layers above the diffusion break.
3l5 Meteorological Data
' The availability of meteorological data varies widely among
prospective modeling domains. Also, there are a variety of
techniques available for developing wind fields, temperature
fields, and mixing heights. Although high resolution and
confidence for all meteorological data are desirable, time and
resource constraints force a compromise between desirable and
acceptable methods. Historically, measured meteorological data
have been interpolated for most UAM applications. More recently,
diagnostic and prognostic meteorological modeling techniques have
been explored as possible mecins to develop input fields
(particularly wind fields) for air quality models.
, Wind fields and mixing heights are two of fche.most important
(
meteorological inputs that significantly affect photochemical model
predictions. Methodologies and recommendations for determining
these inputs are described below.
3.5.1 Wind fields
! Methodologies to construct wind fields for the UAM
applications have historically fallen into three categories:6
Issues
19
-------
1. Objective analyses that interpolate observed surface and
aloft data throughout the modeling domain
2. Diagnostic wind models in which physical constraints are
used in conjunction with objective analyses to determine
the wind field
3. Prognostic models based on numerical solution of the
governing equations for mass, momentum, energy, and
moisture conservation along with numerical solutions for
thermodynamic processes
More recently, an additional methodology has been developed in
areas where the EPA ROM has been applied. Computer software has
been developed to map a ROM diagnostic gridded wind field into a
nested UAM domain.5
Objective analysis - These procedures generally involve
straightforward interpolative techniques. They have the advantage
of being relatively simple and inexpensive to use. The primary
disadvantages are that these analyses contain limited physical
concepts, and results are highly dependent upon the temporal and
spatial resolution of the observed values. Thus, in domains
containing sparse observational data or complex topography, results
may be unsatisfactory.
Diagnostic wind models - These models improve mass consistency
for the flow fields. This may be addressed through pararaeteriza-
tions for terrain blocking effects and upslope and downslope flows,
as in the UAM Diagnostic Wind Model.3 Diagnostic models generally
require minimal computer resources and can produce a three-
dimensional wind field. However, diagnostic models need
Data Issues
20
-------
representative observational data to generate fesitures such as land
and sea breezes.
I Prognostic models - These models simulate relevant atmospheric
physical processes while requiring minimal observational data.
Prognostic models require a specification of the synoptic-scale
flow. Reliability of these approaches is usually enhanced if
sufficient observations are available to "nudge" solutions closer
to observations. Since these models can simulate temperature
fields in addition to the wind field, it is possible to determine
stabilities and mixing heights, thus eliminating the need to
generate these from sparse observational data. Another significant
advantage is that interdependencies of various meteorological
inputs with one another are considered in prognostic models. A
major disadvantage is the extensive computational resources needed
to run a prognostic model. Additionally, the availability of
evaluated models and expertise needed to apply them for general
application with photochemical grid models is limited.
; The ROM-UAM Interface System - This system can develop a UAM
gtidded wind field from a diagnostically derived wind field used in
i
the ROM. Such a ROM-derived wind field can be applied for a UAM
domain that is nested within a ROM domain, provided ROM data are
available for identical episode periods. Use of ROM data has the
advantage of being easy to implement and also provides a
1 . i
consistency between ROM model predictions used to specify UAM
boundary conditions and the corresponding wind j fields. The ROM
data are based on an approximately 18 x 18 km horizontal grid cell
size. Thus, one disadvantage is that ROM gridded wind fields may
not sufficiently describe detailed features such as land/sea
Sata Issues
21
-------
circulations. A more finely resolved wind field may be obtained
by using the ROM-gridded winds as the initial wind field for the
UAM's Diagnostic Wind Model (DWM) preprocessor (see Reference 3).
This provides a means for mass consistency when using" the ROM data
as boundary conditions in conjunction with another wind model.
The selection of a specific technique for generating the
domain wind field depends largely on (1) availability of concurrent
ROM diagnostic wind fields, (2) the spatial and temporal resolution
of surface and upper-air observations, (3) available modeling
expertise in applying alternative meteorological models, and (4)
available computer resources. However, some guidelines on
preferences for generating the wind fields are as follows.
Recommendations
The ROM-UAM Interface System should be used to derive the UAM
gridded wind fields when the UAM domain is nested within a ROM
domain for concurrent time periods and ROM predictions are
used to derive the hourly UAM boundary conditions. If it is
judged by the Technical Committee (and identified in the
Protocol) that a wind field derived from the UAM DWM is more
representative of the domain-scale flow, then this wind field
may be used in lieu of the ROM diagnostic wind field. To
minimize mass inconsistency problems, the ROM-gridded winds
may be used as the initial wind field in the DWM (see
Reference 3) when generating the UAM gridded wind field.
For cases in which concurrent ROM applications are
unavailable, it is recommended that the DWM be used to
generate the UAM gridded wind fields. The use of other
techniques for deriving the wind field, such as prognostic
wind models or other objective techniques, may be employed on
a case-by-case basis, subject to approval from the appropriate
EPA Regional Office.
*The ROM data for use in the ROM-UAM Interface System can be
accessed through the EPA UAM Subsystem of the Gridded Model
Information Support System (GMISS) .
Data. Issues
22
-------
j 3.5.2 Data needs for wind field development:
The development of a wind field for each modeling episode
depends upon ground-level and elevated wind observation data. It
is!preferred that a surface-based monitoring network report wind
speed and direction as hourly averages, because an. hour is the time
period commensurate with most UAM concentration output analyses.
Th4 surface monitoring network should be broad and dense so that
diagnostic models (if that is the technique chosen) can depict
major features of the wind field., Data representing vertical
profiles of wind speed and direction are required in order to
establish upper-level wind fields. Preferably, data should provide
adequate spatial (horizontal) and temporal resolution. Results of
UAM applications are often criticized because of inadequate
meteorological data, and lack of sufficient meteorological data
often prevents definitive diagnostic analyses. Thus, the need for
adequate meteorological data cannot be overstated.
i Time constraints imposed by the 1990 CAAA will probably
preclude consideration of new meteorological monitoring stations.
Thus, it is likely that the base case to be used in the attainment
demonstration will be from an historical episode for which model
performance has been deemed acceptable.
; Recommendations
j Meteorological data routinely available for a UAM modeling
, demonstration usually consist of National Weather Service
(NWS) hourly surface and upper-air observations (for winds
i aloft). If these data are the only data available for use in
| a modeling demonstration, they may have to suffice. However,
the NWS data consist of observations made over very short
I periods rather than hourly averaged values. An assumption
that wind velocity measured over a very short period persists
unaltered over an hour may lead to an overestimate of
transport. Therefore, whenever possible, hourly averaged
meteorological data (e.g., from an intensive field study)
Issues
23
-------
should be used. Additional meteorological data may be
available from other sources in the domain (e.g., an on-site
meteorological monitoring program at an industrial facility).
These data may be used to supplement the NWS data, provided
the data have been adequately quality assured. Additionally,
the EPA guideline entitled On-Site Meteorological Program
guidance for Regulatory Modeling Application"should be
consulted to assess whether the supplementary data reflect
proper siting of meteorological instruments and appropriate
data reduction procedures.
In planning a special field study to provide a more spatially
and temporally dense meteorological data base, the number of
surface meteorological monitoring stations should be
sufficient to describe the predominant wind flow features
within the modeling domain. An experienced meteorologist
familiar with local climatic patterns should be consulted
concerning the location and suitability of the surface
meteorological stations. Vertical sounders or profilers are
highly encouraged in a special field study to resolve winds
aloft and mixing heights. Any special field study and
monitoring program should be planned in consultation with the
appropriate EPA Regional Office before implementing the study.
3.5.3 Mixing heights
Predictions from the UAM have been shown to be sensitive to
the mixing height field.6 Therefore, the temporal variations in
the mixing height field over the UAM domain should be described as
realistically as possible. The UAM modeling system contains a
methodology for deriving mixing heights (diffusion break in the
UAM) based on surface temperatures, vertical sounding measurements
of temperature, and cloud cover (see Reference 2). However,
because of the diversity of techniques and data bases that may be
available on a case-by-case basis, we cannot recommend a specific
procedure for deriving the mixing height field in all cases.
Recommendations
It is recommended that, at a minimum, the techniques
described in Reference 2 be used in establishing the mixing
height field in the domain.
-Data
24
-------
The choice of upper-air stations to be used in the mixing
height calculations should be based on prevailing wind fields
and location of the upper-air stations relative to the UAM
domain. If there are no upper-air stations within the domain,
stations outside the domain may need to be used. A trained
meteorologist should be consulted on the selection of upper-
air stations for use in determining mixing heights.
The techniques for generating the mixing height field should
be described in the Protocol Document. Techniques other than
that described in Reference 2 should be documented and
justified.
3.5.4 Clear-sky assumption for photolysis rate calculations
I For regulatory UAM applications, clear-sky conditions have
typically been assumed for photolysis rate calculations in the
MEJTSCALARS processor. The UAM's current structure does not allow
for spatial variation in cloud cover, so the choice is either
uniformly clear or a uniform cloud cover based on a mean cloud
cover over the domain. Use of mean cloud cover could significantly
understate reaction activity .in "clear" patches of the domain.
Potentially, this could be a more serious error than assuming
cl'ear-sky conditions and simulating an overall excess of "domain-
wi,de" insolation. Additionally, the ROM-UAM Interface System
IMETSCL processor assumes clear-sky conditions for photolysis rate
calculations.
Recomroendat i ons
For applications involving the current regulatory version of
the UAM, it is recommended that clear-sky conditions be
assumed for calculating photolytic rate constants in the
METSCALARS processor.
Data Issues
25
-------
3.6 Air Quality
Ambient air quality data are generally used for two purposes:
(1) to specify initial- and boundary-condition concentrations, and
(2) to assess the model performance for each meteorological episode
to be used in the attainment demonstration. These topics are
addressed in the following two subsections.
3.6.1 Initial and boundary conditions
Three general approaches for specifying boundary conditions
for UAM simulations are as follows: (l) use objective/interpola-
tion techniques with a sufficient amount of measured data (i.e.,
data from an intensive field program), (2) use default background
values and expand the upwind modeling domain and simulation period
to mitigate uncertainties due to paucity of measurements, and (3)
use regional-scale model predictions of ozone and precursor
concentrations. Initial conditions for UAM simulations are handled
in one of two ways: (1) use regional-scale model predictions to
derive initial conditions, and/or (2) begin the UAM simulation
sufficiently far in advance of the primary day to eliminate
sensitivity of results to arbitrary assumptions regarding initial
conditions.
