EPA-600/R-95-004
January 1995
THE DEVELOPMENT AND IMPROVEMENT OF
TEMPORAL ALLOCATION FACTOR FILES
Final Report
By:
Theresa Moody, J, David Winkler, Terry Wilson, and Sharon Kersteter
TRC Environmental Corporation
6340 Quadrangle Drive, Suite 200
Chapel Hill, NC 27514
EPA Contracts 68-D9-0173 (WA 3/314)
and 68-D2-0181 (WA 1/014)
EPA Project Officer: Charles O. Mann
Air and Energy Engineering Research Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Prepared for:
U.S. Environmental Protection Agency
Office of Research and Development
Washington, DC 20460
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before compl',
l, REPORT NO.
EPA-600/R-9 5-004
iiiiiiiiiii inn in
PB95-166153
4. TITLE AND SUBTITLE
The Development and Improvement of Temporal
Allocation Factor Files
5. REPORT DATE
January 1995
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
T.Moody, J. D. Winkler, T.Wilson, and S.- Kerateter
8. PERFORM! NG ORGANIZATION REPORT NO.
CH-94-35
9. PERFORMING ORGANIZATION NAME AND ADDRESS
TRC Environmental Corporation
100 Europa Drive, Suite 150
Chapel Hill, North Carolina 27514
10. PROGRAM ELEMENT NO,
and
68-D2-0181, WA 1/014
12. SPONSORING AGENCY NAME ANO ADDRESS
EPA, Office of Research and Development
Air and Energy Engineering Research Laboratory
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND PERIOD COVERED
Task Final; 3/93-5/94
14. SPONSORING AGENCY CODE
EPA/600/13
AEERL project officer is Charles O. Mann, Mail Drop 62, 919/
15, SUPPLEMENTARY NOTES
541-4593.
'The report gives results of a project to: (1) evaluate the quality and com-
pleteneis of data and methods being used for temporal allocation of emissions data,
(2) identify and prioritize needed improvements to current methods for developing
temporal allocation factors (TAFs), and (3) collect data to improve existing TAF
file's .""The project focused on improving profiles for stationary point sources. Well-
documented, national default allocation profiles were developed. The availability of
an improved TAF file does not eliminate the need to collect actual data for tempor-
ally resolved emissions when these data are needed for an emissions inventory. The
TAF file provides an improved database of default values that should be used when
actual data cannot be collected.'..(NOTE: Traditional emissions inventories produce
annual or daily emissions estimates. For photochemical models, hourly emissions
estimates are required. Ideally, hourly emissions would be measured directly; how-
ever, this approach is usually impractical due to technical and resource restraints.
Alternatively, hourly emissions are estimated using surrogate factors from "tem-
poral profiles" assigned to emissions source categories. This approach has been
followed in previous studies.)
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b. IDENTIFIERS/OPEN ENDEDTERMS
c. COSATI Field/Group
Pollution
Emission
Estimating
Allocation Models
Inventories
Pollution Control
Stationary Sources
Temporal Allocation
F actors
Emissions Inventories
13 B
14G
15E
18. DISTRIBUTION STATEMENT
Release to Public
19, SECURITY CLASS (This Report/
Unclassified
21. NO. OF PAGES
456
20. SECURITY CLASS (This page)
Unclassified
22, PRICE
EPA Form 2220-1 (9-73J
\
\
\
-------
EPA REVIEW NOTICE
This report has been reviewed by the U.S. Environmental Protection Agency, and
approved for publication. Approval does not signify that the contents necessarily
reflect the views and policy of the Agency, nor does mention of trade names or
commercial products constitute endorsement or recommendation for use.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
ABSTRACT
This report gives results of a project to: (1) evaluate the quality and completeness of data
and methods presently being used for temporal allocation of emissions data, (2) identify and
prioritize needed improvements to the current methods for developing temporal allocation
factors, and (3) collect data to improve existing temporal allocation factor (TAF) files. This
project focused on improving profiles for stationary point sources. Well-documented, national
default allocation profiles were developed. The availability of an improved TAF file does not
eliminate the need to collect actual data for temporally resolved emissions when these data are
needed for an emissions inventory. The TAF file provides an improved database of default
values that should be used when actual data can not be collected. (NOTE: Traditional emissions
inventories produce emissions estimates for annual or daily time periods. For photochemical
!
models, hourly emissions estimates are required. Ideally, emissions for specific hourly time
periods would be measured directly; however, this approach is normally impractical due to
technical and resource restraints. Alternatively, hourly emissions are estimated using surrogate
factors from "temporal profiles" assigned to emissions source categories. This approach has
been followed in previous studies.)
ii
-------
TABLE OF CONTENTS
Section/Chapter Page
Abstract ii
List of Figures vii
List of Tables ...... viii
List of Acronyms ix
Preface xi
1.0 INTRODUCTION 1-1
1.1 BACKGROUND 1-1
1.2 OBJECTIVES 1-3
1.3 CONTENTS 1-3
2.0 SUMMARY OF METHODS AND DATA CURRENTLY USED IN
TEMPORAL ALLOCATION OF EMISSIONS DATA 2-1
2.1 CURRENT TEMPORAL ALLOCATION FACTORS 2-1
2.1.1 National Acid Precipitation Assessment Program (NAPAP) 2-1
2.1.2 Flexible Regional Emissions Data System (FREDS) 2-2
2.1.3 Urban Airshed Model (UAM) Emissions Preprocessor System 2-4
2.1.4 Geocoded Emissions Modeling and Projections (GEMAP) System 2-5
2.2 INFORMATION GATHERING 2-6
2.2.1 Literature Search 2-6
2.2.2 Telephone Interviews 2-7
2.2.2.1 Results of Discussions with EPA 2-7
2.2.2.2 Results of Discussions with State/Local Regulatory
Agencies 2-7
2.2.2.3 Results of Discussions with Other Government
and Private Organizations 2-8
2.2.3 Results of Data Collection and Telephone Interviews 2-9
3.0 DATA SOURCES 3-1
3.1 NATIONAL ACID PRECIPITATION ASSESSMENT PROGRAM
(NAPAP) DATA 3-5
3.1.1 Data Description 3-5
3.1.2 Data Collection 3-6
3.1.3 Incorporation of Data into Intermediate File 3-6
3.1.4 Comments 3-12
3.2 ECONOMIC DATA 3-13
3.2.1 Data Description......... 3-14
3.2.2 Data Collection 3-15
3.2.3 Incorporation of Data into Intermediate File 3-16
3.2.4 Comments 3-17
3.3 CALIFORNIA AIR RESOURCES BOARD (CARB) DATA 3-19
iii
-------
TABLE OF CONTENTS (continued)
Section/Chapter Page
3.3.1 Data Description 3-19
3.3.2 Data Collection 3-20
3.3.3 Incorporation of Data into Intermediate File 3-20
3.3.4 Comments 3-21
3.4 TEXAS NATURAL RESOURCE CONSERVATION COMMISSION
(TNRCC) DATA 3-22
3.4.1 Data Description 3-22
3.4.2 Data Collection 3-23
3.4.3 Incorporation of Data into Intermediate File 3-23
3.4.4 Comments 3-25
3.5 SOUTHERN OXIDANT STUDY (SOS) DATA 3-25
3.5.1 Data Description 3-25
3.5.2 Data Collection ; 3-26
3.5.3 Incorporation of Data into Intermediate File 3-27
3.5.4 Comments 3-29
3.6 LAKE MICHIGAN OZONE STUDY (LMOS) DATA 3-29
3.6.1 Data Description 3-29
3.6.2 Data Collection 3-30
3.6.3 Incorporation of Data into Intermediate File 3-30
3.6.4 Comments 3-31
3.7 CONTINUOUS EMISSIONS MONITORING (CEM) DATA 3-31
3.7.1 Data Description 3-32
3.7.2 Data Collection 3-33
3.7.3 Generation of Interim File 3-34
3.7.3.1 Ohio CEM Data 3-34
3.7.3.2 Kentucky CEM Data 3-37
3.7.3.3 Pennsylvania CEM Data 3-39
3.7.4 Comments 3-40
3.8 WASTE-TO-ENERGY SOURCE (WTE) DATA 3-41
3.8.1 Data Description 3-41
3.8.2 Data Collection . 3-41
3.8.3 Incorporation of Data into Intermediate File 3-42
3.8.4 Comments 3-43
3.9 ACID-MODES FIELD STUDY DATA 3-43
3.9.1 Data Description 3-43
3.9.2 Data Collection 3-43
3.9.3 Incorporation of Data into Intermediate File 3-44
3.9.4 Comments 3-47
4.0 SOURCE CATEGORY PRIORITIZATION.. 4-1
4.1 INTRODUCTION ...............4-1
4.2 PRIORITIZATION SEQUENCE ...4-1
iv
-------
TABLE OF CONTENTS (continued)
Section/Chapter Page
5.0 FINAL TEMPORAL ALLOCATION FACTOR FILE DEVELOPMENT...... 5-1
5.1 DATABASE DESIGN 54
5.2 RULE DEFINITION 5-1
5.3 FILE COMBINATION 5-3
5.3.1 Module 1: Create Single Data Set 5-8
5.3.2 Module 2: Flat Profiles 5-8
5.3.3 Module 3; NAPAP Profiles 5-9
5.3.4 Module 4: Economic Data 5-9
5.3.5 Module 5: Six-Digit TNRCC Data............... 5-10
5.3.6 Module 6: Eight-Digit TNRCC Data................. 5-10
5.3.7 Module 7: Six-Digit Composite Daily and Diurnal Profiles 5-11
5.3.8 Module 8: Combine and Use Six-Digit TNRCC, LMOS, CEM
Data 5-11
5.3.9 Module 9: Combine and Use Eight-Digit TNRCC, LMOS,
CEM Data 5-12
5.3.10 Module 10: Other Data 5-13
5.3.11 ' Module 11: SOS Area Source Data 5-13
6.0 QUALITY ASSURANCE AND CONTROL OF TEMPORAL ALLOCATION
FACTOR FILE DATABASES ;. 6-1
6.1 INTRODUCTION 6-1
6.2 QUALITY CONTROL PROCEDURES 6-2
6.3 SUMMARY OF DATABASE INTEGRITY 6-3
7.0 CONCLUSIONS AND RECOMMENDATIONS
7.1 CONCLUSIONS
7.2 RECOMMENDATIONS
8.0 REVIEW AND ANALYSIS OF TEMPORAL ALLOCATION FACTOR FILE
8.1 INTRODUCTION
8.2 SCC SELECTION
8.3 PROFILE ANALYSIS
8.4 STATISTICAL ANALYSIS
8.4.1 Approach I: Measure of Dispersion Using Standard Deviations...
8.4.2 Approach II: Measure of Dispersion Using Absolute Differences
..7-1
..7-1
..7-2
..8-1
..8-1
.,8-1
..8-2
..8-6
8-13
8-14
9.0 CONCLUSIONS AND RECOMMENDATIONS .9-1
9.1 CONCLUSIONS 9-1
9.2 RECOMMENDATIONS 9-2
v
-------
TABLE OF CONTENTS (continued)
Section/Chapter Page
10.0 REFERENCES 10-1
APPENDIX A MAY 27,1993, JUNE 9,1993, AND JULY 7,1993
MEMORANDA... A-l
APPENDIX B DATA SOURCE MATRIX B-l
APPENDIX C BLS DATA C-l
APPENDIX D CARB DATA D-l
APPENDIX E SOS DATA E-l
APPENDIX F LMOS DATABASE FORMAT F-l
APPENDIX G CEM DATA G-l
APPENDIX H WASTE-TO-ENERGY DATA H-l
vi
-------
Number
LIST OF FIGURES
1 -1 Application of a temporal profile
3-1 Intermediate temporal allocation factor file format
3-2 Sample 1985 NAPAP temporal allocation profiles
5-1 Final TAF file format
5-2 File combination flow chart
7-1 Temporal profiles for SCC 10100101 from the TAF file..
vii
-------
LIST OF TABLES
Number Page
3-1 Daily Fractions for Emissions Allocations 3-4
3-2 Hourly Fractions for Emissions Allocations.. 3-4
3-3 Source Categories for Area Sources in the United States 1985
NAPAP Emissions Inventory 3-7
3-4 Fractions Used to Determine Daily Temporal Profiles for CARB Data 3-21
5-1 Tv a. I""1! r*
-1 Data Source Flags 5-3
5-2 Seasonal Profile Codes 5-4
5-3 Daily and Hourly (Diurnal) Profile Codes 5-5
8-1 Priority Ozone SCCs 8-3
8-2 Priority CO SCCs 8-4
8-3 Priority Ozone SCCs Review 8-7
8-4 Priority CO SCCs Review.................... 8-10
viii
-------
LIST OF ACRONYMS
ADOM
Acidic Deposition and Oxidant Model
AEERL
U.S. EPA, Office of Research and Development, Air and Energy Engineering
Research Laboratory,
AES
Atmospheric Environmental Service
AFS
AIRS Facility Subsystem
AIRS
Aerometric Information Retrieval System
AMS
AIRS Area and Mobile Source Subsystem
AS
Area Source
BLS
Business and Labor Statistics
CARB
California Air Resources Board
CEM
continuous emissions monitoring
CES
Current Employment Statistics
CO
carbon monoxide
COF
Control Options File
CRB
Commodity Research Bureau
CRC
Coordinating Research Council
DEM
Department of Environmental Management
DNR
Department of Natural Resources
DOE
U.S. Department of Energy
E/E
Employment and Earnings
EIA
U.S. DOE, Energy Information Administration
EIB
U.S. EPA, OAQPS, Emission Inventory Branch
EIS
Emissions Information System
EPA
U.S. Environmental Protection Agency
EPRI
Electric Power Research Institute
EPS
Emissions Preprocessor System
FCG
Florida Electric Power Coordinating Group
FIN
facility identification number
FREDS
Flexible Regional Emissions Data System
GAA
Governmental Advisory Associates, Inc.
GEMAP
Geocoded Emissions Modeling and Projections
h2s
hydrogen sulfide
HAP
hazardous air pollutant
HC1
hydrogen chloride
HF
hydrogen fluoride
ID
identification
IEPA
Illinois Environmental Protection Agency
Kb
kilobytes
LADCO
Lake Michigan Air Directors Consortium
LMOS
Lake Michigan Ozone Study
Mb
megabytes
MIP
Modeler's Input
MPO
metropolitan planning organization
MW-hr
megawatt-hour
NADB
National Allowance Data Base
ix
-------
LIST OF ACRONYMS (continued)
NAPAP
National Acid Precipitation Assessment Program
NCC
National Computing Center
NECRMP
Northeast Corridor Regional Modeling Project
NEDS
National Emissions Data System
nh3
ammonia
no2
nitrogen dioxide
NOx
oxides of nitrogen
OAIAP
Office of Atmospheric and Indoor Air Pollution
OAQPS
U.S. EPA, Office of Air Quality Planning and Standards
OME
Ontario Ministry of the Environment
PDS
partitioned data set
POTW
publicly-owned treatment works
ppm
parts per million
QA
quality assurance
QA/QC
quality assurance/quality control
RADM
Regional Acid Deposition Model
RAPS
Regional Air Pollutant Study
ROM
Regional Oxidant Model
SAMS
SEP Air Pollutant Inventory Management System
SCAQMD
South Coast Air Quality Management District
see
source classification code
SIC
Standard Industrial Classification (code)
SIP
State Implementation Plan
so2
sulfur dioxide
SOS
Southern Oxidant Study
SURE
Sulfate Regional Experiment
TAF
temporal allocation factor
TAM
Temporal Allocation Module
THC
total hydrocarbon(s)
TNRCC
Texas Natural Resource Conservation Commission
TRC
TRC Environmental Corporation
TSDF
hazardous waste treatment, storage, and disposal facility
TSP
total suspended particulate(s)
TVA
Tennessee Valley Authority
TZF
Time Zone File
UAM
Urban Airshed Model
VMT
vehicle miles traveled
VOC
volatile organic compound(s)
x
-------
PREFACE
This project was completed in two phases. The first phase, completed under EPA Contract
68-D9-0173, covered the detailed development of the national temporal allocation factor file. For
this effort, existing temporal allocation methods and databases were reviewed, additional data were
collected, and the data processing to create the initial temporal allocation factor file was performed.
Minimal quality assurance of the data was performed in the first phase. The results from the first
phase are described in Chapters 1 through 7. The second phase of the project, completed under
EPA Contract 68-D2-0181, provided for additional quality assurance and analysis of the initial
temporal allocation factor file. Revisions to the temporal allocation factor file were made based on
the results of these analyses. The results from the second phase are contained in Chapters 8 and 9.
xi
-------
CHAPTER 1.0
INTRODUCTION
1.1 BACKGROUND
Emission inventories have traditionally been developed to produce estimates of emissions
for annual or daily time periods. However, in order to be used as input to photochemical and
other atmospheric simulation models, hourly emission estimates are usually required. Ideally,
emissions for specific hourly time periods would be measured or calculated directly at the
emissions source; this approach is normally impractical due to technical and resource constraints.
As an alternative, hourly emission estimates can be obtained using surrogate temporal allocation
factors from "temporal profiles" assigned to specific emissions source categories.
A temporal profile is a collection of dimensionless decimal fractions (or factors) that
represent seasonal, daily, or hourly activity from specific source categories. These profiles are
defined as follows:
Seasonal profiles represent relative operating levels during four three-month seasons, defined as
follows:
Winter - December, January, February
Spring - March, April, May
Summer - June, July, August
Fall - September, October, November
4T
Daily profiles represent activity through the week. The following three day types were used:
Weekday - Monday, Tuesday, Wednesday, Thursday, Friday
Saturday
Sunday
Hourly f diurnal) profiles represent relative activity levels at 24 one-hour intervals through the
day.
Estimates of hourly emissions may then be calculated by applying the appropriate temporal
allocation factors to available annual, seasonal, or daily emission rates. This approach has been
CH-94-35
1-1
-------
followed in previous air pollution studies, including the National Acid Precipitation Assessment
Program (NAPAP) and the Northeast Corridor Regional Modeling Project (NECRMP). Because
the performance of atmospheric simulation models is dependent upon the availability of accurate,
temporally resolved emissions rates, suitable methodologies and databases must be available to
personnel responsible for developing the daily emissions estimates needed for model inputs.
Figure 1-1 is a graphical representation of how a temporal profile may be applied to an annual
inventory.
Annual
Annual Emission 1
Inventory
Estimate 1
Seasonal Profile
Daily Profile
Seasonal Emission
Estimates
Daily Emission
Estimates
Hourly Profile
Hourly Emission
Estimates
Hourly Resolved
Inventory
Figure 1-1. Application of a temporal profile.
CH-94-3S
1-2
-------
1.2 OBJECTIVES
The purposes of this project were to evaluate the quality and completeness of data and
methods presently being used for temporal allocation of emissions data, to identify and prioritize
needed improvements to the current methods for developing temporal allocation factors, and to
collect and use data to improve existing temporal allocation factor (TAF) files. The electronic TAF
file will be used as national default allocation factors by the emissions model processing systems
that calculate temporally resolved emissions estimates for model input,
1.3 CONTENTS
This report is divided into three major sections: Chapters 2 through 7 describe the overall
development of the TAF file, Chapters 8 and 9 discuss the review and revision of the TAF file
described in Chapters 2 through 7, and Chapter 10 provides references for the report. In addition,
Chapters 2 through 7 represent work completed under Work Assignment No. 3/314 (Contract No.
68-D9-0173) and Chapters 8 and 9 describe work performed as part of Work Assignment No. 1/014
(Contract No. 68-D2-0181). Chapter 2 presents an overview and evaluation of the methods and
data presently being used for temporal allocation of emissions data and presents the results of a
literature search for developing TAF data. Chapter 3 identifies and describes each of the data
sources from which applicable data were extracted to improve or develop the TAF file. The
approach used for generating an interim file from these data is also described. Chapter 4 discusses
the methodology used to prioritize source categories for TAF file development. Chapter 5
discusses the methodology used for incorporating interim files into the final TAF file. Chapter 6
describes the quality assurance (QA) procedures performed throughout the data generation and TAF
file development. Chapter 7 presents conclusions and recommendations.
Chapter 8 discusses the analysis of the TAF file described in Chapters 2 through 7 and
describes the revision of the TAF file. Chapter 9 presents conclusions and recommendations for
additional analyses.
CH-94-35
1-3
-------
CHAPTER 2.0
SUMMARY OF METHODS AND DATA CURRENTLY USED IN
TEMPORAL ALLOCATION OF EMISSIONS DATA
The quality and completeness of methods and data presently being used for temporal
allocation of emissions data were evaluated to identify and prioritize needed improvements to the
current methods. This chapter presents an overview of the temporal allocation factors developed
for NAPAP, the Flexible Emissions Data System (FREDS), the Urban Airshed Model (UAM),
and the Geocoded Emissions Modeling and Projections (GEMAP) System. The results of other
information gathering efforts are also presented.
2.1 CURRENT TEMPORAL ALLOCATION FACTORS
2.1.1 National Acid Precipitation Assessment Program (NAPAP)
The most comprehensive set of temporal allocation factors to date has been developed for
NAPAP, although other work has focused on some specific aspects of temporal allocation.
Temporal allocation factors were developed for emissions of the following 10 pollutants from the
point and area source categories in the 1980 NAPAP emissions inventory: sulfur dioxide (SO2),
primary sulfate, oxides of nitrogen (NOx), total suspended particulates (TSP), carbon monoxide
(CO), ammonia (NH3), hydrogen chloride (HC1), hydrogen fluoride (HF), volatile organic
compounds (VOC), and total hydrocarbons (THC). Of these, NOx, TSP, and THC were further
resolved into component species or groups of species.
NAPAP temporal allocation factors were updated and applied to the 1980 and 1985
emissions data. Four seasonal, three seasonal-daily (i.e., a typical weekday, Saturday, and
Sunday) and 24 hourly allocation factors were developed for NAPAP point and area sources.
Factors were developed for the 102 source categories (including mobile sources) in the 1985
NAPAP area source data. Depending on the magnitude of emissions within the source category
CH-94-35
2-1
-------
and availability of data, factors were frequently resolved to the regional, state, or local level.
Point source factors were developed1 for electric utility processes.
The NAPAP factors were developed from a variety of sources:
1. Many factors for U.S. sources came from the Northeast Corridor Regional Modeling
Project. Seasonal factors for the temporal distribution of point source emissions were
originally developed on a fuel- and State-specific basis for facilities in the NECRMP
study using power generation statistics from the U.S. Department of Energy.
2. Temporal allocations for point source emissions were based on operating schedule
information included in the National Emissions Data System (NEDS)-based point source
data records. Because of the magnitude of electric utility emissions, process-specific
factors were developed for these sources.
3. Daily factors were developed at the national level from weekly load cycle listings in the
Electric Power Research Institute (EPRI) Regional Systems report. Fuel- and State-
specific weekday hourly patterns for power generation facilities were developed during
the NECRMP effort. Profiles of hourly operation were derived from hourly power plant
fuel use data collected during the development of EPRl's Sulfate Regional Experiment
(SURE) inventory.
A major limitation of the NAPAP factors is that they were developed only for the NAPAP
point and area sources. This includes 102 area source categories and only electric utility
processes for point sources. In addition, NAPAP factors focus primarily on criteria pollutants
that play a role in the formation of acid rain; future focuses will require temporal allocation
factors for hazardous air pollutants (HAPs). A detailed discussion of NAPAP temporal allocation
factors is presented in Section 3.1.
2.1.2 Flexible Regional Emissions Data System (FREDS)
The Flexible Regional Emissions Data System is a software system designed to process
emissions data for input to regional acid deposition and oxidant models. FREDS extracts
emissions data, modeling parameters (e.g., stack height, exhaust gas temperature, ere.), and source
identification information from point and area source data records contained in preproeessed
SAS® files and applies temporal, spatial, and species allocation factors to arrive at a gridded,
speciated, and temporally resolved emissions file.
CH-94-35
2-2
-------
Resolution of annual emissions is accomplished by the execution of a series of
"allocation modules" which produce gridded, speciated and temporally allocated output files. The
majority of the FREDS software is written in SAS and relies heavily on the use of SAS
macros for flexibility in processing different file structures.
Point and area source emissions are temporally resolved during execution of the Temporal
Allocation Module (TAM) of FREDS. Although different versions of TAM have been developed
to process point and area source data on anthropogenic and natural emissions in the United States
and Canada, the overall methodology used to apportion these data is similar. The generalized
TAM structure accepts four main input files. The emissions data file, which has been processed
by preceding FREDS modules, is imported by TAM along with a Control Options File (CDF).
The COF passes file types information to TAM to permit file-specific execution of the module.
The TAF file is a separate file containing the seasonal, daily, and hourly multipliers used for
temporal allocation. Adjustments of local temporally allocated emissions to Greenwich Mean
Time, including an offset for daylight savings time where appropriate, are accomplished by
applying factors from the Time Zone File (TZF).
The TAF file provides allocation factors for up to 12 "typical days" (a weekday, Saturday
and Sunday for each season). These days are commonly referred to as "temporal scenarios."
Individual records in the file are distinguished by one or more identifiers. For emission sources
in the United States, these identifiers include source category, state, plant, and point. For sources
in Canada, a special linking variable is created. TAM uses these identifiers to link temporal
profiles with appropriate emission records. For U.S. sources, profiles are applied to emission
records by starting at the most specific level and proceeding to the most general level. As an
exception, seasonal factors for point sources in the United States are based primarily on seasonal
activity percentages supplied with the emissions records; seasonal factors from the TAF file are
used secondarily.
In the absence of specific temporal allocation factors for point sources in the United
States, TAM provides daily and hourly emissions allocation based on operating schedule data
from emission records (i.e., day per week and hours per day of process operation). (Seasonal
activity percentages on emission records are automatically used when they are present.) As a
final step, both point and area source TAM programs assign uniform default allocation profiles
if temporal factor data were missing or incomplete.
CH-94-35 2-3
-------
2.1.3 Urban Airshed Model (UAM) Emissions Preprocessor System
The Urban Airshed Model has been designated as the preferred model for "photochemical
pollutant modeling applications involving entire urban areas" by the U.S. Environmental
Protection Agency's (EPA's) Office of Air Quality Planning and Standards (OAQPS). The UAM
simulates the hour-by-hour photochemistry occurring for each grid cell in the modeling domain;
consequently the input emissions data must also be at an hourly resolution. To accommodate this
level of resolution in the input data, a system of computer programs has been designed to
perform the intensive data manipulations necessary to adapt a county-level annual or seasonal
emission inventory for photochemical modeling use: the UAM Emissions Preprocessor System
(EPS), Version 2.013, In the EPS, source classification codes (SCCs) are cross-referenced to a
month, day of the week, and diurnal profile code which determines the temporal profiles applied
to emissions being processed for input into the UAM, These monthly, weekly, and diurnal
hourly temporal profiles are available from the EPS. The EPS contains default temporal
allocation files to perform temporal allocation of annual emissions.
In the EPS, TAP files are cross-referenced to point, area, and mobile source processes by
profiles codes. These TAF files exist for an array of different temporal scenarios consisting of
typical monthly distributions, day to week distributions, and hourly distributions for typical
weekday and weekend scenarios. The TAF files were compiled from operating parameters found
in the NAPAP emissions inventory files, as well as data resulting from field studies conducted
by the California Air Resources Board (CARB) approximately 10 years ago.
The EPS consists of six programs that generate input files for the UAM. These programs
can assign temporal distribution profiles (seasonal, hourly, and daily) based on operating
information (e.g., weeks per year of operation) contained in the annual or seasonal inventory.
The EPS contains default values for sources not explicitly reporting operating schedules. For
example, the EPS assigns a flat operating profile (equal season fractions of annual throughput,
52 weeks per year, 1 days per week and 24 hours per day) to all sources not reporting operating
schedules. The UAM develops source-specific allocation factors based on input data on each
source or uses default allocation factors where such data are missing.
CH-94-35
2-4
-------
2.1.4 Geocoded Emissions Modeling and Projections (GEMAP) System
The Geocoded Emissions Modeling and Projections system was developed for use in the
Lake Michigan Ozone Study (LMOS) and modified for use by the State of California. The
model computes gridded, hourly adjusted, pollutant-specific, emission estimates. General data
processing is handled by a SAS®-based data processor, with spatial data processing handled by
an ARC/INFO®-based processor.
The SAS®-based temporal data processor has two primary data files: foundation SAS®
data sets and general temporal lookup tables. The foundation data sets include device-specific
(i.e., process-specific) information including average operating rates such as hours per day and
days of operation per week. The lookup tables allocate these rates to predict daily and weekly
emission cycles.
The foundation data sets are converted from ASCII data files by a SAS® preprocessor.
A foundation data set for each device is included in the model. The data contained in the ASCII
and SAS® foundation files include the following: state, county, facility, stack, and device
number identification numbers (IDs); four-digit Standard Industrial Classification (SIC) codes;
monthly and seasonal throughput; hours of operation per day; days of operation per week; weeks
of operation per year, days of operation per year; hours of operation per year, and emission type
(point, area, mobile, biogenic, etc.).
Two lookup tables allocate the operation rates in the foundation file. The HOURPROF. IN
file assigns operating ratios for the proportion of the daily emissions occurring in any week in
accordance with the number of hours per day indicated in the foundation file. These allocations
are based on assumptions and observations of operating procedure, such as distributing the hourly
operation rate of an eight-hour-per-day device evenly across the hours beginning at 8:00 a.m. and
3:00 p.m. Records with hours-per-day codes ranging from 1 to 24 distribute hourly production
in the modeling day according to the hours of operation per day indicated in the foundation file.
Hours-per-day values over 24 indicate processes or groups of processes with known daily
operation schedules. The WEEKLYOP.IN file assigns operating ratios for the proportion of the
weekly emissions occurring in any day in accordance with the number of days per week indicated
in the foundation file. These ratios are also based on generalizations, such as where a two-day
CH-94-35
2-5
-------
weekly operation rates are distributed evenly across the Monday to Friday work week, GEMAP
also contains lookup table records in WEEKLYOP.IN which describe operation schedules of
some specific processes and groups of processes. These records are marked in foundation files
and the lookup tables as having days-per-week codes greater than seven.
The entries across a record in the lookup tables consist of integers indicating the relative
amount of the week's activities occurring in the day (or of the day's activities occurring in the
hour). Therefore, normalized values for HOURPROF.IN can be calculated by dividing a day's
activities by the sum of the values for all of the days in the record. The same applies for days
in WEEKLYOP.IN. Two digits are allocated for each value in the tables.
2.2 INFORMATION GATHERING
TRC obtained additional information pertaining to the development of temporal allocation
factors through an intensive literature search and through telephone interviews. Generally, time
series data showing operating rate fluctuations on an hourly, weekly, monthly, or quarterly basis
were preferred. A summary of the results of these information gathering efforts is presented in
Sections 2.2.1 and 2.2.2, respectively. More, specific information on the applicable data sources
used for development or improvement of temporal allocation factors is included in Chapter 3 of
this report.
2.2,1 Literature Search
Relevant current literature references were identified, acquired,- and reviewed in order to
evaluate the quality and completeness of data and methodologies presently being used for
temporal allocation of emissions data. Specific information on the data sources identified through
this literature search is provided in Chapter 3, The literature search was documented in a
May 27, 1993 technical memorandum from TRC Environmental Corporation (TRC) to EPA and
is included in Appendix A.1
CH-94-35
2-6
-------
2.2.2 Telephone Interviews
TRC conversed by telephone with technical staff members of EPA, selected state agencies,
universities, and other government or private research organizations to identify and obtain
information pertaining to the development of temporal allocation factors. Copies of the telephone
contact reports summarizing the information found were submitted to EPA in a technical
memorandum dated May 27, 1993 and are also included in Appendix A,2
2.2.2.1 Results of Discussions with EPA
Staff members with the Air and Energy Engineering Research Laboratory (AEERL) were
contacted to discuss the EPA documentation for developing temporal allocation factors from data
contained in the St. Louis Regional Air Pollutant Study (RAPS). It was determined that the data
were outdated and not useful for this project. The EPA Emission Inventory Branch (ELB) was
contacted to discuss data on hourly emissions from storage tanks using their latest emissions
estimation tool, the TANKS program. The EIB reported that hourly emissions data for VOC
storage tanks were not available from the TANKS program.
Additional contacts, identified by the EPA steering committee, were interviewed by TRC
for relevant information. No additional data sources were identified as a result of the interviews.
2.2.2.2 Results of Discussions with State/Local Regulatory Agencies
Representatives from the South Coast Air Quality Management District (SCAQMD),
Virginia Department of Environmental Management (DEM), Ohio EPA, Illinois Environmental
Protection Agency (IEPA), Texas Natural Resource Conservation Commission (TNRCC), and
Wisconsin Department of Natural Resources (DNR) were contacted for information regarding the
development of temporal allocation factors. The SCAQMD uses diurnal codes from their airshed
filing system. For each facility, the emission rate and daily operating hours are given. An
estimate of daily emissions is sometimes determined. SCAQMD determines the emissions on
an hourly basis by the use of a code from the standard UAM EPS. The operating schedule for
each facility is available through the Emissions Information System (EIS).
CH-94-35
2-7
-------
The Virginia DEM is planning to obtain temporal operational data for major point sources
from major industry representatives and is also working with EPA on defaults to the UAM. TRC
was unable to obtain any useful information or data pertaining to this project.
The Ohio EPA maintains information pertaining to operating schedules (i.e., point-by-point
and plant-by-plant) and gives seasonal throughput information on an hours/day, days/week and
weeks/year basis. This information, in a database format, was acquired and evaluated.