Clearly, the nature of case-specific applications will
determine what approaches should be taken for establishing initial
and boundary conditions for particular domains. Ideally, the
preferred technique would be based on an intensive field program
with regional-scale modeling used to fill in spatial and temporal
gaps. This approach is seldom feasible, however, particularly for
historical episodes. Presented below are recommendations for
implementing each of the three techniques identified above for
deriving boundary conditions, including discussion of the
Data Issues
26
-------
advantages and disadvantages of each technique. !Default boundary-
condition values for ozone and precursor concentrations are also
provided. Finally, recommendations are provided on the approach
most likely to be feasible for specifying the initial and boundary
conditions for modeling historical episodes in most locations.
! • j
I ' ' ' : .' ' "
i Use of measured data - All sources of air quality data for a
s
particular modeling domain should be evaluated for applicability in
establishing initial and boundary condition's. Unfortunately, most
ongoing monitoring programs have been designed (xinderstandably so)
with a receptor-based orientation. While available monitoring data
are useful for evaluating model performance, they usually are not
adequate for establishing initial and boundary concentrations.
! I
! Recommendations
i To develop initial and boundary conditions, it is recommended
! that one or more monitoring stations be sited upwind of the
central urban area along prevailing wind trajectories that
j give rise to ozone exceedances.
! The sampling and analysis program should provide data to
! calculate hourly values for ozone, NO, NO2, and speciated
hydrocarbons.
! .
j ,| •
At the inflow boundaries, air quality data at the surface and
I aloft should be used whenever available to specify the
! boundary conditions. Default values (Table 2) may be used
j where necessary.
i !'
i i^
Use of default values - Some urban areas may lack adequate
data suitable for establishing initial and boundary conditions.
Section 3.2 on domain selection and Chapter 4 on diagnostic
analyses recommend constructing domains and simulation periods
large enough to minimize the sensitivity of inner core and downwind
concentrations to assumed initial and boundary conditions.
Data- Issues
27
-------
TABLE 2
DEFAULT BOUNDARY CONDITION CONCENTRATIONS FOR
CARBON-BOND-IV SPECIES (SEE REFERENCES 1 AND 7)
Species
OLE
PAR
TOL
XYL
FORM
ALD2
ETH
CRES
MGLY
OPEN
PNA
NXOY
PAN
MONO
H202
HN03
MEOH
ETOH
03
N02
NO
CO
I SOP
Species Name
Concentration fppbC)*
Olefins
Paraffins
Toluene
Xylene
Formaldehyde
Higher Aldehydes
Ethene
Cresol,Higher Phenols
Methyl Glyoxal
Aromatic ring fragment
acid
Peroxynitric acid
Total nitrogen
compounds
Peroxyacyl nitrate
Nitrous acid
Hydrogen peroxide
Nitric acid
Methanol
Ethanol
Ozone
Nitrogen Dioxide
Nitric oxide
Carbon monoxide
Isoprene
0.60
14.94
1.26
0.78
2.1
1.11
1.02
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0,
0,
40.0 (ppb)
2.0 (ppb)
0.0 (ppb)
350.0 (ppb)
0.1 (ppb)
ppbC, parts per billion Carbon
Data Issues
28
-------
I Initial- and boundary-condition concentrations are influenced
by large- and small-scale weather patterns and emissions distri-
butions that are unique to each modeling domain. Thus, case-
specific attributes should be used in estimating these concentra-
tions whenever feasible. For example, boundary concentrations of
hydrocarbons, particularly those species (or intermediate products)
emitted from vegetation, are likely to be higher in urban areas
! :
surrounded by dense vegetation than in areas surrounded by sparse
vegetation.
Recommendations ' • !
It is recommended that use of default values to establish
boundary conditions be limited to areas surrounded by large
expanses of low-density anthropogenic emissions. Accordingly,
the modeling domain may need to envelop rural areas.
Those choosing to use default values should plan to perform
diagnostic/sensitivity simulations (see Chapter 4) to evaluate
the sensitivity of domain-interior model predictions to the
boundary conditions.
i
Table 2 lists the recommended default boundary values for the
chemical species used in the model. Use of default boundary
values under regional transport conditions should be closely
evaluated. When using default values, the boundary of the
domain should extend as far upwind as practicable.
To diminish dependence on arbitrary specification of initial
conditions, a simulation should begin at least 1 day prior to
the primary day.
i Use of regional model concentration predictions - Output from
regional-scale models such as the EPA ROM provides estimates of
initial and boundary conditions (as well as certain meteorological
inputs) for urban-scale models. This is especially important under
regional transport conditions. The ROM-UAM Interface System
i '
referred to in Section 3.5.1 can use ROM concentration predictions
to develop UAM input files of initial and boundary conditions.
This interfacing software should be used for UAM domains nested
Data
29
-------
within more extensive ROM domains. Using the ROM is the
recommended approach for generating boundary conditions. It is the
most technically defensible approach for estimating future boundary
conditions for the attainment year.
Certain considerations arise when using interfacing methods.
First, selection of historical episodes is limited to those that
have been modeled on a regional scale. Second, there may be
inconsistencies in mass conservation when applying ROM-derived
initial and boundary conditions in conjunction with wind fields not
derived from the ROM wind field (see Section 3.5.1). The
combinations of concentrations and wind velocities produced by the
ROM-UAM Interface System represent mass fluxes passing through the
urban-scale modeling domain. In cases where ROM-derived initial
and boundary conditions are applied without ROM-generated wind
fields, locally developed wind fields may need to be evaluated for
mass consistency throughout the urban-scale domain. Methods for
addressing this problem will need to be chosen on a case-by-case
basis. A general procedure for enhancing mass consistency is
described in Section 3.5.1. Additionally, initial and boundary
conditions derived from the ROM data should be compared with
corresponding monitoring data wherever available. This will ensure
that the ROM wind fields adequately represent the transport of
ozone and precursors into the domain region.
Recommendations
It is recommended that, whenever feasible, the ROM-UAM
Interface System be applied to derive the initial and boundary
conditions for the episode(s) being modeled. If the Interface
System is used to derive the initial and boundary conditions,
it is also recommended that it be used to derive the UAM
gridded wind field, unless there is sufficient justification
that other techniques for deriving the wind field are more
accurate.
Data Issues
30
-------
I In cases for which ROM predictions are not available, it is
j generally recommended that measured data be used to establish
! the initial and boundary conditions, provided the Technical
I Committee identified in the Protocol determines the data are
adequate. If measured data are not adequate, the default
i values may be used. To diminish sensitivity of results to
i assumed initial conditions, simulations should begin 1 day
prior to each primary day.
,
i ...
3.6.2 Performance Evaluation Data !
.
Air quality data are needed to diagnose problems in setting up
model applications and assessing model performance for the
meteorological episodes being considered inj the attainment
demonstration. A lean air quality data base may introduce
significant uncertainties in characterizing model performance.
I Under Title I, Section 182 of' the CAAA of 1990, the EPA is
required to develop regulations for enhanced monitoring of ozone
and ozone precursors in serious, severe, and extreme ozone
nonattainment areas. When promulgated, these regulations will
specify criteria for network design, monitor siting, monitoring
methods, operating schedule, quality assurance, and data
submittal.11 The enhanced ozone monitoring system is designed to
•I
provide a more comprehensive data base for model input and to
improve model performance evaluation.
Recommendations [
It is recommended that the data base used in the attainment
demonstration modeling meet the requirements for the
enhanced ozone monitoring system to be promulgated by the EPA.
However, the EPA recognizes that some historical episodes that
will be used in the attainment demonstration modeling for the
November 1994 ozone SIP submittals may have data bases that
would not meet the requirements for an enhanced ozone
monitoring system. Under these conditions, the data bases
should be scrutinized in detail by the Technical Committee to
help ensure that model performance tha^. appears to be
Data Issues
31
-------
acceptable has not actually resulted from compensating errors
in the data bases. Additional diagnostic analyses may be
necessary for lean data bases from historical episodes.
If it is determined that the existing air quality monitoring
program does not meet the requirements for the enhanced ozone
monitoring system, responsible regulatory agencies should
begin planning for development of an enhanced ozone monitoring
system for potential future modeling studies.
3.7 Emissions
The credibility of UAM applications is directly tied to
formulating the best possible emission inputs. Model performance
may hinge on how well emissions are estimated. Also, in the
attainment demonstration, modeling results are used to determine
emission scenarios that lead, to improved air quality levels
consistent with the NAAQS. A faulty emission inventory could lead
to erroneous conclusions about the extent of needed controls and,
in some cases, errors in judgment about the need to control certain
classes of precursors (e.g., NOX).
Developing photochemical model emission input data is the most
intensive task of model applications, and requires consideration of
many issues. The source of the UAM modeling emission inventory
will be the 1990 SIP nonattainment base year inventory required
under the CAAA of 1990 for all ozone nonattainment areas. A
further discussion of the 1990 base year inventory is contained in
Emission—Inventory Requirements for Ozone State Implementation
Flaps.12 It is important to note that the 1990 modeling inventory
will not be identical to the 1990 nonattainment area inventory
required for reasonable further progress (RFP) tracking under
Section 182 of the CAAA. For example, the modeling inventory will
probably have to cover a larger geographical area than that
required for the nonattainment area inventory. The discussion of
Data Issues
32
-------
modeling domain and boundary-condition issues in Sections 3.2 and
3.6 makes it clear that the modeling inventory must encompass a
larger area than the nonattainment MSA. A complete description of
relationships between the modeling inventory and the nonattainment
area inventory is provided in Procedures for the Preparation of
Emission Inventories for Volatile Organic Compounds. Volume II:
Emission Inventory Reauir-ements for Photochemical—Air Quality
simulation Models (Revised").13 Additional guidance for developing
the modeling emission inventory is found in Reference 4.
| _ | ...
i ' i
For use in regulatory applications of the UAM, the 1990
modeling inventory will have to undergo several adjustments.
First, the inventory needs to be adjusted to be consistent with
meteorological conditions during each selected episode (i.e., "1990
day-specific emissions"). Second, the resulting "1990 day-specific
emissions" should be adjusted to reflect control programs and
activity levels prevailing during the year(s) of selected episodes.