The emissions inventory system used by IEPA contains information pertaining to throughput
and scheduling data for 1990 through 1992. The IEPA was contacted to determine what information
was available, but no data were collected.
The TNRCC maintains an electronic database system which contains information on
operating schedules (along with facility contact information for obtaining operating schedules),
enforcement, emissions inventories, inspections, emission permits, and New Source Review permits.
This database was obtained and imported into a dBASE® format for data manipulation.
The Wisconsin DNR has conducted a study (Lake Michigan Ozone Study) of source-specific
hourly emissions from over 200 companies in the United States. Facilities contained in the study
include power plants and paper mills. The draft data were obtained in ASCII format and imported
into a SAS® format for data manipulation. The data were not publicly available in final format at the
writing of this report.
2.2.2.3 Results of Discussions with Other Government and Private Organizations
Many of the major metropolitan planning organizations (MPOs) throughout the United States
are studying mobile source emission estimates (many on an hourly basis) based on percentages of
vehicle miles traveled (VMT) over various traffic segments and links. The Denver Regional Council
of Governments and the Baltimore Metropolitan Council provided daily and hourly mobile source
activity data. The Delaware Valley Regional Planning Commission also indicated that they maintain
an extensive vehicle activity database. Other major MPOs were contacted to determine what
information was available for developing temporal allocation factors for mobile sources.
2-8
-------
Radian Corporation (Sacramento, CA) provided the following information from the GEMAP
system: (1) lookup files; (2) point and area source data; (3) weekday/diurnal data; (4) the data
dictionary; and (5) user's manual.
The U.S. Department of Energy (DOE) Energy Information Administration (EIA) produces
energy consumption and fuel production reports with monthly statistics. The DOE electronic
publication bulletin board system also contains some of the weekly production statistics. Copies of
these data were also obtained.
2.2.3 Results of Data Collection and Telephone Interviews
Personnel interviewed by telephone were also asked for input on ways in which temporal
allocation factors could be improved. Responses are summarized below:
* The Ohio EPA suggested that a sensitivity analysis of air quality models be performed in
order to determine how modeling responds to a change in a temporal allocation factor.
• The Wisconsin DNR stated that there are problems with the classification of the ozone season
since the season does not coincide with quarterly data used by the State agency. DNR urged
EPA to amend the guidance to allow location-specific classification of the ozone season.
• Lake Michigan Air Directors Consortium (LADCO) suggested that day specific data, as
opposed to average data, are needed to obtain the best results from modeling efforts.
* The General Motors Auto Oil Program had two suggestions for improving temporal
allocation factors for mobile sources: (1) improving diurnal behavior of evaporative
emissions and resting losses with respect to ambient temperatures; and (2) improving the
magnitude, diurnal behavior, and utilization profiles for off-road vehicle emissions.
2-9
-------
CHAPTER 3.0
DATA SOURCES
This section summarizes information on temporal allocation factors identified through
intensive literature searches and telephone interviews with technical staff members of EPA,
selected state/local regulatory agencies, universities, and other government or private research
organizations. In summary, the review showed that the most comprehensive set of temporal
allocation factors has been developed for NAPAP, although other work has focused on some
specific aspects of temporal allocation. The following data sources were identified as providing
sufficient information to support TAF file development:
• Business and Labor Statistics (BLS) data
• DOE data pertaining to production/consumption from various energy industries
• CARB AB-2588 "Hot Spots" pooled source test reports
• TNRCC stationary source operating schedule data
• Southern Oxidant Study (SOS) data
• LMOS data
• Continuous Emissions Monitoring (CEM) data
• Waste-to-energy (WTE) data
• Acid-Modes field study data
Appendix B presents all data sources used to generate the final TAF and the applicable
SCCs associated with each data source. Two distinct types of data were identified; first, national
economic statistics were identified as potential surrogate indicators of production activity
applicable at a major category level where process-specific data were absent or incomplete;
second, specific category or plant data were identified through a variety of sources, including
state-of-the-art emissions inventories, continuous emissions monitoring, and industrial and federal
reports, surveys, and databases.
CH-94-35
3-1
-------
Temporal allocation profiles created from the various data sources were stored in data
source-specific "intermediate files" of like format and level of resolution. Several of the data
sources were comprised of large databases which required some data manipulation to create an
"intermediate file." A file format for the "intermediate files" is shown in Figure 3-1. The
remainder of this section presents more specific information on incorporating data into
intermediate files for each data source.
In order to achieve consistency between data sources, some conventions were adopted for
assigning profiles. One convention used to allocate time periods is the NAPAP rule for selecting
specific days or hours in each profile based on available operating data. This rule (referred to
in this document as the "NAPAP rale") may be summarized as follows:
Daily Profiles
Days of operation per week Emissions are allocated as follows
1 Saturdays Only
2 Equally on Saturdays and Sundays
3-5 Equally on weekdays only
6 Equally on weekdays and Saturdays
7 Equally on all days of the week
Diurnal Profiles
Hours of operation per day Emissions are allocated as follows
1-17 Zero for midnight to 7:00 a.m.; equally for
x hours beginning with 7:00 a.m.; and zero
for hours remaining before midnight
>17 Equally among 24 hours of the day
[It should be noted that the part of the NAPAP rule stating that sources operating one or two
days are operating on weekends may be misleading. This part of the rule is only an assumption
and does not have any supporting documentation.]
To ensure that all emissions are accounted for, the fractions must sum (as nearly as
possible) to one. Daily and hourly fractions contained in Tables 3-1 and 3-2 were used. The
"Total" columns in these tables indicate the errors in summing the assigned fractions. Since these
fractions are used to allocate emissions, summation errors will result in slightly over- or
CH-94-35
3-2
-------
I 1
NAME
FIELD
FORMAT
FIELD
WIDTH
DESCRIPTION/VALIDATION
sec
Numeric
10
Source Classification Code (Point, Area, or Mobile)
DAY_CODE
Numeric
1
Day Scenario Identifier
(1=Weekday, 2=Saturday, 3=Sunday)
SEA_CODE
Numeric
1
Season Scenario Identifier
(l=Spring, 2=Summer, 3=Fall, 4=Winter)
DAY_FRAC
Real
7.5a
Day Scenario Fraction
(65 x weekday fraction)+(13 x Saturday fraction)
+(13 x Sunday fraction)=L0
SEA_FRAC
Real
7.5a
Seasonal Scenario Fraction
(Spring fraction + Summer fraction + Fall fraction +
Winter fraction)=1.0
HOUR1-
HOUR24
Real 24
7.5*
Diurnal Profile for Scenario Indicated
OBS
Integer
4
Number of observations used in creating profile.
SOURCE
Character
2
Data Source Identifier
Data Source
Identification
Data Source
1
LMOS
2
TNRCC
3
BLS
4
CARB
5
SOS
6
CEM
7
Wastewater (Not Used)
8
WTE
9
Acid-Modes
10
EPS (Not Used)
11
NAPAP
"Field width for field type "Real" is defined as X.Y, where X represents the field width,
and Y represents the allowable number of digits reported to the right of the decimal.
Figure 3-1. Intermediate temporal allocation factor file format.
CH-94-35
3-3
-------
TABLE 3-1. DAILY FRACTIONS FOR EMISSIONS ALLOCATIONS
(WITHIN A SEASON)
Scenario (No. Operating Days)
Weekday
Saturday
Sunday
Total
Weekday Only
0.01538
0
0
0.9997
Saturday Only
0
0.07693
0
1.00009
Sunday Only
0
0
0.07693
1.00009
Weekday and Saturday
0.01282
0.01282
0
0.99996
Weekday and Sunday
0.01282
0
0.01282
0.99996
Saturday and Sunday
0
0.03846
0.03846
0.99996
Weekday, Saturday and Sunday
0.01099
0.01099
0.01099
1.00009
TABLE 3-2. HOURLY FRACTIONS FOR EMISSIONS ALLOCATIONS
Scenario (No. of Daily Operating
Hours)
Hours
Total
1
1.00000
1.00000
2
0.50000
1.00000
3
0.33333
0.99999
4
0.25000
1.00000
5
0.20000
1.00000
6
0.16667
1.00002
7
0.14286
1.00002
8
0.12500
1.00000
9
0.11111
0.99999
10
0.10000
1.00000
11
0.09091
1.00001
12
»
0.08333
0.99996
13
0.07692
0.99996
14
0.07142
0.99988
15
0.06666
0.99990
16
0.06250
1.00000
, 17-23
0.05882
0.99994
24
0.04167
1.00008
CH-94-35
3-4
-------
underestimating emissions. The maximum error shown in these tables is 0.03 percent (weekday
only scenario, Table 3-1). For a weekday only source operating 15 hours per day, the error is:
0.9997 (weekday only, Table 3-1) * 0.99990 (15 hours, Table 3-2) = 0.9996 or 0.04 percent error
3.1 NATIONAL ACID PRECIPITATION ASSESSMENT PROGRAM (NAPAP) DATA
The 1985 NAPAP inventory developed a national base-year 1985 emissions inventory of
acid-deposition precursors. One prime objective was to support acid deposition modeling efforts
such as the Regional Acid Deposition Model (RADM). This objective required that emissions
data be resolved temporally, spatially, and by chemical species. A comprehensive software
processing system, FREDS, was developed to produce the necessary gridded, hourly, and
speciated emissions data. One necessary component of the processing systems was a set of
temporal profiles which could be associated with the point and area emission sources in the
inventory. Supporting documentation are The 1985 NAPAP Emissions Inventory: Overview of
Allocation Factors (EPA-600/7-89-010a, October 1989)3 and The 1985 NAPAP Emissions
Inventory: Development of Temporal Allocation Factors (EPA-600/7-89-01 Od, April 1990).4
3.1.1 Data Description
The 1985 NAPAP TAF files are contained in a SAS® data set representing 24 hourly
fractions for typical weekday, Saturday and Sunday operations in each of the four seasons. Each
profile is tagged with an eight-digit point source SCC or a three-digit NAPAP area source SCO.
Because NAPAP primarily relied on source-specific operating data submitted with the 1985
emissions data through NEDS, the TAF file development for NAPAP contained only a subset of
point source categories and all (1985) area source categories.
These 1985 profiles were based on the TAF files developed for the 1980 NAPAP
inventory. The largest single source of data was the 15-state NECRMP data set (Northeast
Corridor Regional Modeling Project Annual Emission Inventory Compilation and Formatting,
EPA-450/4-82-013a-r, 1982).5 Improvements and enhancements were made by incorporating
CH-94-35
3-5
-------
additional 1985 area source categories and 1985 point source data. Specifically, temporal factors
for electric utilities were updated to include 58 unit-specific profiles in the Tennessee Valley
Authority, and profiles from an analysis of EIA Form 759 and DOE/EIA report RO80. The
underlying 1980 TAF file for utilities had been developed from the NECRMP study.
Profiles developed for area source categories for the 1985 NAPAP inventory were derived
from a number of data sources. Two hundred and twelve unique profiles were developed for the
102 area source categories listed in Table 3-3. The table also shows what data or defaults were
used in the preparation of the area source profiles. These area source categories include both
stationary and mobile sources.
3.1.2 Data Collection
The 1985 NAPAP TAF file was extracted from the TAM of the FREDS software. These
GO
profiles are in SAS format and were compatible with the TAF file format used in this project.
An example of the data format of the 1985 NAPAP TAF file is illustrated in Figure 3-2.
3.1.3 Incorporation of Data into Intermediate File
The 1985 NAPAP TAF file required no format or data conversion to be used as an
interim TAF file for this project. Its format and content were reviewed for consistency with the
1985 documentation. The file contains seasonal, daily and diurnal profiles for weekdays,
Saturdays and Sundays for selected point and NEDS area source SCCs.
Because the Aerometric Information Retrieval System (AIRS) area source classification
scheme differs significantly from that used by NAPAP, the three-digit NAPAP area source codes
were mapped to the corresponding ten-digit AIRS Area Mobile Source (AMS) source
classification codes. This mapping does leave some gaps in the AMS category profiles developed
from the 1985 data due to both the expanded source coverage of the AMS codes and the greater
detail present in these codes. In most cases, each three-digit NEDS code encompasses many
individual ten-digit AMS codes. This complete mapping from the NEDS codes, including mobile
sources, had not been accomplished previously.
CH-94-35
3-6
-------
TABLE 3-3. SOURCE CATEGORIES FOR AREA SOURCES IN THE UNITED
STATES 1985 NAPAP EMISSIONS INVENTORY
NAPAP
Category
Description
Temporal Data Source
1
Residential Fuel - Anthracite Coal
1980 Heating Degree Day Data;
NOAA intra-day temperature
data
2
Residential Fuel - Bituminous Coal
3
Residential Fuel - Distillate Oil
4
Residential Fuel - Residual Oil
5
Residential Fuel - Natural Gas
6
Residential Fuel - Wood
7
Commercial/Institutional Fuel - Anthracite Coal
EPA Guidelines Defaults
8
Commercial/Institutional Fuel - Bituminous Coal
9
Commercial/Institutional Fuel - Distillate Oil
10
Commercial/Institutional Fuel - Residual Oil
11
Commercial/Institutional Fuel - Natural Gas
12
Commercial/Institutional Fuel - Wood
13
Industrial Fuel - Anthracite Coal
EPA Guidelines Defaults and
NECRMP
14
Industrial Fuel - Bituminous Coal
15
Industrial Fuel - Coke
16
Industrial Fuel - Distillate Oil
17
Industrial Fuel - Residual Oil
18
Industrial Fuel - Natural Gas
19
Industrial Fuel - Wood
20
Industrial Fuel - Industrial Process Gas
NECRMP
21
Incineration - Residential
22
Incineration - Industrial
23
Incineration - Commercial/Institutional
24
Open Burning - Residential
25
Open Burning - Industrial
26
Open Burning - Commercial/Institutional
27
Light Duty Gas Vehicles - Limited Access Roads
USDOT
28
Light Duty Gas Vehicles - Rural Roads
29
Light Duty Gas Vehicles - Suburban Roads
30
Light Duty Gas Vehicles - Urban Roads
31
Medium Duty Gas Vehicles - Limited Access Roads
32
Medium Duty Gas Vehicles - Rural Roads
33
Medium Duty Gas Vehicles - Suburban Roads
34
Medium Duty Gas Vehicles - Urban Roads
(Continued)
CH-94-35
3-7
-------
TABLE 3-3. SOURCE CATEGORIES FOR AREA SOURCES IN THE UNITED
STATES 1985 NAPAP EMISSIONS INVENTORY (continued)
NAPAP
Category
Description
Temporal Data Source
35
Heavy Duty Gas Vehicles - Limited Access Roads
U.S. Trucking Association and
NECRMP
36
Heavy Duty Gas Vehicles - Rural Roads
37
Heavy Duty Gas Vehicles - Suburban Roads
38
Heavy Duty Gas Vehicles - Urban Roads
39
Off-Highway Gas Vehicles
USDOT Highway Statistics;
EPA Guideline Defaults
40
Heavy Duty Diesel Vehicles - Limited Access Roads
US Tracking Association
NECRMP
41
Heavy Duty Diesel Vehicles - Rural Roads
42
Heavy Duty Diesel Vehicles - Suburban Roads
43
Heavy Duty Diesel Vehicles - Urban Roads
44
Off-Highway Diesel Vehicles
USDOT Highway Statistics;
EPA Guideline Defaults
45
Railroad Locomotives
EPA Guideline Defaults
46
Aircraft - Military
EPA Guideline Defaults; Civil
Aeronautical Board
47
Aircraft - Civil
48
Aircraft - Commercial
49
Vessels - Coal
EPA Guideline Defaults
50
Vessels - Diesel
51
Vessels - Residual Oil
52
Vessels - Gasoline
53"
Solvents Purchased (not used)
54
Gasoline Marketed
Uniform Pattern
55
Unpaved Road Travel
N/A
56
Unpaved Airport LTOs
EPA Guidelines Default
57
(Not used)
58
(Not used)
59
(Not used)
60
Forest Fires
Independent Assessment
(continued)
CH-94-3S
3-8
-------
TABLE 3-3. SOURCE CATEGORIES FOR AREA SOURCES IN THE UNITED
STATES 1985 NAPAP EMISSIONS INVENTORY (continued)
NAPAP
Category
Description
Temporal Data Source
61
Managed Burning - Prescribed
NECRMP
62
Agricultural Field Burning
63
(Not used)
64
Structural Fires
Uniform Pattern
65
(Not used)
66
Ammonia Emissions - Light Duty Gasoline Vehicles
N/A
67
Ammonia Emissions - Heavy Duty Gasoline Vehicles
N/A
68
Ammonia Emissions - Heavy Duty Diesel Vehicles
N/A
69b
Livestock Waste Management - Turkeys
N/A
70b
Livestock Waste Management - Sheep
N/A
71b
Livestock Waste Management - Beef Cattle
N/A
72b
Livestock Waste Management - Dairy Cattle
N/A
73b
Livestock Waste Management - Swine
N/A
74b
Livestock Waste Management - Broilers
N/A
75b
Livestock Waste Management - Other Chickens
N/A
76
Anhydrous Ammonia Fertilizer Application
N/A
77
Beef Cattle Feed Lots
Uniform Pattern
78
Degreasing
Bureau of Labor Statistics
79
Drycleaning
80
Graphic Arts/Printing
81
Rubber and Plastics Manufacturing
82
Architectural Coating
83
Auto Body Repair
84
Motor Vehicle Manufacture
85
Paper Coating
86
Fabricated Metals
87
Machinery Manufacture
88
Furniture Manufacture
89
Flat wood Products
90
Other Transportation Equipment Manufacture
91
Electrical Equipment Manufacture
92
Ship Building and Repairing
93
Miscellaneous Industrial Manufacture
94c
(Not used)
95c
Miscellaneous Solvent Use
(continued)
CH-94-35
3-9
-------
TABLE 3-3, SOURCE CATEGORIES FOR AREA SOURCES IN THE UNITED
STATES 1985 NAPAP EMISSIONS INVENTORY (continued)
NAPAP
Category
Description
Temporal Data Source
96
Minor Point Sources - Coal Combustion
N/A
97
Minor Point Sources - Oil Combustion
N/A
98
Minor Point Sources - Natural Gas Combustion
N/A
99
Minor Point Sources - Process Sources
N/A
100
Publicly-Owned Treatment Works (POTWs)
1985
NAPAP
Point Source
Data
101
Cutback Asphalt Paving Operation
102
Fugitives from Synthetic Organic Chemical Manufacture
103
Bulk Terminal and Bulk Plants
104
Fugitives from Petroleum Refinery Operations
105
Process Emissions from Bakeries
106
Process Emissions from Pharmaceutical Manufacture
107
Process Emissions from Synthetic Fibers Manufacture
108
Crude Oil and Natural Gas Production Fields
109
Hazardous Waste Treatment, Storage, and Disposal Facilities
(TSDFs)
NECRMP STATES: Connecticut, Delaware, District of Columbia, Maine, Maryland, Massachusetts, New
Hampshire, New Jersey, New York, Ohio, Pennsylvania, Rhode Island, Vermont, Virginia,
West Virginia,
""Category 53 is disaggregated into process categories 78 to 95.
These categories formerly referred to as "manure field application."
'Formerly "miscellaneous industrial solvent use" (94) and "miscellaneous non-industrial solvent use" (95); now
combined into one category (95).
Because some of the point source profiles for utilities had been keyed to specific units,
multiple profiles existed for some utility SCCs. These profiles were combined at the eight-digit
level using arithmetic averaging of the individual eight-digit seasonal, daily and diurnal profiles
to produce single, national, eight-digit SCC profiles used in the interim file. Each profile was
subsequently normalized to ensure that the seasonal, daily and diurnal profiles again summed
to unity.
CH-94-35
3-10
-------
Day
Hr
Hr
Hr
Hr
Hr
Hr
Hr
Hr
Hr
Hr
Hr
Hr
Hr
Hr
Hr
Hr
Hr
Hr
Hr
Hr
Hr
Hr
Hr
Hr
Profile
NBR
SEA
Day
01
02
03
04
05
06
07
08
09
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
A001
1
621
0110
051
051
051
057
057
057
048
048
048
047
047
047
028
028
028
016
017
016
038
037
038
047
048
047
2
621
0110
051
051
051
057
057
057
048
048
048
047
047
047
028
028
028
016
017
016
038
037
038
047
048
047
3
621
0110
051
051
051
057
057
057
048
048
048
047
047
047
028
028
028
016
017
016
038
037
038
047
048
047
4
201
0110
083
083
083
110
110
110
127
127
127
013
013
013
000
000
000
000
000
000
000
000
000
000
000
000
5
201
0110
083
083
083
110
110
110
127
127
127
013
013
013
000
000
000
000
000
000
000
000
000
000
000
000
6
201
0110
083
083
083
110
110
110
127
127
127
013
013
013
000
000
000
000
000
000
000
000
000
000
000
000
7
000
0000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
8
000
0000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
9
000
0000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
10
178
0110
067
067
067
106
106
106
120
120
120
000
000
000
000
000
000
000
000
000
000
000
000
040
040
040
11
178
0110
067
067
067
106
106
106
120
120
120
000
000
000
000
000
000
000
000
000
000
000
000
040
040
040
12
178
0110
067
067
067
106
106
106
120
120
120
000
000
000
000
000
000
000
000
000
000
000
000
040
040
040
Note: Allocation fractions are left-registered to a decimal point (e.g., 051 is 0.051)
Figure 3-2. Sample 1985 NAPAP temporal allocation profiles.
CH-94-35
-------
The NAPAP-derived TAP file was used as a secondary or tertiary data source for both
point and area source TAF profiles, including seasonal, daily and diurnal profiles. An exception
was mobile sources, where NAPAP served as the primary data source due to the absence of other
available data.
3.1.4 Comments
The 1985 NAPAP data, including the TAF file, represented the most complete and current
information available at that time. These data included actual unit-specific electric utility data
and were the only area source temporal data available for that period. Along with the actual
operating profiles generated from the 1985 inventory data, this information supported the gridded,
allocated and speciated inventories used in the 1985 and subsequent acid deposition modeling
studies using RADM and the Acidic Deposition and Oxidant Model (ADOM). It also supported
oxidant modeling using the Regional Oxidant Model (ROM) and formed the basis of the current
EPS default temporal profiles. Due to its concentration on acid deposition precursors, utility
profiles received the greatest attention while other point source categories were assigned
essentially flat default profiles where source-specific data were not available in the 1985
inventory. The design and priorities of the NAPAP inventory do not correspond to the current
emphasis on ozone precursors and other air pollutants.
The NAPAP profiles were developed from data available from 1980 through 1987 for
base years 1980 and 1985. As such, changes in processes, technological advances and economic
shifts since that period are not reflected in the data. Insofar as the source types have been
unaffected from a temporal operating perspective, these TAF file data remain appropriate defaults
for current inventories. Where operating characteristics have changed (e.g., increased
productivity due to additional shifts at fewer plants in an industry), the NAPAP profiles may not
accurately reflect actual, current operations.
Finally, a potentially large source of NAPAP operating data could not be included in this
effort. The actual source-by-source temporal profiles from the 1985 operating data associated
with the inventory were not used, primarily because no existing intact archive files of these
ancillary NAPAP files were located locally. The on-line files had been removed from EPA's
CH-94-35
3-12
-------
National Computer Center (NCC) system and were irretrievable. Intact archive files were located
but could not be obtained, loaded and processed within the timeframe of the study. In addition,
these are 1985 operating data and, in most cases, would have been superseded by the other data
sources used to develop the TAF file. Therefore, these data would have had only a small impact
on the final TAF file. However, these data do represent an excellent benchmark for quality
assurance and comparison to the TAF file developed for this project and any subsequent
enhancements.
3.2 ECONOMIC DATA
Industrial activity and output are monitored by trade associations, private organizations,
and government agencies. The types of statistical information compiled by these groups include
number of employees, labor hours worked, sales, production, capacity, energy consumption, peak
demand, and other operating and economic indicators, such as production rate per employee. The
information often is specific to SIC groups.
Labor and economic statistics can be used to develop a default TAF file for emissions
modeling purposes. The statistics are published on varying temporal resolutions: seasonally,
monthly, and weekly. Data may be supplemented by industry survey or CEM data for further
temporal resolution to an hourly basis.
The general methodology used to develop these data into a TAF file involved calculating
seasonal fractional proportions over the temporal basis of the data by dividing the sum of the
monthly activities for a given season by the total activity for a given year. The basic assumption
was that operating or economic statistics are surrogate indicators of industrial processes releasing
pollutants. For example, the number of hours worked by employees or the industry's production J
rate are assumed to be directly related to that industry's potential emissions during that time
frame.
Information sources and frequency of publication can be documented for easy retrieval
during future TAF file updates. The three publications discussed below provide data covering
a wide range of industrial source categories. These data were compiled on a level suitable for
CH-94-33
3-13
-------
SIC code or third-level SCO assignment. Data contained in each of these publications were used
to develop seasonal temporal profiles.
3.2.1 Data Description
Several publications were used in developing temporal profiles: Business Statistics -
1963-91,6 obtained from the Bureau of Economic Analysis, U.S. Department of Commerce; and
Employment and Earnings,7 obtained from the Bureau of Labor Statistics, U.S. Department of
Labor. A third publication, which contains statistical reference sources, was obtained through
a search of a comprehensive index entitled Statistical Reference Index, 1991 Annual,8 Published
annually by the Southwest Research Institute, this index contains statistical reference sources,
arranged by subject and names, and references document abstracts for more detailed information.
Based on the review of this document, TRC obtained the 1991 CRB Commodity Year Boolfc9
published by the Commodity Research Bureau (CRB) in New York City.
Business Statistics - 1963-91 provides production data for individual durable and
nondurable goods on a monthly/quarterly basis for 1988-91, while Employment and Earnings
provides average weekly production labor hours and average weekly overtime labor hours by
SIC. Employment and Earnings (E/E) is published on a monthly basis and the reported data were
collected in cooperation with State employment security agencies under the Current Employment
Statistics (CES) Program. Reported statistics include number of employees (grouped by totals,
production and non-supervisory categories), average weekly labor hours worked, average weekly
overtime hours worked, and average hourly and weekly earnings. The average weekly and/or
overtime labor hours were of most interest in the development of the TAF file since hours
worked correspond to variations in plant production. The data were reported on a monthly basis
and published annually. Statistics for mining, construction, manufacturing, transportation and
public utilities, wholesale trade, retail trade, finance, insurance and real estate, services, and
government facilities are generally available to a four-digit SIC resolution. These data may be
used to develop seasonal temporal profiles. The data may be used without permission from the
publishers. Additionally, the data were available in electronic format at a modest subscription
fee.
CH-94-35
3-14
-------
The 1991 CRB Commodity Year Book contains dependable and readily available temporal
data for a number of commodities. Monthly statistics such as production, stocks, consumption,
shipments, exports, and imports are reported for several years. For example, U.S. production of
ethyl alcohol and spirits is presented on a monthly basis in units of millions of tax gallons. The
commodity data accessed from this publication were used to generate seasonal temporal profiles
which were assigned to an eight-digit SCC. The data were copyrighted; permission for use has
been requested from the publishers.
3.2.2 Data Collection
Economic data were used to develop temporal profiles for SCCs having no other data
sources and in cases where geographical biases in seasonal profiles were possible. Production
and/or consumption data extracted from either Business Statistics - 1963-91 or the 1991 CRB
Commodity Year Book were assigned to applicable SCCs whose processes matched the type of
product produced or consumed. Economic indicator data related to producer commodity prices
(i.e., producer price index) were not used since such indicators may not be directly related to the
quantity produced.6
SIC-level employment data contained in Employment and Earnings were matched to SCCs
by comparing SCCs with their major SIC groupings as referenced in the AIRS Facility Subsystem
Source Classification Codes and Emission Factor Listing For Criteria Air Pollutants (EPA 450/4-
90-003), March 1990.10 In some cases, a given SCC or family of SCCs was tied to at least two
SIC groupings. In these cases, the average weekly hours worked and overtime hours worked
pertaining to both SICs were summed. In all cases, the monthly production/consumption data
as well as the average production worker weekly hours and/or overtime hours worked for each
month were averaged over a three-year period to account for variations in economic recessionary
activity. Finally, it should be noted that the surrogate indicator data contained in these
publications could only be matched to a six-digit SCC. Consequently, families of SCCs were
assigned to a single set of surrogate data.
CH 94-35
3-15
-------
3.2.3 Incorporation of Data into Intermediate File
Once the applicable surrogate data were assigned to applicable SCCs, the following
relevant information was entered into spreadsheets: (1) the applicable SCC; (2) the reference from
which the data were obtained; (3) the surrogate data (i.e., data from Employment and Earnings,
Business Statistics - 1963-91, or the 1991 CRB Commodity Year Book) used to generate the
temporal seasonal profile including the units for production/consumption type data; (4) the data
used to generate the profiles; and (5) the calculated seasonal profiles along with the calculated
average profiles for each season, An example of the type of business/labor statistics data used
for calculating seasonal temporal profiles is contained in Appendix C.
In most cases, the most recent three years of monthly data were entered into spreadsheets
for each of the SCCs where economics/labor statistics data were applicable. Data were entered
for SCCs that were not covered by other, more preferred, data sources (i.e., SOS, LMOS).
Depending on the number of years of data available for each SCC, seasonal temporal winter
profiles were calculated using either one of the following two methods:
• In situations where only three years of data were available (e.g., in the case of
employment statistics data), data for January, February and December of the same year
were added together to generate the winter seasonal profile
• In situations where fours years of data were available, December of the previous year was
added to data for January and February of the subsequent year to generate the winter
profile.
The algorithm used to generate each of the seasonal profiles was calculated by dividing
quarterly data by cumulative yearly data. The seasons are defined as follows: (1) spring season:
March-May; (2) summer season: June-August; (3) fall season: September-November; and (4)
winter season: December-February. If four years of data were available, a year of data was
defined as the monthly data for December of the previous year through November of the
subsequent year. This definition allows consistency with pollutant-season definition. In all cases,
seasonal fractions were calculated for each season for all three years of data contained in the data
set. The average seasonal fraction for the years represented in the data set was then calculated.
CH-94-35
3-16
-------
The average seasonal temporal values generated for each SCC represented in this data source
were inserted into the intermediate TAF file. Computing a given seasonal temporal profile can
be represented as follows:
QJQT = a seasonal fraction
where:
Qs = sum of the monthly activity for a given season
Qt = sum of monthly activities for the entire year
The average seasonal fraction for a given season can be shown in the following equation:
q _ Qsi + ^S2 *
where: Qavg =average seasonal fraction for a given season
QS1 = seasonal fraction for a given season for year 1
QS2 = seasonal fraction for a given season for year 2
QS3 = seasonal fraction for a given season for year 3
It should be noted that for several SCCs, only two years of monthly data were available. In such
instances, the methods described above were used to compute the average seasonal fraction,
however, only two years of data were averaged.
3.2.4 Comments
The economics data may provide adequate seasonal indicators of pollutant emissions for
a variety of SCCs at the national level. In general, the economics data were used to generate
seasonal temporal profiles for several hundred SCCs for which either no other surrogate was
available or where the data would potentially present regional biases to emissions modeling
applications (as with the TNRCC database discussed in Section 3.3). The data from these sources
used to generate the temporal seasonal profiles were based on economics/business labor statistics
CH-94-35
3-17
-------
data collected across the entire nation. The data contained in the Business Statistics - 1963-91,
Employment and Earnings, and the 1991 CRB Commodity Year Book are updated on a yearly
basis.
In addition, since production and/or consumption statistics are indicators of market
conditions, averaging these types of data over a three-year period to generate the seasonal profiles
may provide a more accurate indicator of emissions from a given industry since this method
accounts for variances in market demand over longer time periods.
One of the most pronounced weaknesses in using economics data in developing temporal
fractions for a given SCC is the assignment of data to families of SCCs. As previously stated,
it was difficult to assign the data to an SCC beyond the sixth digit. Therefore, in many cases,
it was necessary to use the same surrogate data to generate seasonal temporal profiles for
families of SCCs which produce the same or similar commodities. Families of SCCs with
different processes at a given facility will not necessarily produce uniform emissions.
Furthermore, caution is necessary when assuming that a direct linear relationship exists between
production/consumption of a given commodity and emissions of various pollutants resulting from
activities which produce or consume a particular commodity.
Although employment statistics such as average weekly production worker hours worked
in a given industry can be directly related to production/consumption quotas, the data contained
in the TAF file reflect average weekly production hours and do not indicate the "data spread" of
hours worked across an entire industry nationwide. As previously discussed, where labor
employment data were used as the surrogate indicator, such data were related to a specific SIC
code(s). As such, the corresponding SCC(s) had to be determined. Although this task was
simplified by the use of the AIRS Facility Subsystem Source Classification Codes and Emission
Factor Listing For Criteria Air Pollutants, EPA 450/4-90-003, March 1990,10 such data were still
assigned to families of SCCs with either the same or different six digit codes. As noted above,
the different processes represented within SCC families may not produce uniform emissions.
Finally, since these data could only be used to calculate temporal profiles to a seasonal
resolution for a given SCC, default profiles or profiles from other data sources for both daily and
hourly temporal profiles must be used.
CH-94-33
3-18
-------
3.3 CALIFORNIA AIR RESOURCES BOARD (CARB) DATA
In accordance with the California Air Toxic "Hot Spot" Inventory and Assessment Act
of 1987, many types of industrial sources were required to submit air toxic emissions data to
local air pollution control agencies in the State. Collectively, the program is called the Assembly
Bill (AB)-2588 program. These facilities were required to undergo source testing in order to
determine the quantities of toxic air pollutants emitted from their operations. The AB-2588
regulations allowed similar facilities or industries to perform pooled source testing. These test
results would then be applied to all sources or facilities of the same type.
J
3.3.1 Data Description
The CARB data consist of mostly single observations (source tests) which contain both
operating schedules and monthly production activity percentages. As such, the data may be
resolved seasonally, daily and hourly. For daily and hourly resolution, the recommended default
fractions are used. Fifteen SCCs are represented in this data source.
In general, the CARB source test data may be used to provide the necessary surrogate
indicator information for priority SCCs for which other information sources are insufficient
(so-called "gap fillers"). The data contained in the CARB reports include emissions of various
pollutants determined from actual source test measurements taken at each facility during a certain
time of the year while the facility was operating at a normal capacity. For many of the reports,
these emission rates were given in pounds per hour. In addition to the measured emission rates,
many of the CARB reports also contained various process operating parameters which were used
as surrogate indicators in developing the profiles. Operating information included operating
schedules (i.e., hours per day, days per week, weeks per year) provided at the time of the source
test. In addition, these particular reports gave relative monthly activity percentages for the
various processes tested. These numbers represent the amount of production activity during a
particular month.
CH-94-15
3-19
-------
3.3.2 Data Collection
TRC reviewed available C ARB-pooled source test reports and summarized the information
on source/processes tested, useful surrogate data (i.e., plant activity or throughput data), emissions
data (if given on an hourly basis), testing period, and control equipment (including control
equipment efficiencies where available). The extracted information was used to determine the
applicability of these data to the development of profiles for the various source categories
represented. For many of the reports, SCCs and SIC codes were not given; in these cases, TRC
made a determination of applicable SCCs for each of those source processes.
Once the information was reviewed and summarized, the seasonal, daily and diurnal
profiles for each of the 15 SCCs were calculated using computer spreadsheets. An example of
the information contained in the spreadsheet is attached in Appendix D.
3.3.3 Incorporation of Data into Intermediate File
The temporal profiles for each of the SCCs represented in the CARB data were calculated
from the source operating schedules and relative percent monthly activity data given for the
various processes. The data were entered manually into spreadsheets and the appropriate
formulas were entered to calculate the seasonal, daily and diurnal profiles.
Seasonal profiles were determined by calculating the fraction of annual activity occurring
during each season, and were normalized where necessary to ensure summation to unity. The
daily and diurnal profiles were determined based on the operating schedules given. The
normalized percent activities were then used to compute each of the seasonal fractions for each
of the source processes contained in the CARB source test reports. Depending on whether the
source operated five, six or seven days per week, the daily fractions for each of the SCCs were
calculated as shown in the Table 3-4.
If the facility operated six days/week, it was assumed that the sixth day of operation was
conducted on a Saturday. Hourly fractions were assumed to be uniformly distributed and were
calculated by dividing the number hours of operation for a given day by one. With the exception
CH-94-35
3-20
-------
TABLE 3-4. FRACTIONS USED TO DETERMINE DAILY TEMPORAL
PROFILES FOR GARB DATA
Days/Week Operation
Weekday Fraction
Saturday Fraction
Sunday Fraction
5
1/65 (0.01538)
N/A
N/A
6
1/78 (0.01282)
1/78 (0.01282)
N/A
7
1/91 (0.01099)
1/91 (0.01099)
1/91 (0.01099)
of continuous operation (i.e., 24 hours), hourly fractions were placed in hourly time slots around
the noon hour beginning at 7:00 a.m.
Once the seasonal, daily and hourly temporal profiles were calculated in Lotus, the
profiles were uploaded to the NCC. Intermediate TAF files were created from the uploaded data.
3.3.4 Comments
It should be noted that the CARB data should provide an accurate assessment of the
source's operating characteristics (e.g., operating schedules) since the information was compiled
by source owners/operators during the actual source test period.
The fraction operating capacity (e.g., capacity utilization) reported at the time of the test
should be typical of the capacity at which the plant operates during the period (e.g., month or
season) of the year that the test was performed. Consequently, for many facilities, such operating
parameters may change throughout the year. Therefore, with the realization that these data
represent only a "snap-shot" in time, uniform daily and seasonal temporal fractions inserted into
the final TAF file for SCCs represented in this data source may prove to be somewhat erroneous.
This limitation may be compounded by the fact that temporal profiles were developed for only
a few SCCs and that more data may be necessary to make an accurate assessment of the actual
operating parameters for each of these SCCs.
Since the information contained in the CARB reports includes sources located only in
California, temporal profiles developed from these data may present geographical biases to
photochemical or other modeling applications in other areas of the United States. In this respect,
augmenting this data source with seasonal temporal factors developed from the national
CH-94-35
3-21
-------
economics/labor statistics data source may prove to be beneficial since these data represent
production/consumption quotas and/or production worker hours from around the entire country.
3.4 TEXAS NATURAL RESOURCE CONSERVATION COMMISSION (TNRCC)
DATA
The Texas Air Control Board, now the TNRCC, emissions inventory database, which is
maintained by the Technical Services Division of the TNRCC, includes an abundance of
information pertaining to source activity for individual businesses and their respective processes.
The purpose of the TNRCC database is to track all relevant source information for emissions
inventories, enforcement, and permitting. The TNRCC database is routinely updated when
relevant source information pertaining to permitting new sources, renewing operating permits, or
inspection results is received. [Note; During the course of this project, the TACB became the
TNRCC. Some of the computer files generated as part of this project retain the TACB name.]
3.4.1 Data Description
Data from the TNRCC database may be downloaded into various user formats. TRC
obtained a sample of the 1992 TRNCC Nonattainment Emissions Inventory Questionnaire from
the Emissions Inventory Branch of the TNRCC. This report provided information for a single
facility and included SIC codes and SCCs for each individual process as well as individual
process descriptions. As a result of this review, TRC requested the following information from
the TNRCC:
* Company name and SIC code(s)
• Contact person/phone number
• Business description
• Site operating schedule
* Seasonal operating percentages
* Facility identification number (FIN) information consisting of the
following for each individual facility process(es):
- facility name
- operating schedule
CH-94-35
3-22
-------
- seasonal operating percentages
- SCC and facility (SIC) description
The database requested from the TNRCC contains both SIC codes and SCCs for each
respective facility. The plant-specific process information includes the facility operating status,
individual process descriptions and their respective operating schedules, and seasonal operating
percentages. Throughput data were considered confidential and therefore cannot be accessed.
Data were maintained for over 6,000 plant sites within the State of Texas. Site operating
schedules are given for each process in hours per day, days per week, and weeks per year.
Seasonal operating percentages are provided as a percentage for each of the four seasons of the
year. Data were collected through the permitting, State Implementation Plan (SIP) inventory, and
enforcement processes. For each SCC given, facility (i.e., process-level) descriptions are also
given. The TNRCC data element "facility" descriptions are synonymous with the AIRS data
element "process."
3.4.2 Data Collection
The TNRCC database was requested on non-labelled 9-track tape for uploading onto a
mainframe system. After TRC received the database, the database was uploaded onto EPA's
National Computer Center's IBM 3090 mainframe system. The database contained 66,396
records representing 2,095 individual SCC values. The database was then retrieved into a SAS®
data set for statistical manipulation and subsequent assignment of temporal fractions.
3.4.3 Incorporation of Data into Intermediate File
Using FOCUS®, the data were first arranged into a table consisting of three separate fields
and three columns with 25,140 rows of data. The three fields were (1) SCC; (2) a day code; and
(3) a seasonal code. As stated above, the SCC contained 2,095 individual SCC values. For each
SCC, 12 observations were inserted into an intermediate FOCUS® database. The day code and
seasonal code columns contained all possible permutations of these two columns and their
respective domains. In summary, the domain for each of these three codes is as follows:
CH-94-35
3-23
-------
• SCC - 2,095 individual SCCs
• Day Code - the domain of this field has three values: l=weekday, 2=Saturday, 3=Sunday
• Season Code - four values: l=spring, 2=summer, 3=fall, and 4=winter
• Day Code - three values: l=weekday, 2=Saturday, and 3—full
The TNRCC database contains a day/week variable from which the following day
fractions were assigned.
Normal Operating
Days/Week (NODW)
Description
Day Fraction
1
Saturdays Only
0.07693
2
Equally on Saturday/Sunday
0.03846
3-5
Equally on Weekdays Only
0.01538
6
Equally on Weekdays/Saturday
0.01282
7
Equally All Days of the Week
0.01099
In determining which day fractions to insert into the intermediate FOCUS® file, the
average operating days per week by SCC was first calculated. Corresponding daily fractions
included in Table 3-2 were inserted into the intermediate FOCUS® file.
Seasonal fractions were determined by normalizing each seasonal percentage field in the
TNRCC database. Next, an average seasonal percentage for each individual SCC was
determined. The average seasonal percentages per SCC were then inserted into the intermediate
FOCUS® file.
Finally, the average number of hours of operation for SCCs represented in the TNRCC
database was determined. The following assumptions were made in executing this procedure:
(1) each day contains 24 hours, and (2) each work day begins at 7:00 a.m.
The appropriate hourly fractions were then inserted into the hourly field contained in the
intermediate FOCUS® database for each of the SCCs.
CH-94-35
3-24
-------
3.4.4 Comments
The TNRCC database provides an abundance of source data from industries in Texas.
Data are updated on a routine basis and are available electronically, making future TAF file
improvements possible.
Since the TNRCC database provides state-specific industry activity data, developing
temporal profiles using this data may tend to bias the resultant TAF file such that usefulness for
emissions modeling applications in other states may be reduced. Although the database contains
information collected for enforcement and/or permitting purposes, the information contained in
the database may also be based on information obtained from a source's operating permit and
may reflect the source's permitted, not actual operating characteristics. For example, many of
the seasonal percentages provided in the database may represent the source's rated performance
for a given season instead of its actual throughput and/or production output.
3.5 SOUTHERN OXIDANT STUDY (SOS) DATA
The SOS point and area source inventory was developed during the 1992 Atlanta, Georgia
summer intensive program from July 15 to August 31, 1992. This study was sponsored by EPA
and jointly conducted by the Georgia Department of Natural Resources, Air Protection Branch,
and the Georgia Institute of Technology. This study targeted NO, and YOC emitters in northern
Georgia. NOx and VOC emitters with releases greater than 100 tons per year and 25 tons per
year, respectively, were included in the point source inventory, NOx and VOC emitters from the
auto-body painting and printing operations categories with emissions less than 100 tons per year
and 25 tons per year respectively were included in the area source inventory.
3.5.1 Data Description
The survey requested that facility operators report production parameters for each
operating day in the six-week period. Collected data include the following: process code and
description, normal operating schedule (including normal start and end time for processes), stack
CH-94-35
3-25
-------
height, type of raw material, specific gravity or density, percent by weight VOC, industrial input,
and emissions. All data not involving time measure were reported in units chosen by those
surveyed. The general survey data were entered into the SIP Air Pollutant Inventory
Management System (SAMS) where the units were standardized.
For each process, the following data were recorded for each day in the six-week intensive
ozone season: actual operating schedule (hours per day, start time, and end time); input or
production rate; and VOC emissions. An example of schedule and production data from the SOS
point source survey is included in Appendix E. For boilers and fuel-burning equipment,
information on the design and maximum Btu capacity, fuel type and usage, stack height, and NOx
emissions was solicited. The Air Protection Branch used the SAMS program to convert some
data obtained in the survey into electronic format for revising the AIRS Facility Subsystem (AFS)
data. Quarterly throughput ratios, typical days per week, typical daily point-level start time, and
typical daily end time ranges are found in the SAMS database, but have data in less than half
of the records in any of these data fields. These summaries were not used for developing profiles
because they do not distinguish between weekday and weekend operations.
The SOS area source survey requested that facility operators report production information
for each operating day in the six-week period. Collected data included normal operating schedule
and emissions. For each facility, the actual operating schedule (hours per day, day per week),
facility name, facility address, and emissions information was recorded. The facilities represented
in this area source survey were smaller facilities than those contained in the SOS point source
survey. Information on the type of process, type of equipment, and other detail process-related
information was not included in this survey. These data were only used to calculate diurnal
fractions, since data were available for only one season.
3.5.2 Data Collection
The Georgia Department of Natural Resources, Air Protection Branch provided their
primary data for reproduction. Photocopies of the survey forms received by the Air Protection
Branch were made, and disk copies of the Branch's SAMS summary data were also provided.
CH-94-35
3-26
-------
Upon the request of TRC, the Georgia Department of Natural Resources supplied electronic files
of the autobody painting and printers survey.
3.5.3 Incorporation of Data into Intermediate File
The daily operating data were used to calculate daily and diurnal profiles for SCCs in the
SOS point source database. Since the survey period occurred over the summer, no seasonal
profiles could be derived. The daily and diurnal data from SOS were applied to all four seasons
in the interim TAF file. The daily start and stop times were used to determine diurnal profiles,
whereas daily throughput numbers were used as activity indicators for the daily profiles. SCCs
were assigned to SOS processes by referencing the SAMS data entered by the Air Protection
Branch.
Diurnal profiles were interpolated from the daily start and stop times provided by the
facilities surveyed. The data were entered from the survey forms into spreadsheets with separate
Start and Stop columns for Weekdays, Saturdays, and Sundays. An example of the spreadsheets
that were created from the survey data is included in Appendix E. These columns were averaged
to determine the mean start and stop times for the six-week period. The mean values were
rounded to integers for each of the three day types. Activity levels were then assigned to the
hours indicated by the average start and stop times, (e.g., a start time in the eighth hour and a
stop time in the fifteenth hour would indicate eight hourly activity levels of 0.125 from hour 8
until hour 15, inclusive). Some facilities provided only the hours of operation per day, instead
of specific start and stop times. The NAPAP rule, which assumes that facilities operating fewer
than 17 hours begin production at 7:00 a.m. and facilities operating 17 or more hours operate for
the entire day, was used for these facilities. The process spreadsheet contains a checksum
calculation that was used to verify data quality and normalization.
Daily profiles were derived from the production levels provided by facilities in the survey.
The data were entered from the survey forms into spreadsheets with separate columns for
weekday, Saturday, and Sunday values. These Saturday and Sunday columns were simply
averaged to obtain the mean activity level for those days. The weekday values were averaged
and divided by five to determine the mean activity level for a single weekday. Temporal profiles
CH 94 35
3-27
-------
were then calculated by dividing each day type's total by the sum for all three weeks. The
spreadsheet contains a checksum calculation that was used to verify data quality and
normalization.
The spreadsheets containing process-specific daily and diurnal profiles were then copied
into SCC-specific disk directories. The daily and diurnal profiles were combined and totalled
into a single file. These profiles were then divided by the number of observations to create a
temporal profile compatible with the interim file format requirements. SCC-specific profiles were
then aggregated into a single spreadsheet. A tab-delimited text file was generated from the
spreadsheet and submitted as the SOS point source interim TAF file.
For data from the area source survey, ASCII files for autobody painting and printing
operations were imported from ASCII files into electronic spreadsheets. Once the files were
imported into electronic spreadsheets, the following steps were performed for each file to develop
diurnal profiles for a weekday, Saturday, and Sunday and to prepare for the interim TAF file
format:
(1) The files were sorted and grouped by the number of operating days per week. Groups
were established for 0 to 5, 6, and 7 operating days per week.
(2) For each operating day group, the average number of hours per day was computed.
Resulting averages were rounded to the nearest whole number.
(3) Using the NAPAP rule, the average operating hours per day for each operating day group
were distributed. For averages, production was assumed to begin at 7:00 a.m. for
facilities with less than 17 operating hours per day. Averages greater than 17 hours per
day were assumed to operate for the entire day. However, with these data, all averages
were less than 17 hours per day and consequently all production was assumed to begin
at 7:00 a.m.
(4) For each operating day group, average operating hours per day, hourly fractions were
assigned based on the assumption that activity uniformly occurs throughout the operating
hours. Therefore, every hourly fraction is equal to every other hourly fraction.
The resulting diurnal fractions for each day-type were then entered into a temporal fraction data
entry system that created the interim data files.
CH-94-35
3-28
-------
3.5.4 Comments
The Georgia Department of Natural Resources, Air Protection Branch conducted a survey
of local facilities to develop the SOS point and area source inventories. This provides a data
source with high temporal resolution that is based on actual operating data. However, because
the data were only collected for the summer intensive ozone season, they do not provide any
measure of production variation. Facility responses to the point source survey varied in detail,
but most provided the requested daily start time, stop time, production/fuel use, and emissions.
There are some data irregularities inherent to industrial processes, such as shutdowns and
"maintenance only" periods. These were found in two SOS facilities. Such aberrations were not
incorporated in profiles, thus reducing the number of observations supporting some of these
facilities' profiles by 20 to 50 percent.
SCC assignment of the two source categories represented in the area source survey were
made from the AIRS/AMS SCC listing. These diurnal profiles supersede the default diurnal area
source profiles in the final TAF.
The limited number of observations for the Saturday and Sunday diurnal profiles raises
some question about the representativeness of these data for these source types. However, these
data were still much higher quality than the defaults that currently reside in the NAPAP profiles
for the applicable area source SCCs.
3.6 LAKE MICHIGAN OZONE STUDY (LMOS) DATA
Data gathered by the Wisconsin DNR to improve estimates of emission rates for CO,
NO,, and VOC for the Lake Michigan Ozone Study were obtained by TRC as a potential source
of data for developing the TAF file.
3.6.1 Data Description
The LMOS data include emission rates calculated from production data received by the
Wisconsin DNR for approximately 200 of the largest facilities in the 21-county LMOS region
CH-94-35
3-29
-------
of Wisconsin. Production data, including operating rates, fuel use, and solvent use, were
collected from facilities on a process-by-process basis from June 10 through August 24, 1991.
The emissions estimates were calculated on a process level basis and were identified by SCCs
to obtain the most accurate estimates possible. Before emission calculations were performed for
a facility, quality assurance procedures were conducted on the data to identify and correct
missing, suspect, or erroneous data. Naming conventions for emission sources and stacks were
analyzed, as well as a number of other validation and range checking procedures. Approximately
65,000 daily emission estimates are included in the LMOS database, representing 560 unique
SCCs. These data were roughly 15 megabytes (Mb) in size. The LMOS database file structure
is presented in Appendix F.
3.6.2 Data Collection
LMOS data were requested from the LADCO. Due to the current draft form of the data,
personnel at LADCO were unable to release these data to TRC, but referred TRC to the
Wisconsin DNR. TRC received data from the Wisconsin DNR, along with related "foundation
files," also in draft form. [Note: As of March 1994, the LMOS data have been finalized. There
were no changes to the Wisconsin data included in this study.]
3.6.3 Incorporation of Data into Intermediate File
Even in the current draft form, the LMOS data represent a reliable, documented source
of information for the development of a TAF file. The hourly emission estimates were associated
to SCCs found in the LMOS "foundation files." These "foundation files" relate process-level
information with source and process identification codes. The correlation of the SCCs with
hourly emission estimates was accomplished by match-merging the LMOS data file with the
"foundation file" on the device and process identification codes. Algorithms were developed to
construct temporal allocation factor distributions from the hourly emission estimates related to
the various processes.
CH-94-35
3-30
-------
Since the LMOS data were received in three independent database files, the files required
extensive data manipulation to create the intermediate profiles. The following procedure was
followed:
1. Merge all data files
a. Read in each record in the LMOS database, excluding observations that contained
missing value(s) in the 24 hourly readings or where the 24 hourly readings
totaled 0.
b. Merge the database with an ancillary database to match SCCs to all records.
c. Merge the database with another ancillary database to add seasonal profiles,
d. Store the final data set.
2. Produce intermediate code
a. Read and normalize data.
b. Sort the data set SCC and DAY_CODE.
c. For each combination of SCC and DAY_CODE, calculate the mean activity for
each hour,
3.6.4 Comments
The LMOS database consists of recorded data from an intensive study. The database was
greater than 90 percent complete with minimal missing entries within a given recorded period.
The quantity and completeness of the data set allowed for accurate, reproducible statistical
analyses. The limitations in using these data were recognized because the data set is still in draft
form and has not been released by LADCO to the public, and the recorded observations are for
only one season.
3.7 CONTINUOUS EMISSIONS MONITORING (CEM) DATA
Real-time hourly CEM emissions data provide an ideal basis from which to estimate
source activity levels. This is based on the assumption that hourly emission data are indicative
of activity levels. Subsequent evaluation of the CEM data collected for this project showed this
assumption not to be true. See Chapter 8.0 for discussion. For pollutants such as S02, source
activity is usually proportional to emissions. For pollutants such as NOx and CO, emissions may
CH-94-35
3-31
-------
also vary with other operating parameters such as combustion temperature and excess air.
Activity levels from CEM data can be derived from calculating the percent contribution of each
hour to the day's total emissions. These hourly fractions are then used as activity indicators of
a process's operation. Therefore, these data are very useful toward providing a detailed activity
profile of a process. The CEM data were used to create diurnal and daily fractions. Seasonal
fractions were not developed from these data.
Hourly emissions data obtained from CEM systems are a potential source of data for
developing TAF files. CEM systems are installed at facilities to monitor pollutant emissions on
a continuous basis for demonstrating compliance with permit conditions and/or state or federal
regulations. The most commonly monitored pollutants are S02, NO,, and CO,
CEM data are reported at various levels of temporal resolution, depending on the pollutant
and regulatory requirements. In some cases, the data may be reported on a temporal scale as
small as five-minute averages or as large as three-hour averages. Therefore, temporal profiles
may be determined diurnally, by day of the week, or by season of the year. For this project, the
CEM data were aggregated to develop temporal profiles for each hour of the day for a typical
weekday and weekend day in each season.
The data used to develop the TAF file(s) are the facility-specific CEM data for each
source. CEM data were maintained at the individual facilities in most states because there were
no regulatory requirements that the data be submitted to the state.
3.7.1 Data Description
Hourly monitored emissions data from the States of Ohio, Kentucky, and Pennsylvania
were used in TAF file development. These data represented pollutant emissions measured on a
continuous hourly basis for selected facilities in those states. The majority of the CEM data
obtained represented electric utilities. Additional source categories represented in the CEM data
collected from these states were as follows:
CH-94-35
3-32
-------
• Municipal Waste Incinerators
• Oil Refineries
• Sulfuric Acid Plants
• Commercial/Institutional Boilers
• Medical Incinerators
• Household Video Products Manufacturing
• Primary Lead Production
All of the collected CEM data were compiled on a hourly basis by day of the week for
a complete calendar year. The major pollutants represented in the CEM data were S02, CO, and
NO,, with S02 being the most prevalent.
The Ohio CEM data were generated from two municipal waste incinerators between July
1992 and June 1993. Hourly emissions data by month were reported for each of the facilities
in a hard copy format. S02 was the only pollutant measured (in pounds/hour) and these data
were used as the hourly parameter from which the temporal fractions were created.
The Kentucky CEM data represented a full calendar year from January 1992 to
January 1993. The Kentucky data were received in an electronic format, and consisted of
monthly diurnal measurements recorded for emission points. Measured pollutants included S02,
NOx, hydrogen sulfide (H2S), and CO; therefore, a single facility in the Kentucky database could
have multiple emission points.
The Pennsylvania CEM data consisted of 226 electronic files, each containing data for a
separate, unique emission point. Measured pollutants included S02, NOx, and CO.
3.7.2 Data Collection
State environmental agencies for the 48 contiguous states and the District of Columbia
were initially contacted in the spring of 1993 to determine the availability of hourly CEM data.
As a result of these contacts, CEM data were obtained from Ohio, Connecticut, New Hampshire,
Kentucky, Pennsylvania, California (selected facilities), Oregon and Nebraska. Example records
are included in Appendix G.
CH-94-35
3-33
-------
3.7.3 Generation of Interim File
The development of the interim TAF files from each of the CEM data sources is
discussed in the following sections.
3,7.3.1 Ohio CEM Data
The Ohio CEM data from two municipal waste incinerators were received in hard copy
format and converted to electronic format. The data were reviewed to determine if all
observations had 24 hourly measurements with no missing values. If a day had missing hourly
measurements for any reason (i.e. shutdown, invalid measurement, etc.), then that day's data were
not used. This data completeness approach was consistent with all the other data sources used
in the creation of TAF profiles. Once the data completeness step was finalized, the data were
reduced from 196 kilobytes (Kb) of data to approximately 65 Kb of data (approximately a 67
percent reduction).
The profiles were calculated as follows:
(1) All hourly measurements by day-type and season were summed and then averaged for
each season. This is represented by the following expression:
Equation 1:
n
Etj
where:
Ms d x = Average hourly measurement for a hour (jc) by day-type (d) [d = day-types
of weekday (wd), Saturday (sat), and Sunday (sun)] and season (s)
n = Number of observations
H = Hourly measurement emissions value
(2) For each season, the average hourly measurements were then summed for each day-type
to arrive at a daily sum for that season. This is represented by the following expression:
CH-94-35
3-34
-------
Equation 2:
= ^s4.\ + ^s.d,2 + + "" + ^M,24
where:
DSs d = Daily sum by season and day-type
(3) Hourly fractions were then calculated for each observation from the following expression:
Equation 3:
H,
p =
h 24
E»
where:
Fh = Hourly fraction
Hj = Hourly measurement value
(4) The average hourly fraction for each season, day-type, and source category was then
calculated by summing all hourly fractions by hour number and dividing by the number
of observations in a season for the day-type and source category as represented by the
following expression:
Equation 4:
r _ /=1
where:
Fh avg = The average hourly fraction for each season, day-type, and hour number
Fh = Hourly fraction
n = Number of observations
(5) A season sum was calculated from the following equation:
Equation 5:
SS = 65 x DS , + 13 x DS +13 x DS
S S.wd MBf SJUH
where:
SSs = Seasonal sum for a given season
CH-94-35
3-35
-------
(6) Daily fractions were then calculated from the following expression:
Equation 6;
DS A
j? _ sA
4 ~ss~
Sample Calculation:
• For hour 1, summer weekday;
Weekday values = 9.39, 20.56, 8.65, 7.93, 6.74, 5.69, 4.22, 2.55, 9.63, 4.27,
5.77, 6.74, 6.02, 10.45, 6.59, and 6.41
EAU weekday hour 1 values
_ = 7.6 = m
• Avg weekday hour 1 value = 16
* Calculate the other Msd, values for summer weekday (wd), hours 2 through 24.
• From Equation 2, sum all MsdtX values for summer weekday, hours 2 through 24 to derive the
Daily Sum (DSsd) for the summer weekday.
Other Msdx values = 7.4, 11.45, 10.43, 9.5, 11.11, 9.26, 8.32, 7.88, 7.49, 6.56, 8.17, 6.34,
5.39, 4.87, 5.07, 5.75, 7.23, 12.55, 11.73, 9.99, 8.89, 8.27, and 8.19
Add total to hour 1 average weekday value to get daily sum.
DS_wd = 7.6 + 191.8 = 199.4
~ Compute the seasonal sum for summer;
Other Daily Sum values are: DSsummerSat = 200.6 and DSsumxnerSun = 167.1
SS = 65 x 199.4 +13 * 200.6 + 13 x 167.1= 17,741
summer
Calculate the daily fraction for the weekday:
199.4
Kj 17,741
0.0112395
For observation dated August 24, 1992, hour 1 value = 5.77.
CH-94-35
3-36
-------
* Hour 1 fraction —
„ 5.77
Fh} =
Y, all hourly values for August 24
* The average hourly fraction for summer season, weekday, hour 1 is then calculated from
Equation 4.
The incinerator reporting the most complete information was used to calculate and
generate the temporal profiles for the source category represented by this data source. Once all
the hourly and daily fractions were calculated, the fractions were arranged in the spreadsheet into
the format dictated by the interim file guidelines. Seasonal fractions could not be calculated
since two seasons of data were discarded according to the data completeness protocol. Once the
daily and hourly fractions were calculated for the remaining two seasons, those fractions were
then averaged and used as the hourly and daily fractions for the missing two seasons. Seasonal
distribution was assumed at 25 percent for each season within the interim file. However, during
the final file processing, the 25 percent seasonal distribution was overwritten if actual seasonal
data from another data source were available.
3.7.3.2 Kentucky CEM Data
The Kentucky CEM data were received in an electronic file format from the Kentucky
Division of Air Quality. This electronic file consisted of 12 ASCII files, representing a month
of CEM readings for all facilities reported. The source categories represented in the Kentucky
CEM data included electric utilities, sulfuric acid plants, and an oil refinery.
Processing these data for temporal fraction development required available information
on hourly emissions measurements, plant name, and pollutant and units of measure. When the
text editing was complete, the files were combined into one text file for uploading onto the NCC
mainframe for additional data manipulation.
Once uploaded, the following approach was used in the data manipulation of this file:
(1) All (19,090) observations from the uploaded file were read into a SAS® data set.
Observations having missing hourly readings were discarded. This resulted in 9,899
observations being retained for further analysis (a 48 percent reduction of data).
CH-94-35
3-37
-------
(2) The data set was sorted by source category, season, and day-type to create subsets of data
organized by source category, season, and day-type. For example, a subset including
spring season, weekday, and the electric utility category would represent one subset. This
step resulted in 36 subsets being created (4 seasons x 3 day-types x 3 source categories).
(3) Each subset was evaluated using SAS® to determine the validity of and assess suspicions
and observations in the data for each subset. For each subset, the mean, standard
deviation, minimum value, maximum value, range, and variance were calculated. As a
result of this analysis, three additional observations were discarded that contained outliers.
(4) Day and seasonal fractions were then calculated from the following expressions:
Weekday fraction:
V All WD measurements in a season
(_^ ) 1 65
22 All daily measurements in a season
Weekend fraction:
, T All Sat/Sun measurements in a season x
{±± ) / 13
$>// daily measurements in a season
(Saturday and Sunday fractions were calculated separately.)
Seasonal fraction:
Y. measurements in a season
J] measurements in a year
(5) Hourly fractions for hour (/) were then calculated for each observation from the following
expression:
H.
F>°^-
E"
1—1
where:
Fh = Hourly fraction
H, = Hourly measurement value for hour (i)
(6) The average hourly fraction for each season by day-type and source category was then
calculated by summing all hourly fractions by hour number and dividing by the number
GH-94-35
3-38
-------
of observations in a season for the day-type and source category. This is represented by
the following expression;
t".
where:
Fh avg = The average hourly fraction for each season, day-type, and hour number
n = Number of observations
(7) A final data set containing the hourly, daily, and seasonal fractions for each source
category, season, and day-type was produced.
(8) This data set was then downloaded from the NCC mainframe for importation into an
electronic spreadsheet for final manipulation into the interim TAF file format.
3.7.3.3 Pennsylvania CEM Data
The Pennsylvania CEM data were received in an electronic file format that represented
15 Mb data, 60,000 observations, and 226 CEM sites. Each emission point was reported in a
separate file that contained a full year of CEM measurements. Information provided in the
Pennsylvania data included hourly CEM measurement values and the date of hourly
measurements. The source categories represented in the final Pennsylvania data include the
following:
• Electric Utilities
• Municipal Waste Incinerators
• Commercial/Institutional Boilers
• Medical Waste Incinerators
• Household Video Products Manufacturing
• Primary Lead Production
The Pennsylvania CEM data files were each assigned an unique CEM H> number and
appended as one file. This file was uploaded to the NCC mainframe. The following approach
was then followed:
CH-94-35
3-39
-------
(1) All (60,000) observations from the uploaded file were read into a SAS® data set.
Observations having missing hourly readings were discarded. This resulted in 22,047
observations being retained for further analysis (a 63 percent reduction of data). This step
also eliminated four source categories due to incomplete data.
(2) The data set was sorted by source category, season, and day-type to create subsets of data
organized by source category, season, and day-type. For example, a subset with spring
season, weekday, and the electric utility category would represent one subset. This step
resulted in 72 subsets being created (4 seasons x 3 day-types x 6 source categories).
The data set was sorted by source category, season, and day-type; the same approach was
used for the Kentucky and Pennsylvania data. Refer to the Kentucky CEM section (Section 3.2)
for additional details.
3.7.4 Comments
CEM data would be useful for the development of temporal profiles if hourly (or more
refined) emissions measurements are available which can provide good indicators of a source
activity. This data collected for this project were subsequently found not to be good indicators
of source activity. See Chapter 8.0 for discussion. Because the CEM data are reported at the
hourly level, they are best used for development of hourly, diurnal profiles; therefore, CEM data
are considered a high priority data source in the development of diurnal profiles. The CEM data
that were obtained and used for this project are of reasonably good quality. The data
completeness checks reduced the number of observations significantly, but also improved the
defensibility of the resulting fractions that were calculated,
Limitations associated with these data include geographic biases and a limited number of
observations for some source categories. In an ideal scenario, obtaining all the available CEM
data from all States and individual facilities would further improve the quality and
representativeness of temporal profiles. All CEM data identified were not obtained.
The CEM data that were obtained had significant amounts of missing data, sometimes
resulting in a reduction of useful data by up to 62 percent. The missing data either were not
reported for various reasons (shutdowns, equipment malfunctions, etc.) which were used by CEM
monitors to indicate a malfunction in the CEM equipment or a process-related malfunction
CH-94-3S
3-40
-------
impacted the hourly emissions measurements or represented invalid data codes. It is not known
whether this is representative of CEM monitors or was coincidental with the data that were used
for this effort.