For example, if a selected episode occurred in 1988, the "1990 day-
spjecif ic emissions" would be further adjusted to reflect controls
and activity levels prevailing in 1988. This latter adjustment is
needed to provide an estimate of emissions most suitable for
evaluating performance of the UAM.
As noted in Chapter 1, once the UAM's performance has been
evaluated and the model has been determined to perform
satisfactorily, it is used to derive control strategies to attain
the NAAQS. This requires another adjustment to the "1990 day-
specific emissions" described above. This adjustment entails use
of growth factors, ongoing control programs and retirement rates
for obsolete sources of emissions to project "1990 day-specific
emissions" to the years by which the CAAA specify that the NAAQS
must be attained. Reference 14 describes the appropriate
methodology for making emission projections.I The resulting
33
-------
"attainment year modeling inventory" is used as a starting point
from which to construct "strategy inventories." A "strategy
inventory" is obtained by superimposing additional control measures
on sources of emissions in the "attainment year modeling
inventory."
In summary, a 1990 modeling inventory is first adjusted to
evaluate UAM performance. The 1990 modeling inventory is then
readjusted to reflect emissions most likely at the time the CAAA
require attainment of the NAAQS.
Two emission files drive the UAM, a file of emissions that are
injected into the first, surface-based vertical layer, and an
elevated point source file of emissions that are injected into
vertical layers above ground level. The UAM Emissions Prepro-
cessing System (EPS)4 reads county-level area- and point-source
files and performs four major functions: (1) area sources and point
sources are allocated to grid cells; (2) temporal profiles are
assigned to source categories; (3) hydrocarbon speciation profiles
are assigned to source categories, and (4) point sources with
effective plume heights greater than a prescribed cutoff level are
assigned to the elevated point source file and the remaining point
sources are assigned to the surface-layer emissions file.
Addressed below are the following issues that arise in
developing emission input data: (1) use of speciation profiles,
(2) use of surrogate factors to grid area sources, (3) treatment of
mobile sources and top/down versus bottom/up approaches, (4)
episodic adjustment of inventories to day-specific modeling inputs,
(5) treatment of biogenic emissions, (6) cutoff levels for NOX
point sources, and (7) consistency with national inventories.
Data Issues
34
-------
3.7.1 VOC speciation
i , i
I The EPA provides "default" nationwide VOC speciation profiles
for various source category codes (SCCs).13 Use of local speciation
information, especially for major emitters, is preferable to
national default profiles. If feasible, major VOC point- and area-
source categories should be surveyed to determine appropriate VOC
composition profiles. In many cases, both the quantity and the
composition of emissions change as process operations are modified.
To,the extent feasible, this should be accounted for when deriving
local speciation,profiles and in simulating control strategies.
The emissions inventory guidance document" provides details on
developing local speciation profiles. |
Most current-year applications are likely to rely on existing
default data for speciating mobile-source emissions. Projected
future-year mobile-source emissions files may be based on different
formulations of gasoline and use of alternative fuels. Speciation
guidance for these fuels will be provided by the EPA Office of
Mobile Sources (OMS) through the appropriate EPA Regional Office.
I
' Recommendations I
; It is recommended that local speciation profiles for point-
I source and area-source categories be used whenever feasible.
The Technical Committee should determine the appropriateness
I of using local or national default speciation profiles.
! Profiles used in the modeling demonstration must be
! documented, and any changes assumed in profiles as the result
! of control strategies must be identified and justified.
3.7.2
aroa sources
i Area-source emission data, including motor vehicle emission
data, are often supplied on a county basis. Spatial allocation of
coynty-level emission estimates to grid cells is performed for each
35
-------
identified area-source category; the allocation requires use of
surrogate distribution factors such as population distribution,
land use, and road type. The UAM EPS4 contains a program that uses
gridded surrogate factors to allocate county-level emissions data
to the grid cell size of the modeling domain.
Recommendations
It is recommended that the emission inventory guidance
document13 be consulted for alternative surrogate factor-
choices and sources of information for assimilating surrogate
data. The EPA is currently developing a utility to provide
gridded surrogate data. States will be notified of the
availability of gridded surrogate data through the EPA
Regional Offices.
3.7.3 Mobile sources
Development of gridded, time-variant mobile-source inputs
raises several concerns and often represents the largest fraction
of effort when assimilating mobile-source emissions inputs.
Mobile-source emissions have been compiled from original data or
from existing county-level emissions.13 Developing gridded mobile-
source emissions from original data requires aggregating sub-grid-
cell-level components. This may require exercising transportation
models that produce inputs for the mobile-source emissions model
(i.e., the latest EPA MOBILE model), and then performing the
necessary spatial allocations to grid cells and temporal
distribution over every hour. This practice is far from
standardized. Also, in certain areas, execution of transportation
models is restricted by lack of appropriate traffic count and speed
data.
The emission inventory guidance document13 provides direction
for developing mobile-source inputs from original data (referred to
Data Issues
36
-------
as a bottom/up method) or from an existing county-level inventory
(referred to as a top/down method).
Recommendations
Bottom/up methods are the preferred approach for estimating
vehicle activity levels and emission factors because these
methods have potential for resolving variations in speed and
vehicle miles traveled (VMT) among different grids over hourly
time slices. Bottom/up approaches are most appropriate for
addressing the inner urban core of modeling domains.
Peripheral, less dense traffic areas can be treated with
top/down methods. Exceptions to these recommendations should
be considered by the Technical Committee on a case-by-case
basis. Justification for more extensive use of top/down
methods should be sought in discussions with the appropriate
EPA Regional Office.
3.7.4 Episode-specific adjustments
i j
i • i
[ Several source categories of VOC emissions are sensitive to
meteorological conditions. Thus, it is important for modeling
inventories to reflect episode-specific meteorological conditions.^
For example, biogenic emissions, mobile-source evaporative
emissions, and solvent categories will need to reflect specific
modeling days. In addition, known episode-specific events such as
changes in process operations for point sources affect emissions
rates and should be reflected in the episode modeling inventory.
Recommendations
Mobile-source emissions should be adjusted for episode-
specific temperatures. This is done by running the latest EPA
MOBILE model using episode-specific maximum and minimum
temperatures. Chapter 7 of the emission inventory guidance
document13 describes the procedures for deriving episode-
specific mobile-source emissions using the latest MOBILE
model. Use of models other than the latest EPA MOBILE model
should be reviewed by the Technical Committee on a case-by-
case basis, and is subject to approval by the EPA Regional
Office.
Issues
37
-------
Biogenic emissions must reflect episode-specific conditions
(see Section 3.7.5).
If available, episode-specific operating rates for point
sources are preferable for estimating temporal point-source
emissions. Procedures for temporally adjusting point and area
sources are also provided in the emission inventory guidance
document.li
3.7.5 Bioqenic emissions
Biogenic emissions can be a significant portion of the overall
VOC emission inventory for a given domain, particularly in areas of
high vegetative density. The EPA provides the Biogenic Emissions
Inventory System (BEIS), which can develop day-specific, hourly,
gridded, speciated inputs (see Reference 4), and also provides a
national data base of land use distributions with this system.
Spatial variability is limited to the county level (i.e., emissions
are evenly spread throughout the grids within a specific county).
The EPA is currently modifying the BEIS to allow users to
input user-derived and possibly more up-to-date land use
distribution data. Users will be advised of the expected delivery
date of the modified processor via the SCRAM BBS and EPA Regional
Offices.
Recommendations
Biogenic emissions must be included in the emission inventory
developed for each model simulation (i.e., base case and
control strategy). The biogenic emission processor (BEIS)
that is part of the EPA Emissions Preprocessor System4 should
be used to derive the inventory. Use of alternative land use
factors in the BEIS should be described and documented in the
Protocol Document.
Also, methods other than the BEIS may be considered for
deriving the biogenic emissions. These methods must be
described in the Protocol Document along with justification
for using them.
Data Issues
38
-------
! 3.7.6 Point-source and plume-rise cutoff levels
•t • • •
j Guidance for initiating ozone/CO SIP emission inventories
pursuant to the 1990 CAAA15 specifies point-source cut-off levels
of [10 tons/yr and 100 tons/yr for VOCs and NOX/ respectively. Any
source may be treated as a point source as long as stack data are
specified that allow derivation of effective plume height, and
source location is provided.
4
In some cases, the Technical
Committee may wish to treat selected smaller sources as point
sources.
I The UAM EPS4 requires the specification of a plume-rise cutoff
level for delineating elevated point sources from area sources. If
the plume rise that the EPS calculates for a given point source is
below the user-specified level, then the point-source emissions are
placed in the area-source emissions file. If the plume rise is
above the level, the emissions are treated as coming from elevated
point sources and are then placed within the appropriate UAM
vertical layer.
Recommendations
Point-source cutoff levels of 10 tons/yr for VOCs and no
greater than 100 tons/yr for N(X are recommended for inclusion
in the modeling emission inventory. Point sources must have
the stack data needed to calculate effective plume height, so
that the heights at which emissions are injected into the
modeling system can be determined.
The Technical Committee may consider using a lower plume-rise
cutoff level, particularly in areas where there may be a high
density of point sources. Additionally, the CAAA specifies
"major source" definitions that have lower cutoff limits for
purposes such as application of reasonably available control
technology (RACT), new source review (NSR) and creation of
Emission Statements.15 The Technical Committee may consider
using these lower cutoff limits in the modeling inventory.
The Technical Committee should specify the plume-rise cutoff
level to be used in delineating point-source and area-source
39
-------
emissions, and the level should be identified in the Protocol
Document.
3.7.7 Consistency with national inventories
Comparisons should be made between the modeling inventory and
the 1990 SIP and RFP tracking emission inventories reported in the
EPA Aerometric Information Retrieval System (AIRS).16 Although
these inventories will not be identical, such a check can be
considered part of the quality assurance process. Major
inconsistences should be noted and documented. It is especially
important that those planning to use ROM-derived air quality data
in the model simulations follow applicable guidance/regulations for
reporting statewide emissions data to AIRS. These national
inventories are used in the ROM modeling. As noted previously,
using the ROM is the preferred procedure for estimating UAM
boundary conditions and meteorological inputs. Attainment
demonstrations will be less consistent if the ROM and the UAM use
significantly different emissions data bases.