3.8 WASTE-TO-ENERGY SOURCE (WTE) DATA
The 1993-94 Resource Recovery Yearbook, Directory and Guide [published by
Governmental Advisory Associates, Inc. (GAA)]11 was identified as a source of waste-to-energy
data. This data source contains information on the WTE for waste disposal facilities in the
government, commercial/institutional, and industrial sectors.
The Resource Recovery Yearbook provides data on the number and types of waste
recovery facilities located throughout the United States. Information is provided for conceptually
planned facilities, advance planned/existing facilities, and facilities which have been shut down
on either a temporary or permanent basis. The facility-specific data can be used as gap fillers
for developing a TAF file specific to the waste recovery industry.
3.8.1 Data Description
The report provides operating schedule and throughput data for specific waste recovery
facilities located throughout the United States. Based on conversations with GAA, the applicable
SCCs represented in this data source were determined. These SCC codes were 5-01-001-01,
5-02-001-01, 5-02-001-02, 5-03-001-01, and 5-03-001-02.12
Operating scheduling data were given in hours per week, hours per shift, shifts per day
and days of operation per year. The GAA collects information from the WTE recovery industry
on an ongoing basis.
3.8.2 Data Collection
Electronic data for planned and existing facilities were collected and used to generate the
intermediate TAF file for these source categories. The data were available in dBASE® format.
CH-94-35
3-41
-------
3.8.3 Incorporation of Data into Intermediate File
Operations data were imported into an electronic spreadsheet to allow the intermediate
profile to be created. These data include the source facility name, source location by state and
region, process code, and operating schedule. An example of the spreadsheet developed for WTE
data is included in Appendix H. For each facility represented in the data set, the applicable SCC
was determined based on the type of boiler process and whether the boiler had a single or
multiple chambers. Typical operating schedules were determined for each SCC. With the
exception of the sources represented by SCC 5-01-001-01, the typical operating schedule was
determined from the average operating schedule calculated for each SCC.
For the facilities represented by 5-01-001-01, the number of operating days per week was
determined based on the highest frequency of occurrence. Approximately two thirds of reported
facilities operate seven days per week. Therefore, it was assumed that seven days per week
operation was typical of this SCC.
Generally, the operating scheduling data can be divided among the days using the reported
number of operating shifts. The daily operating schedules were assumed to be constant through
the year. The total number of hours per day can be determined by multiplying the hours per shift
by shifts per day. For all SCCs represented in the data set, the typical operating schedule was
determined to be 7 days per week and 24 hours per day. Therefore, the seasonal, daily and
diurnal profiles were the same for each of the SCCs represented in this data set. The profiles
determined to be acceptable default profiles are shown below:
seasonal = 1/4 (0.25)
daily = 1/91 (0.01099)
hourly = 1/24 (0.04167)
These temporal profile fractions were uniformly distributed over each season (spring, summer,
fall, winter), day (weekdays/Saturdays/Sundays), and hour (1 through 24).
CH* 94-35
3-42
-------
3.8.4 Comments
The true national scope of these data avoids the regional biases that many data sources
posses. The data appear to contain true operating characteristics of a large number of data
sources.
3.9 ACID-MODES FIELD STUDY DATA
3.9.1 Data Description
The hourly emissions inventory developed for the Acid-Modes Field Study provides real-
time hourly emissions data for large sources of S02 and NOx in the eastern United States and
Canada for the period extending from May 1988 to June 1990. The real-time data were used
with the 1985 NAPAP Emissions Inventory as the emissions input for the verification runs of two
acidic deposition computer simulation models. Because the goal of the project was to provide
actual hourly emissions estimates rather than desegregated "typical" estimates, the megawatt-
hours generated on either a gross or net basis were converted to total heat input to each boiler
for each hour using heat rate data supplied by the companies.
3.9.2 Data Collection
The data collection efforts were sponsored by the EPA, the Ontario Ministry of the
Environment (OME), the Atmospheric Environmental Service (AES) of Environment Canada, the
Electric Power Research Institute (EPRI), and the Florida Electric Power Coordinating Group
(FCG).
The database was comprised of fixed design and hourly operating data from 31 major
utilities, with a combined total of 382 power-generating units. The temporal resolution varied
by facility, but in most cases, hourly load data were submitted. For facilities not submitting
hourly data, hourly estimates were made using the facility-provided weekly or monthly data.
CH-94-35
3-43
-------
CEM data were submitted by nine utilities, either hourly or as a daily average. Daily averages
were applied uniformly to the entire 24-hour period. For missing or unreadable data, averages
were taken of values directly preceding and following the point in question. A number of
variations of the basic methodology were necessary due to the variety of formats that were used
by the participating companies.
3.9.3 Incorporation of Data into Intermediate File
The data files for the project were stored at EPA's NCC. In the three years since the
project was completed, these data have been archived and warehoused. An electronic copy of
the data was obtained and uploaded by NCC for use in this project. These data were in SAS®
format data files and are readable and usable.
The SAS® data tapes were labelled as follows:
IVSRADM.EMISS.SAS5.JUN88.DATA
IVSRADM.EMISS.SAS5.JUL88.DATA
IVSRADM.EM1SS.SAS5.AUG88.DATA
IVSRADM.EMISS.SAS5.SEP88.DATA
IVSRADM.USCAN.EMISS.OCTDEC88.SASDAT
IVSRADM.USEM1SS.SAS.JANMAR89.DATA
IVSRADM. USEMIS S. S AS. APRJUN 89.DAT A
IVsradm.USEMISS.SAS.JULSEP89.DATA
rVSRADM.USEMISS.SAS.OCTDEC89.DATA
IVSRADM.USEMISS.SAS.JANFEB90.DATA
IVSRADM.USCAN.EMISS.MARMAY90.SASDAT
After a thorough examination of the data tapes, it was concluded that the tapes include
duplicate data resulting from mislabelling. In fact, instead of 24 months of data, the tapes
included data for the following months only:
» June 1988 (three identical files)
• August 1988 (three identical files)
• February 1989 (four identical files)
• January 1990 (two identical files)
• July 1988 (three identical files)
• January 1989 (four identical files)
• March 1989 (four identical files)
• February 1990 (two identical files)
CH-94-35
3.44
-------
Duplicate files were eliminated and the tapes were merged into a single file.
Derivation of Seasonal Fractions:
Hourly operating data in megawatt-hours (MW-hr) were divided into categories corresponding
to the four seasons. For each season, the total MW-hrs generated were calculated. Dividing the
MW-hrs generated in each season by annual MW-hrs (obtained by combining the MW-hrs of all
four seasons) results in seasonal throughputs. However, since no MW-hrs data were available
for the fall, it was assumed that MW-hrs data for the spring are applicable to the fall,
Mathematically, for each season i (i = 1, 2, 3 or 4),
(Seasonal Fraction)^ = (Sum of Hourly MW-hrs). I Total Hourly MW-hrs)i
Derivation of Daily Fractions:
Hourly MW-hrs data were divided into 12 categories corresponding to the 12 combinations of
Season/Day of Week (e.g., spring/Saturday; spring/Sunday; spring/weekday; summer/Saturday;
etc.). For each season i, hourly MW-hrs data were combined for each 24-hour period resulting
in daily generation of MW-hrs for each day of the week j (i.e., weekdays, Saturday, Sunday).
Mathematically:
(Daily Generation of MW-hrs)^ = Sum (All Hourly MW-hrs).
MW-hrs generated in each day of the week were then combined resulting in total MW-hrs
generated in Saturdays, total MW-hrs generated in Sundays, and total MW-hrs generated in
weekdays for each of the four seasons. Mathematically:
Total MW-hrs generatedj = Sum (Daily Generation of MW-hrs)}
Dividing the total MW-hrs generated in each of the three categories of days of week (i.e.,
Saturday, Sunday, and weekday) for each season by the season's total MW-hrs generated resulted
in fractional MW-hrs generated in each season in Saturdays, Sundays, and weekdays.
Mathematically:
(Fractional MW-hrs Generated).j = (Total MW-hrs generated). / (Sum of Hourly MW-hrs)^
Based on the fact that there are 65 weekdays, 13 Saturdays, and 13 Sundays in a particular
season, dividing the fractional MW-hrs generated in each season in Saturdays, Sundays, and
CH-94-35
3-45
-------
weekdays by the corresponding number of days results in the daily fractions of MW-hrs
generated for each season. Mathematically:
(Saturday Fraction of MW-hrs). = (.Fractional MW-hrs Generated)liSamrday / 13
(Sunday Fraction of MW-hrs). = (Fractional MW-hrs Generated)iSunday / 13
(Weekday Fraction of MW-hrs) { = (Fractional MW-hrs Generated) t Weekday / 65
Derivation of Hourly Fractions:
As described above, hourly MW-hrs data were divided into 12 categories corresponding to the
12 combinations of season/day of week (e.g., spring/Saturday, spring/Sunday, spring/weekday,
summer/Saturday, etc.). Mathematically:
(Daily Generation of MW-hrs)u = (Sum of All Hourly MW-hrs),
For each day of each combination of season/day of week, hourly MW-hrs were normalized by
dividing each hourly MW-hrs (k = 1 to 24) by the sum of all MW-hrs generated over the 24-
hours period of that same day. Mathematically:
(Normalized MW-hrs)ijk = Hourly MW-hrs)iJk / (Sum of Hourly MW-hrs).jX
Next, the diurnal profiles of each combination of season/day of week are created by averaging,
for each combination of season/day of week, the normalized MW-hrs for each of the 24-hour
period of a day.
(Hourly Fraction of MW-hrs)ij k = Sum of Normalized MW-hrs) ^ / (Total Number of Normalized MW-hrs)(JJi
Data were submitted to the aggregate database in electronic format.
CH-94-35
3-46
-------
3.9.4 Comments
The database did not contain entries for all days for some boilers. The modelers therefore
used the 1985 typical values where those missing days were to be modeled. This situation
occurred only when data were not submitted for large blocks of time (i.e., at least one month).
For some plants, data were often missing for between several hours and several days. Various
methods were used to estimate the missing data for these periods.
The Acid-Modes data were used to replace or augment the current allocation profile for
electric utility emission sources. The database includes actual hourly profiles, and the
documentation identifies the sources that submitted actual hourly data. The current profile for
electric utilities is based on similar work performed under NAPAP, but is based on a sample of
power plants in only the northeastern United States. The Acid-Modes Field Study database
contains data from a larger number of sources in a wider variety of areas (eastern United States
and Canada). The hourly MW load data were used to develop an improved profile of a typical
operating day for specific utility SCCs. These operating profiles were then used to replace
existing profiles for a few selected utility SCCs.
The Acid-Modes database was compiled between 1988 and 1990. Although data were
collected for a continuous two-year term, the data have not been completely maintained. The
available Acid-Modes database does not contain a full, continuous 12-month data set.
Accordingly, some key assumptions were made to allow a complete temporal profile to be
developed.
The following data were available:
* Winter — winter operations (December, January, February) are represented by January and
February data only.
* Spring ~ only March data were available for the spring months (March, April, May).
• Summer — the summer months (June, July, August) are completely represented.
~ Fall - no fall (September, October, November) data were available.
CH-94-35
3-47
-------
To develop a complete temporal profile, it was assumed that March operations were
typical of spring activity. It was also assumed that fall and spring activity were similar (i.e.,
identical for the purposes of this study). January and February data are assumed to appropriately
represent winter activity.
CH-94-35
3-48
-------
CHAPTER 4,0
SOURCE CATEGORY PRIORITIZATION
4.1 INTRODUCTION
A prioritized list of source categories was developed to focus on major national
contributors to air pollution and on efforts to develop the final TAF file. The prioritized source
categories reflected only data available from AIRS/AFS and focused on ozone precursors (VOC,
NO,, and CO), since these pollutants will be targeted in pending ozone/CO SIP modeling efforts.
Although the development process initially focused on specific priority SCCs, it was
decided that the final TAF file should offer profiles for as many source categories as possible
within the limitations of the data sources. Therefore, the priority SCC list functioned in three
ways:
~ Directed the search for additional data sources or additional data within a data set (e.g.,
economic data) towards high priority SCCs
Served as a benchmark to measure progress towards an adequate set of profiles
* Ensured that high-priority source categories received additional review and quality checks
The following section briefly describes the prioritization process. A more detailed
discussion is available in a June 9, 1993 memorandum and is included in Appendix A.14
4.2 PRIORITIZATION SEQUENCE
The prioritized list of SCCs and data sources used in the final TAF file development is
included in Appendix B. The SCCs contained in AFS were prioritized in the following sequence.
CK-S4-35
4-1
-------
1. Emissions of VOC, NOx, and CO by SCC for the current AFS database (including 1990
base year and SIP data) were retrieved from AFS using the AFS 650 report (Emission by
SCC Report). This report contains the following information:
• A national listing of source categories sorted by emissions of each pollutant
• Total estimated emissions in tons per year by SCC
• Number of records (a record in AFS is considered one emissions generation
process) for each SCC contributing to the total estimated emissions
2. An initial ranking of source categories from the AFS 650 report was performed to identify
the top 100 SCCs emitting each of the criteria pollutants. This list was presented in the
memorandum dated June 9, 1993 from Theresa Moody and David Winkler of TRC to
Chuck Mann, AEERL,14 The rank-ordered scale ranged from 1 to 100 with 1 being the
highest total emissions and 100 being the lowest total emissions.
3. Emissions-per-record for each criteria pollutant were calculated for the top 100 SCCs,
These SCCs were ranked in descending order. Emissions per record was selected as the
criterion that determines rank for the following benefits:
• Simplifies identifying true outliers (i.e., processes emitting relatively small
amounts per process may have significant total emissions by virtue of their
numbers)
• Provides a clearer picture of emissions by process
• Focuses attention on the higher emission processes that are most likely to be
included in a point source inventory and have the greatest impact on modeling
results
4. These top 100 SCCs were then scrutinized for plausibility of the reported emissions. As
a result of this quality control measure, some SCCs were discarded for each of the ozone
precursor pollutant SCC lists. The SCCs discarded were assumed to contain faulty data,
possibly representing misassigned SCCs.
5. SCCs representing poorly characterized sources (such as miscellaneous industrial fugitive
losses) were also omitted.
6. The remaining SCCs were re-ranked from highest to lowest emissions per record.
7. An overall rank order value (an average rank order value across all the ozone precursors
pollutants) was then calculated for each ozone precursor pollutant SCC (VOC, NO,, and
CO) based on the following expression:
CH-94-35
4-2
-------
E Rankpolluumt
Overall Rank Order = Pallmmt
Number of pollutants
If a category was not represented among all three ozone precursors pollutants, it was
assigned a rank order of 100 to provide consistency in the calculations and to simulate
the numerical averaging impact of the lowest ranked SCC,
CH-94-35
4-3
-------
CHAPTER 5.0
FINAL TEMPORAL ALLOCATION FACTOR FILE DEVELOPMENT
Once intermediate temporal profiles were complete (as described in Chapter 4) the files
were combined into the single final TAF file. This process required four major steps:
• Database design
* Rule definition
~ File combination
* Quality assurance
Each of these steps is discussed in this chapter.
5.1 DATABASE DESIGN
The final TAF file is a collection of temporal profiles, resident on EPA's NCC as a flat
text file. A temporal profile is constructed of three sets of multipliers: four seasonal fractions,
three day type fractions (i.e., weekday, Saturday, and Sunday), and 24 hourly fractions. The final
TAF file contains 12 lines (records) for each SCC. Each line represents a different season-day
type combination. The database format is shown in Figure 5-1.
5.2 RULE DEFINITION
Having defined the database structure and completed the intermediate TAF files, a set of
rules was defined to structure how the data were combined from the applicable data sources.
These rules allowed data to be combined in a consistent, well documented manner. As a result,
the data sources were assigned profile codes and were prioritized for their applicability to
developing seasonal, daily, and diurnal profiles. Profile codes identify the profile(s) that were
used to develop the TAF file for an SCC; these codes are presented in Table 5-1. The profiles
were based on eight-digit SCC assignments or based on all data in the relevant six-digit SCC
CH-94-35
5-1
-------
Variable Name
Field
Format
Field
Width
Description/Validation
see
Numeric
10
Source Classification Code (Point, Area,
or Mobile)
DAY_CODE
Numeric
1
Day Scenario Identifier
(l=Weekday, 2=Saturday, 3=Sunday)
SEA_CODE
Numeric
1
Season Scenario Identifier
(l=Spring, 2=Summer, 3=Fall, 4=Winter)
DAY.FRAC
Real
7.5a
Day Scenario Fraction
(65 weekday fraction)+(13 Saturday
fraction) + (13 Sunday Fraction) = 1.0
SEA_FRAC
Real
7.5a
Seasonal Scenario Fraction
(Spring fraction + Summer fraction + Fall
fraction + Winter fraction) = 1.0
SEA_FLAGb
Character
4
Seasonal Profile Data Source or
Combination Method Identifier
DAY_FLAGb
Character
4
Daily Profile Data Source or Combination
Method Identifier
HOUR_FLAGb
Character
4
Hourly Profile Data Source or
Combination Method Identifier
HOUR 1-HOUR24
Real
7.5a
Diurnal Profile
"Field, width for field type "Real" is defined as X.Y, where X represents the field width,
and Y represents the allowable number of digits reported to the right of the decimal.
^Table 5-1 defined the origin of the profile.
Figure 5-1. Final TAF file format.
group. Quarterly profile codes are listed in Table 5-2 in rank order of preference; Table 5-3 lists
profile codes for daily and hourly profiles in rank order or preference. Figure 5-2 documents the
established method for combining intermediate profiles.
CH-94-3S
5-2
-------
TABLE 5-1. DATA SOURCE FLAGS
Code
Description
FLAT
Flat profiles assigned (all seasons, all days and all hours equally
distributed.
NAPA
Profiles assigned from 1985 NAPAP TAF file.
ECON
Profiles assigned from economic data.
T6
Eight-digit profiles created from TNRCC data were averaged to the
six-digit level and assigned to all eight-digit SCCs in the respective
six-digit family.
T8
Eight-digit profiles created from TNRCC data were assigned to
appropriate SCCs.
B6
Eight-digit profiles created from SOS Point Source, CEM, LMOS,
Acid-Modes and TNRCC data were averaged to the six-digit level
and assigned to all eight-digit SCCs in the respective six-digit family.
A6
Eight-digit profiles created from SOS Point Source, CEM, LMOS,
Acid-Modes data were averaged to the six-digit level and assigned to
all eight-digit SCCs in the respective six-digit family.
A8
Eight-digit profiles created from SOS Point Source, CEM, LMOS,
Acid-Modes data were averaged and assigned to appropriate SCCs.
OTHR
Eight-digit profiles taken from CARB and WTE
B8
Eight-digit profiles taken from SOS Area Source.
5.3 FILE COMBINATION
Combining the intermediate profiles into a single final TAF FILE was accomplished using
an 11-step approach. The data were maintained on the NCC and manipulated using SAS.® The
SAS® programs were contained as members in a partitioned data set (PDS). SAS® files linking
CH-94-35
5-3
-------
TABLE 5-2. SEASONAL PROFILE CODES
Code
Corresponding Profile Data sets
A_8
TNRCC, CEM, LMOS at the eight-digit level All data from these data
sets should be combined using the number of sources represented as a
weighting factor to create a weighted average profile.
A_6
TNRCC, CEM, LMOS at the six-digit level. Six-digit profiles will have
to be created from the available eight-digit SCCs for the individual data
sets. The all data from these data sets should be combined using the
number of sources represented as a weighting factor to create a
weighted average profile.
TACB_8
TNRCC data at the eight-digit level. Sufficient CEM and LMOS data
unavailable.
TACB_6
TNRCC data at the six-digit level CEM and LMOS data unavailable
and no or too few TNRCC sources at the eight-digit level Six-digit
profiles will have to be created from the available eight-digit SCCs for
the individual data sets.
ECON
Economics data available, including BLS, E/E, or CRB. It needs to be
determined if TRC will combine these data sources or simply develop
an order of application.
NAPAP
NAPAP data available at the eight-digit level.
OTHR
To be defined on an ad hoc basis as a gap-filler, profile fixer, from
among TNRCC (<5), CARB and WTE. It needs to be determined how
these data will be used where other data sets, deemed insufficient due to
number of sources, arc also available.
FLAT
To be implemented, 0.25 for seasons; 0.01099 for daily fractions (i.e., 7
equal days); 0.04167 for hourly fractions (i.e., 24 equal hours)
CH-94-35
5-4
-------
TABLE 5-3. DAILY AND HOURLY (DIURNAL) PROFILE CODES
Code
Corresponding Profile Data sets
A_8
SOS, CEM, LMOS and Acid_Modes at the eight-digit level. All data from
these data sets should be combined using the number of sources represented as
a weighting factor to create a weighted average profile.
A_6
SOS, CEM, LMOS and Acid_Modes at the six-digit level. Six-digit profiles
will have to be created from the available eight-digit SCCs for the individual
data sets. The all data from these data sets should be combined using the
number of sources represented as a weighting factor to create a weighted
average profile.
B_6
SOS, CEM, LMOS and Acid_Modes and TNRCC at the six-digit level. Since
these SCCs failed at the A_6 level, this screens the SCCs based on adding
TNRCC data. Six-digit profiles will have to be created from the TNRCC eight-
digit SCCs. The all data from these data sets should be combined using the
number of sources represented as a weighting factor to create a weighted
average profile.
TACB_8
TNRCC data at the eight-digit level. CEM and LMOS data unavailable.
TACB_6
TNRCC data at the six-digit level. CEM and LMOS data unavailable and no
or too few TNRCC sources at the eight-digit level. Six-digit profiles will have
to be created from the available eight-digit SCCs for the individual data sets.
NAPAP
NAPAP data available at the eight-digit level.
OTHR
To be defined on an ad hoc basis as a gap-filler, profile fixer, from among
TNRCC (<5), CARB and WTE. It needs to be determined how these data will
be used where other data sets, deemed insufficient due to number of sources,
are also available.
FLAT
To be implemented. Certain source types have been identified (tank breathing
losses, storage piles) where a flat profile may be justified as a realistic profile.
0.25 for seasons; 0.01099 for daily fractions (i.e., 7 equal days); 0.04167 for
hourly fractions (i.e., 24 equal hours)
CH-94-35
5-5
-------
V
Dataset
containing
1 1 S s ,
point,
area, and
mobile,
(ALLPRO)
Apply
rolled-up
TAC3, and
LMOS data to
SCC8 level.
SOS,AMODES
TACB, and
LMOS data
to SCC6.
Roll-up
TACB to
SCCS level
(TACB 6}
IT £3
Apply the
Economic data
for seasonal
profiles
Only..
(ECON) .
IECON]
Apply SCCS
level data
(TACB 8) ,
[T I]
Roll-up
TACB, and
LMOSdata to
SCCS level.
Apply a
¦FLAT
Profile.?
days and 24
hours.
(FLATSCC)
[FLAT]
Final TAFF
(TAFF.DATA)
Make
case-by-case
adjustments
for "OTHER"
data sources.
(OTHER).
I OMR]
Apply SOS
AS to SCO.
Apply the
NAPAP
profiles.
(NAPAP).
[NAPA]
Figure 5-2. File combination flow chart
CH-94-35
5-6
-------
the steps of this approach comply with the naming convention WTILEIA.TAFF.xxxxxxxx.SAS
DAT, where xxxxxxxx is
ALLPRO for file containing profiles from all data sources
FLAT for file containing flat distributions for all SCCs
NAPAP for FLAT overlaid with NAPAP profiles
ECON for NAPAP overlaid with ECON profiles
TACB6 for ECON overlaid with profiles created from rolling-up TNRCC profiles
to the six-digit SCC level
TACB8 for TACB6 overlaid with TNRCC eight-digit SCC profiles
B6 for TACB6 overlaid with SOS, Acid-Modes, TNRCC, and CEM profiles
combined at the six-digit SCC level (for. daily and diurnal profiles only) s
A6 for B6 overlaid with profiles created from TNRCC, LMOS, and CEM
profiles combined at the six-digit SCC level
A8 for A6 overlaid with profiles created by combining TNRCC, LMOS, and
: CEM profiles at the eight-digit SCC level
OTHR for A8 manipulated for special case profiles,
B8 for AREA, replacing area source profiles with data from SOS area sources
FINAL Final TAF file
The stepwise approach used to combine the files is presented graphically in Figure 5-2. A
characteristic of the construction of the final TAF file is that for approximately five percent of
the SCCs, some season-day combinations exist for which no daily activity is suggested, yet
hourly profiles are provided. Due to the structure of the file, there are season-day profile
combinations where there are neither daily nor hourly profiles (e.g., sources operating in the
summer which do not operate on Saturdays). This phenomenon is created due to the independent
creation of seasonal, daily, and hourly profiles.
CH-94-35 5-7
-------
5.3.1 Module 1: Create Single Data Set
This module appended profiles from all data sources into a single SAS data set. This
was accomplished with a SAS SET statement incorporating data sets for all data sources. The
output file (ALLPRO) was sorted by SCC, DAY_CODE, SEA_CODE and SOURCE and was in
the intermediate file format.
PDS MEMBER:
INPUT FILE(S):
OUTPUT FILE;
ALLPRO
Name Format
WTIT ,EIA. ACID-MODES.SASDAT Intermediate
WTILEIA.CARB.SASDAT Intermediate
WTDLEIA.LMOS.SASDAT Intermediate
WTILEIA.NAPAP.TAFF. AS.SASDAT Intermediate
WTILEIA.NAPAP.TAFF.PS.SASDAT Intermediate
WTILEIA.TACB.SASDAT Intermediate
WTDLEIA. SOS .AS. SASDAT Intermediate
WTnJEIA.SOS.PS.SASDAT Intermediate
WTHJEIA.WASTE.TO.ENERGY.SASDAT Intermediate
WT1LELA.CEM.SASDAT Intermediate
WTILEIA.BLS.SASDAT Intermediate
WT1LEIA.TAFF.ALLPR0.SASDAT Intermediate
5.3.2 Module 2: Flat Profiles
This module assigned flat distributions (i.e., 0.25 for seasonal fractions, 0.01099 for daily
profiles, and 0.04167 for all 24 hours) to the universe of SCCs for each day type and season (i.e.,
3 day types and 4 seasons, 12 records). The output file (FLAT) was sorted by SCC,
DAY_CODE, and SEA_CODE and was in the final TAF file format. The source flags were set
as SEA_FLAG = 'FLAT', DAY_FLAG='FLAT'S and HR_FLAG='FLAT.*
PDS MEMBER:
FLAT
INPUT FILE(S): WTILEIA.ALL.SCCS,SASDAT
OUTPUT FILE(S): WTILEIA.TAFF.FLAT. SASDAT
VARIABLES: (SCC)
FORMAT: Final
CH-94-35
5-8
-------
5.3.3 Module 3: NAPAP Profiles
This module replaces flat profiles contained in FLAT with NAPAP profiles were called
for in the rules outlined in Section 5.2. NAPAP profiles were extracted from ALLPRO file
(SOURCE=ll). Specifically, NAPAP profiles were merged with FLAT, using SCC,
DAY_CODE, and SEA.CODE as key fields. The output file (NAPAP) was sorted by SCC,
DAY_CODE, and SEA..CODE and was in the final TAF file format. Source flags were set as
follows: SEAJFLAG = 'NAPA'; DAYJFLAG='NAPA'; and HR_FLAG-'NAPA\
PDS MEMBER: NAPAP
INPUT FILES: WTILEIA.TAFF.FLAT.SASDAT FORMAT: Final
WTELEIA.TAEF.ALLPRO.SASDAT FORMAT: Intermediate
OUTPUT FILE(S): WTILELA.TAFF.NAPAP.SASDAT FORMAT: Final
5.3.4 Module 4: Economic Data
This module replaced seasonal profiles created in the NAPAP module with seasonal
distribution derived from economic data. Economic profiles were extracted from ALLPRO
(SOURCE = 03). These economic profiles were merged with the file created in the NAPAP
module, using SCC, DAY_CODE, and SEA_CODE as key fields. The output file was sorted by
SCC, DAY_CODE, and SEA_CODE and was in the final TAF file format. The source flag was
set as SEAJFLAG = "ECON".
PDS MEMBER: ECON
INPUT FILE(S): WTILEIA.TAFF.NAPAP.SASDAT FORMAT: Final
WTILEIA.TAFF.ALLPRO.SASDAT FORMAT: Intermediate
OUTPUT FILE(S): WTILEIA.TAFF.ECON.SASDAT FORMAT: Final
CH-94-35
5-9
-------
5.3.5 Module 5; Six-Digit TNRCC Data
This module calculated and inserted six-digit composite TNRCC profiles into the working
file. TNRCC profiles were extracted at the six-digit SCC level from ALLPRO (SOURCE=02).
Averages [weighted by the number of observations (OBS)] were calculated for SEA_FRAC,
DAY_FRAC, and HR_1 through HR_24. Rounding errors were eliminated by normalizing the
TACB profiles (to ensure summation to unity) prior to insertion in the output file. The ECON
file and six-digit TNRCC profiles were merged, using SCC, DAY_CODE, and SEA_CODE, as
key fields. The output file (TACB6) was in the fmal TAF file format. The following source
flags were set: SEA_FLAG = T6% DAY_FLAG='T6\ and HR_FLAG='T6'.
PDS MEMBER:
TACB6
INPUT FILE(S): WTILEIA.TAFF.ECON .SASDAT
WTILEIA.TAFF.ALLPRO.SASDAT
OUTPUT FELE(S): WT1LEIA.TAFF.TACB6.SASDAT
FORMAT: Final
FORMAT: Intermediate
FORMAT: Final
5.3.6 Module 6: Eight-Digit TNRCC Data
This module replaces TACB6 profiles for which eight-digit profiles are available,
increasing the SCC resolution upon which the profiles are based. Initially, eight-digit SCC
profiles from TNRCC were extracted from ALLPRO (SOURCE =02). These profiles were merged
with TACB6, using SCC, DAY_CODE, SEA_CODE as key fields. The output file (TACB8) was
sorted by SCC, DAY_CODE, and SEA_CODE; the file was in the final TAF file format. The
source flags were set as follows: SEA_FLAG = 'T8\ DAY_FLAG=' T8 \ and HR_FLAG='T8\
PDS MEMBER:
TACB8
INPUT FILE(S): WTILEIA.TAFF.TACB6.SASDAT
WTILEIA.TAFF. ALLPRO.SASDAT
OUTPUT FILE(S): WTILEIA.T AEF.TACB 8. SASDAT
FORMAT: Final
FORMAT: Intermediate
FORMAT: Final
CH-94-35
5-10
-------
5.3.7 Module 7: Six-Digit Composite Daily and Diurnal Profiles
This module dealt solely with daily and diurnal profiles, substituting profiles based only
on TACB and economic data with composite profiles. Profiles from SOS Point Source
(SOURCE=05)» Acid-Modes (SOURCE=09), TNRCC (SOURCE=02), LMOS (SOURCE=01),
and CEM (SOURCE=06) were combined at the six-digit SCC level (i.e., lQlOOlxx), The profiles
were combined by calculating averages for seasonal, daily, and diurnal profiles (SEA_FRAC,
DAY_FRAC, and HR_1 through HR_24) weighted by the OBS variable. The composite profiles
were normalized to eliminate rounding errors. The normalized profiles were merged with
TACB8, using SCC, DAY„CODE, and SEA_CODE as key fields. The output file (B6) was
sorted by SCC, DAY_CODE, and SEA CODE and was in the final TAF file format. Source
flags were set as follows: DAY_FLAG='B6'; HR^FLAG='B6\
PDS MEMBER:
B6
INPUT FILE(S): WTILEIA.TAFF.TACB8.SASDAT
WTILEIA.T AFF. ALLPRO. S ASD AT
FORMAT: Final
FORMAT: Intermediate
OUTPUT FILE(S): WTILEIA.TAFF.B6.SASDAT
FORMAT: Final
5.3.8 Module 8: Combine and Use Six-Digit TNRCC, LMOS, CEM Data
This module inserted six-digit level combined TNRCC, LMOS, and CEM profiles into
the database. Profiles from TNRCC (SOURCE=02), LMOS (SOURCE=01), and CEM
(SOURCE=06) were extracted from ALLPRO. Incomplete seasonal data from CEM sources
predicated that CEM data not be included for creating seasonal profiles. Thus, two data sets were
created, one containing profiles from the TNRCC, LMOS, and CEM; the other containing only
TNRCC and LMOS data. Daily and diurnal profiles for each S CC/DA Y_CODE/HR_ 1 through
HR_24 combination were created from the three sources from averaging DAY_FRAC, and HR_1
through HR_24. These averages were weighted by OBS. Seasonal profiles for each
SCC/DAY_CODE/SEA_CODE combination were created by calculating average SEA_FRAC
values, weighted on OBS. These two working data sets were then merged, using SCC,
CH-94-35
5-11
-------
DAY_CODE and SEA_CODE as key fields. The resulting profiles were normalized to eliminate
rounding error. This file was merged with Bd, using SCC, DAY_CODE and SEA_CODE as key
fields. The output file (A6) was sorted by SCC, DAY_CODE, and SEA_CODE and was in the
final TAF file format. The source flags were set accordingly: SEA_FLAG='A6',
DAY_FLAG='A6\ and HR_FLAG='A6\
PDS MEMBER: A6
INPUT FILE(S): WTILEIA.TAFF.B6.SASDAT FORMAT: Final
WTILEIA.TAFF. ALLPRO. SASDAT FORMAT: Intermediate
OUTPUT FILE(S): WTILEIA.TAFF.A6.SASDAT FORMAT: Final
5,3.9 Module 9: Combine and Use Eight-Digit TNRCC, LMOS, CEM Data
This module performed the same tasks as described in 5.3.8, except on the eight-digit
level. This replaces the less source-specific six-digit data on the working file with more specific
eight-digit data. Profiles from TNRCC (SOURCE=02), LMOS (SOURCE=01), and CEM
(SOURCE=06) were extracted from ALLPRO. Incomplete seasonal data from CEM sources
predicated that CEM data not be included for creating seasonal profiles. Thus, two data sets
were created, one containing profiles from the TNRCC, LMOS, and CEM; the other containing
only TNRCC and LMOS data. Daily and diurnal profiles for each SCC/DAY_CODE/HR_ 1
through HR_24 combination were created from the three sources from averaging DAY_FRAC,
and HR_1 through HR_24, These averages were weighted by OBS. Seasonal profiles for each
S CC/D A Y_CODE/S EA_CODE combination were created by calculating average SEA_FRAC
values, weighted on OBS. These two working data sets were then merged, using SCC,
DAY_CODE and SEA_CODE as key fields. The resulting profiles were normalized to eliminate
rounding error. This file was merged with B8 file, using SCC, DAY_CODE and SEA_CODE
as key fields. The output file (AS) was sorted by SCC, DAY_CODE, and SEA_CODE and was
in the filial TAF file format. The source flags were set accordingly: SEAJFLAG = 'A8\
DAY_FLAG='A8\ and HRJFLAG='A8\
CH-94-3S
5-12
-------
PDS MEMBER: A8
INPUT FILE(S): WTILEIA.TAFF.A6.SASDAT FORMAT; Final
WTILEIA, TAFF. ALLPRO. SA SDAT FORMAT: Intermediate
OUTPUT FDLE(S): WTILEIA .T AFF. A8. S A SD AT FORMAT: Final
5.3.10 Module 10: Other Data
This module addresses profiles on a case-by-case basis for data sources tagged as
OTHER. These data sources are generally very specific, covering only a few SCCs (e.g., waste-
to-energy data). Data from these sources were extracted from ALLPRO, then merged with A8,
using SCC, DAY_CODE and SEA_CODE as key fields. The output file (OTHR) was sorted by
SCC, DAY_CODE, and SEA. CODE, then normalized to resolve any remaining rounding errors.
The source flags were set as SEA_FLAG=' OTHR', DAY_FLAG='OTHR\ and
HRJFL AG=' OTHR'.
OTHR
WTTLEIA.TAFF.A8.SASDAT FORMAT: Final
WTILEIA.TAFF.ALLPRO.SASDAT FORMAT: Intermediate
WTILEIA.TAFF.OTHR.SASDAT FORMAT: Final
SOS Area Source Data
PDS MEMBER:
INPUT FILE(S):
OUTPUT FILE(S):
5.3.11 Module 11:
This module replaced area source ten-digit SCC profiles with SOS Area Source (AS)
profiles. The SOS AS (SOURCE=05) profiles were extracted from ALLPRO (SOURCE = 05).
These extracted profiles were merged with AREA using SCC, DAY_CODE, and SEA_CODE as
key fields. The resultant file (FINAL) was sorted by SCC, DAY_CODE, and SEA_CODE. The
resulting data set is the final TAF file, and is in the format specified in Figure 5-1. Source flags
were set as SEAJFLAG = *B8\ DAYJFLAG='B8\ and HRJFLAG='B8'.
/
CH-94-35
5-13
-------
PDS MEMBER: B8
INPUT FILE(S): WTEJEIA.TAFF.AREA.SASDAT FORMAT: Final
WTILEIA.TAFF.ALLPRO.SASDAT FORMAT: Intermediate
OUTPUT FILE(S): WTILEIA.TAFF.FINAL.SASDAT FORMAT: Final
CH-94-3S
5-14
-------
CHAPTER 6.0
QUALITY ASSURANCE AND CONTROL OF TEMPORAL ALLOCATION FACTOR
FILE DATABASES
6.1 INTRODUCTION
The final TAF file was created in various stages that began with source-specific files (raw
files); and progressed through profiles generated into intermediate SAS® files, through various
stages of working module files. These complex operations required quality assurance/quality
control (QA/QC) procedures to ensure the integrity of the final file. Data integrity is used here
to indicate that data were complete (including no unintended deletions or additions); and that the
procedures used to merge, roll up, average, and normalize the data produced reasonable, correct,
and reproducible results. Because of the variety of data sources and the large amount of data
from certain sources, both manual and computerized sampling and review were used.
The following general approach was used to review the data and data files. First, source-
specific files were reviewed for completeness and consistency. Very incomplete or inconsistent
results (such as certain CEM data) were eliminated from data sets. Data obtained from
information sources that did not contain SCCs were reviewed for appropriate SCC designations
and were modified or deleted. This review was applied to economic data and waste-to-energy
data. In addition, spreadsheets and data file printouts of the profiles generated from source-
specific files were examined and calculations checked to confirm that the proper equations and
formulas were used to calculate the various factors, and that the results were reasonable.
Discrepancies or errors were corrected and the new results reviewed.
The review of various files (including the SAS® intermediate data files, working module
files, and the final TAF file) depended, in part, on the reliability of certain SAS® procedures to
produce correct results (such as weighted means) and to provide diagnostic information (such as
the number of records and the number of missing records). The review of the SAS® data
concentrated on processing steps that utilized the SAS® merge procedure because this was more
unpredictable than the other SAS® procedures used and more likely to produce unanticipated
results. Sample data were reviewed both before and after merge procedures for errors and/or
inconsistencies. When discrepancies were found, the procedure or data set was corrected as
CH-94-35
6-1
-------
necessary; the data were reprocessed and the new results reviewed. In addition, approximately
one percent of the records in the final temporal allocation factor file and ten percent of the area
source records were reviewed.
6.2 QUALITY CONTROL PROCEDURES
Quality control procedures were applied to the TAF data and files in four phases:
« Review of data source specific files and resultant profiles
* Initial assessment of SAS® procedures and resultant intermediate and working module
files
* Ongoing assessment of corrected SAS® files and procedures
* Assessment of final TAF file ~
In the first phase, the source-specific files were reviewed for completeness, consistency, and.
variability. Judgements were made to delete or limit the use of some data. For example, certain
CEM data were eliminated because insufficient data were available to produce representative
profiles. In other cases, CEM data were used to produce hourly profiles and not seasonal
allocations. For each data source, the equations, formulas, and procedures used to calculate
profiles were verified for correctness and the data spot-checked for consistency and
reasonableness. For data that were derived from other sources, such as seasonal allocation factors
derived from economic indicators (production, employment, or commodities) or temporal profiles
for waste-to-energy facilities, accurate SCC assignments were also reviewed. The revised data
output was rechecked. The profiles generated from the data source specific files were then
converted to SAS® intermediate files.
The second phase involved an initial assessment of the SAS® intermediate files and data
before and after processing to create several working module files. For reasons mentioned in
Section 6.1, this assessment focused on those steps in the process that used the SAS® merge
procedure. SAS® summary procedures and diagnostic messages were also used to verify data
completeness, duplicates, and missing records. Initial observations of input and output data files
CH-94-J5
6-2
-------
were reviewed. The data review indicated that the procedures were producing reasonable and
consistent profiles.
The deficiencies in the initial assessment were corrected in the third phase, which
involved an ongoing assessment, again concentrating on the working module files produced by
the SAS® merge procedure. For this phase, additional printouts were obtained in which records
were sorted by season, day, and hour flags. There were 20 observations for each combination.
This assessment provided assurance that the procedures were producing reasonable data from a
variety of intermediate files and in a variety of combinations of data sources.
The final TAF file was examined in the fourth phase. Approximately one percent of all
observations was reviewed for reasonableness. The data records were found to be complete and
all observations appeared reasonable.
6.3 SUMMARY OF DATABASE INTEGRITY
There is a definite distinction between data quality and data integrity. Data quality is
related to the correctness and reliability of data from various data sources; data integrity is related
to how well the final data are representative of completeness, traceability, and correctness. In
this study, the quality of most of the data could not be controlled by TRC. Some original source
data were removed by TRC from the data source-specific files because of incompleteness, lack
of internal consistency, or homogeneity. In addition, TRC tried to ensure that the economic data
used to create certain seasonal profiles were representative of the assigned SCCs. However,
ultimately the quality of the final TAF file depends primarily on the quality of the data's original
sources - NAPAP, LMOS, SOS, CEM, TNRCC, CARB, Acid-Modes, or WTE. Based on TRC's
review of the data source specific and intermediate files, each of the original data sets had some
associated discrepancies. These discrepancies included missing values for seasonal or daily
fractions, values that may represent permit allowables rather than typical values, and hourly
values from CEM data that may represent abnormal operating spikes. Resolving these
discrepancies was beyond the scope and resources of this work assignment.
There are some inconsistencies within the final TAF file that are inherent in the creation
process. No attempt was made to provide an overall rationalization of the data and factors.
However, a concerted effort was made to use the best data available for each particular allocation
CH-94-35-
6-3
-------
(seasonal, daily, or hourly). The process stages were arranged to ensure that the best available
data were included in the final TAF file. Thus, economic data or TNRCC data may be the basis
for a seasonal fraction while a combination of CEM, TNRCC, and LMOS data may be used for
the hourly allocation. Under these circumstances, it is possible to have a set of hourly allocation
factors for a seasonal or daily fraction that is zero. TRC made every effort to identify and
remove those discrepancies and inconsistencies that would lead to incorrect temporal allocations.
CH-94-35
6-4
-------
CHAPTER 7.0
CONCLUSIONS AND RECOMMENDATIONS
7.1 CONCLUSIONS
The work described in these chapters provides a significant improvement to the existing
temporal allocation factors. Specific areas of improvement include extensive use of real, facility -
level data, and improvement of the profile documentation methodology. The improved factors
should have a direct impact on air quality planning due to the broad use of photochemical grid
modeling and the models' requirement for hourly emissions input. Although the final TAF file
developed has a different file structure from that of other existing TAF files, this TAF file is
neutral (flat), allowing access by any model.
The methodology used to develop this TAF file is well documented, allowing future
updates and improvements. The profiles are based on activity, rather than emissions, making
them equally valid for allocating emissions of any pollutant. The TAF file is broad in scope,
including all point, area, and mobile SCCs. At the request of the steering committee, all NAPAP
profiles were either used or replaced, avoiding any loss of the coverage or detail in this file.
TRC has selected data sources on a three-tiered approach. Tier I data are national
economic statistics. These data are routinely updated, and future updates to profiles based on
these data would be relatively straightforward. Tiers II and HI data are, with the exception of
TNRCC data (updated annually), primarily one-time survey data and could not be updated. To
the extent possible, TRC has used "real" data, representing true activity rather than surrogates.
The sources of real data include (1) CEM data; (2) survey data from TNRCC, SOS, and LMOS;
and (3) WTE data. Using these "real" data helps avoid errors caused by judgement in identifying
appropriate surrogates and operating practices.
These profiles may be used to temporally resolve emissions estimates for a variety of
applications as identified in this document. As discussed, the TAF profile database may be used
in conjunction with any data sets and for any application. The conceptual action of the profiles
was introduced in Chapter 1 as a series of SCC-level allocations that may begin with an annual
inventory and conclude with 24 hourly estimates for three day types over four seasons for each
CH-94-35
7-1
-------
SCC in an inventory. The following example illustrates how one of these TAF profiles might be
used to allocate an annual emission estimate to an hourly emission value. In this example, only one
hour for one day and one SCC is considered, an actual inventory would contain many such SCCs,
days and hours.
Assume that we are interested in emissions from an utility boiler burning anthracite coal
(SCC 10100101), emitting 100 tons VOC (200,000 lbs), in 1992. The time of interest to the
modeler is Tuesday, July 24, 1990, between the hours of 8:00 am and 9:00 am. The inventory is an
annual inventory and no source-specific operating parameters are available for this source. The
actual profiles in the TAF for this SCC are presented in Figure 7-1. From these, the appropriate
single profile is selected for the Summer (SEA. CODE=2), weekday (DAY_CODE=l) and hour
(HOUR=9). These three profile components are applied to the annual estimate:
Emissionshour = Annual Emissions x Seasonal Factor x Daily Factor x Hourly Factor
= 200,000 lbs x SEA_ FRAC x DAY__ FRAC x HOUR 9
= 200,000 lbs x 0.22260 x 0.01099 x 0.03884
= 19.003 lbs
7.2 RECOMMENDATIONS
Although the work performed under this Work Assignment provides an improved TAF file,
work should continue to provide further improvements. Some potential opportunities for
refinement are as follows:
• Develop software to allow access from EPS2.0, FREDS, and GEMAP (leading
photochemical models) to this TAF file.
• Further analyze profiles assigned to high-priority SCCs, possibly leading to telephone
surveys to develop profiles from facility-level data.
• Update and refine temporal profiles for area and mobile sources
• Compare new and existing profiles
• Incorporate data that were not received in time to be included in this work assignment (e.g.,
wastewater data)
• Develop a PC-based TAF file for use with workstation-based models (notably GEMAP)
CH-94-33
7-2
-------
see
D S D»y_frac Sc*_frtc Hourl Hour2 Hour3 Haui4 Hour5 Houxti Houi7 Hour8 Haui9 HaurlO Hour 11 Hour 12 Hourl3 Hourl4 Hourl5 Hour 16 Haurl7 Hourl8 Hour 19 Hour20 Hour21 Hour22 Haur23 Hour24
i
10100101
1
1
0.01099
0.24858
0.02843
0.02759
0.02666
0.02697
0.02782 0.02928
10100101
1
2
0.01099
0.22260
0.02748
0.02633
0.02572
0.02556
0.02610 0.02848
10100101
1
3
0.01100
0.19406
0.03076
0D3015
0.03030
0.03084
0.03776 0.03699
10100101
1
4
0.01099
033476
0.03479
0.03371
0.03317
0.03287
0.03333 0.03494
10100101
2
1
0.01098
0.24858
0.03600
0.03400
0.03200
0.03100
a03200 0.03200
10100101
2
2
0.01098
0.22260
0.03808
0.03507
0.03307
0.03206
0.03106 0.03206
10100101
2
3
0.01096
0.19406
0.03600
0.03400
0.03200
0.03100
0.03200 0.03200
10100101
2
4
0.01098
033476
0.00593
0.03293
0.03293
0.03293
0.03293 0.03293
10100101
3
1
0.01098
024858
0.03896
0.03696
0.03596
0.03596
0l03596 0.03596
10100101
3
2
0.01097
0.22260
0.03896
0.03596
0.03397
0.03397
0.03297 0.03197
10100101
3
3
0.0i095
0.19406
0.03896
0.03696
0.03596
0.03596
0.03596 0.03596
10100101
3
4
0.01097
033476
0.03604
0.03403
0.03303
0.03203
0.03203 0.03303
0.03304 0.03996
0.00347 0.03884
0.04291 0.04629
0.03832 0.04224
0.03400 0.03800
0.03307 0.03607
0.03400 0.03800
0.03593 0.03892
0.03696 0.03996
0.03197 0.03377
0.03696 0.03996
0.03504 0.03704
0.04373
0.04299
a04790
0.04462
0.04300
a04108
0.04300
0.04291
0.04196
0.03796
0.04196
0.04004
0.04519
0.04537
0.04806
0.04585
0.04700
0.04509
0.04700
0.04691
0.04396
0.04096
0.04396
0.04204
0.04534
0.04667
0.04806
0.04585
0.04800
0.04810
0.04800
0.04790
0.04396
0.04496
0.04396
0.04404
0.04542
0.04682
0.04736
0.04600
0.04900
0.04910
0.04900
0.04790
0.04496
0.04695
0.04496
0.04505
0.04457
0.04644
0.04721
0.04554
0.04700
0.04910
0.04700
0.04790
0.04496
0.04795
0.04496
0.04605
0.04434
0.04659
0.04660
0.04515
0.04700
0.04810
0.04700
0.04491
0.04396
0.04795
0.04396
0,04505
0.04342 0.04280
0.04629 0.04621
0.04583 0.04598
0.04469 0.04446
0.04500 0.04500
0.04709 0.04709
0.04500 0.04500
0.04391 0.04391
0.04196 0.04096
0.04695 0.04695
0.04196 0.04096
0.04404 0.04505
0.04303
0.04613
0.04660
0.04554
0.04600
0.04709
0.04600
0.04790
0.04196
0.04595
0.04196
0.04805
0.04304
0.04506
0.04836
0.04708
0.04600
0.04709
0.04600
0.05090
0.04296
0.04595
0.04296
0.05105
0.04327
0.04345
0.04898
0.04746
0.04600
0.04609
0.04600
0.04890
0.04396
0.04595
0.04396
0.05105
0.04503
0.04352
0.04806
0.04700
0.04700
0.04509
0.04700
0.04691
0.04695
0.04595
0.04695
0.05005
0.04465
0.04337
0.04521
0.04554
0.04800
0.04609
0.04800
0.04491
0.04995
0.04795
0.04995
0.04805
0.04173
0.03946
0.03983
0.04385
0.04600
0.04409
0.04600
0.04291
0.04795
0.04795
0.04795
0.04605
0.03650
0.03409
0.03460
0.04055
0.04300
0.04208
0.04300
0.03992
0.04396
0.04496
0.04396
0.04304
0.03127
0.02863
0.03038
0.03747
0.03800
0.03707
0.03800
0.03593
0.03896
0.04096
0.03896
0.03904
Figure 7-1. Temporal profiles for SCC 10100101 from the TAF file.
-------
CHAPTER 8.0
REVIEW AND ANALYSIS OF TEMPORAL ALLOCATION FACTOR FILE
8.1 INTRODUCTION
The purpose of Work Assignment 1/014 was to perform selected analyses of the data
collected and documented under Work Assignment No, 3/314, EPA Contract No, 68-D9-0173.
These analyses were to include evaluations of temporal profiles for selected high priority source
categories and comparisons of the new default profiles with previously available profiles. The
results of these analyses will be used in improving the new national default file of TAFs. One
objective of Work Assignment 1/014 was to review, analyze, and revise (if necessary) the TAF
files developed under the initial TAF project (Work Assignment 3/314) and described in Chapters
2-7. Chapters 8 and 9 describe the analysis and results (Chapter 8), and provide conclusions and
recommendations based on the results (Chapter 9).
8.2 SCC SELECTION
The final product of the initial TAF project was a file of new TAFs for all SCCs, Under
this Work Assignment, a comparative analysis of new and existing TAFs was performed for
selected SCCs. SCCs to be reviewed were selected to represent the "most significant" source
categories, in terms of emissions and number of occurrences, for the ozone season and for the
CO season. Specifically, the following categories of SCCs were chosen:
(1) All twelve temporal scenarios (four seasons, three day types) were reviewed for SCCs
with priority rank 1 through 10, as defined in Work Assignment 3/314, This selection
constituted a detailed review of a few important source categories.
(2) Summer weekday profiles were reviewed for SCCs ranked 11 through 25 in the above
priority scheme. Reviewing a single temporal scenario rather than all season-day types
allowed for a larger number of SCCs to be analyzed, while reviewing activity patterns of
sources of ozone precursors during the ozone season. This effort, combined with the
detailed analysis of the "top ten" SCCs, allowed profile review for the top 25 ozone-
causing SCCs for the ozone season.
CH-94-35
8-1
-------
(3) Winter weekday profiles were reviewed for SCCs ranked 1 through 25 in a CO-only
prioritized SCC list to review activity patterns of sources of CO during the CO season.
Two rank-ordered lists of SCCs were used in this review. The first list, generated under
Work Assignment No. 3/314, was designed to identify SCCs emitting precursors of ozone. SCCs
for this list were rank-ordered based on AFS reported emissions of nitrogen dioxide (N02)
[oxides of nitrogen (NOJ are reported as N02 in AFS], VOC, and CO. As summarized in the
July 7, 1993 memorandum (see Appendix A), SCCs were ranked as follows:15
• emissions (tons per year) per record were calculated for each SCC/pollutant combination
• SCCs were rank-ordered for N02, VOC, and CO, based on emissions per record
• Ranking indices for the three pollutants were averaged to develop a combined ranking
index for each SCC
• SCCs were rank-ordered based on the combined ranking index
Table 8-1 presents the SCCs selected for this study (Nos. 1 and 2 above).
A second rank-ordered list, also developed under Work Assignment No. 3/314, was
designed to identify important sources of CO emissions. This list contained several SCCs that
appear to be miscoded data sources. These SCCs were removed from the list and the remaining
SCCs were reranked. The top 25 SCCs for CO from this revised list are shown in Table 8-2.
8.3 PROFILE ANALYSIS
All selected profiles were compared to analogous profiles from the 1985 NAPAP FREDS
Modelers Input (MIP). Graphic representations comparing the TAF and NAPAP were generated
as line charts for this analysis. A visual analysis of the profiles was performed comparing the
new TAF profiles to the existing profiles and evaluating the profiles based on knowledge of the
process or activity represented by the SCC. The purpose of the two-pronged analysis was to
CH-94-35
8-2
-------
TABLE 8-1. PRIORITY OZONE SCCs
Rank
see
SCC Description
1
3-07-001-04
Kraft Pulp: Recovery Furnace/Direct Contact Evaporator
2
3-06-002-01
Fluid Catalytic Cracking Unit
3
3-01-005-04
Carbon Black Production: Oil Furnace Process: Main Process Vent
4
3-01-031-02
Terephthalic Acid: Reactor Vent
5
3-03-008-13
Iron Sintering: Windbox
6
3-01-006-03
Charcoal Manufacturing: Batch Kiln
7
1-01-003-01
Utility Boiler: Pulverized Coal: Lignite
8
1-01-002-02
Utility Boiler: Pulverized Coal: Dry Bottom: Bituminous Coal
9
1-01-002-01
Utility Boiler: Pulverized Coal: Wet Bottom: Bituminous Coal
10
1-01-002-22
Utility Boiler: Pulverized Coal: Dry Bottom: Subbituminous Coal
11
1-01-002-26
Utility Boiler: Dry Bottom: Tangential Fired: Subbituminous Coal
12
1-01-002-03
Utility Boiler: Cyclone Furnace; Bituminous Coal
13
3-06-004-01
Blowdown System with Vapor Recovery System
14
1-01-002-12
Utility Boiler: Dry Bottom: Tangential Fired: Bituminous Coal
15
1-01-006-01
Utility Boiler > 100 mmBtu/Hr except Tangential
16
1-01-004-04
Utility Boilers: Grade 6 Oil: Tangential Firing
17
1-01-006-04
Utility Boilers: Tangentially Fired Units
18
3-05-007-06
Wet Cement: Kilns
19
3-01-025-01
Cellulosic Fiber: Viscose (e.g., Rayon)
20
1-01-003-03
Utility Boilers: Cyclone Furnace: Lignite
21
1-01-003-02
Utility Boilers: Pulverized Coal: Tangential Firing: Lignite
22
1-01-002-23
Utility Boilers: Cyclone Furnace: Subbituminous Coal
23
3-01-025-05
Cellulose Fiber: Acetate
24
3-02-010-03
Whiskey Fermentation: Aging
25
3-01-060-03
Pharmaceutical Preparations: Distillation Units
CH-94-35
8-3
-------
TABLE 8-2. PRIORITY CO SCCs
Rank
see
SCC Description
' 1
3-03-008-26
Iron Blast Furnace Slips
2
3-03-009-13
Steel Mfg: Basic Oxygen Furnace; Open Hood - Stack ¦¦
3
3-03-009-14
Steel Mfg: Basic Oxygen Furnace: Closed Hood - Stack
4
3-03-008-12
Iron Sintering: Raw Material Transfer/Handling
5
3-04-003-15
Iron/Steel Foundry: Gray Iron Foundry: Charge Heating
6
3-03-001-01
Aluminum Ore/Electro-Reduction; Prebaked Reduction Cell
7
3-03-001-02
Aluminum Ore/Electro-Reduction: Horizontal Stud Soderberg Cell
8
3-01-035-03
Inorganic Pigment: Ti02 Chloride Process: Reactor
9
3-03-012-01
Titanium: Chlorination
10
3-01-005-03
Carbon Black Production: Gas Furnace Process: Main Process
• Vent
11
3-03-001-03
Aluminum Ore/Electro-Reduction: Vertical Stud Soderberg Cell
12
3-01-005-09
Carbon Black Production: Furnace Process: Fugitive Emissions
13
3-01-031-01
Terephthalic Acid: HN03 - Paraxylene - General
14
3-03-001-05
Aluminum Ore/Electro-Reduction: Anode- Baking Furnace
15
3-01-158-01
Cyclohexanone/Cyclohexanoi Production: General
16
3-01-019-01
Phthalie Anhydride: o-Xylene Oxidation: Main Process Stream
17
1-01-013-01
Utility Boilers: Liquid Waste
18
1-02-014-02
Industrial Boilers: CO Boiler: Process Gas
19
3-01-001-03
Adipic Acid: Cyclohexane Oxidation
20
3-03-009-07
Primary Metals: Iron and Steel: Steel Furnaces: Tapping-Electric
Arc
21
3-07-001-10
Pulp and Paper: Kraft Pulping: Recovery Furnace: Indirect Contact
Evaporator
22
3-03-009-08
Steel: Electric Arc Furnace: Carbon Steel
23
3-01-003-05
NHj Production: Feedstock Desulfurization
24
3-03-001-07
Aluminum Ore: Electric Reduction: Roof Vent
25
3-90-007-02
Process Gas: Coke Oven Gas
CH-94-35
8-4
-------
determine if the profiles were "reasonable" based on knowledge of the process and to identify
possible anomalies in the data.
Analysis of the data revealed profiles that were based on CEM data, showing no
discernable diurnal pattern. In addition, there were several profiles with a pronounced spike at
2:00 a.m. Because of these problems, the original CEM data were reviewed to identify potential
errors.
Review of the archived CEM files showed that the CEM data were reported in units of
pounds per million Btu (lbs/mmBtu) or parts per million (ppm). These data indicate fuel
characteristics, not mass emissions. State agency personnel from Kentucky and Pennsylvania
were contacted to confirm the units of the data and to clarify what data are available from CEM
records. Both indicated that their respective states collect temporal data in terms of lbs/MBtu
or ppm, depending on the source, consistent with the standards regulating boiler/incinerator
activity. These states do not collect the fuel consumption/energy output or flow data required
to convert to units of mass emissions (i.e., they are measuring against what is in the existing
operating permit, not intending to estimate cumulative emissions).
These CEM data were considered insufficient to be the sole basis for generating emissions
profiles, as the profiles reviewed essentially showed the random scatter around the average
emissions rate. Because neither state collected the fuel consumption/energy output or flow data
required to convert to emissions, these CEM data could not be converted for use as emissions
profiles. For this reason, the profiles were regenerated without the CEM data. As before, TAF
profiles for the selected SCCs were compared to analogous profiles from the 1985 NAPAP
FREDs MIP file. Line charts were generated for visual analysis of the data.
This review still resulted in some questions and concerns about the profiles. For example,
if data from several plants with different operating schedules were averaged to produce a profile,
the resulting profile had a step-wise appearance. This type of profile is not truly representative
of any one plant, but may represent the industry as a whole if the plants used to develop the
profile are truly representative (in size and frequency) of the industry. If this cannot be verified,
the existing MIP profile may be preferable to the new TAF profile.
CH-94-35
8-5
-------
Tables 8-3 and 8-4 provide comments on and suggested actions for each SCC profile
reviewed. These tables were presented to the EPA Work Assignment Manager and Review Team
in a March 29, 1994 project review meeting. The suggested actions noted in comments in
Tables 8-3 and 8-4 were approved by the EPA Work Assignment Manager and Review Team,
The Review Team included representatives from the Atmospheric Research and Exposure
Assessment Laboratory (AREAL) and the Office of Air Quality Planning and Standards
(OAQPS).
8.4 STATISTICAL ANALYSIS
Among the concerns voiced at the March 29, 1994 project review meeting was the need
for some statistical analyses of the data used to create the profiles. The following paragraphs
describe methodologies that could be incorporated in a comparative analysis of the diurnal TAFs
created under Work Assignment No. 3/314. This analysis will attempt to identify outliers within
the new TAFs. The fundamental definition of an outlier for this analyses will be based on an
individual hourly factor. After an outlier has been identified, all the profile's hourly factors will
be inspected.
Two main approaches for this analysis were discussed. Both approaches are based on a
measurement of dispersion. Approach I is based on standard deviation. Approach II uses a
process developed specifically for the project and is based on absolute differences. These
approaches were designed to provide quality ratings instead of an approach in which hourly
factors would either meet or fail a set criterion. This type of analysis was considered more
beneficial for obtaining an overall picture of the quality of the new TAFs.
Both approaches are based on comparison between eight-digit SCC profiles and SCC-
family aggregate profiles. These SCC-family aggregate profiles would consist of either eight-
digit SCC profiles being grouped to the six-digit SCC level, or grouped according to EPA's new
tier category reporting system. Grouping eight-digit SCC profiles using the new tier system
should prove more appropriate for this analysis than grouping eight-digit SCC profiles at the six-
digit SCC level due to fundamental differences in operational schedules found within six-digit
SCC groupings.
8-6
CH-94-35
-------
TABLE 8-3. PRIORITY OZONE SCCs REVIEW
Rank
see
SCC Description
Comment
1
3-07-001-04
Kraft Pulp: Recovery Furnace/Direct
Contact Evaporator
Continuous process. Spike at hours 22-24, weekdays, all seasons. Profile based on
LMOS data. Suggested action: Compare with other similar SCCs. If profile
pattern holds, accept new TAF profile.
2
3-06-002-01
Fluid Catalytic Cracking Unit
Catalyst regeneration can be continuous or cyclic, however, if it is cyclic,
regeneration can occur at any time during the day, not necessarily at the same time
every day. Suggested action: Accept new TAF profile.
3
3-01-005-04
Carbon Black Production: Oil
Furnace Process: Main Process Vent
Approximately 90 percent of the industry uses a process which is a continuous
operation. Suggested action: Accept new TAF profile.
4
3-01-031-02
Terephthalic Acid: Reactor Vent
No new data for new TAF; MIP is flat. Dominant process in the industry is a
continuous process. Suggested action: Use MIP profile.
5
3-03-008-13
Iron Sintering: Windbox
No new data for new TAF; MIP profile suggests that data from several plants with
different operating schedules were averaged to produce this profile. Suggested
action: Use MIP profile.
6
3-01-006-03
Charcoal Manufacturing: Batch Kiln
This is a batch kiln, but the process is continuous throughout the batch cycle
(approximately a three-week cycle). Therefore, emissions profile should be flat.
Suggested action: Accept new TAF profile.
7
1-01-003-01
Utility Boiler: Pulverized Coal:
Lignite
New TAF is flat and is based on TNRCC data, which ;ue most likely normal
operating data. These data will only show thai activity occurs daily, but will not
show diurnal patterns. Suggested action: Use MIP profile.
8
1-01-002-02
Utility Boiler: Pulverized Coal: Dry
Bottom: (Bituminous Coal)
New TAF is based on LMOS data and may show a regional bias. Profile appears to
be driven by industrial activity. Suggested action: Discuss with EPA personnel.
9
1-01-002-01
Utility Boiler: Pulverized Coal: Wet
Bottom: Bituminous Coal
New TAF = NAPAP. Profile is similar to other utility boiler profiles. Suggested
action: Accept new TAF.
10
1-01-002-22
Utility Boiler: Pulverized Coal: Dry
Bottom: Subbituminous Coal
New TAF is based on LMOS data ;uid may show a regional bias. MIP profile is
flat, suggesting that there were no data for this SCC in NAPAP. Flat profile is not
reasonable for utility boilers. Sunday spike is real, based on the LMOS data.
Suggested action: Discuss with EPA personnel.
(Continued)
-------
TABLE 8-3. PRIORITY OZONE SCCs REVIEW (Continued)
Rank
see
SCC Description
Comment
11
1-01-002-26
Utility Boiler: Dry Bottom:
Tangential Fired: Subbituminous
Coal
New TAF is flat and is based on TNRCC data, which are most likely normal
operating data. These data will only show that activity occurs daily, but will not
show diurnal patterns. M1P profile is flat, suggesting that there were no data for this
SCC in NAPAP. Flat profile is not reasonable for utility boilers. Suggested action:
Develop a generic subbituminous profile (e.g., 1-01-002-22) for cases similar to this.
Discuss with EPA personnel.
12
1-01-002-03
Utility Boiler: Cyclone Furnace:
Bituminous Coal .
New TAF is based on LMOS data and may show a regional bias. MIP file may be
more representative as a national default. Suggested action: Discuss with EPA
personnel.
13
3-06-004-01
Blowdown System w/ Vapor
Recovery System
New TAF is flat and may be based on normal operating data from TNRCC. Thus, it
may not show diurnal fluctuations. MIP profile suggests that data from several
plants with different operating schedules were averaged to produce this profile.
Emissions profile may show a spike anytime during the day, but not necessarily at
the same time every day. Flat profile may be reasonable in this case. Suggested
action: Discuss with EPA personnel.
14
1-01-002-12
Utility Boiler: Dry Bottom:
Tangential Fired: Bituminous Coal
New TAF is based on LMOS data and may show a regional bias. MIP is flat,
suggesting that there were no data in NAPAP for this SCC. Suggested action:
Discuss accepting new TAF with EPA personnel.
15
1-01-006-01
Utility Boilers > 100 MBtu/Hr
except Tangential
New TAF is flat and is based on TNRCC data, which are most likely normal
operating data. These data will only show that activity occurs daily, but will not
show diurnal patterns. Hierarchy of data used to develop TAF results in the use of
the TNRCC flat profile. Suggested action: Use MIP profile; discuss data hierarchy
with EPA personnel.