Recommendations
For an acceptable attainment demonstration, documentation
should be provided that shows that the modeling emission
inventory is consistent with the emission inventory being
reported in AIRS in accordance with applicable guidance and
regulations.16
Data Issues
40
-------
CHAPTER 4
DATA QUALITY ASSURANCE AMD MODEL DIAGNOSTIC ANALYSES
This chapter provides general guidance for quality assurance
testing of component data input fields and diagnostic testing of
base case episodes. These analyses are designed to establish and
improve reliability of the input data and proper functioning of the
model.
Although the UAM has been evaluated on a number of historical
data bases, measures of model behavior with respect to observed
data are necessary for new applications. Model developers and
users perform diagnostic tests to uncover potential input data gaps
that, when corrected, may lead to improved treatment of model
processes. Regulators need some indication that the model captures
the key features of the base meteorological episodes being applied
inlthe model simulations in order to have confidence in a model's
1
ability to predict future ozone (1) after applying projected growth
and planned emission controls and (2) after applying alternative
emission control strategies.
j Important prerequisites for a model performance evaluation
(see Chapter 5) are (1) quality assurance testing of model inputs
and (2) diagnostic testing of the base meteorological episode
simulation to ensure that the model is functioning properly and
that apparently accurate model results are being obtained for the
right reasons. For example, quality assurance testing of input
data helps to ensure that apparently good model results have not
resulted from compensating errors in input data.
'"Quality
41
-------
An excellent compilation of model performance evaluation
techniques, including diagnostic tests and related issues, is
contained in Reference 17. This reference serves as the basis for
this chapter and for the model performance evaluation described in
Chapter 5. Various graphical and numerical measures described
below are treated in detail in Reference 17.
Two useful graphical displays for both quality assurance and
diagnostic testing are mapping and time-series plots.
Mapping is a two- or three-dimensional spatial display of
values illustrated with various contouring and tiling methods.
These displays may depict political boundaries and monitoring site
locations as well. Mapping capability is a multipurpose tool
applicable for all forms of gridded data, such as future-year
emission control strategy results and most input data fields (e.g.,
gridded wind fields, temperatures, and emission densities). Point-
source locations may also be depicted to ensure that they are
properly located. Spatial displays of predicted and observed ozone
patterns are particularly useful as part of a model performance
evaluation.
Time-series plots displciy hourly measured and predicted ozone
values for specific locations such as monitoring sites. Time-
series plots provide an overview of the temporal performance of the
model predictions. Comparison of time-series plots across multiple
monitoring sites provides an indication of spatial response. Even
though measured VOC or NOX species data may be limited, it may
still be useful to plot time-series plots for some of these
species, particularly for cases where ozone predictions do not meet
expectations. Such plots may provide insights to the ozone
prediction patterns and also to data base inconsistencies requiring
further investigation.
Quality
Assurance
42
-------
i The following sections describe recommended steps for
conducting diagnostic testing of each base case meteorological
episode simulation.
l
4.1 Step 1: Quality Assurance Testing of Component Fields
Starting with initial, quality-assured data, input data are
developed for use in various UAM preprocessors. The first stage of
diagnostic testing should focus on assessing the accuracy of major
UAM input fields produced by the UAM preprocessors. Generally,
the testing is qualitative in nature and based on. comparing visual
displays, of preprocessor outputs with patterns exhibited by the
observed data. Prior to conducting a base case meteorological
episode simulation, individual air quality, meteorological, and
emissions fields should be reviewed for consistency and obvious
omission errors. Both spatial and temporal characteristics of the
data should be evaluated. These checks may be only cursory, but
errors uncovered by this component testing might be extremely
difficult to diagnose later in the modeling process, when errors
could arise from any subset of the data inputs. Examples of
i •
co'mponent testing include the following:
Air Quality:
Emissions:
Check for correct order of magnitude,
especially when using background values;
assure reasonable speciation
Plot various source types by grid cell and
review major source locations with local
emissions patterns; check major highway
routes; generally, look for obvious omission
errors; plot VOCs, NOX and CO by grid cell and
cross-check with source distribution for
- Duality
43
-------
logical patterns, such as high NOX levels
associated with major power plants
Meteorology:
Plot surface and elevated wind vectors and
compare with monitoring stations and weather
maps for consistent patterns; compare mixing
height fields with sounding data; check
temperature fields
In quality assurance testing of component input fields, the
emphasis is on capturing rather large errors before performing
model simulations.
Recommendations
It is recommended that quality assurance testing of the air
quality, emissions, and meteorological data input files be
conducted before proceeding to diagnostic testing of the base
case meteorological episodes. At a minimum, emissions data
should be quality assured by looking at emission distribution
maps and known source locations and emission strengths.
4.2 Step 2: Diagnostic Testing of the Base Case Meteorological
episodes
After confidence has been achieved in producing UAM input
fields, the UAM should be exercised for each base case
meteorological episode. The initial run is termed a diagnostic
simulation because review of initial base case simulations usually
uncovers additional input errors requiring correction before an
acceptable set of base case inputs can be derived. During this
stage of the process, emphasis is placed on assessing the model's
ability to correctly depict plume orientation and the timing of
observed ozone maxima. Accordingly, visual methods such as mapping
and time-series plotting, using measured data as reference marks,
may be used to assess model behavior.
Quality
Assurance'
44
-------
Recommendations
To aid in interpreting simulation results, it is recommended
that predicted and observed ozone concentration maps be
constructed for each base meteorological episode simulation.
Concentration maps present spatial information on the
structure of the ozone plume.
Maps at 1- or 2-hour intervals should be constructed over the
periods of most interest. While a typical period might be
defined as early morning to late afternoon for the day of
highest ozone, it is also useful to look at most time
intervals under recirculation, stagnation, and transport
conditions.
Consideration should also be given to constructing a map that
depicts the highest predicted daily maximum ozone value for
each grid cell. Examples of various mapping techniques are
described in Reference 17.
It is also recommended that the predicted concentrations used
in the time-series plots be consistent with the method for
deriving predicted concentrations for the model performance
evaluation described in Chapter 5. This method is based on
Reference 17 and produces a weighted average using bilinear
interpolation of the predictions from the four grid cells
nearest to the monitor location.
Other methods for deriving predicted concentrations for time-
series comparisons may be judged appropriate by the Technical
Committee; some suggestions are contained in Chapter 5. These
methods should be described in the Modeling Protocol.
If suitable data are available, time-series plots should be
developed for NO and NO.,, and for VOC species at selected
locations, particularly for cases in which ozone time-series
or mapping results are not consistent with expectations.
Comparisons of ozone precursors should be done for
concentration levels above the detection limits "for the
monitoring equipment.
45
-------
4•3 Step 3;—Additional Base Meteorological Episode Sensitivity
Testing
In addition to running the base meteorological episode
diagnostic simulation, other episode diagnostic simulations that
perturb levels of emissions, initial and boundary conditions, and
meteorological inputs may provide valuable information for
identifying critical input areas and ensuring proper domain and
episode selection. The following sampling of simulations, which
are equivalent to sensitivity tests on major model inputs,
illustrate the utility of this exercise.
!• Zero emissions - To indicate levels of sensitivity to
emissions, all emissions are ' set to zero and the
resulting predicted concentrations are compared with the
base meteorologiccil episode predictions that include
emissions. A lack of substantial sensitivity may
indicate a need to reexamine the selection of episodes or
domains. Variations can be performed by zeroing out
emission subsets, such as biogenic emissions, mobile-
source emissions, and individual source categories.
2« Zero boundary concentrations - Inflow concentrations at
the lateral boundaries and top of the modeling domain are
reduced to zero or low background levels. Sensitivity of
concentrations in the inner core and downwind portions of
the modeling domain provide a measure of the boundary
conditions' influence. This simulation can identify
transport-dominated episodes and provide assurance that
the upwind extent of the domain is adequate for episodes
where intradomain emissions dominate. In minimum trans-
port conditions, the second- and third-day concentrations
Quality
Assurance
46
-------
I (inner core and downwind locations) should be relatively
I • •
i insensitive to boundary-condition concentration changes.
i . •
i 3. Mixing height and wind speed variations - Much
i
j uncertainty is associated with mixing heights and wind
| speeds, and simulated concentrations are often sensitive
| to these inputs. Simulations that test the sensitivity
t .
j of model estimates to variations in wind speed and/or
1
i mixing height provide bounds on some of the uncertainty
i resulting from these parameters. Large sensitivity may
' suggest that future model applications will need
. t ' "
improvement in the meteorological data bases. Also,
large sensitivity may suggest a need to consider
1 alternative wind field generation techniques.
i Certain numerical measures, which are recommended in the
discussion of model performance evaluation in Chapter 5, are also
useful diagnostic tools. For example, consistent underpredictions
usually produce bias values less than zero. This phenomenon could
be- due to various factors, such as overstatement of wind speeds or
mixing heights, or low emission estimates. Modelers are encouraged
toi use numerical as well as graphical techniques in the diagnostic
prbcess.
' The diagnostic analyses described in this chapter are
considered to be a starting point for a specific modeling study.
Diagnostic tests discussed in Reference 17 should be considered
whenever possible. Also, the EPA is developing a UAM Post-
processing System18 to assist in diagnostic testing of the base
meteorological episodes. Availability of this software will be
announced through the SCRAM BBS.
Assurance
47
-------
Recommendations
Diagnostic testing of the model should begin with quality
assurance testing on input data files (Section 4.1).
Diagnostic testing of each base meteorological episode should
follow (Section 4.2). Additional diagnostic sensitivity tests
for the base episode should also be considered (Section 4.3),
including using zero emissions and/or zero boundary condi-
tions, and varying mixing height and wind speed estimates.
Agreement should be obtained among members of the Technical
Committee concerning input field modifications arising from
the quality assurance testing. These modifications should be
based on scientific or physical reasoning and not just on what
will improve model performance. All changes to the data that
result from the diagnostic testing should be documented and
justified.
In addition, all diagnostic steps should be documented to
avoid misinterpretation of model performance results. After
confidence is gained that the simulation is based on
reasonable interpretations of observed data, and model
concentration fields generally track (both spatially and
temporally) known urban-scale plumes, a performance evaluation
based on numerical measures is conducted for each base
meteorological episode (see Chapter 5).