16
1-01-004-04
Utility Boilers: Grade 6 Oil:
Tangential Firing
New TAF = NAPAP: Profile is similar to other utility boiler profiles. Suggested
action: Accept new TAF profile.
17
1-01-006-04
Utility Boilers: Tangentially Fired
Units
New TAF is flat and is based oil TNRCC data, which are most likely normal
operating data. These data will only show that activity occurs daily, but will not
show diurnal patterns. Hierarchy of data used to develop TAF results in the use of
the TNRCC Hat profile. Suggested action: Use MIP profile; discuss data hierarchy
with EPA personnel.
(Continued)
-------
TABLE 8-3. PRIORITY OZONE SCCs REVIEW (Continued)
Rank
see
SCC Description
Comment
18
3-05-007-06
Wet Cement: Kilns
New TAF is flat and is based on TNRCC data, which are most likely normal
operating data. These data will only show that activity occurs daily, but will not
show diurnal patterns. MIP profile suggests that data from several plants with
different operating schedules were averaged to produce this profile. Suggested
action: Discuss using MIP profile with EPA personnel.
19
3-01-025-01
Cellulosic Fiber: Viscose (e.g.,
Rayon)
No new data for new TAF. MIP is flat. Fiber production is most likely a
continuous operation. Suggested action: Use MIP profile.
20
1-01-003-03
Utility Boilers: Cyclone Furnace:
Lignite
New TAF = NAPAP. Profile is similar to other utility boiler profiles. Suggested
action: Accept new TAF profile.
21
1-01-003-02
Utility Boilers: Pulverized Coal:
Tangential Firing: Lignite
New TAF is flat and is based on TNRCC data, which are most likely normal
operating data. These data will only show that activity occurs daily, but will not
show diurnal patterns. MIP profile is flat, suggesting that there were no data for this
SCC in NAPAP. Flat profile is not reasonable for utility boilers. Suggested action:
Develop a generic lignite profile for cases similar to this. Discuss with EPA
personnel.
22
1-01-002-23
Utility Boilers: Cyclone Furnace:
Subbituminous coal
New TAF is flat and is based on TNRCC data, which are most likely normal
operating data. These data will only show that activity occurs daily, but will not
show diurnal patterns. MIP profile is Hat, suggesting thai there were no data for this
SCC in NAPAP. Flat profile is not reasonable for utility boilers. Suggested action:
Develop a generic subbiluminous profile for cases similar to this. Discuss with EPA
personnel.
23
3-01-025-05
Cellulose Fiber: Acetate
No new data for new TAF. MIP is flat. Fiber production is most likely a
continuous operation. Suggested action: Use MIP profile.
24
3-02-010-03
Whiskey Fermentation: Aging
No new data for new TAF. MIP is flat. Whiskey fermentation is most likely a
continuous operation. Suggested action: Use MIP profile.
25
3-01-060-03
Pharmaceutical Preparations:
Distillation Units
No.new data for new TAF; MIP profile suggests that data from several plants with
different operating schedules were averaged to produce this profile. Suggested
action: Use MIP profile.
-------
TABLE 8-4. PRIORITY CO SCCs REVIEW
Rank
see
SCC Description
Comment
1
3-03-008-26
Iron Blast Furnace Slips
No new data for new TAF; MIP is flat. SCC is most likely a continuous process.
Suggested action: Use MIP profile.
2
3-03-009-13
Sleel Manufacturing: Basic Oxygen
Furnace: Open Hood-Stack
No new data for new TAF; MIP profile suggests that data from several plants with
different operating schedules were averaged to produce this profile. Suggested action:
Discuss using profile for SCC 30300914 (#3 below) with EPA personnel.
3
3-03-009-14
Steel Manufacturing: Basic Oxygen
Furnace: Closed Hood-Stack
New TAF suggests that this SCC operates two shifts per day. MIP profile suggests that
data from several plants with different operating schedules were average to produce this
profile. Reason for the spike at 8 a.m. is unknown. Suggested action: Accept new
TAF.
4
3-03-008-12
Iron Sintering: Raw Material
Transfer/Handling
SCC not applicable for CO (no CO emission factors available for this SCC).
5
3-04-003-15
Iron/Steel Foundry: Gray Iron
Foundry: Charge Heating
No new data for new TAF. MIP profile suggests that data from several plants with
different operating schedules were averaged to produce this profile. Suggested action:
Discuss using mode rather than mean with EPA personnel or use MIP profile as is.
6
3-03-001-01
Aluminum Ore/Electro-Reduction:
Prebaked Reduction Cell
No new data for new TAF; MIP profile is flat. SCC is most likely a continuous process.
Suggested action: Use MIP profile.
7
3-03-001-02
Aluminum Ore/Eleclro-Recliiction:
Horizontal Slud Soderberg Cell
No new data for new TAF; MIP profile is Hat. SCC is most likely a continuous process.
Suggested action: Use MIP profile.
8
3-01-035-03
Inorganic Pigment: TiO, Chloride
Process: Reactor
No new data for new TAF; MIP profile is flat. SCC is most likely a continuous process.
Suggested action: Use MIP profile.
9
3-03-012-01
Titanium: Chlorinaiion
No new data for new TAF; MIP profile is flat. SCC is most likely a continuous process.
However, no CO emission factors were found for this SCC. Suggested action: Check
the number of occurrences of (his SCC in AIRS. Discuss with EPA personnel.
10
3-01-005-03
Carbon Black Production: Gas
Furnace Process: Main Process
Vent
New TAF is Hat, may be based on TNRCC normal operating data (which would not show
diurnal patterns) or may suggest continuous operation. MIP profile suggests that data
from several plants with different operating schedules were averaged to produce this
profile. Suggested action: Accept new TAF.
11
3-03-001-03
Aluminum Ore/Eleclro-Reduclion:
Vertical Stud Soderberg Cell
No new data for new TAF; MIP profile is Oat. SCC is most likely a continuous process.
Suggested action: Use MIP profile.
(Continued)
-------
TABLE 8-4. PRIORITY CO SCCs REVIEW (Continued)
oo
Rank
see
SCC Description
Comment
. 12
3-01-005-09
Carbon Black Production: Furnace
Process: Fugitive Emissions'
New TAF is flat, may be based on TNRCC normal operating data (which would not show
diurnal patterns) or may suggest continuous operation. MIP profile suggests that data
from several plants with different operating schedules were average to produce this
profile. Suggested action: Discuss using mode rather than mean with EPA personnel.
13
3-01-031-01
Terephlhalic Acid: HN03 -
Paraxylene - General
No new data for new TAF; MIP profile is Hat. SCC is most likely a continuous process.
Suggested action: Use MIP profile.
14
3-03-001-05
Aluminum Ore/Electro-Reduclion:
Anode Baking Furnace
New TAF is flat, may be based on TNRCC normal operating data (which would not show
diurnal patterns) or may suggest continuous operation. MIP profile suggests that data
from several plants with different operating schedules were average to produce this
profile. Suggested action: Discuss using mode rather than mean with EPA personnel.
15
3-01-158-01
Cyclohexanone/Cyclohexanol
Production: General
New TAF is flat, may be based on TNRCC normal operating data (which would not show
diurnal patterns) or may suggest continuous operation. MIP profile is flat. SCC is most
likely a continuous process. Suggested action: Accept new TAF profile.
16
3-01-019-01
Phthalic Anhydride: o-Xylene
Oxidation: Main Process Stream
New TAF is flat, may be based on TNRCC normal operating data (which would not show
diurnal patterns) or may suggest continuous operation. MIP profile is flat. SCC is most
likely a continuous process. Suggested action: Accept new TAF profile.
17
1-01-013-01
Utility Boilers: Lit[iiid Waste
New TAF is Hal, may be based on TNRCC normal operating data (which would not show
diurnal patterns) or may suggest continuous operation. MIP profile is flat, suggesting a
continuous process. However, this is a utility boiler and one would expect to see the
usual diurnal profile. Suggested action: Determine from AIRS where these facilities are
located and how many there are. Discuss with EPA personnel.
18
1-02-014-02
Industrial Boilers: CO Boiler:
Process Gas
New TAF profile is very unusual. TNRCC normal operating data could not give this type
of profile. MIP profile is flat, suggesting a continuous process. Suggested action:
Determine origin of TAF profile and check data. If no data entry errors are discovered,
use MIP profile. If data entry errors are discovered, attempt to correct and plot TAF
profile again.
19
3-01-001-03
Adipic Acid: Cyclohexane
Oxidation
New TAF is flat, may be based on TNRCC normal operating data (which would not show
diurnal patterns) or may suggest continuous operation. MIP profile is flat. SCC is most
likely a continuous process. Suggested action: Accept new TAF profile.
(Continued)
-------
TABLE 8-4.
PRIORITY CO SCCs REVIEW (Continued)
00
i
to
Rank
see
SCC Description
Comment
20
3-03-009-07
Primary Metals: Iron & Steel:
Steel Furnaces: Tapping - Electric
Arc
New TAF is flat, may be based on TNRCC normal operating data (which would not show
diurnal patterns) or may suggest continuous operation. MIP profile suggests that data
from severed plants with different operating schedules were average to produce this
profile. Suggested action: Discuss using mode rather than mean with EPA personnel or
accept new TAF.
21
3-07-001-10
Pulp & Paper: Kraft Pulping:
Recovery Furnace, Indirect Contact
Evaporator
New TAF is flat, may be based on TNRCC normal operating data (which would not show
diurnal patterns) or may suggest continuous operation. MIP profile is flat. SCC is most
likely a continuous process. Suggested action: Accept new TAF profile.
22
3-03-009-08
Steel: Electric Arc Furnace:
Carbon Steel
New TAF is flat, may be based on TNRCC normal operating data (which would not show
diurnal patterns) or may suggest continuous operation. MIP profile suggests that data
from several plants with different operating schedules were average to produce this
profile. Suggested action: Discuss using mode rather than mean with EPA personnel or
accept new TAF.
23
3-01-003-05
NH, Production: Feedstock
Desulfurization
No new data for new TAF. MIP profile suggests that data from several plants with
different operating schedules were averaged to produce this profile. Suggested action:
Discuss using mode rather than mean with EPA personnel or use MIP profile as is.
24
3-03-001-07
Aluminum Ore: Electric
Reduction: Roof Vent
New TAF is flat, may be based on TNRCC normal operating data (which would not show
diurnal patterns) or may suggest continuous operation. MIP profile is Hat. SCC is most
likely a continuous process. Suggested action: Accept new TAF profile.
25
3-90-007-02
Process Gas: Coke Oven Gas
No new data for new TAF. MIP profile suggests that data from several plants with
different operating schedules were averaged to produce this profile. Suggested action:
Discuss using mode rather than mean with EPA personnel or use MIP profile as is.
-------
8.4,1 Approach I: Measure of Dispersion Using Standard Deviations
The standard deviation is a descriptive statistic that measures scatter in a distribution from
its mean. The measure of dispersion based on the standard deviation principle is often used to
compare two groups of data (i.e., experimental and control) to determine significant differences.
For the purposes of this analysis, the control group would be the SCC-family aggregate profile,
and the experimental groups would be each of the individual eight-digit SCC profiles used in the
creation of the SCC-family aggregate profile. First, the mean of each hourly factor would be
calculated within the SCC-family. Second, the standard deviation of each hourly factor would
be calculated. Third, the standard deviation of each hourly factor would be compared to the
hourly factors of eight-digit SCC profiles that are in the respective SCC-family. Fourth, each
hourly factor from an eight-digit SCC profile would be compared to hourly factors from the
SCC-family aggregate profile, and the number of standard deviations necessary to envelop the
eight-digit SCC hourly factor would be calculated. An overall profile rating would be calculated
as follows;
24 U. - fl.1
Profile Rating - ^ L
\ i m I O-
where:
i is an indicator of the hours of a day
Ai is the mean of the hourly factor from the SCC-family profile
Bi is the hourly factor from the eight-digit SCC profile
CTj is the standard deviation from the mean of the SCC-family profile
This method produces a value which could be used as a profile quality rating for how well
an eight-digit SCC profile conforms to the SCC-family profile. However, it should be recognized
that these ratings are only indicative of how well one eight-digit SCC profile reflects the SCC-
family profile in relationship to another. In other words, these ratings are not benchmarked by
actual temporal data for each SCC-family. Therefore, this rating system would not only identify
CH-94-35 8-13
-------
outliers within SCC-families but would actually indicate where the profile is not indicative of the
process.
It should be noted that one hourly factor outlier could cause an entire SCO profile to fall
as much as four or five standard deviations from the family profile although all remaining 23
hourly factors may remain within one standard deviation. As a result, the unit of analysis should
be individual hourly factors and not an entire SCC profile.
A problem with any approach created for this analysis which includes the use of standard
deviations is that standard deviations have not been calculated at any level during the aggregation
of data from the individual data source to date-source-specific profiles, or from data source-
specific profiles to national default profiles. Without having the standard deviations calculated
during the process of creating these national defaults, the error in the standard deviation can not
be correctly propagated for the SCC-family aggregate profiles. Therefore, it would be inaccurate
to calculate and use a standard deviation calculated from the consolidation of national defaults
into SCC-family aggregate profiles. Without retracing all of the steps involved in the creation
of the national default profiles, the standard deviations calculated by this method would produce
fraudulent results.
8.4,2 Approach II: Measure of Dispersion Using Absolute Differences
Approach II, although not a standardized statistical method, will supply an accurate
representation of the quality of TAFs under analysis. This approach is similar to Approach I in
that eight-digit SCC profiles will be compared to SCC-family aggregate profiles on an hour-by-
hour basis. The key difference between Approach II and Approach I is that absolute differences
would be used as profile ratings. Approach II would simply calculate the absolute difference for
each hour between the eight-digit SCC profile and the SCC-family aggregate profile. The
absolute value of those differences would be summed to obtain a profile rating. An overall rating
for a profile would be calculated as follows:
CH-94-35
8-14
-------
24
Profile Rating = |Aj - B,\
i = 1
where:
i is an indicator of the hours of a day
Aj is the hourly factor from the SCC-family profile
Bj is the hourly factor from the eight-digit level SCC profile
The advantages of Approach II are its simplicity and its ability to furnish profile-quality
ratings without duplicating previously performed work. One disadvantage to Approach II is that
it is not a standardized statistical method, and therefore, has not been documented and/or tested.
It should also be noted that, as with Approach I, Approach II is only an indicator of agreement
to the SCC-family profile, and not benchmarked with actual data. Although this could be viewed
as a problem if these values are to be permanently attached to the profiles as quality ratings, or
quality ratings derived from them, the values will prove useful in evaluating where to direct
future efforts.
While Approach II is the recommended approach for the analysis, neither statistical
analysis was performed due to limited funds remaining in this Work Assignment. However,
since the current profiles do not lend themselves to the calculation of standard deviations, it is
felt that Approach II would prove to be more useful in applying quality ratings to the TAFs.
CH-94-35
8-15
-------
CHAPTER 9.0
CONCLUSIONS AND RECOMMENDATIONS
9.1 CONCLUSIONS
Three technical tasks were completed during this Work Assignment: (1) the prioritized list of
source categories used to identify TAFs for review was refined; (2) temporal allocation profiles for
high-priority source categories were reviewed; and (3) an analysis plan for reviewing the complete
TAF file was developed.
Inconsistencies and unexpected source categories found in the prioritized SCC list developed
under the previous TAF work assignment were generally attributed to incorrectly coded emissions
sources. The presence of emissions from unexpected source categories in AIRS [for example, iron
production - sintering - raw material transfer/handling (SCC 3-03-008-12), a process that does not
include fuel combustion, was ranked fourth for CO emissions per reported source] is a principal
reason why AIRS is not completely reliable for providing data for prioritizing emissions sources.
However, AIRS is EPA's best source of emissions data. After the list was reviewed and the suspect
SCCs removed, the resulting prioritized list should provide a useful tool for focusing efforts on this
and other projects.
Review of graphical representations of high-priority SCC profiles identified unexpected
profiles for many SCCs for electric utilities. Specifically, the bimodal profile expected for electric
utilities which corresponds to peak electricity consumption during morning and evening was absent.
This problem was traced back to the CEM data that were the basis for most electric utility profiles.
After determining that the CEM data were inappropriate for calculating allocation factors, the TAF
file was regenerated without using CEM data. The current electric utility profiles are still based on
real data, and should prove more representative of actual electric generation activity.
The development of the analysis plan for reviewing the complete TAF file focused on
selecting an appropriate, reasonable statistical analytical tool. Reviewing data using variance or
standard deviation is the most common method to identify outlying data points; however, this
approach was discarded because of the unequal error inherent in the data comprising the TAF file.
The approach recommended is based on comparison of TAF profiles to benchmark profiles. This
9-1
-------
may be accomplished by assigning SCCs to groups, then comparing individual SCC profiles to the
group averages. Deviations from group averages will be used as an analytical tool.
9.2 RECOMMENDATIONS
Based on the results of the TAF development work to date, the following activities may be
considered to continue developing high quality temporal allocation profiles.
Implement statistical analysis: Execute the quantitative analysis approach developed under
Work Assignment No. 1/014. The benefits of this activity would include identifying source
categories with the most questionable allocation profiles, and assigning confidence factors to
the profiles in the TAF file.
Develop temporal profiles for area and mobile sources: Little effort has been dedicated to
evaluating and revising allocation factors for these important source categories. Although the
TAF file contains temporal profiles for area and mobile sources, these profiles are based
almost entirely on previous NAPAP profiles. For this project few data sources were found
that could be used to improve the temporal profiles for point and area sources. Only profiles
for auto body painting and printing operations were updated, based on the results of survey
data collected for the Atlanta area. As a result, nearly all of the improvements made to
temporal profiles under this project were for point source categories. To improve temporal
profiles for area source data, appropriate data must be obtained. The only approach for this
may be to conduct surveys for the most important categories. This approach was beyond the
scope of this project. The survey approach would require substantial resources to design the
survey, obtain Office of Management and Budget (OMB) clearance, and to perform the actual
surveys.
Incorporate remaining LMOS data: During the data collection effort, the LMOS database
included Wisconsin data only. The current TAF file reflects that shortcoming. The final
LMOS database released in Fall 1993 includes Wisconsin, Illinois, Indiana, and Michigan
data. Incorporating the final LMOS database into the TAF file would broaden the scope and
reduce the regionality of the TAF file.
Investigate additional data sources: Several additional data sources were suggested at the
March 29, 1994 meeting. These include: northeast utility data from Sigma Research
(currently in draft form); data collected and used by states running UAM for their modeling
efforts (probably available after November 1994); and Tennessee Valley Authority (TVA)
facility data (probably available 1995 or later).
Develop regional TAF profiles: The large databases used during TAF development
represent distinct geographic regions. EPA may want to investigate using these databases to
9-2
-------
develop regional default TAF profiles. Specifically, LMOS would represent midwest
activity, SOS would represent the southeast, TRNCC would represent the southwest, and
CARB would represent the west. A current lack of northeast data is a significant obstacle to
this effort. This task would be preceded by a feasibility study to determine if the databases
listed provide sufficient data for TAF development.
Incorporate National Allowance Data Base (NADB) data: The NADB maintained by
EPA's Office of Atmospheric and Indoor Air Pollution (OAIAP) is a collection of detailed
utility emissions data under Title IV of the Clean Air Act. Although no data were available
for the initial data collection effort, more data may be available in the future. The possibility
of using these data in 1995 and beyond may be studied to determine whether this emissions
database would be useful for developing temporal operating profiles.
Expand visual analysis: During this analysis, performing quality assurance using temporal
profiles in a graphical format proved to be a cost-effective method of identifying problems
with high-priority source categories. This effort was constrained to fifty SCCs. Expanding
this list to include the top 100 categories may identify other problems with important source
categories.
Develop a PC-based TAF file viewer: This software would allow individuals using the
TAF profiles to view the profile graphically.
Develop access software: This software would allow access from UAM, FREDS, GEMAP,
and other leading models to the TAF file.
9-3
-------
10. REFERENCES
1. TRC Environmental Corporation. Summary of Identified Literature References Relevant to
Methodologies and Data Used for Temporal Allocation of Emissions Data. Technical
Memorandum under EPA Contract No. 68-D9-0173, Work Assignment No. 3/314,
U.S. Environmental Protection Agency, Research Triangle Park, NC, May 27,1993.
2. TRC Environmental Corporation. Results of Telephone Interviews Conducted for the
Identification of Allocation Methodologies and Available Data, Technical Memorandum
under EPA Contract No. 68-D9-0173, Work Assignment No. 3/314, U.S. Environmental
Protection Agency, Research Triangle Park, NC, May 27,1993.
3. Walters, R.A., L.G. Modica, and D.B. Fratt. The 1985 NAPAP Emissions Inventory:
Overview of Allocation Factors, EPA-60Q/7-89-010a (NTIS PB90-126012), U.S.
Environmental Protection Agency, Research Triangle Park, NC, October 1989.
4. Fratt, D.B., D.F. Mudgett, and R.A. Walters. The 1985 NAPAP Emissions Inventory:
Development of Temporal Allocation Factors, EPA-600/7-89-010d (NTIS PB90-237181),
U.S. Environmental Protection Agency, Research Triangle Park, NC, April 1990.
5. U.S. Environmental Protection Agency. Northeast Corridor Regional Modeling Project
Annual Emission Inventory Compilation and Formatting, EPA-450/4-82-013a-r, Research
Triangle Park, NC, 1982.
6. Bureau of Economic Analysis. Business Statistics 1963-91, U.S. Department of Commerce,
Washington, DC, June 1992.
7. Bureau of Labor Statistics. Employment and Earnings, U.S. Department of Labor,
Washington, DC, August 1991.
8. Southwest Research Institute. Statistical Reference Index, 1991 Annual, San Antonio, TX,
1992.
9. Commodity Research Bureau. 1991 CRB Commodity Year Book, New York, NY, 1991.
10. U.S. Environmental Protection Agency. AIRS Facility Subsystem Source Classification
Codes and Emission Factor Listing For Criteria Air Pollutants, EPA-450/4-90-003 (NTIS
PB90-207242), Research Triangle Park, NC, March 1990.
11. Bercnyi, E. and R. Gould. 1993-94 Resource Recovery Yearbook, Directory and Guide,
Governmental Advisory Associates, Inc., New York, NY. 1993.
10-1
-------
12. Telecon. Buschow, Ritchie, TRC Environmental Corporation, Chapel Hill, NC, with Robert
Gould, Governmental Advisory Associates, Inc., New York, NY. Waste-to-Energy Source
Data, September 3,1993.
13. Causley, M., G. Wilson, M. Jimenez, L.A. Gardner, and A. Noda. User's Guide for the
Urban Airshed Model, Volume IV, Part A, Emissions Preprocessor System {Version 2):
Core FORTRAN System and Part B, Emissions Preprocessor System (Version 2): Interface
and Emissions Display System, EPA-450 /4-90-007d (NTIS PB93-122380), U.S.
Environmental Protection Agency, Research Triangle Park, NC, June 1992.
14. TRC Environmental Corporation. Improvement of Temporal Allocation Factor Files Project.
Technical Memorandum under EPA Contract No. 68-D9-0173, Work Assignment No. 3/314,
U.S. Environmental Protection Agency, Research Triangle Park, NC, June 9, 1993.
15. TRC Environmental Corporation. Action Plan for the Development and Improvement of
Temporal Allocation Factor File(s). Technical Memorandum under EPA Contract
No. 68-D9-0173, Work Assignment No. 3/314, U.S. Environmental Protection Agency,
Research Triangle Park, NC, July 7, 1993
10-2
-------
APPENDIX A
MAY 27,1993, JUNE 9,1993, AND JULY 7,1993
MEMORANDA
Page
Memorandum, TRC (Moody and Winkler) to C. Mann, May 27,1993 A-2
Memorandum, TRC (Buschow and Moody) to C. Mann, May 27, 1993 A-95
Memorandum, TRC (Moody and Winkler) to C. Mann, June 9,1993 A-158
Technical Memorandum, TRC to C, Mann, July 7, 1993 A-168
A-l
-------
TRC
Environmental Solutions through Technology
MEMORANDUM
TO: Charles Mann
Air and Energy Engineering Research Laboratory, MD-62
FROM; Theresa Kemmer Moody and David Winkler
TRC Environmental Corporation
DATE: May 27, 1993
SUBJECT: Summary of Identified Literature References Relevant to Methodologies and Data
Used for Temporal Allocation of Emissions Data ;
EPA Contract No. 68-D9-0173, WA No. 3/314, Task 1
TRC Reference Number 1-637-314-3
TRC Environmental Corporation
100 Europa Drive, Suite 1-50
Chopel Hill NC 275 U
« (919) 968-9900 Fax (9191 968-7557
1.0 INTRODUCTION
This memorandum documents the results of TRC Environmental Corporation's (TRC's)
efforts to identify, acquire, and review current literature references relevant to methodologies and
data used for temporal allocation of emissions data. This is the first step in an effort to evaluate
the quality and completeness of data and methodologies presently being used for temporal
allocation of emissions data and to identify improvements to the current methods.
There are two attachments to this memorandum. Attachment A lists the articles and
reports reviewed for this task, and Attachment B consists of summaries of these articles and
reports. As indicated in Attachment A, only those items procured prior to May 21, 1993 will
have summaries in Attachment B. The summaries focus on information relevant to temporal
allocation.
. 2.0 LITERATURE SEARCH
In identifying current literature, TRC first acquired the following three specific documents
identified in the work plan.
The 1985 NAPAP Emissions Inventory: Development of Temporal Allocation Factors,
EPA-600/7-89-010d, April 1990
* Procedures for the Preparation of Emission Inventories for Carbon Monoxide and
Precursors of Ozone, Volume II: Emission Inventory Requirements for Photochemical Air
Quality Simulation Models, EPA-450/4-91-014, May 1991
CH-93-49
Offices in California, Colorado, Connecticut, Illinois, Louisiana, Massachusetts, New Jersey, New York, North Carolina, Pennsylvania, Texas,
Washington, Washington, D.C., and Puerto Rico A TRC Company
Printed on Rscydeii Paper A - 2
-------
User's Guide for the Urban Airshed Model, Volume TV: User's Manual for the Emissions
Preprocessor System 2.0, EPA-450/4-90-007D(R), June 1992.
The first of these documents is particularly important because some of the most comprehensive
and frequently used temporal allocation factors were developed in conjunction with the National
Acid Precipitation Assessment Program (NAPAF). The second and third documents describe the
temporal allocation procedures used in the Urban Airshed Model (UAM).
In addition to these documents, TRC conducted a key word literature search of on-line
databases accessed through-DIALOG to identify other current available information. The on-line
search accessed information printed in English from 1985 through 1993. This search used the
following key words;
* seasonality
• temporal allocation
* seasonal adjustment
temporal adjustment
seasonal allocation
~ seasonal emissions
emissions inventory
emissions allocation
* allocation factors
temporal allocation factors
hourly emissions data
short-term emissions
* average emissions
• peak emissions
• continuous emissions monitoring
TRC searched the following databases through, DIALOG: National Technical Information
System (NTIS), Engineering Information Compendex-Plus (EI), Meteorological and
Geoastrophysical Abstracts (MET/GEO ASTRO), Enviroline, and Pollution Abstracts.
TRC reviewed the list of documents, articles, and abstracts and selected 28 to read and
review in depth. The bibliography provided as Attachment A lists the documents reviewed.
Many of the documents identified in the literature search described various aspects of the NAPAP
program. How.ever, because of the magnitude of published reports pertaining to NAPAP, TRC
limited review to final report summaries and relied on in-house knowledge of NAPAP to provide
requested summaries and information for this task.
CH-93-49 2
A-3
-------
3.0 RESULTS OF THE .LITERATURE REVIEW
This section summarizes information on temporal allocation factors from the reports and
articles identified in the literature search, in addition to the three specific reports listed in Section
2.0. In summary, TRC's review of the literature showed that the most comprehensive set of
temporal allocation factors has been developed for NAPAP, although other work has focused on
some specific aspects of temporal allocation. In addition, significant work has been completed
for the UAM on developing a methodology for temporal allocation and a set of default factors.
NAPAP Allocation Factors Development
Temporal allocation factors were developed'for emissions of the following ten NAPAP
pollutants from the point and area source categories in the 1980 NAPAP data: sulfur dioxide
(S02), primary sulfate, oxides of nitrogen (NOJ, total suspended particulates (TSP), carbon
monoxide (CO), ammonia, hydrogen chloride, hydrogen fluoride, volatile organic compounds
(VOC), and total hydrocarbons (THC). Of these, NO,, TSP, and THC were further resolved into
component species or groups of species.
NAPAP temporal allocation factors were updated and applied to the 1985 data, as well
as the 1980 data. Four' seasonal, three daily factors per season (i.e., a typical weekday, and
Saturday and Sunday) and 24 hourly allocation factors were developed for NAPAP point and area
sources.
Data for the NAPAP factors came from a variety of sources, and several of the acquired
references described projects supporting the NAPAP factors development. The major data
sources for developing the NAPAP factors are described as follows:
1. Many factors for United States' sources came from the Northeast Corridor Regional
Modeling Project (NCRMP). Seasonal factors' for the temporal distribution of point
source emissions were originally developed on a fuel- and state-specific basis for facilities
in the NCRMP study using power generation statistics from the U.S. Department of
Energy.
2.' Temporal allocations for point source emissions were based on operating schedule
information included in the National Emissions Data System (NEDS)-based point source
data records. Because of the magnitude of electric utility emissions, process-specific
factors were developed for these sources.
3. Daily factors were developed at the national level from weekly load cycle listings in the
Electric Power Research' Institute (EPRI) Regional Systems report. Fuel- and state-
specific weekday hourly patterns for power generation facilities were developed during
the NCRMP effort. Profiles of hourly operation were derived from hourly power plant
CH-53-»9 3
A-4
-------
fuel use data collected during the development of the EPRI's Sulfate Regional Experiment
(SURE) inventory.
Factors were developed for the 102 area source categories (including mobile sources) in
the 1985 NAPAP area source data. Depending on the magnitude of emissions within the source
category and availability of data, factors were frequently resolved to the regional, state or local
level. As noted previously, point source factors were developed for electric utility processes.
A major limitation of the NAPAP factors is that they were developed only for the NAPAP
point and area sources. This includes 102 area source categories and only electric utility
processes for point sources. In addition, NAPAP factors focus primarily on criteria pollutants
that play a role in the formation of acid rain, but, in the future, temporal allocation factors for
hazardous air pollutants will be an important focus.
EPA Procedures for Preparation of Emission Inventories and the UAM User's Guide
The second document listed in the work plan was Procedures for the Preparation of
Emission Inventories for Carbon Monoxide and Precursors of Ozone, Volume II: Emission
Inventory Requirements for Photochemical Air Quality Simulation Models. This report is part
of a two-volume set that provides assistance in preparing and maintaining emissions inventories.
Volume II provides technical assistance for developing inventories of VOC, CO, and NO,. for use
in photochemical air quality simulation models, with special emphasis on the input requirements
of the UAM. Volume II describes an approach for temporal distribution of emissions from point,
area and mobile sources. Some of the methodology (e.g., area source methodology) is based on
the NAPAP work.
The User's Guide for the Urban Airshed Model, Volume IV: User's Manual for the
Emissions Preprocessor System 2.0 describes how to use the UAM features mentioned below.
Because both of these documents focus on the UAM, they are discussed together.
The UAM's Emission Preprocessor System (EPS) consists of six programs that generate
input files for the UAM. These programs can assign temporal distribution profiles (seasonal,
hourly, and daily) based on operating information (e.g., weeks per year of operation) contained
in the annual or seasonal inventory. The UAM EPS contains default values for sources not
explicitly reporting operating schedules. For example, the UAM EPS assigns a flat operating
profile (equal season fractions of annual throughput, 52 weeks per year, 7 days per week and 24
hours per day) to all sources not reporting operating schedules.
While this guidance document presents an approach for calculating temporal allocation
factors, it does not actually develop factors. The UAM develops source-specific allocation
factors based on input data on each source or uses default allocation factors where such data are
missing.
CH-93-49 4
A-5
-------
Other Sources of Information
Additional articles and reports offered some specific information that could be useful,
mainly in regard to temporal allocation of certain pollutants and/or specific sources. Many of
these references are described briefly below. Further detailed information is provided in the
summaries in Attachment B.
The Lake Michigan Ozone Study (LMOS) final report is scheduled for publication
shortly. Preliminary review of area source temporal allocations indicated that daily and
weekly allocations are based on data assigned by states in the LMOS or on default
profiles. Temporal profile documentation is found in Appendix D to the LMOS report.
The electronic database containing point and area source data is presently under revision
and should be forwarded to TRC around June 8, 1993,
A preliminary review of the FEDWORLD bulletin board system indicated that the only
potentially useful information was secondary information such as consumption and labor
statistics that could be used in factor generation.
* Research on the temporal variability of aerosol acidity and sulfate concentration indicated
that the acidity and composition of sulfate aerosol are highly variable in both time and
space. However, for a given time period, the relationship between LP and SOj" tends to
be more consistent at a specific site or region than it is among sites, especially when the
sites involve large cities. Since a high correlation between H+ and SO^" does not ensure
reliable predictions of acidity values based on measured sulfate, conclusions about the
effects of H+ should not be inferred from studies in which only SO}" was measured. (See
"On Spatial and Temporal Variability of Aerosol Acidity and Sulfate Concentration,"
Journal of the Air and Waste Management Association, April 1993)
A study of ozone episodes near the south-central California coast during the fall of 1985
documents seasonal and diurnal behavior of ozone concentrations and meteorological
conditions in the south-central California coast. No temporal allocation approaches were
used in this study because most of the diurnal data were from air monitoring stations
which recorded hourly ozone concentrations. (See Analysis of Historical Ozone Episodes
in the South-Central Coast Cooperative Aerometric Monitoring Program (SCCCAMP)
Region and Comparison with SCCCAMP 1985 Field Study Data, American
Meteorological Society, May 1991)
• A study of temporal distribution of atmospheric nitric acid (HN03) and particulate nitrate
concentrations in the Los Angeles area indicated that the highest nitrogen dioxide
concentrations in the study area occurred overnight and during the early morning hours.
Inorganic nitrate production almost always is in great excess of the amount of HN03 in
the atmosphere, with average fine-particle nitrate concentrations in winter often higher
than in summer. Coarse particulate nitrate concentrations display approximately the same
seasonal variations as HN03 concentrations, with a flat seasonal distribution near coastal
CH-93-49 5
A-6
-------
areas and a summer seasonal peak inland. (See "Spatial and Temporal Distribution of
Atmospheric Nitric Acid and Particulate Nitrate Concentrations in the Los Angeles Area,"
Environment, Science and Technology, Vol. 26, No. 8, 1992)
A modeling study of atmospheric emissions and atmospheric concentrations of VOCs
from the Exxon Valdez oil spill addressed temporal distribution of emissions of 15
specific compounds. The model predicted hourly concentrations. Emission estimates
were based on evaporation rates from a liquid spill which are controlled by mass transfer
from the surface of the spill or diffusion through the liquid layer. (See "Modeling VOC
Emissions and Air Concentrations from the Exxon Valdez Oil Spill," Journal of Air and
Waste Management, 43:298-309, 1993)
A study of temporal variations of 18 polycyclic aromatic hydrocarbons (PAH) and lead
emissions on the coast of Sweden, near Stockholm, documented PAH fluxes during the
winter-spring and the summer. (See "A Multi-Sediment-Trap Study on the Temporal and
Spatial Variability of Polycyclic Aromatic Hydrocarbons and Lead in an Anthropogenic
Influenced Archipelago," Environment, Science and Technology, vol. 22, no. 10, 1988.)
The temporal distributions of SOx, NOx, VOC, and CO emissions in eastern European
• countries were addressed in a study focusing on the construction of a database to be used
for model calculations of long-term ozone levels. The temporal distribution addressed
utilities and road transport, but not industry and solvent losses. The paper presented
month, day and hour temporal allocation factors. (See "Emissions of SO,, NOx, VOC, and
CO from Eastern European Countries", AWMA, 1991).
4.0 ADDITIONAL SOURCES OF INFORMATION
Based on the initial literature review, TRC is pursuing additional sources of information
on temporal allocation factors in order to determine where improvements can be made to support
atmospheric simulation models. Specifically, TRC is pursuing the following sources:
The United States Department of Commerce, Bureau of Labor Statistics will provide TRC
with data (on paper and on disk) on hours worked by production workers and physical
output from selected industries. TRC should have this information by June 1, 1993. TRC
has reviewed some similar data in published sources.
• The available California Toxic "Hot Spots" Assembly Bill 2588 Inventory Program test
reports will be reviewed for source operating schedules and. parameters. TRC has
identified and acquired about 17 test reports (via Radian Corporation, Research Triangle
Park) requiring review.
* The 1983-1984 Nationwide Personal Transportation Study developed by the Federal
Highway Administration's (FHA's) Office of Highway Information Management includes
tables listing the distribution of vehicle miles traveled (VMT) by seven daily time
CH-93-19 6
A-7
-------
increments. Distributions are also provided for weekday and weekend diurnal patterns.
TRC is currently contacting the FHA for updated information. Related information on
traffic volume in Virginia has also been located, and similar information is available from
metropolitan planning organizations throughout the country.
TRC is currently reviewing a document entitled Locomotive Emission Inventory:
Locomotive Emissions by County Study (January 1991), and its supplement, prepared by
the California Air Resources Board and supplied by the Office of Mobile Sources. The
study covers six air basins in California and includes hours of operation, travel miles,
average daily emissions, emission factors, and seasonal and weekday activity variations.
The. Manufacturing Energy Consumption Survey Report, published by the Energy
Information Administration, lists electricity, fuel oil, natural .gas, coal, and other annual
consumption statistics by Standard Industrial Classification (SIC) code. The document
refers to a Form EIA-810, Monthly Refinery Report which may provide greater temporal
detail for the project. .Additionally, the Department of Energy's electronic publication
bulletin board system contains monthly and weekly fuel production data that TRC are
currently reviewing.
The Southern Oxidant Study (SOS), conducted through the Georgia Institute of
Technology and the State of Georgia, generated two databases of seasonally and
temporally allocated data. One subset of the database containing data from 36 different
highway locations is scheduled for delivery to TRC on June 1, 1993. TRC has been
unsuccessful in acquiring the point source database from the SOS study. However, TRC
has made additional contacts and located the database, and is optimistic that the point
source database will be acquired by June 4, 1993 from the Georgia Department of Natural
Resources, Environmental Protection Division.
The Southeast Michigan Council of Governments has published a survey of regional
traffic volume patterns by hour of day for southeastern Michigan. This may be useful in
temporal allocation of mobile source emissions.
TRC is investigating the use of operating data from the Aerometric Information Retrieval
System (AIRS) to improve the default temporal allocation values in the UAM. The
preprocessor for the UAM, the Emission Processing System (EPS), contains default
temporal allocation files to perform temporal allocation of annual emissions. However,
AIRS operating data contained in the AIRS Facility Subsystem (AFS) and the Area and
Mobile Subsystem (AMS) may be used to improve the EPA default temporal allocation
values by a data sequence outlined in the UAM User's Manual, Volume IV. The AIRS
system allows users to create workfiles of data extracts for direct input into EPS. These
workfiles contain operating or emissions data at the Source Classification Code (SCC)
level. Although these workfiles could be directly uploaded into EPS for processing, these
workfiles could also be downloaded to personal computers and statistically analyzed to
CH-93-»9
A-8
-------
assess data quality and completeness. Therefore, no special programming is required to
perform these data extracts.
The Texas Air Control Board has an electronic database containing information on
operating schedules, enforcement, emission permits, emission inventory, inspections, and
new source review which TRC is reviewing to determine if some of the information will
be useful.
The Clean Air Act Amendments of 1990 required more extensive use of continuous
emission monitoring (CEM) than sources have been required to do in the past. Over the
next 5 to 10 years, this CEM data could provide information for improvement of temporal
allocation factors,
TRC has requested Geocoded Emissions Modeling and Projections (GEMAP) systems
information from Ron Dickson at Radian Corporation, Sacramento, California. The
requested information provided will allow TRC to compile the newly generated data into
a format suitable for input to the GEMAP system.
CH-93-19
8
A-9
-------
ATTACHMENT A
BIBLIOGRAPHY
Adams, D.F. et. al. Biogenic Sulfur Emissions in the SURE Region. EA-1516. Washington
State University for Electric Power Research Institute. Palo Alto, CA. September 1980.
Aircraft and Lake Vessel Special Measurements: The Lake Michigan Ozone Study 1991 Summer
Field Program, Volume 11, North American Weather Consultants for Lake Michigan Air
Directors Consortium. December 1991.*
Appendix D: Area Source Spatial, Temporal and Speciation Profile Assignment Documentation;
Draft Final Report for Development of Modeling Inventories Under the Lake Michigan Ozone
Study (LMOS). RCN 264-144-09-00/DCN 93-264-114-04, April 26, 1993.
Base Year On-Road Mobile Source Emissions Inventory for the Ozone State Implementation Plan
for the Southeast Michigan Nonattainment Area, Southeast Michigan Council of Governments.
February 1993.*-
Bowman, D. et. al. Combustion Area Sources: Data Sources. EPA-60C/R-92-197. TRC
Environmental Corporation under EPA Contract No. 68-D9-0173. U.S. Environmental Protection
Agency, Research Triangle Park, NC. October 1992.
Broman, D. et. al. "A Multi-Sediment-Trap Study on the Temporal and Spatial Variability of
Polycyclic Aromatic Hydrocarbons and Lead in an Anthropogenic Influenced Archipelago."
Environment, Science and Technology, Vol. 22, No. 10, 1988.
Buckman, F. et. al. Evaluation of Continuous Emission Monitoring Data from Municipal Sold
Waste Incineration. Air and Waste Management Association. 82nd Annual Meeting and
Exhibition. Analheim, CA. June 25-30, 1989.
Business Statistics, 1961-88. U.S. Department of Commerce, Bureau of Economic Analysis, 26th
Edition.*
Employment and Earnings. U.S. Department of Labor. Bureau of Labor Statistics. December
1992.
Episodic Emissions Data Summary - Final Report. EPA-450/3-87-016, U.S. Environmental
Protection Agency, Office of Air Quality Planning and Standards. Research Triangle Park, NC.
June 1987,
Fairley, D. "Photochemical Model Bias: Is It Real or Is It a Statistical Artifact?" Air and
Waste, Vol 43. March 1993.
CH-9!W
-------
Flexible Regional Emissions Data System (FREDS) for the 1985 NAPAP Emissions Inventory.
EPA-600/9-89-047.
Fratt, D. et. al. The 1985 NAPAP Emissions Inventory: Development of Temporal Allocation
Factors, Final Report. Alliance Technologies Corporation under EPA Contract No. 68-02-4274,
U.S. Environmental Protection Agency. Research Triangle Park, NC.
Gardner, L. et. al. Procedures for the Preparation of Emission Inventories for Carbon Monoxide
and Precursors of Ozone, Volume II: Emission Inventory Requirements for Photochemical Air
Quality Simulation Models. EPA-450/4-91-014. Systems Applications International under EPA
Contract No. 68-DO-0124. U.S. Environmental Protection Agency. Research Triangle Park, NC.
May 1991.
Hanna, S.R. and P.J. Drivas. "Modeling VOC Emissions and Air Concentrations from the Exxon
Valdez Oil Spill." Journal of Air Waste Management Association, Vol. 43. March 1993.
Heisler, S.L. et. al. Interim Emissions Inventory for Regional Air Quality Studies. EA-6070.
ENSR Consulting and Engineering for Electric Power Research Institute. Palo Alto, CA,
November 1988.
Hilst, G. et. al. Time-Variable Air Pollutant Emission Strategies for Individual Power Plants.
EA-418. TRC - The Research Corporation of New England for Electric Power Research
Institute. Palo Alto, CA. April 1977.
Hodgson, A.T. et. al. "Emissions of Volatile Organic Compounds from New Carpets Measured
in a Large-Scale Environmental Chamber." Air & Waste, Vol. 43, March 1993.
Inventory of Organic Emissions from Fossil Fuel Combustion for Power Generation. E A-1394.
GCA Corporation for Electric Power Research Institute. Palo Alto, CA. April 1980.
Klemm, H.A. and R.J. Brennan. . Emission Inventory for the SURE Region. EA-1913. GCA
Corporation for Electric Power Research Institute. Palo Alto, CA. April 1981.
Lebowitz, L. and A.S. Ackerman. Flexible Regional Emissions Data System (FREDS)
Documentation for the 1980 NAPAP Emissions Inventory. EPA-600/7-87-025a. Alliance
Technologies Corporation under EPA Contract No. 68-02-3997. U.S. Environmental Protection
Agency. Research Triangle Park, NC. November 1987,
Lipfert, F.W. and R.E. Wyzga. "On the Spatial and Temporal Variability of Aerosol Acidity and
Sulfate Concentration. Air & Waste. Volume 43. No. 4. April 1993.
Modica, L.G. and D.R. Dulleba, The 1985 NAPAP Emissions Inventory: Development of Spatial
Allocation Factors. EPA-600/7-89-010b. Alliance Technologies Corporation for U.S.
Environmental Protection Agency. April 1990.
A-11
-------
Moe, J. A. Survey of Regional Traffic Volume Patterns in Southeast Michigan. September 1985.*
Ohio Trends Analyses Graphs of the Cirteria Pollutants (except PM10). Memo (with disks) from
Gary L. Engles, State of Ohio Environmental Protection Agency to Phil Marsosudixo, TRC
Environmental Corporation. May 14, 1993.
Pechan, E.H, An Air Emissions Analysis of Energy Projections for the Annual Report to
Congress. DOE/E1A-01G2/16. U.S. Department of Energy, Energy Information Administration.
September 6, 1978.
Ryan, R. et, al. User's Manual for the Hourly Emissions Database for the Acid-Modes Field
Study. Final Report. Alliance Technologies Corporation for U.S. Environmental Protection
Agency under Contract No. 68-09-0M3. Research Triangle Park, NC. October 1990.
Scire, J.S. and J. Chang. "Analysis of Historical Ozone Episodes in the SCCCAMP Region and
Comparison with SCCCAMP 1985 Field Study Data." Journal of Applied Meteorology.
American Meteorological Society. Volume 30. May 1991.
Sellars, F.M. et. al. National Acid Precipitation Assessment Program Emission Inventory
Allocation Factors. EPA-600/7-85-035. GCA Corporation under EPA Contract No. 68-02-3698.
U.S. Environmental Protection Agency. Research Triangle Park, NC. September 1985.
Solomon, P.A. et. al. . "Spatial and Temporal Distribution of Atmospheric Nitric Acid and
Particulate Nitrate Concentrations in the Los Angeles Area/' Environment, Science and
Technology, Vol. 26, No. 8, 1992.
Users Guide for the Urban Airshed Model, Volume IV: Users Manual for the Emissions
Preprocessor System 2.0. EPA-450/4-90-007D. U.S. Environmental Protection Agency.
Veldt, C. Emissions ofSOx, NOx, VOC and CO From East European Countries. Air and Waste
Management Association, 1991.
Wilkerson, G. SF6 Tracer Studies. The Lake Michigan Ozone Study 1991 Summer Field
Program. North American Weather Consultants, Salt Lake City, Utah, for Lake Michigan Air
Directors Consortium. December 1991.
* Received May 21, 1993: review ongoing.
CM-93-49
A-12
-------
ATTACHMENT B
SUMMARIES
-------
Adams, D.F, et. al. Biogenic Sulfur Emissions in the SURE Region. EA-1516. Washington
State University for Electric Power Research Institute. Palo Alto, CA. September 1980.
SUMMARY
This report presents the results of estimating biogenic (natural) sulfur emissions from the
northeastern United States defined as the EPRI SURE study area. Biogenic sulfur emission
estimates were derived from representative sampling of 21 sites in the SURE study area during
the fall of 1977 and the spring and summer of 1978. The objective of this study was to develop
a detailed emissions inventory of biogenic sulfur emissions. The results of this study indicated
that biogenic emissions are less than 1 % of the man-made sulfur emissions from utilities and
other industries in the SURE region. Diurnal effects on biogenic sulfur emissions are also
discussed in this report but only related to the meteorological conditions present during sample
collection (e.g, cloudy and cool, clear and warm, morning, night-time). Therefore, temporal
allocation information in this report is neither presented or discussed.
A-14
-------
Appendix D: Area Source Spatial, Temporal and Speciation Profile Assignment Documentation;
Draft Final Report for Development of Modeling Inventories Under the Lake Michigan Ozone
Study (LMOS). RCN 264-144-09-00/DCN 93-264-114-04, April 26, 1993,
These appendices document the various procedures and assumptions used to develop the
allocation factors for the LMOS area source categories. The appendix references the.
classification scheme and present codes for the daily and weekly allocation schemes.
CATEGORIZATION
It appears that the inventory utilized the AIRS AMS area source categorization scheme
(10-digit). These codes are standard and are therefore already in the format necessary for any
further analyses,
DAILY TEMPORAL ALLOCATION
The "Area Source Model" used by Radian operates with information supplied by each
State in the LMOS or assigns default profiles. The "codes" themselves are those used by
California ARB in their Emission Data System (EDS). Radian states that the assignment of
codes by the model are subjective. (Radian makes no mention as to how the States made profile
assignments, however.) I have attached the code list for reference.
WEEKLY TEMPORAL ALLOCATION
Weekly codes and assignments were developed in the same manner as daily allocations.
A list is attached for reference.
AREA SOURCE SPREADSHEET
By category by State assignments (codes) are included in a spreadsheet covering close to
200 10-digit SCCs. Unfortunately, little information on the derivation of the values is provided.
Apparently, Michigan conducted a survey on temporal factors for degreasers; Wisconsin provided
pleasure craft temporal data which seems to have been based on number of anrests by time of
day!
CONCLUSION
The document addressed only area sources and provided little information on the
derivation of the factors. The point source data may be more important and should be sought.
In both cases, more information on the derivation of the factors will be necessary'to properly
evaluate the data's utility. Area source codes may be compared with 1985 NAPAP data to
discover similarities and differences.
A-15
-------
Bowman, D. et. al. Combustion Area Sources: Data Sources. EPA-600/R-92-197. TRC
Environmental Corporation under EPA Contract No. 68-D9-0173. U.S. Environmental Protection
Agency. Research Triangle Park, NC. October 1992.
SUMMARY
The purpose of this report was to identify, document, and evaluate data sources for
stationary combustion area sources. Data sources were identified for three major categories of
area source emissions: fuel combustion, solid waste combustion, and other combustion (forest
fires and agricultural burning). For each source reviewed the following information was
identified and recorded: title, author, date published, frequency of update, document number, type
of data, timeliness of data, available data media, resolution of data, comprehensiveness and
accuracy of the data, and a brief description of the contents of the document.
Section 2 of this report describes the methodology used to identify data sources. Sections
3, 4, and 5 summarize fuel combustion, solid waste, and other combustion (forest fires and
agricultural burning), respectively. Section 6 is an evaluation of satellite data and potential
sources. In addition, Appendix A of this report contains information data sheets for each data
source, and Appendix B contains a list of relevant trade and professional associations.
Fuel Combustion
Eighteen area source fuel combustion data sources were identified. The following seven
sources of these eighteen sources identified contain the most useful information:
State Energy Data Report - data on energy consumption at the State and national levels
Annual Energy Review - energy data on the census and national levels
Gas Facts - natural gas consumption data On the national and State levels
¦ Gas Househeating Survey - fuel consumption and costs for natural gas househeating
Petroleum Supply Annual , 1989, Volumes I and II - data on petroleum products
consumption
Fuel Oil and Kerosene Sales - data on sales (consumption) of fuel oil and kerosene
products
Natural Gas Annual - data on regional, State, and national levels
A-16
-------
Solid Waste Combustion
Seventeen solid waste data sources and seven census data sources were identified. The
following sources of these seven census data sources contain the most useful information:
Census of Retail Trade - the quantity and types of retail establishments resolved to
metropolitan areas
Summary of Population and Housing Characteristics, 1991 - data resolved to county and
most current data
The County and City Data Book, 1988 - comprehensive data resolved to cities
Statistical Abstract of the U.S. - data on the social, political, and.economic organization
of the United States
Ten non-census data sources were reviewed. The following sources of these ten non-
census sources contain the most useful information:
Characterization of Municipal Solid Waste in the United States: 1990 Update - national
MSW data
The 1991 Resource Recovery Yearbook, Directory & Guide - all waste-to-energy projects
in the United States, broken down by region and State
Municipal Waste Combustion (MWC) Study - MWC characterization by technology and
emission control systems
A Comprehensive Report on the Status of Municipal Waste Combustion - MSW annual
status update
Other (Forest Fires and Agricultural Burning) Combustion
Thirty-two sources for the other combustion data sources were identified. The following
sources of these thirty-two sources contain the most useful data:
Wildfire Statistics - state and geographic area data for state and privately owned lands
Annual Fire Report - compilation of southeastern U.S. forest fire data"
Structural Fire Statistics - data on housing fires
Estimates of U.S. Biofuels Consumption - energy produced from wood and alcohol by
industrial sector and region
A-17
-------
Global Biomass Burning - compilation of satellite data on biomass burning
Local Climatological Data: Monthly Summary - climatological data from weather stations
around the United States
Climatic Averages and Extremes for U.S. Cities - annual compilation of monthly
climatological data
Keyguide to Information Sources in Remote Sensing - summary of institutions and
individuals involved with satellites and remote sensing
Remote Sensing Yearbook - annual update of satellite and satellite related industries
Remote Sensing
Remote sensing is the process of deriving information through systems not in direct
contact with the objects or phenomena of interest. Remote sensing can address a, variety of
spatial and temporal scales. When flight conditions and visibility permit, aerial photography can
record detailed information over a limited' area. The two primary remote sensing data providers
are the Earth Observation Satellite Company (EOS AT) and Systeme Probatoire de la Observation
de la Terre (SPOT).
Remote sensing systems can be classified as passive or active systems. Passive systems
read energy emitted by or reflected from other sources; active systems provide their own energy
source for reflectance and recording, similar to using, a flash attachment for snapshots.
Remote sensing systems can also be classified as photographic and non-photographic
systems. Photographic systems use cameras, and the photographs are made when electromagnetic
radiation passes through the lens, strikes the emulsion, and creates a latent image. Non-
photographic systems read other portions of the electromagnetic spectrum. Non-photographic
recordings of infrared, ultraviolet, and radio frequencies can be recorded with a radiation detector
or antenna system and a recording medium for the electronic signal.
The two methods of scanning a wide band of information on a single pass over a site are
whisk-broom and push-broom scanners. The whisk-broom scanner uses a rotating mirror to
reflect the series of individual pixels from across each row to a single sensor which encodes the
data for reading as a larger image. The push-broom scanner uses a row of sensors to
simultaneously detect the series of pixels across the band for each row of information.
Advantages of the push-broom scanner,are higher speed and the lack of moving parts.
Image Processing
Image processing describes the manipulation of the raw data yielded by remote sensing
systems. Computer interpretation of digital data can be advantageous. While the human eye can
only distinguish between sixteen or seventeen gray tones, an image recorded with 8-bit
A-18
-------
quantization will have 256 gray levels which can be analyzed using a computer. Comparison and
overlay of different images of the same space can model change over time.
A-19
-------
{
Broman, D. et. al. "A Multi-Sediment-Trap Study on the Temporal and Spatial Variability of
Polycyclic Aromatic Hydrocarbons and Lead in an Anthropogenic Influenced Archipelago."
Environment, Science and Technology, Vol. 22, No. 10, 1988.
{This summary contains applicable segments directly lifted from the original paper. This should
be remembered when referencing the given information).
This paper is discusses the spatial and temporal variations of 18 PAH compounds and lead
in terms of concentrations and fluxes, in the Stockholm archipelago (Baltic Proper) in general
and along the urban-influenced transect in particular. PAH profiles from the different sampling
areas and seasons are compared with one another, with profiles from other studies and with those
from different specific emission sources.
Eighteen PAH were quantified in 28 samples in order to estimate the PAH levels and
fluxes in deposited seston collected in sediment traps moored at 16 places in the archipelago of
Stockholm. The results from the analyses of the seston from the 10 sediment traps anchored
along the transect from the city of Stockholm toward the open Baltic Proper made it possible to
determine a PAH gradient. Both the concentrations and the fluxes exhibit the expected
logarithmic declines with increasing distance from the urban area, with a much more rapid
decline for the fluxes.
A steep decrease in concentration with distance from the source has also been found in
other studies, which show that most PAH axe rapidly incorporated and/or decomposed in the
sediments near the source. Furthermore, in this study the logarithmic decline is smoother and
less pronounced during the summer period compared with that of the winter-spring period. The
former period is characterized by a stable water regime with a pronounced thermocline and no
turnover, and a low runoff from land and from Lake Maleren (with a monthly mean waterflow
of 109 rrr/s for this period).
Concentrations of PAH for the winter-spring period are higher than those of the summer
period at each sampling station. A possible explanation of these higher concentrations during the
winter is that total PAH discharges axe greater than during the summer due to the larger amount
of fossil fuels (in this area mainly oils) used for household heating. Furthermore, PAH deposited
on land during the summer should not reach the aquatic environment to the same extent as during
the spring when the urban runoff and the runoff from the land is much more extensive and the
scavenging due to melting snow and washout of PAH is more efficient. A complimentary
explanation is that lower rates of microbial degradation and photo-oxidation of the PAH during
the winter will contribute to relatively higher PAH concentrations during this period compared
to the summer.
The PAH fluxes during the winter-spring are also higher than during the summer period,
especially for the stations near central Stockholm, whereas the PAH fluxes t the station- further
out aie approximately the same for the two sampling periods. This can be explained by the close
localization of the traps to the turbidity maximum area, i.e. the area of the outstreaming water
of Lake Malaren. This results in resuspension and flocculation (aggregation) of earlier deposited
PAH-rich bottom material.
A-20
-------
The concentrations of PAH for the three reference stations are lower than most of the
transect stations. Small variations in PAH concentration between the periods at this station and
the location of all the outskirts stations indicate that these levels reflect the constant background
and load of PAH, in the open coastal area of the Baltic Proper. One of the station located near
an oil refinery did not reflect an elevated PAH level. The effluent water from the sewage plant
of the refinery does not frequently pass the sampling station, but the main reason for the low
concentrations of PAH is probably the very good water exchange in the area.
The PAH fluxes show a somewhat different picture with relatively high values at the
reference stations compared to those at the outer stations of the transect. This is explained by
their hydrographically exposed position in the outskirts of the archipelago close to the open
Baltic, where they are subjected to resuspension from the surrounding shallow areas, especially
during storm events during the non stratified winter-spring period.
In Figure 4, a number of profiles of all analyzed PAH are presented from the inner (A-C),
middle (D-F), and outer (Gand K) sections -of the transect for the two periods, showing the
relative appearance of the 18 PAH. These profiles show that they all have a similarity save
perhaps the winter-spring profiles of first three stations (A-C) located nearest to Stockholm. In
addition, a correlation analysis of the profiles from the summer period shows that they all
correlate well to each other. During the winter period, differences between the profiles are
greater, and the only significant correlation coefficient is found between D-F and G and K'
stations. The reason for the poor correlation is the exposed position during snow unlike the outer
stations, directly subjected to runoff particulate PAH and also to the above-mentioned heavy
resuspension.
Analysis of all the above mentioned 18 PAH has been made in very few other studies,
so when comparing with PAH profiles from different emission sources from other studies, is
necessary to make comparisons between the most commonly analyzed PAH which are Flu, Pyr,
BghiP, BaA, Chr, BkF, BeP, BaP, Ind, BghiP, and Cor. This group of PAH is also considered
to be typically combustion-derived being normally found in the highest concentrations. These
PAH are also represented as the main PAH of the profiles of this study and this becomes" more
evident in the profiles from the outer stations (D-F and G and K).
Flu, Pyx, BghiP, and Cor often dominate the PAH profile of one of the dominating
sources in the area, namely gasoline automobile particle (filter condensed) exhausts. For diesel
automobile exhausts, the high particle concentration and the high adsorptivity of those particles
result in a proportionally high amount of low molecular weight PAH in the particle phase.
Therefore, Flu and Pyr often dominate the particle profiles from both light-heavy -duty diesels
automobile exhausts. It is assumed that the pattern is the same for emissions from other similar
diesel engines, e.g., those often found in leisure boats and ships.
Consequently, it might be possible to obtain an indication of the relative importance of
the emissions from the different engine types by calculating the ratios between Flu and Pyr
against BghiP and Cor. For gasoline automobile particle exhausts, the mean ratio is lower than
a factor of 2 and for light-duty diesels and for heavy-duty diesels the ratios are 12.3 and 9.7,
respectively. The seston ratios from this study vary between 4.9 and 1.1, the highest values
A-21
-------
being found at the stations close to the city. The mean ratio of the winter-spring period is 2.8
and during the summer period 3,6.
Another possible source previously described as important in urban waters is used
crankcase oils. Typical PAH profiles of used crankcase oils from gasoline and diesel engines
illustrate the relative similarity of these to the automobile particulate exhausts. Crankcase oils
from gasoline automobiles (in the diesel automobile samples Cor was not detected) show a Flu
sand Pyr against BghiP and Cor ratio of 2.8, which is also in the same order as in the seton
samples.
Other possibly important PAH emission sources are domestic heating (diesel oil) and
central heating boilers (diesel oil). Because of the low amount of available information, the
authors have assumed that these emit a PAH profile similar to that of heavy-duty diesels engines
on the basis of the fuel and burning conditions. These emissions should be of importance during
the winter-spring period, although this is not apparent form the seasonal seston variation of the
Flu and Pyr . against BghiP and Cor ratio. One reason to this could be that the higher traffic
intensity with large ships during the summer, all facilitated with heavy-duty diesel engines;
diminishes the importance of heating. The relatively high concentration of Chr in the emissions
from heavy-duty diesel engines is another indication of the impact from these-engines since Chr
is the third dominant PAH after Flu and Pyr in almost all seston samples. However, Chr has also
been found to be of importance in emissions from municipal incinerator plants. This might also
be valid for the domestic heating and central heating boilers if the similarity between them and
the heavy-duty diesels engines are correct.
The waters near urban areas are also directly subjected' to so-called road-wear particles,
which presumably constitute a contribution to the PAH load and the profile.Other researchers
have analyzed weathered asphalt taken from a road and quantitatively identified all the 11 above-
mentioned combustion derived PAH giving a PAH profile similar to urban sediment samples
taken in the same area. These are in turn similar to the seston sample PAH profiles of this study.
Another contributory emission source of PAH might be aircraft traffic, since one of the
starting and landing flight routes to the airport of Stockholm passes over the archipelago area.
From another study it has been shown that qualitative PAH profile from aircraft turbine
particulate emissions is also dominated by the 11 typical combustion-derived PAH from Flu to
Cor.
Comparisons of the seston PAH profiles are of course dependent on the PAH being
equally decomposed during their transport from the source to the environment, i.e. that the profile
remains constant. When comparing the PAH profile along the transect the authors did not find
a degradation pattern. The authors found that the similarity and good correlation between the
PAH profiles on the basis of all analyzed PAH, save the inner stations, especially during the
winter-spring period. This is confirmed when considering all the separate stations on the basis
of the 11 most commonly identified PAH, which also are the ones that are found in the highest
concentration in general. The PAH profiles from all the stations show surprisingly high
coefficients of correlation between the two seasonal periods and also between the different
stations for the same period. On the basis of the PAH emission source profiles, the distribution
A-22
-------
forms of PAH, the insusceptibility to degradation in both air and water, the unavailability of the
particulate PAH to dynamic exchange with the water, and the fact that all seston PAH profiles
are very alike, the authors concluded that aerosol soot particles (black carbon) from the complex
mixture of emission sources that constitute the significant PAH carriers and distributors in the
area. However, the stations in the urban area probably also receive directly runoff carried
particles from the hard surface of the city. For air-transported PAH, Beymer and Hites have
found that PAH adsorbed on fly ash and black carbon have much longer half-lives (especially
for black carbon) than when adsorbed to other substrates and that the half-lives become more
uniform for the different PAH. Butler and Crossley found that the photo-reactivity for PAH
adsorbed on soot (black carbon) to be of no great significance and that PAH present in urban
areas are primarily associated with aerosol soot particles in the submicron range.
Analyses of the lead concentrations in 27 seston samples show that the concentrations for
both periods have the same logarithmic decline as for PAH but also that the two periods are
much more similar. When the concentration of lead are corelated to the total PAH
concentrations, the linear correlation for both sampling occasions is very striking. A coefficient
of correlation of 0.996 (p < 0.001) for the summer period and 0.987 (p < 0.001) for the winter-
spring period is exhibited. Prahl et al suggested that a strong covariance between PAH and lead
indicates that automobile exhaust emissions are important sources of PAH. Even if a strong
correlation could be simply be a result of PAH and lead not being emitted only from the same
source but merely from the same areas, the authors believe that the theory that the automobiles
are highly responsible is valid for this area, since there is no other significant source of lead other
than from traffic. Further, the relative concentrations of total PAH to lead in the seston samples
is equivalent to the relative concentrations of total PAH to lead in particulate matter from the
street air in Stockholm. Also, the relationship between total PAH concentrations and total lead
concentrations of particulate automobile exhaust (leaded gasoline) emissions is in the same order
of magnitude.
Fluxes for the two seasons show approximately the .same pictures as the concentrations
but with somewhat higher values during the summer. Since the commercial boat traffic is made
up of diesel engine facilitated ships, emissions from leisure boats are the only source of lead that
differ during the seasons and might therefore be responsible for the somewhat higher summer
fluxes. The similar lead concentrations and the fluxes between the two seasonal periods and the
constant automobile traffic intensity throughout the year further support the earlier stated theory
that the PAH flux and concentration differences between the two periods are to a great deal
dependent upon the additional load from the domestic heating and central heating boilers during
the winter-spring.
A-23
-------
Buckman, F. et. al. Evaluation of Continuous Emission Monitoring Data from Municipal Sold
Waste Incineration. Air and Waste Management Association. 82nd Annual Meeting and
Exhibition. Analheim, CA. June 25-30, 1989.
(This summary contains applicable segments directly lifted from the original paper. This should
be remembered when referencing the given information).
Municipal solid waste incineration is a complex combustion process requiring specialized
understanding of the relationships among material input characteristics, process operational
controls, and discharges. Control agencies use various regulatory indicators to establish uniform
enforcement of permitted stack emission limits. Process rate standards are used in order to base
actual performance on throughput. Typical units arc pounds pollutant per ton of refuse burned.
Operating -stack gas analyzers which respond to the changing concentrations of various flue gas
components to measure ongoing compliance requires an evaluation, processing, and reporting of
data.