Quality
Assurance
48
-------
CHAPTER 5
MODEL PERFORMANCE EVALUATION
; At some point in the modeling process, an assessment of model
performance is required. Specifying rigid rejection/acceptance
criteria has not been supported by model developers nor by decision
makers participating in previous modeling efforts. Instead,
performance measures derived from previous photochemical model
applications may provide a reasonable benchmark for model perform-
ance . Also, graphical procedures reveal qualitative relationships
bejtween predicted and observed concentrations that can be used in
model performance evaluation.
j Poor performance may necessitate (1) delaying model
applications until further diagnostic testing and quality assurance
checks are reflected in the input data base, or (2) selecting
another meteorological episode for modeling. However, this is not
a valid reason for delaying SIP attainment demonstration submittals
beyond the dates required in the 1990 CAAA. Also, cases where good
model performance is shown should .be reviewed as well, because
compensating errors can induce spurious agreement among observed
and predicted values. •
5.1 Performance Measures
This section describes recommended graphical and statistical
performance measures for ozone predictions. These measures should
be applied for modeling results beginning on the second day of the
modeled episode. As described in Section 3.1, the first day is
' Evaluation:
49
-------
eliminated to mitigate the effects of specifying initial conditions
arbitrarily. Performance measures should also be considered for
ozone precursors wherever possible, based on availability of
monitored data. Obvious problems exist in comparing model
predictions with observed values. The UAM output represents
volumetric (e.g., 25 km3), l-hour average concentrations, but air
quality data represent point locations with various sampling
periods. This "incommensurability" may lead to considerable
uncertainty in the comparisons, especially for precursor species
that are not buffered chemically and may have been sampled at
locations not representative of areawide concentrations.
As part of the UAM Postprocessing System, the EPA is currently
developing a model performance utility that will contain the
performance measures listed below. Users will be able to access
this utility for model performance evaluation testing. This
utility is expected to be available in late 1991. Model users will
be advised on its availability through the EPA SCRAM BBS.
The measures used in the performance evaluation should include
both qualitative (e.g., graphical) and quantitative (e.g.,
statistical) analyses. Statistical measures may provide a meaning-
ful test of model performance for dense monitoring networks, such
as those for special field studies. However, for some routine
monitoring networks where coverage may be sparse, statistical
measures may provide a distorted view of model performance,
especially for paired values.
Reference 17 provides detailed descriptions of graphical and
statistical measures available for assessing the performance of
photochemical grid models. The Technical Committee should consult
this reference when formulating model performance evaluation
methods, and may want to use it for developing additional perform-
Eval&ation;
50
-------
ance evaluation procedures other than those recommended in this
Guidance Document.
5.1.1 Graphical performance procedures
i
i Graphical, displays can provide important information on
qualitative relationships between predicted and observed concentra-
tions. At a minimum, the following graphical displays should be
developed for each meteorological episode: time-series plots,
ground-level isopleths, quantile-quantile plots, and scatterplots
of predictions and observations.
Time-series plots - The time-series plot, developed for each
monitoring station in the modeling domain, depicts the hourly
predicted and observed concentrations for the simulation period.
Thfe time series reveals the model's ability to reproduce the peak
prediction, the presence of any significant bias within the diurnal
cycle, and a comparison of the timing of the predicted and observed
ozone maxima.
17
; Ground-level isopleths or tile maps - Ground-level isopleths
or tile maps display the spatial distribution of predicted
concentrations at a selected hour.. Isopleths of predicted daily
maxima may also be constructed. The isopleths provide information.
on the magnitude and location of predicted pollutant "plumes."
Superimposing observed hourly or daily maximum concentrations on
the predicted isopleths reveals information on the spatial
alignment of predicted and observed plumes. Subregional biases of
predictions versus observations may result from spatial
misalignments.
Scatterplots of predictions and observations - Scatterplots
depict the extent of bias and error in the ensemble of hourly
Evaluation
51
-------
prediction-observation pairs. Bias is indicated by the systematic
'positioning of data points above or below the perfect correlation
line. The dispersion of points is a measure of the error in the
simulation. The scatterplot also reveals outlier prediction-
observation pairs.
Ouantile-quantile plots - Quantile-quantile plots compare the
frequency distributions of rank-ordered observed and rank-ordered
predicted concentrations. The observed and predicted concentra-
tions are sorted from highest to lowest then plotted on an x-y
plot. This graphically depicts any model bias over the frequency
distribution.
"Paired" predictions of daily maxima - In attainment
demonstrations, particular interest is focused on daily maximum
ozone concentrations. One test that may provide insight into model
performance is to consider model predictions occurring within
±1 hour of the observed daily maxima at each monitoring site in the
nine grid squares surrounding and including the monitor. The
"prediction," for purposes of this pairing, would be the one that
agrees most closely with the observed daily maximum for each site.
This method may be useful for sparse meteorological and air quality
networks, because it recognizes that both the inputs and air
quality observations have some attendant uncertainty. Resulting
comparisons can be superimposed on a map depicting emissions and
monitors to help assess model performance.
Recommendations
At a minimum, the following graphical displays are recommended
in the evaluation of each meteorological episode:
Evaluation
52
-------
Time-series plots of predicted and observed hourly ozone
values should be constructed for each simulation period for
each monitoring station where data are available.
Ground-level isopleths or tile maps of the spatial distribu-
tion of predicted concentrations should be constructed for
selected hours. Also, ground-level isopleths or tile maps of
the daily ozone maxima should be constructed. The correspond-
ing observed concentrations should be superimposed on the
predicted concentration isopleths to analyze spatial plume
patterns and ozone magnitudes.
Scatterplots should be constructed for all hourly prediction-
observation pairs for each simulation; quantile-quantile
plots are also recommended for each simulation.
The development of additional graphical displays, such as the
paired predictions of daily maxima, is encouraged. The
graphical displays to be used in the model performance
evaluation should be described in the Protocol.
5.1.2 Statistical performance measures
i
I • ' . , • , ,
Statistical performance measures can provide meaningful
measures of model accuracy for dense monitoring networks, such as
those for special field studies. However, statistical measures may
give a distorted view of model performance in cases of routine
monitoring networks, where coverage may be sparse. The Technical
Cpmmittee should evaluate the adequacy of the existing monitoring
network for conducting statistical tests for performance
evaluation.
Recommendations
It is recommended that, at a minimum, the following
mathematical formulations be applied as measures for model
For this purpose, the predicted value is the weighted
average of the predictions from the four grid cells nearest to
the monitoring station. The four-cell weighted average is
derived from bilinear interpolation as described in Reference 17.
53
-------
performance evaluation.
Appendix C.
These formulations are detailed in
Unpaired highest-prediction accuracy - This measure quantifies
the difference between the highest observed value and highest
predicted value over all hours and monitoring stations.
Normalized bias test - This'test measures the model's ability
to replicate observed patterns during the times of day when
available monitoring and modeled data are most likely to
represent similar spatial scales.
Gross error of all pairs above 60 ppb - In conjunction with
bias measurements, this metric provides an overall assessment
of base case performance and can be used as a reference to
other modeling applications. Gross error can be interpreted
as precision.
Additional measures may include the following:
Average station peak prediction accuracy - This is a measure
of peak performance at all monitor sites, using pairings based
on time and space.
Bias of all pairs above (50 ppb - This bias metric measures the
overall degree to which model predictions overestimate or
underestimate observed values. Note, however, that a zero
bias for several observation-prediction pairs can be caused by
a canceling effect of overprediction and underprediction in
different subregions.
Bias of all station peak£5 - For this metric, bias calculations
are performed on observation-prediction pairs associated with
peak ozone values for each monitoring station. This metric
provides information on the model's ability to replicate peak
ozone observations.
Fractional bias for peak concentration - Fractional bias is
calculated for both the mean and the standard deviation of
peak predicted and observed values. This metric provides
additional information on the model's ability to replicate
peak ozone observations.
Evaluation
54
-------
5.2 Assessing Model Performance Results
Both graphical and statistical performance measures should be
used for the performance evaluation. However, although the
recommended statistical measures should be applied for all
meteorological episodes and monitoring networks, caution is
suggested for interpreting these measures in cases of sparse
r
monitoring network coverage. The Technical Committee should
consider the monitoring network design in interpreting statistical
measures.
, In assessing model simulation results for the performance
evaluation, there is no rigid criterion for model acceptance or
rejection (i.e., no pass/fail test). Reference 17 states that,
ba|sed on past photochemical model evaluations, this type of
modeling "generally produces peak (unpaired) prediction accuracy,
ov.erall bias, and gross error statistics in the approximate ranges
of ±15-20 percent, ±5-15 percent, and 30-35 percent, respectively."
In general, performance results that fall within these ranges would
be acceptable. However, caution is urged in using these ranges as
th|e sole basis for determining the acceptability of model perform-
ance. These ranges were derived from past iiodel performance
1 •
evaluations with varying densities of air quality and meteorologi-
cal monitoring networks and corresponding variations .in the quality
arid quantity of aerometric model input data. In some cases, they
reflect use of earlier versions of the UAM. Thus, these ranges
should be used in conjunction with the graphical procedures to
assess overall model performance.
If statistical results are worse than the above ranges and
graphical analyses also indicate poor model performance, users
should consider choosing an alternative meteorological episode for
modeling. Performance evaluations should be* done on other
55
-------
candidate episodes to identify those that might result in better
model performance.
If statistical results are worse than the above ranges for any
of the three statistics, but graphical analyses generally indicate
acceptable model performance, simulation results used for attain-
ment demonstration should be applied with caution. Users may
consider conducting performance evaluations on other candidate
episodes to identify any that might yield improved model
performance.
A minimum of 3 primary episode days is required for use in the
model simulations for attainment demonstration (Section 6.4). If
fewer than 3 primary episode days can be identified that have
acceptable model performance for the attainment demonstration, the
responsible regulatory agencies are strongly encouraged to take
steps that will improve model performance for any future attainment
demonstrations. For serious and above nonattainment areas, this
may require short, intensive field studies to supplement installa-
tion of the enhanced monitoring network required under the CAAA of
1990.
Recommendations
It is recommended that the model performance for each
meteorological episode be assessed as follows:
1. The graphical performance procedures specified in Section
5.1.1 should be conducted for each meteorological episode. To
assess model performance, the Technical Committee should
analyze the time-series plots r ground-level isopleths,
quantile-guantile plots, and scatterplots. Use of "paired"
predictions of daily maxima may also be considered.
2. The statistical performance measures specified in Section
5.1.2 should also be derived and evaluated for each meteoro-
logical episode. When interpreting these measures, the
monitoring network density and design should be considered.
Evaluation ;!:
56
-------
Caution is urged when interpreting the statistical measures
for a sparse monitoring network.
It is recommended that the statistical performance measures be
compared with the following ranges:
• Unpaired highest prediction accuracy: ±15-20 percent
• Normalized bias: ±5-15 percent
• Gross error of all pairs >60 ppb: 30-35 percent
If all of these statistical measures are within the ranges
shown, and the graphical performance procedures also are
interpreted to yield acceptable results, then the model is
judged to be performing acceptably.
If any of the statistical measures are worse than the above
ranges, or the graphical procedures are interpreted to yield
unacceptable performance, users should consider choosing an
alternative highly ranked meteorological episode for the
attainment demonstration. Performance evaluations should be
conducted on a prospective alternative episode to determine
whether it yields improved model performance.
Additional model performance measures are encouraged.
should be described in the Modeling Protocol.
These
* Evaluation.
57
-------
Evaluation
58
-------
j CHAPTER 6
; ATTAINMENT DEMONSTRATION
i . • ' ' I
i - •
[ This chapter provides guidance on using modeling simulations
for attainment demonstrations. The primary reason for conducting
f '
photochemical modeling is to demonstrate the effectiveness of
alternative control strategies in attaining the NAAQS for ozone
throughout the modeling domain. This demonstration consists of
four main parts: (1) developing attainment-year modeling emission
inventories, (2) developing alternative-control strategy emission
inventories, (3) performing model simulations for the attainment
year with and without alternative control strategies, and
($) comparing attainment year and control strategy simulation
results with the ozone NAAQS. Attainment year and control strategy
simulations are conducted for each selected meteorological episode
(see Section 3.1).
6;. 1 Developing Attainment—Year Model Inputs
; The attainment-year modeling inventory must be derived from
the 1990 SIP nonattainment base year inventory, adjusted for
episode-specific meteorology, and then projected to the attainment
year. Also, to the extent possible, initial- and boundary-
condition ozone and precursor concentrations must be projected to
the attainment year. The attainment year is determined by the
npnattainment area designation and the attainment dates specified
in the 1990 CAAA. Projections of emission inventories reflect the
net effect of mandated controls and growth projections for various
source categories. Guidance for projecting inventories is
" Demonstration
59
-------
available in Procedures for Preparing Emissions Projections.14 The
most direct method for projecting initial- and boundary-condition
precursor concentrations is by applying ROM simulation results for
which the UAM domain is nested within the ROM domain (see
Chapter 3). In the absence of available ROM data, the projection
of ozone precursor concentrations used for initial conditions
typically has been done by linear scaling based on emission changes
projected to take place from the 1990 base year to the future year.
For initial ozone concentrations, there is little basis for doing
anything other than assuming initial ozone remains constant. In
the absence of regional modeling results or better information, the
guidance in Reference 7 for specifying future boundary conditions
may be followed.
Recommendations
It is recommended that the EPA guidance document entitled
Procedures for Preparing Emissions Projections14 be consulted
for developing attainment-year inventories. The guidance
document provides procedures for projecting point-source,
area-source, mobile-source, and biogenic emissions, and
addresses projections of spatial, temporal, and chemical
composition changes between the 1990 SIP inventory and the
attainment-year inventory.
Also, if regional modeling predictions for the attainment year
are available, it is recommended that these be used to derive
the attainment-year initial and boundary conditions for the
attainment-year model simulations (see Chapter 3).
6.2
;tion of Attainment
Many possible attainment-year emission control strategies can
be set up and simulated. Eventually, a modeling analysis must be
submitted for approval as the basis of a SIP demonstration. The
effectiveness of a given set of control measures in reducing ozone
(and perhaps other pollutants) is a major factor in selecting the
final emission control strategy.
Demonstration
60
-------
i Prior studies have typically used a progression of control
strategy scenarios in the modeling to ascertain an effective
i - • • - -.
strategy for attainment. A suggested logical progression is the
following:
Simulate the CAAA and other mandated control measures for
the attainment year to determine if these measures are
sufficient to demonstrate attainment of the ozone NAAQS.
If mandated controls are insufficient to demonstrate
attainment, superimpose a series of additional, across-
the-board reductions in VOCs-only, VOCs-plus-NOv, and
A
NOx-only strategies, relative to the mandatory CAAA
controls, to identify a suitable emission-reduction
target range. I
3. Once an approximate target range is ascertained in
steps 1 and 2, simulate control strategies that reflect
source-specific or source-category-specific control
measures and that realize the approximate emission
reductions identified as sufficient to reduce daily
maximum ozone to 0.12 ppm or less.
4. Adjust the strategy chosen in step 3 until it is
sufficient to demonstrate attainment of the NAAQS, as
described in Section 6.4. Adjustments; may be needed in
VOC controls, or NOX controls, or both.
Recommendations
The procedures for deriving control strategies for evaluation
in the attainment demonstration must be specified in the
Modeling Protocol.
61
-------
Emission control strategies for linked urban-area modeling
domains (e.g., northeastern U.S. Corridor) should be coordi-
nated among State agencies having lead responsibility for
respective domains to ensure consistency among the domains.
6.3 performing Attainment-Year Simulations to Assess Various
Control Strategies
Many graphical display and numerical procedures are available
for illustrating the effects of alternative emission control
strategies on predicted concentrations of ozone and other species.
For example, the emission levels in the control strategies are
often compared with the attainment-year base emissions. Also of
interest are comparisons with the inventory derived for purposes of
model performance evaluations and corresponding base-case UAM
results. Difference maps are extremely useful for illustrating
changes in daily maximum ozone predictions throughout the modeling
domain.
Recommendations
The focus of any ozone attainment demonstration is on the
daily maximum 1-hour concentration predicted at each location
in the modeling domain. However, it is recommended that
responsible parties broaden the scope of an attainment demon-
stration to examine the impact on other important metrics,
such as different concentration averaging times, population
exposure, subdomain and temporal impacts, effects on other
pollutant species, and other important measures that are
sensitive to emission control strategies.
For deriving initial and boundary conditions for a particular
urban-area domain, using appropriate regional model predic-
tions that reflect control measures applied in other urban-
area domains within the regional modeling domain is
recommended.
Demonstration
62
-------
6.4 Using Modeling Results in the Attainment: Demonstration
j As described in Section 3.1, at least 3 primary episode days
shpuld be modeled for the attainment demonstration. Also, a
minimum of 1 primary day should be modeled from each identified
meteorological regime. Therefore, for example, if there are three
meteorological regimes, at least 1 primary episode day from each
regime should be modeled; if there are only two meteorological
regimes, at least 2 primary episode days should be modeled from one
of I the regimes and at least 1 primary episode day modeled from the
other regime. Note that the episodes simulated would generally be
at least 48 hours long (i.e., the first day would be an initial
modeling day and the second day would be the primary episode day).
This would count as simulation of 1 primary episode day.
To demonstrate attainment of the ozone NAAQS, there should be
f
no! predicted daily maximum ozone concentrations greater than
0.12 ppm anywhere in the modeling domain for each primary episode
day modeled. Alternative methods for demonstrating attainment must
be! approved by the appropriate EPA Regional Office on a case-by-
case basis.
i The attainment test described in the preceding paragraph is
consistent with the flexibility allowed in the choice of episode
days (see Section 3.1) and reflects concerns over the difficulty of
accurately estimating emissions inputs to the model.
Recommendations
To demonstrate attainment of the ozone NAAOS. there should be
no predicted daily maximum ozone concentrations greater than
0.12 t>t>m anvwhere in the modeling domain for each primary
episode day modeled.
be modeled.
At least 3 primary episode days should
63
-------
States may opt to conduct more comprehensive statistical
testing of the modeling results for the attainment demonstra-
tion. Any alternative methods for attainment demonstration
must be approved by the appropriate EPA Regional Office on a
case-by-case basis. Any optional methods should be agreed
upon during the development of the Modeling Protocol.
6.5 libeceptions to Guidance Document
It is not possible in a general guidance document like this to
anticipate all contingencies associated with developing an
attainment demonstration study. The Modeling Policy Oversight and
Technical Committees responsible for a specific modeling study may
propose an alternative photochemical modeling approach provided
that (1) the Modeling Protocol requires consensus on the proposed
alternative approach within the Technical Committee, and (2)
justification for the proposed approach is documented. Application
of any alternative photochemical modeling approach must first
receive concurrence in writing from the responsible EPA Regional
Office(s).
Demons tr ertioti'; I.
64
-------
REFERENCES
i.;
2.
3.
4.
5.1
6.
7.1
Morris, R.E., T.C. Myers and J.L. Haney, 1990. User's Guide
for the Urban Airshed Model. Volume I: User's Manual for UAM
(CB-IV) . EPA-450/4-90-007A, U.S. Environmental Protection
Agency, Research Triangle Park, NC (NTIS No.: PB91-131227).
Morris, R.E., T.C. Myers, E.L. Carr, M.C. Causley, S.G.
Douglas and J.L. Haney, 1990. User's Guide for the Urban
Airshed Model. Volume II; User's Manual for the UAM fCB-IV)
Modeling System. EPA-450/4-90-007B, U.S. Environmental
Protection Agency, Research Triangle Park, NC (NTIS No.: PB91-
131235).
Douglas, S.G., R.C. Kessler and E.L. Carr, 1990. User's Guide
for the Urban Airshed Model. Volume III; User's Manual for the
Diagnostic Wind Model. EPA-450/4-90-007C, U.,S. Environmental
Protection Agency, Research Triangle Park, NC (NTIS No.: PB91-
131243). i
Causley, M.C., J.L. Fieber, M. Jimenez and L. Gardner, 1990.
User's Guide for the Urban Airshed Model. Volume IV; User's
Manual for the Emissions Preprocessor System. EPA-450/4-90-
007D, U.S. Environmental Protection Agency, Research Triangle
Park, NC (NTIS No.: PB91-131250).
Tang, R.T., S.C. Gerry, J.S. Newsom, A.R. Van Meter, R.A.
Wayland, J.M. Godowitch and K.L. Schere, 1990. User's Guide
for the Urban Airshed Model. Volume V; Description—and
Operation of the RQM-UAM Interface Program System. EPA-450/4-
90-007E, U.S. Environmental Protection Agency, Research
Triangle Park, NC (NTIS No.: PB91-131268).
DaMassa, J. , 1990. Technical Guidance Document: Photochemical
Modeling. California Air Resources Board, Technical Support
Division, Sacramento, CA.
U.S. Environmental Protection Agency, 1991. Criteria for
Assessing the Role of Transported Ozone/Precursors in Ozone
Nonattainment Areas. EPA-450/4-91-015, Office of Air Quality
Planning and Standards, Research Triangle Park, NC (NTIS No.:
PB91-195958).
65
-------
8.
9.
10,
11,
12,
13,
14,
15,
16,
17,
U.S. Environmental Protection Agency, 1991. Gridded Model
Information Support System (GMISS1 User's Guide. Volume II;
UAM Subsystem, EPA-450/4-91-009/ Office of Air Quality
Planning and Standards, Research Triangle Park, NC (NTIS No.:
PB91-206268).
U.S. Environmental Protection Agency, 1987. On-Site Meteoro-
logical Program Guidance for Regulatory Modeling Applications.
EPA-450/4-87-013, Office of Air Quality Planning and Stan-
dards, Research Triangle Park, NC (NTIS No.: PB87-227542).
U.S. Environmental Protection Agency, 1989. Procedures for
Applying City-Specific EKMAf EPA-450/4-89-012, Office of Air
Quality Planning and Standards, Research Triangle Park, NC
(NTIS No.: PB90-256777)„
U.S. Environmental Protection Agency, 1991. Enhanced Ozone
Monitoring Regulations: Proposed Rule (Drafts , 40 CFR Part 58,
Office of Air Quality Planning and Standards, Research
Triangle Park, NC.
U.S. Environmental Protection Agency, 1991. Emission Inven-
tory Requirements for Ozone State Implementation Plans, EPA-
450/4-91-010, Office of Air Quality Planning and Standards,
Research Triangle Park, NC. ,
U.S. Environmental Protection Agency, 1991. Procedures for
the Preparation of Emission Inventories for Volatile Organic
Compounds. Volume II; Emission Inventory Requirements for
Photochemical Air Quality Simulation Models, EPA-450/4-91-014,
Office of Air Quality Planning and Standards, Research
Triangle Park, NC (NTIS No.: PB91-216176).
U.S. Environmental Protection Agency, 1991. Procedures for
Preparing Emissions Projections, Office of Air Quality
Planning and Standards, Research Triangle Park, NC (Draft).
U.S. Environmental Protection Agency, 1991. Guidance for
Initiating Ozone/CO SIP Emission Inventories Pursuant to the
1990 Clean Air Act Amendments, Office of Air Quality Planning
and Standards, Research Triangle Park, NC.
U.S. Environmental Protection Agency, 1989. Aerometric
Information Retrieval System (AIRS'). Volume I. Office of Air
Quality Planning and Stemdards, Research Triangle Park, NC.
Tesche, T.W., P. Georgopoulos, F.L. Lurmann and P.M. Roth,
1990. Improvement of Procedures for Evaluating Photochemical
Referenced
66
-------
Models. Draft Final Report. California Air Resources Board,
Sacramento, CA. :
18. U.S. Environmental Protection Agency, 1991. Urban Airshed
Model Postprocessing System User's Guide, Office of Air
Quality Planning and Standards, Research Triangle Park, NC
(Draft).
Inferences
67
-------
References
68
-------
APPENDIX A
RECOMMENDED MODELING PROTOCOL CONTENTS
! Table 1 of Chapter 2 lists recommended contents for a Modeling
Protocol. This appendix gives a general description of each compo-
nent, to aid in the development of the Protocol. As stated in
Chapter 2, the contents presented here are adopted from the GARB
Technical Guidance Document: Photochemical Modeling.6
UAM MODELING STUDY DESIGN
Background and Objectives
t ' ' -
The Protocol Document should describe the policy and technical
objectives of the study and pertinent background, information such
as the legislative mandate under which the study is being done.
Schedule
; Development of a complete schedule for all phases of the
project is needed. The critical paths and deadlines should be
identified and discussed, as should a schedule for addressing
critical issues that require special attention, such as air quality
and meteorological data preparation and quality eissurance, episode
selection, and emission inventory preparation and quality
assurance.
69
-------
Deliverables
A list of the interim and final deliverables for the modeling
study should be specified.
Modeling Policy
rersiaht/Technical Committees
The composition and responsibilities of the Modeling Policy
Oversight and Technical Committees should be specified to the
extent possible. Meeting frequency and circumstances for
convening a meeting should be identified. Because technical
conflicts may arise, a resolution process for handling them should
be included.
Participating Organizations
The organizations that are sponsoring the modeling study and
those that may contribute to it should be identified.
^Relationship to Regional Modeling Protocols
Procedures for coordinating development of the urban-area
Modeling Protocol with the regional Modeling Protocol should be
described. This would include a description of control strategies,
emission inventories, projection years, modeling episodes, etc.
The coordination of urban-area Modeling Policy and Technical
Committees with their regional counterparts should be described.
Relationship to Other Urban Area Modeling Protocols
In some cases, such as the Northeast U.S., nonattainment
MSA/CMSAs required to do attainment demonstrations may be linked to
other nonattainment MSA/CMSAs. It is important that procedures be
Appendix A
70
-------
established for coordinating the Modeling Protocols among these
areas, and that these procedures be clearly specified in each
nonattainment area Modeling Protocol. It is likely that Modeling
Policy Oversight and Technical Committees will include some joint
membership among the nonattainment areas.
t '
i .
Relationship to Planning/Strategy Groups
I Key planning agencies and others responsible for emission
projections or other model inputs should be identified, and the
means by which these groups interact to obtain realistic growth
prpjections and control strategies should be discussed.
• i . •
DOMAIN AND DATA BASE ISSUES
i
Preprocessor Programs
The preprocessor programs to be used in constructing any of
the model input fields should be identified and described.
Data Bases
: The proposed air quality and meteorological data bases should
be described. The completeness of the data base, techniques for
filling in missing data, and quality assurance procedures should be
discussed.
Base Meteorological Episode Selection
i
i The episode selection criteria should be detailed, including
the methodology to group candidate episodes into meteorological
I
regimes. How the episodes will be used in the modeling study
should also be described.
Appendix
71
-------
Modeling Domain
The Protocol should describe the criteria for selecting the
size and location of the modeling domain. This would include a
description of the MSA/CMSA area size, locations of major sources
outside the MSA/CMSA that may affect it, sensitivity analyses that
may be conducted to assess boundary effects on domain predictions,
relationship of domain size to the episodes selected for use in the
modeling study, etc.
Horizontal Grid Resolution
The Protocol should describe the horizontal grid resolution to
be applied to the modeling domain. If a resolution coarser than
5 x 5 km is chosen, justification for this choice should be
provided.
Number of Vertical Layers
The Protocol should specify the number of vertical layers to
be used in the UAM simulations. If a layering scheme other than
the one recommended in Chapter 3 is chosen, justification for using
the alternative layering should be given.
Emission Inventory
The assumptions, methodologies, and appropriate guidance
references to be used in constructing the modeling emission
inventory should be described. Quality assurance procedures should
also be described. /
Appendix A
72
-------
Specification of Initial and Boundary Conditions
i The techniques to be used to specify the initial and boundary
conditions for the base meteorological episodes and the attainment
year should be described. The assumptions to be used in forecast-
ing attainment-year conditions should be documented. If a nested
grid approach is used (e.g., using predictions from the ROM through
the ROM/UAM Interface System), the details for implementation
should be described (see Chapter 3).
Wind Field Specification
The proposed techniques for.specifying the wind fields should
be described. The procedures to be used to determine the represen-
tat:iveness of the simulated wind fields should be technically
justified and documented (see Chapter 3).
Miking Heights
! The techniques to be used for deriving the mixing height for
the modeling domain should be described.
Sources of Other Input Data
! The Protocol Document should describe the data "and techniques
to| be used to specify other input data, such as cloud cover, water
vapor, UV radiation, surface temperature, terrain, and land use and
surface characteristics.
73
-------
QUALITY ASSURANCE AND DIAGNOSTIC ANALYSES
Quality Assurance Tests of Input. Components
The specific quality assurance tests to be used on the data
input fields should be described (see Chapter 4).
Diagnostic Tests of Base Case Simulation
The specific diagnostic tests to be used for the base-case
meteorological episode simulations should be described. As
discussed in Chapter 4, these should include, at a minimum, time-
series plots, observed and predicted ozone maps, zero emissions and
zero boundary conditions tests, and tests on the mixing height
variations and wind fields. Additional diagnostic tests are
encouraged and should be described in the Protocol.
MODEL PERFORMANCE EVALUATION
Performance Evaluation Tests
The graphical, statistical, and other measures to be used in
the model performance evaluation should be specified. At a
minimum, the tests recommended in Chapter 5 should be included.
Additional measures may also be considered and should be described
if they are to be used.
Appendix A
74
-------
ATTAINMENT DEMONSTRATIONS
Identification of Attainment-Year Mandated Control Measures
i The Protocol Document should include a description of the 1990
CAAA control measures and other measures mandated to be implemented
by the attainment year.
Methodologies for Generating Control Strategy Emission Inventories
1 The procedures for deriving alternative-control-strategy
emission scenarios to meet the study objectives should be de-
scribed. A description of how the control scenarios would relate
to applicable control strategies for areas adjacent to the modeling
domain (particularly upwind areas) should be included.
Procedures for Attainment Demonstration
i • •
i Procedures for using the model simulation results in
demonstrating attainment of the ozone NAAQS should be included.
SUBMITTAL PROCEDURES
i The documentation and analyses that will be submitted for EPA
Regional Office review should be described. Also, any
documentation other than the Modeling Protocol requiring EPA
Regional Office approval should be described.
Appendix A.
75
-------
Appendix
76
-------
APPENDIX B
IDENTIFICATION OF METEOROLOGICAL REGIMES
CORRESPONDING WITH HIGH OBSERVED OZONE
The following is a procedure that may be used to assist in
selecting modeling episodes. Other techniques may be considered on
a case-by-case basis; they should be described in the Modeling
Protocol and approved by the appropriate EPA Regional Office.
! Identification of meteorological regimes for a given area
under review begins with constructing a climatological windrose of
high ozone days. The windrose is constructed by first selecting
all days from the period 1987 to present during which at least one
ozone monitor within the area recorded an exceedance of the ozone
NAAQS or some other cutoff level (e.g., 100 ppb). Additional years
of: data are encouraged in constructing the climatological windrose
(e.g., 1980-1991). Next, for each exceedance day, calculate the
morning (i.e., 7:00 a.m. - 10:00 a.m.) resultant wind velocity.
Then group the resultant wind velocities for all of the exceedance
days into eight compass directions plus calm, to establish a
climatic windrose of high-ozone days for the area under review.
Calm winds are defined as those with speeds less than 1.5 m/s and
referred to as the null wind direction. The windrose will include
nine bins (0-8); place the wind directions corresponding to the
eight compass points into bins 1-8, and the calm or null wind
direction into bin 0. The bins with frequencies significantly
higher than the average frequency for all bins should be defined as
the "predominant wind directions" (PWD).
Appendix B
77
-------
Next, compare the morning (i.e., 7:00 a.m. - 10:00 a.m.)
resultant wind velocity for each exceedance day during 1987-89 and
more recent years with the climatic windrose of high-ozone days.
Categorize exceedance days with wind directions corresponding to
previously identified climatic PWD's as belonging to that PWD.
Lump all other exceedance days occurring during 1987-89 and later
into a category called "other." Rank each exceedance day within
each PWD category and within the "other" category according to its
areawide daily maximum ozone observation. Within each category,
the day with the highest areciwide daily maximum concentration is
ranked first.
After the steps described in the two preceding paragraphs are
completed, meteorological regimes can be defined for use in the
attainment demonstration test described in Section 6.4. This may
be done as follows:
1. Choose the two PWD's which contain the highest areawide
daily maximum ozone values from 1987 to the most recent
year with data available. These represent two of the
meteorological regimes to consider in the attainment
test.
2. The third "meteorological regime" to be considered in the
attainment test is comprised of all exceedance days
previously categorized as "other" plus those belonging to
any PWD not chosen in step 1.
f
3. Identify the top 3-ranked exceedance days from within
each of the three meteorological regimes identified in
steps 1 and 2. These days are candidates for modeling in
the attainment test. Final choice from among these
Appendix B
78
-------
candidates is based on criteria identified in
Section 3.1.
!
It may happen that one or more of the meteorological
regimes identified in step 1 contains fewer than 3
exceedance days. If this occurs, exceedance days
included within PWD's which have been lumped in the third
meteorological regime (see step 2) may be added to one or
both of the first two regimes. If this proves necessary,
selection of days to supplement those in one or both of
the first two regimes needs to be decided on a case-by-
case basis keeping in mind the goal of this exercise: to
provide a choice of exceedance days reflecting high ozone
concentrations with meteorological conditions which
frequently coincide with observed exceedances.
79
-------
;•- Appendix
80
-------
; APPENDIX C
i PERFORMANCE MEASURE FORMULATIONS
t • • ;
|
RECOMMENDED PERFORMANCE MEASURES1
i
j
1. ! Unpaired Highest-Prediction Accuracy (Au)
i ' !
[
[ , '
[ . • :
i r* ( \ — C* ( } i
A = °°{-' ^ £p*'"; X100%
where
cft(.,.)
unpaired highest-prediction accuracy
(quantifies the difference between the
magnitude of the highest2 1-hour observed
value and the highest 1-hour predicted value
maximum 1-hour observed concentration over
all hours and monitoring stations
maximum 1-hour predicted concentration over
all hours and surface grid squares
on Reference 17.
I 2"Highest" refers to the maximum 1-hour concentration across
all hours and monitoring stations;
.. Appendix C
81
-------
2. Normalized Bias Test CD 'I
where
N
normalized bias obtained from all hourly
prediction-observation pairs
number of monitoring stations
number of hourly prediction-observation pairs
for monitoring station i
Nn,
total number of station-hours
N
observed value at monitoring station i for
hour j
predicted value at monitoring station i for
hour j
Predicted value derived from bilinear interpolation of the
predicted values at the four grid cells nearest to station
station ± for the given hour.
82
-------
3. Gross Error of All Pairs >60 ppb
,J) - Cp(i,j)
iKL J=l
where
N
cp(i,j)4
normalized gross error for all hourly
prediction-observation pairs for hourly
observed values >60 ppb
i
total number of station hours (defined
previously)
i
number of monitoring stations?
number of hourly prediction-observation pairs
for monitoring station i
observed value >60 ppb at monitoring station
i for hour j
predicted value at monitoring station i for
hour j
Predicted value derived from bilinear interpolation of the
predicted values at the four grid cells nearest to station i for
the given hour.
'
Appendix C
83
-------
OTHER SUGGESTED PERFORMANCE MEASURES
1. Average Station Peak Prediction Accuracy A
1 = 1
N
.tj -cd.tj
, .
'
where
mean paired peak5 prediction accuracies
averaged over all monitoring stations
N
number of monitoring stations
peak observed value at monitoring station i
for hour tj
predicted value at monitoring station i for
hour t<
5 "Peak" refers to the daily maximum 1-hour concentration at
a particular monitoring station.
6For these "Other Suggested Performance Measures,"
"predicted" can be interpreted in one of several ways: (1) as
the result of bilinear interpolation described in footnote 4;
(2) using the procedures described for paired predictions of
daily maxima (described on page 52 of the text); (3) using the
prediction for the grid square containing the monitor site only.
The Modeling Protocol should document the procedure used to
determine "predicted" values in these tests.
84
-------
t:
hour of peak observed value at monitoring
station i
2. : Bias of All Pairs >60 ppbfDgQ1
N
where
D
60
N
non-normalized bias from all hourly
prediction-observation pairs for observed
values >60 ppb
I
total number of station-hours (defined
previously)
number of monitoring stations;
number of hourly prediction-observation pairs
for monitoring station i
observed value >60 ppb at monitoring station
i for hour j
85
Appendix C
-------
cp(i,j)7 =
predicted value at monitoring station i for
hour j
3. Bias of All Station Peaks (D,
N
±, tj
where
Dpeak
non-normalized bias from all prediction-
observation pairs for peak8 observed values
at all monitoring stations
N
number of monitoring stations
peak observed value at monitoring station i
for hour t)
'For these "Other Suggested Performance Measures,"
"predicted" can be interpreted in one of several ways: (1) as
the result of bilinear interpolation described in footnote 4;
(2) using the procedures described for paired predictions of
daily maxima (described on page 52 of the text); (3) using the
prediction for the grid square containing the monitor site only.
The Modeling Protocol should document the procedure used to
determine "predicted" values in these tests.
8"Peak" refers to the daily maximum l-hour concentration at
a particular monitoring station.
Appendix C
86
-------
predicted value at monitoring station i for
hour t:
I tj • . = hour of peak observed value at monitoring
1 station i
[
4. ; • Fractional Bias for Peak Concentration j
f ' .
•; The fractional bias is calculated for both the mean and
standard deviation of peak ozone values, as follows:
, .
where
fractional bias of means
fractional bias of standard deviation
9For these "Other Suggested Performance Measures,"
"predicted" can be interpreted in one of several ways: (1) as
the result of bilinear interpolation described in footnote 4;
(2|) using the procedures described for paired predictions of
daily maxima (described on page 52 of the text); (3) using the
prediction for the grid square containing the monitor site only,
Th<= Modeling Protocol should document the procedure used to
determine "predicted" values in these tests.
Appendix C
87
-------
m,.
mean maximum observed concentration
m = mean peak predicted concentration
s0 = standard deviation of peak observed
concentrations
Sp = standard deviation of peak predicted
concentrations
The means and standard deviations of predicted and observed
concentrations are determined by each of two methods:
Peak station values:
maximum observed concentration at monitoring
station i across all hours
cp(i,O
10 _
maximum predicted concentration at monitoring
station i across all hours
where i
1,...,N monitoring stations
Peak hourly values:
these "Other Suggested Performance Measures,"
"predicted" can be interpreted in one of several ways; (1) as
the result of bilinear interpolation described in footnote 4;
(2) using the procedures described for paired predictions of
daily maxima (described on page 52 of the text); (3) using the
prediction for the grid square containing the monitor site only.
The Modeling Protocol should document the procedure used to
determine "predicted" values in these tests.
Appendix C
88
-------
maximum observed concentration at hour j
across all monitoring stations
,11 -
maximum predicted concentration at hour j
across all monitoring stations
where j
1,...,H hours
i The fractional bias of the mean and standard deviation
I s
varies from -2 to +2. Negative values indicate c>verprediction
and positive values indicate underprediction.
nFor these "Other Suggested Performance Meassures,"
"predicted" can be interpreted in one of several ways: (1) as
the result of bilinear interpolation described in footnote 4;
(2) using the procedures described for paired predictions of
daily maxima (described on page 52 of the text); (3) using the
prediction for the grid square containing the monitor site only,
The Modeling Protocol should document the procedure used to
determine "predicted" values in these tests.
Appendix C
89
-------
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
BEPOBT N.O,
EPA-450/4-91-013
3. RECIPIENT'S ACCESSION NO.
HTLE AND SUBTITLE
Guideline for Regulatory Application
of the Urban Airshed Model
5. REPORT DATE
July, 1991
6. PERFORMING ORGANIZATION CODE
AUTHOR(S)
8. PERFORMING ORGANIZATION REPORT NO.
PERFORMING ORGANIZATION NAME AND ADDRESS
10. PROGRAM ELEMENT NO.
U.S. Environmental Protection Agency
Office of Air Quality Planning and Standards
Technical Support Division
Research Triangle Park, NC 27711 _
11. CONTRACT/GRANT NO.
2. SPONSORING AGENCY NAME AND ADDRESS
13. TYPE OF REPORT AND PERIOD COVERED
14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
16. ABSTRACT
This document provides guidance on application of the U.S. EPA Urban Airshed
Model for ozone nonattainment area State Implementation Plan development as
required by the Clean Air Act Amendments of 1990. The Urban Airshed Model is an
urban-Scale, grid-based photochemical dispersion model. The model provides a means
for studying the relationship of volatile organic compound and nitrogen oxide emission
to amb(ient levels of ozone in urban areas. This document provides information on
data requirements and model application strategy.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b. IDENTIFIERS/OPEN ENDED TERMS
c. COSATI Field/Group
Air Pollution
Atmospheric Dispersion Models
Photochemical Model
ozone
18. DISTRIBUTION STATEMENT
Release Unlimited
19. SECURITY CLASS (TillsReport!
Unclassified
21. NO. OF PAGES
20. SECURITY CLASS (This page/
Unclassified
22. PRICE
EPA Form 2220-1 (Rev. 4-77) PREVIOUS EDITION is OBSOLETE
-------
-------
-------
U.S.EPAOAQPS
RESEARCH TRIANGLE PARK, NC
AEPA
------- |