Before reaching a conclusion to confirm if the appearance of excessive emissions justifies
additional corrective action, reported data need to be examined and validated. The criteria which
must be followed to produce a quality decision requires a resource intensive protocol which
includes adherence to federal code and implementation of rigorous quality control checks needed
to utilize continuous emission monitoring (CEM) data. This information can only be used as an
indicator of the need to perform additional manual compliance testing if readings are in doubt
or flawed.
Establishing the equations describing combustion and heat calculations for incinerators
along the with stack concentration measurements provides the database needed to characterize
emissions relative to refuse feed rate and composition, thermal efficiency, and burning conditions.
The methods and formulae for describing the quantities of combustion air, refuse, residue,
moisture, fly ash involve the relationships between flue gas properties, composition, and steam
generated. To provide an ongoing analysis of the flue composition, calorific value, and feed rate,
software routines are needed to automatically calculate and identify periods,of excessive
emissions based on the actual amount of refuse burned.
Four methods were presented to determine ongoing process rates. The grapple weight
method is based on actual weight "measurement of crane bucket loads feeding the refuse chutes.
The ratio of steam production rate to baseline conditions were also used to calculate refuse feed
rate. By knowing the enthalpy added to the boiler, combustion efficiency, and heat content by
refuse, the amount of steam generated for designed conditions is determined. The third technique
used by the facility computer to convert ppm to lb/ton developed by Rust Engineering and is
based on rationing to the base case of full load on the boiler. The base case parameters are given
in the original paper.
EPA has proposed use of the fuel factor method (F) which allows for conversion of
continuous monitoring concentration data to process rate standards without additional monitoring
of stack gas velocity, temperature, fuel input rate, etc. Proper derivation of F factor requires an.
analysis of the exact chemical composition and heating value of the fuel. Obtaining this
A-24
-------
information by ultimate analysis is very resource intensive. Alternative procedures for calculating
the stack gas volumetric flow rate based on heat output (units of ftVlO^tu) were proposed to
consider changes in the measured composition of combustion products and quantity of steam
produced. In this manner, the resource recovery boiler represents a calorimeter, and information
from plant operation sensors may be utilized to provide data for solving heat and mass balance
equations. These relationships provide information for the determination of refuse feed rate, heat
content, and composition.
Combustion gas analysis is used to determine the percentages of the principal components
in the flue gas, namely; oxygen, nitrogen, water vapor and carbon dioxide. Traditional Orsat
analysis data is less reliable and use of continuous analyzers will cause deviations in calculating
the F factors. The trend to require continuous monitoring of combustion gases provides a
suitable data base if provision is made for both in-situ wet oxygen and extractive dry oxygen
readings. The rational for basic assumptions and formulas along with sample calculations for a
hypothetical municipal incinerator have been previously established.. Using combustion gas
analysis along with readings directly obtained from process sensitive sensor located at the boiler
economizer exit measuring temperature and % oxygen, the process steam generation rate, and
total combustion air flow, a- variety of combustion and heat calculations can be performed on a
ongoing basis to monitor actual changes from baseline calculations.
ENERGY AND'MASS BALANCES .
Flue Gas Composition
Obtain flue gas composition from process wet oxygen readings (%0;w) and perodic Orsat
analysis for percent dry oxygen (%02d) and carbon dioxide (%C02d). Determine percent dry
nitrogen (%N2d) and percent moisture (Bws) by calculation:
%Nm = 100 - (C02d + 0^)
B„ = (1 - % O J% OJ 100%
Combustion Air and Fuel Mix
The percentage of combustion air in excess [%EA] of the theoretical or stoichiometric air
required to burn all carbon [C] and available hydrogen [(H)] contained in the waste fuel along
with the ratio of carbon to available hydrogen are calculated using Orsat data;
%EA = 100 x [02 - (CO/2)]/[0.264 N2 - (02 - CO/2)]
C ; (H) ratio = CO^S.B - 0.421 (CO. + 0,)1
Stoichiometric Air Requirements
Stoichiometric air requirements (lb air/lb com) = [%C (11,53) + %(H)(34.34)]/100
A-25
-------
Combustible Mixture Analysis
Dry combustible matter found in MSW is derived mainly of either from cellulose or oil
bases. The average makeup is assumed and presented to be
For cellulose (Cel); HHVfapprox.) 7, 500 Btu/lb
For oil: HHV(approx.) 17, 000 Btu/lb
C/(H) = [44.4 (% Cel) + 77.4 (% Oil)]/[10 (% Oil)J
% Oil = 4440/ [10(C/(H) - 33]
% C = [44.4 (% Cel) + 77.4 (%Oil)]/100
% 02 = % H2G 16/18
% (H) = 10 (% Oil)/100
% H = 100 - (%C + %0)
% HzO = 100 - (%C + %(H))
GCV = 7, 500 (% Cel) + 17, 000 (% Oil)
F Factor Determination
From the ultimate analysis of combustible matter, the amount of gaseous products for each
type of fuel undergoing perfect combustion is known and related to heat input rate by fuel (F)
factors. Determine dry F factors from fuel composition by formulas for English Units (SCF/106
Btu):
Fd = 106 [ 3.64 x (%H) + 1.53 (%C) + 0.57 (%S) + 0.14 (%N) - 0.46 (%0)]/GCV
Stack Gas Volumetric Flow rate
Because of the difficulty of measuring the stack gas flow rate (QJ on a routine basis,
correlation with total air flow rate can be used. Data sets from the stack test reports along g with
the indicated values of total combustion air supplied from process readings of primary and
secondary air flows were used correlation coefficients. The values determined by using linear
regression should only vary slightly from actual flows reported by velocity transverses.
Heat Input rate
The stack gas volumetric flow rate (Q.) an F factor are used to find heat input rate (O^)
using oxygen readings for excess air correction.
For dry stack gas:
Qh (Units of 106 Btu/min) = (QJ/ {F, [20.9/(20.9 - fcO.J]}
The gross heat input rate (Qh) corresponds to the actual stack gas flow (Q.) divided by the
appropriate F factor stack flow term. The gross heat input rate is also related to the steam rate.
The enthalpy (H) added to the boiler is determined and if the process thermal combustion
efficiency (C>E) is known, the net heat needed for steam can be calculated. Facility computer
uses 1141.4 Btu/lb steam for this factor.
A-26
-------
Composition of Refuse Feed ,
In order to compare the F factor method for providing heat input rate with the steam
generation data, boiler efficiency calculations require ultimate analysis data. This can be done
by using a moisture balance to find the percentage of water in the MSW. The percentage of non-
combustibles must also be obtained and included in the ultimate analysis. , The feed rate of
combustibles in refuse [com] can-be calculated by dividing the heat input rate by the Btu content:
Com (units lb/min) = Qu/GCV
Values for Btu content, percentages of carbon, hydrogen, and oxygen for the refuse are
proportionally lowered by the percentages of combustibles in the total feed. Once this is
accomplished, an ultimate analysis is available to use with boiler efficiency.
Boiler Efficiency Report
The heat loss method (ASME, PTC) is used to compute % boiler efficiency (C>E>) by
using commercially available software for processing fuel analysis and operating conditions data.
Flue gas temperature is taken at the economizer exit. The net heat rate needed obtained from
steam production can be divided by the combustion efficiency to determine gross heat rate
' needed. The net heat available is determined from the gross heat available times the combustion
efficiency.
A-27
-------
Employment and Earnings. U.S. Department of Labor. Bureau of Labor Statistics. December
1992.,
SUMMARY
All information in this document was presented in a tabular format. A table of the
number of employees on nonfarm payrolls was given for all employees and for production
workers for October 1991, November 1991, September 1992, October 1992, and November 1992
for the following categories:
Total
Total Private
Mining
Construction
Manufacturing (durable and nondurable goods)
Transportation and Public Utilities
Wholesale Trade
Retail Trade
Finance, Insurance, and Real Estate
Services
Government
A table of average weekly hours and average overtime hours for production or
nonsupervisiory workers on private nonfarm payrolls was given for October 1991, November
1991, September 1992, October 1992, and November 1992 for the following industry categories:
Total Private
Mining
Construction
Manufacturing (durable and nondurable goods)
Transportation and Public Utilities
Wholesale Trade
Retail Trade
Finance, Insurance, and Real Estate
Services
A table of average weekly hours of production or nonsupervisory workers on private
nonfarm payrolls was given for November 1991, December 1991, and January through November
1992 for the following major industries and manufacturing groups, seasonally adjusted:
Total Private
Mining
Construction
Manufacturing (durable and nondurable goods)
A-28
-------
Transportation and Public Utilities
Wholesale Trade
Retail Trade
Finance, Insurance, and Real Estate
Services
Indexes of aggregate weekly hours of production or nonsupervisory workers on private
nonfarm payrolls were given for November 1991, December 1991, and January through
November 1992 for the following major industries and manufacturing groups, seasonally adjusted:
Total Private
Goods-producing
Mining
Construction
Manufacturing (durable and nondurable goods)
Service-producing
Transportation and Public Utilities
Wholesale Trade
Retail Trade
Finance, Insurance, and Real Estate
Services
Indexes of productivity, hourly compensation, unit costs, and prices, seasonally adjusted
were given on an annual average and quarterly index for 1990 and 1991 and on a quarterly index
for the first three quarters in 1992 for the following categories:
Business Sector
Nonfairn Business Sector
Manufacturing
Durable Goods
Nondurable Goods
Nonfinancial Corporations
A-29
-------
Episodic Emissions Data Summary - Final Report. EPA-450/3-87-016. U.S. Environmental
Protection Agency, Office of Air Quality Planning and Standards. Research Triangle Park, NC.
June 1987,
SUMMARY
This .document does not provide information directly pertinent to temporal allocation of
emissions. The document is primarily concerned with "episodic" (i.e., accidental or otherwise
irregular) releases rather than increases or decreases in emissions that follow a regular temporal
pattern. The document addresses emissions of nine pollutants considered potential air toxics, and
assesses the emissions of these pollutants from thirteen source categories.
One interesting idea presented in- this document is a review of events that are either
mutually exclusive or necessarily paired. For example, one facility might report that no more
than one specie would be produced on the facility at any given time; therefore, there could never
be simultaneous releases of more than one specie. As another example, a facility might report
that two compounds were always mixed in a particular process; therefore, any accidental releases
from that process would necessarily release both of the compounds. This episodic release
information is not readily useful to the temporal allocation project, but does introduce the
potentially relevant idea that we may be able to some events or "emission drivers" as mutually
exclusive or necessarily paired. Knowing which processes or events fall into these categories
may help us assess temporal allocation factors with "surrogate" data when directly applicable data
is not available.
DOCUMENT CONTENTS
The document addresses both episodic and background (i.e.,- "steady-state" emissions
expected during typical conditions) emissions of the nine potential air toxics from fourteen
industries. The nine pollutants are:
methylene chloride (MC),
~ carbon tetrachloride (CT),
• chloroform (CF),
perchloroethylene (PCE),
~ trichloroethylene (TCE),
• ethylene dichloride (EDC),
* butadiene (BD),
ethylene oxide (EO), and
epichlorohydrin (EPI).
A-30
-------
Emissions data was collected for major producers and users of each pollutant. The thirteen
source categories, along with the pollutants examined for each category, axe;
\
* ethylene dichloride production (EDC);
* ethylene oxide production (EO);
chlorinated hydrocarbon production (CT, MC, PCE, TCE, CF);
chlorinated hydrocarbon use {CT, MC, PCE, TCE, CF);
* butadiene production (BD);
polybutadiene production (BD);
* neoprene/chloroprene production (BD);
* styrene-butadiene rubber production (BD);
* miscellaneous butadiene users (BD);
* epichlorohydrin producers and users (EPI);
pesticides manufacture (CT, MC);
pharmaceutical manufacture (MC); and
chlorofluorocarbon production (PCE, CT, CF).
Types of episodic releases include mechanisms such as:
equipment openings,
process vent discharges,
* pressure relief discharges,
handling operations,
accidental liquid releases, and
* accidental gas releases.
The Emission Standards and Engineering Division (ESED) [now the Emission Standards
Division (ESD)] collected data from 1982 to 1987 under the authority of Section 114 of the
Clean Air Act EPA sent surveys to each facility requesting the following information for each
episodic event:
* type of event;
* emission source;
* nature of event;
number of events per year;
duration (largest event, if more than one per year);
* height of release;
* release diameter;
* discharge velocity;
discharge temperature;
heat capacity of released gas;
total pollutant emission (largest release, if more than one per year); and
emission rate (largest release, if more than one per year).
A-31
-------
Of the twelve items listed above, the first four items were used to identify the event and emission
source; the remaining items axe emissions parameters generally necessary to form dispersion
modeling.
In addition to the episodic data listed above, each facility provided background
information data that set "steady-state" or "baseline" emissions expected to be released during
normal operating conditions. Types of background release mechanisms include storage tank
emissions, fugitive emissions, and "secondary" emissions.
Lastly, each facility was asked to indicate types of episodic releases that could not happen
at,the same time, or releases that would always happen at the same time, For example, the
process flow at one facility might preclude the opening of more than one device at the same
time; therefore, no more than one episodic release due to equipment opening could occur at the
same time.
A-3 2
-------
Fairley, D. "Photochemical Model Bias: Is It Real or Is It a Statistical Artifact?" Air and
Waste, Vol 43, March 1993.
INTRODUCTION
The purpose of this paper is to explain why photochemical models seem to underpr edict
the highest ozone values. The paper includes a plot (Figure 1) of ozone values as predicted by
the Urban Airshed Model (UAM) for the San Francisco bay Area versus measured ozone values.
The figure shows an apparent tendency for the model to underpredict the highest ozone values,
A STATISTICAL ANALOGUE TO PHOTOCHEMICAL MODELS
The author interprets the phenomenon of underpredicting the highest ozone values purely
on a mathematical basis and by a statistical concept known as "regression toward the mean."
Since photochemical models utilize several kinds of smoothing or averaging techniques for both
inputs and outputs, the resulting predictions will always be more uniform, less variable, than the
measured actual concentrations which are affected by localized conditions at a particular monitor.
Thus, photochemical models, like statistical models, exhibit less variability than the monitored
values, and the highest monitored values will be higher than the highest predictions. Similarly,
the lowest monitored values will be lower than the lowest predictions.
POSSIBLE SOLUTIONS
By using appropriate statistical plots, such as plots of observed values versus predicted
values, or the residuals (defined as the differences- between observed and predicted values) versus
the predicted values, the modeler will be better able to differentiate between real bias and
statistical anomalies. If a model is appropriate (no bias in the model), the residuals should
fluctuate randomly around a horizontal line. If there is bias in the model for a particular range
of predicted values, the bias will show up as a preponderance of points either above or below the
horizontal line, with features of nonlinearity such as u-shaped or inverted u-shaped cluster of
points.
A-33
-------
Flexible Regional Emissions Data System (FREDS) for the 1985 NAPAP Emissions Inventory.
EPA-600/9-89-047.
The Flexible Regional Emissions Data System (FREDS) is a software system designed
to process emissions data for input to regional acid deposition and oxidant models. FREDS
extracts emissions data, modeling parameters (e.g., stack height, exhaust gas temperatures, etc.),
and source identification information from point and area source data records contained in
preprocessed SAS files and applies temporal, spatial, and species allocation factors to arrive at
a gridded, speciated, and temporally resolved emissions file. FREDS consists of seven modules
which reformat the data and apply allocation factors to the annual .emissions data. For this
project, only the Temporal Allocation Module (TAM) is of interest.
FREDS modules are written in the SAS language and are installed on the National
Computer Center's IBM 3090. FREDS allows user definition of up to 15 pollutants and the
temporal scenario (e.g., winter weekday) to be processed.
TAM
Annual point and area source emissions can be temporally resolved to hourly emissions
using TAM. Input to TAM includes annual point or area source emissions, a temporal allocation
factor file (TAFF), a timezone file (TZF), and a control options file (COF).
The TAFF is a SAS dataset which contains fractional multipliers representing the
seasonal, daily, and hourly percentages of emissions for each source category. A detailed
description of the development of the temporal allocation factors and the TAFF will be provided
with the review of the Temporal Allocation Factor report. A brief discussion of the temporal
factors and their use by TAM is provided below.
The TAFF contains allocation factors for up to 12 "typical" days (a weekday, Saturday,
and Sunday in each season). Each allocation factor record has one or more identifiers (e.g., SCC,
state, plant, point) which axe used to match emissions records to temporal allocation factors.
TAM match merges emissions records and allocation factors starting with the most specific level
of resolution (point for point sources; state for area sources) and deferring in steps to the least
specific (SCC).
TAM first reads and validates a COF which contains options for TAM processing, such
as the temporal scenario. The processing options are stored as SAS macro variables for use
during TAM execution. A detailed description of the TAM processing and macro subroutines
is provided in the FREDS documentation. A description of the temporal allocation process for
point and area sources is provided below.
For point sources, TAM first searches the emissions record for seasonal throughput
percentages; seasonal percentages from the TAFF are deferred to as a second choice. If the
TAFF has no profile for a given SCC, TAM will generate a temporal pattern based on operating
schedule data contained in the emissions record. If the operating schedule data are missing or
incomplete, a uniform pattern is generated as the default option. For area sources, the TAM code
A-34
-------
first searches the TAFF for a temporal profile. If no temporal pattern is contained in the TAFF,
a uniform pattern is generated as the default.
For point source processing, the temporal factors are merged with emissions records,
offset to GMT based on the temporal scenario and information contained in the TZF. The data
are output for further processing in FREDS. For area sources, emissions records are merged with
temporal fractions and offset to GMT using the TZF and based on the temporal scenario
processed, A Fortran subroutine (MTPREP.FORT) then multiplies the area source emissions by
their temporal factors, producing gridded or county level hourly emissions. For both point and
area sources, a diagnostic report summarizing TAM processing is generated.
PERIPHERAL SOFTWARE
Peripheral software to support the temporal factors includes the Temporal Allocation
Factor Preprocessor (TAFP). TAFP reads the temporal allocation factors, stored in 3 to 4 column
fixed fields, from EBCDIC dataset and stores them in a SAS dataset as 8 byte decimal fractions.
The storage of the temporal factors as 8 -byte SAS variables results in increased accuracy of the
factors as well as reduced processing time of TAM. (Note that the storage of the temporal
factors in fixed fields resulted in emissions losses due to rounding errors; i.e., when reaggregated
the fractions did not sum to 1.0.)
TAFP is also used to ensure the validity of the temporal factors prior to input to TAM.
TAFP performs five checks on the temporal factors:
The presence and proper sequence of scenarios 1 through 12 for each SCC/state/county
temporal allocation factor group;
The urn of the hourly fractions over a day is equal to 1.0;
The weighted sum of the daily factors over a season is 1.0: (65 x weekday factor) + (13
x Saturday factor) + (13 x Sunday factor) = 1.0;
The sum of the seasonal factors over a year is 1.0;
• Seasonal factors do not vary over the three days within one season.
After performing the above checks, TAFP formats point and area source temporal
allocation factors for input to TAM. TAFP processes point and area source temporal factors due
to the differences in TAM processing requirements.
A-35
-------
Fratt, D. et. al. The 1985 NAPAP Emissions Inventory: Development of Temporal Allocation
Factors, Final Report. Alliance Technologies Corporation under EPA Contract No. 68-02-4274.
U.S. Environmental Protection Agency. Research Triangle Park, NC.
OBJECTIVE
This report documents the developments and processing of temporal allocation factors
Version 2 of the 1985 NAPAP inventory. The allocation factors described herein are based upon
those developed and applied in the 1980 NAPAP emission inventory development effort. These
factors and the methodologies used to apply them to emissions data were refined and modified
to meet the expanded needs of the 1985 inventory. The report describes both the data sources
and methods used to develop temporal profiles and the data processing techniques employed to
generate temporal allocation factor files. The discussion emphasizes temporal factor development
for the U.S. anthropogenic emission data bases. However, overviews of Canadian and natural
source temporal factors development are also included.
The current version of the NAPAP emission inventory is based on emissions and facility
data obtained from the NEDS, Environment Canada and other sources for the base year 1985.
The final 1985 NAPAP emission inventory includes emission from sulfur dioxide, primary
sulfate, oxides of nitrogen, total suspended particulates, carbon monoxide, ammonia, hydrogen
chloride, hydrogen fluoride, volatile organic compounds and total hydrocarbons. Three of these
pollutants (N'Ox, TSP and THC) are further resolved into component species or groups of species.
In addition all emission are spatially apportioned into grid cells approximately 20 x 20 km square
and temporally resolved to the hourly level. Allocation factors have been developed from a
combination of new data sources as well as methodologies used for the 1980 inventory. FREDS
has also been modified and enhanced to support the expanded requirements of the 1985 effort.
DEVELOPMENT OF THE 1985 TEMPORAL ALLOCATION FACTORS
Temporal Factor Overview
For modelling applications, annual emissions totals are resolved temporally into 24 hourly
totals for a typical weekday, Saturday and Sunday in each of the four seasons of the year. To
accomplish this task, seasonal, daily, and hourly allocation factors were developed and applied
to the NAPAP point and area source. The temporal allocation factors take the form of three sets
of fractional multipliers, applied to the NAPAP annual emissions records. These are four
seasonal factors, three daily factors per season and twenty-four hourly factors. Details of this can
be found in The 1985 NAPAP Emission Inventory: Development of Temporal Allocation
Factors.(Original document)
Temporal allocation factors were originally developed for the point and area source
categories represented in the 1980 NAPAP inventory. The factors reflected data from a variety
of sources, including previous modeling studies. Many of the factors for the sources in the
United States were based on data from the Northeast Corridor Regional Modeling Project
(NECRMP). Point source temporal allocation factors were developed for only a subset of the
U.S. point source categories. Specifically temporal profiles were derived for electric utility
A-36
-------
processes, as they are among the most significant emitters of acid rain precursors. In most cases,
however, the temporal allocation patterns of point source emissions were estimated using
operating schedule information provided as part of the NEDS annual emissions records.
This section describes the derivation of the temporal allocation factors used to create the
resolved 1985 NAPAP emissions inventory. Temporal allocation factors were developed for the
102 source categories in the 1985 NAPAP area source file, although only 97 area source
categories were included in the resolved 1985 NAPAP emission inventory. Depending on the
magnitude of emissions within the category and availability of-data, factors were frequently
resolved to the regional, state or county level. A complete list of area source categories used in
the 1985 NAPAP inventory, including the level of resolution for each category-specific temporal
pattern is provided in Table 2-1 of the original document.
U.S. Area Source Temporal Factors
Residential Fuel Combustion
The residential fuel categories account for emissions from the residential water heating,
space heating, and cooking. The temporal variability in emission was approximated using
heating degree days (reflective of the variability in average temperature) as a surrogate indicator.
To develop state-specific seasonal factors, monthly average heating degree days for each state
for 1980 were obtained from State, Regional, and National Monthly and Seasonal Heating
Degree Days weighted by Population. These monthly averages were calculated according to the
method described in the original document. The hourly variations in residential fuel were
developed with data from NOAA and this method is described in the original document.
Commercial!Institutional Fuel Combustion
Area source emission from fuel use by commercial and institutional sources consists of
emissions from all fuel burning stationary sources that are not included under residential sources,
industrial sources, power plants or commercial point sources. Seasonal allocation factors were
developed from the Procedures for the preparation of emission Inventories for Volatile Organic
Compounds, Volume II hereafter referred to as the "EPA Guidelines". Hourly patterns for
commercial/ institutional fuel for NECRMP states were taken from the NECRMP study data.
Outside the NECRMP area, hourly allocation factors were taken from the EPA Guidelines.
Industrial Fuel Combustion
These categories include combustion emissions from the industrial sector which are not
accounted for by point sources, national seasonal patterns were developed from the EPA
Guidelines, in which a uniform distribution is recommended. The daily patterns was based on
U.S. Bureau of Labor Statistics on average overtime at manufacturing facilities. The hourly
patterns was taken from the NECRMP.
A-37
-------
Onsite Incineration and Open Burning
For the purposes of determining quantities of solid waste generated, onsite incineration
is defined as disposal in a small incinerator. Using this definition, incineration encompasses the
following types of disposal units: backyard burners, industrial incinerators.a nd incinerators used
by food and department stores, hospitals, and schools. For the purposes of estimated open
burning practices, open burning refers to uncombined burning of wastes such as leaves, landscape
refuse, and other rubbish. A single temporal profile set, developed in the NECRMP effort, was
assigned to these categories on a national level.
Highway Vehicles, Light and Medium Duty
Eight categories account for exhaust and tire wear emissions from automobiles and light
trucks on limited access, rural, suburban and urban roads. National level factors were developed
for each of these categories, using data obtained from the U.S. Department of Transportation.
This agency collects continuous traffic count data from all 50 states, and uses a subset of these
data covering 12 states as a basis for estimating national traffic patterns.
Heavy Duty Highway Vehicles
These categories account for emissions from gasoline and diesel trucks weighing more
than 8,500 lb GVW. Emissions from these vehicles are inventoried separately for limited access
roads, rural roads and urban roads. A uniform seasonal pattern has been assigned to ail
categories, after its applicability was confirmed by the U.S. Tracking Association A lack of daily
factor data necessitated a uniform split for the heavy Duty Gasoline vehicle (HDGV) and Heavy
Duty Diesels Vehicle (HDDV) truck categories. The hourly profile for HDGV and HDDV was
derived from NECRMP data.
Off-Highway Vehicles
The off-highway vehicle area source categories account for gasoline and diesel emissions
generated by farm equipment, construction equipment, industrial equipment, motorcycles, lawn
and garden equipment, and snowmobiles. Seasonal patterns for the off-highway vehicle
categories were derived from data contained in the Department of Transportation's Highway
Statistics. Daily patterns were derived from the EPA Guidelines. The NECRMP documentation
describes the development of separate hourly patterns for gasoline and diesel vehicles.
Railroads
This category includes emissions from the locomotives and fuel used by railroad stations
and workshops for space heating. Within NECRMP region, temporal patterns for locomotives
were based on information provided by Conrail for the Philadelphia AQCR inventory effort.
Outside the NECRMP region, temporal patterns were developed according to EPA Guidelines.
A-38
-------
Aircraft
Aircraft are divided into civil, military and commercial subgroups in the NAPAP
inventory. For civil aircraft, seasonal daily and hourly patterns were derived from the EPA
Guidelines. For military aircraft, patterns were derived from the Philadelphia AQCR Inventory.
For commercial aircraft, seasonal patterns were derived from data presented in the U.S. Civil
Aeronautical Board's Seasonally Adjusted Traffic and Capacity,
Vessels
Emissions from the vessels are split into four area source categories on the basis of fuel
type. Coal-, diesel-, and residual fuel-powered vessels include large river- and ocean-going
barges. Gasoline-powered vessels are mostly for recreational use. Temporal patterns for all
vessels were developed from the EPA Guidelines. The seasonal patterns for pleasure craft were
developed using state-level average temperature data compiled by the NOAA.
Gasoline Marketing
The area source category accounts for emissions from tank truck loading and transit,
gasoline station loading, storage tank breathing and vehicle fueling. Source investigated in the
development of seasonal patterns were refiner sales of end users, and consumption data form the
U.S. Department of Energy's (DOE's) Petroleum Marketing Monthly and the U.S. Department
of Transportation's Highway statistics 1980. From the analyses of these data, it was concluded
that the evidence was insufficient to justify using other than a uniform seasonal distribution.
Daily and hourly patterns were based on the EPA Guidelines.
Unpaved Roads
This area source category was used in the 1980 NAPAP inventory to account for
emissions of particulate matter from the unpaved roads. For the 1985 inventory, a separate data
base of natural source particulate emissions was developed, and the unpaved road dust category
was removed from the anthropogenic area source file. However, the temporal profile developed
for the 1980 inventory was retained and used to temporally allocate the new natural source
category.
Forest Fires
Forest Fires assumed to occur randomly, 7 days per week, 24 hours per day. It was
estimated that 90 percent of forest fires occur during the summer of fall, and that the remaining
10 percent are split evenly between winter and spring.
Agricultural Burning
Both of these categories were assigned the patterns developed for field/slash burning in
the NECRMP study.
A-39
-------
Structural Fires
Structural fires are assumed to occur randomly throughout the year. Thus a uniform
seasonal, daily and hourly pattern was assumed for this category.
. Ammonia Emissions for Vehicular Sources
These categories were established to account for ammonia emissions for three vehicle
types: light-duty gasoline, heavy-duty gasoline, and heavy-duty diesel. Temporal profiles were
taken directly from corresponding urban vehicular categories for which emissions of other
pollutants were reported.
Livestock Waste Management
Ammonia emissions from livestock waste management include those resulting from field
application of different types of livestock manure. Emission from these categories were estimated
on the basis of numbers of each types of livestock, weighted to account for confined and
unconfined animals. National-level temporal patterns were generally uniform for seasonal, daily,
and hourly allocation. To establish state-level profiles, Agricultural extension Agents in 12 states
across the country were contacted; they provided information on the seasonal pattern of manure
application in their states.
Ammonia Fertilizer Application .
The 12 Agricultural Extension Agents contacted to provide information on manure-
spreading practices were also asked to asked their knowledge of ammonia fertilization practices.
Beef Cattle Feed Lots
Emissions were allocated according to uniform seasonal, daily, and hourly patters.
Solvent Use Categories
Early versions of the NAPAP inventory reported a large composite category for organic
solvent evaporation based on data from NEDS. Certain NEDS-generated reports, however,, split
this aggregate into 18 individual; categories of solvent use. Beginning with version 5 of the
1980 NAPAP inventory, emissions previously reported were desegregated into these individual
categories. U.S. Department of Labor statistics on 1980 working hours were consulted to derive
national-level seasonal and daily temporal variation for each of the 18 categories. The length of
a working day was used to generate hourly factors for each category.
Minor Point Sources
These area source emissions categories were originally created for the 1980 NAPAP
emissions inventory to account for small point sources of S02, NOx, and TSP. For Version 2 of
A-40
-------
the 1985 inventory, these area source categories contain data extracted from the point source
inventory based on the following emission criteria;
- Ail individual point-level records with emissions of less than 5 TPY of each criteria
pollutants
- All points at any plant with emissions of less than 100 TPY of each pollutant
Categories 96-98 account for emissions from minor coal, oil, and gas combustion sources,
respectively. The seasonal factors for these categories were derived from electric generation data
contained in the EPRI Regional Systems. The uniform daily pattern is based upon the
assumption that utility activity is constant throughout the week. The hourly profile indicates that
most activity takes place during normal working hours. Category 90 includes combustion sources
for fuels such as coke and process gas, evaporative emissions, and all process emissions. Owing
to the diversity of source types within this category, the temporal pattern reflected a general
operating schedule of 52 week per year, 5 days per week, and 8 hours per day.
Miscellaneous VOC Categories
The 1985 NAPAP inventory provides count-level VOC emissions estimates for several
categories which have not previously been included as area source categories (see document).
To develop temporal profiles for these categories, operating schedule data for similar Source
Classification Codes (SCCs) in the NAPAP point source data base were analyzed. For most
categories, there was no justification for assigning anything other than a uniform profile.
However for category 101 (Cutback asphalt paving operations) and category 105 (bakeries),
factors were derived from mean operating schedule data of the analogous point sources.
U.S. Point Source Temporal Factors
Review of 1980 Factors
Beginning with the 1980 NAPAP emissions inventory, temporal allocation of point source
data was accomplished by tow methods. In most cases, operating schedule data included in the
NEDS-based point source data records served as the basis for point source factors. However,
given the magnitude of emissions from electric utility sources, process-specific factors were
developed to more accurately characterize these sources. Most NAPAP point source emission
records contain operating schedule data which make possible the point-specific temporal
allocation of emissions, these data consist of information on seasonal throughput percentages;
days/week of process operations; and hours/day of process operation. The operating data were
use to temporally apportion point source data for which specific temporal factors were no
developed., seasonal factors were taken &om seasonal throughput percentages. Daily factors were
derived from the number of days per week which a process operates.
Because of the importance of electric power plants to total U.S. emissions'of S02 and
NOx, and because detailed data are available for these sources, special process-level (fuel and
state specific)temporal factors were developed for utilities during the 1980 NAPAP effort, factors
were originally developed during the NECRMP study for power plants within the Northeast
A-41
-------
Corridor. In development of the 1980 NAPAP inventory, most of the NECRMP factors were
retained, while some were replaced by new factors derived from additional data. For sources
ouwide the NECRMP region, some factors were based on NECRMP figures; however, most were
developed from other data sources. Seasonal factors foT the temporal distribution of point source
emissions were originally .developed on a fuel- and state-specific basis for facilities in the
NECRMP study area using power generation statistics form the U.S. DOE 19879 Energy Data
reports. Daily factors were developed at the national level from weekly load cycle listings in the
EPRI Regional Systems. Fuel- and state-specific weekday hourly patterns for utility operating
were developed during the NECRMP effort Profiles were derived from hourly power pant fuel-
use data collected during the development of the SURE inventory.
1985 Point Source Temporal Factors
Temporal allocation of point source emissions for the 1.985 NAPAP inventory was based
on the methodology used for the 1980 inventory. These factors were also updated to
incorporate new information. In addition to updates to operating schedule data and utility
temporal factors, the hierarchy of their use was changed. Updates to seasonal throughputs were
derived from three separate studies. Most were based on a study of plant-specific fuel use data
conducted by NAPAP's Task Group I on Emissions and Controls. The major source of data for
this analysis was the U.S. DOE/energy Information Administration (EIA) Form 759, which
contains information on monthly fuel use by plant and fuel type. The two ether studies used to
update information was the unpublished computer printouts from DOE/E1A titled "R080 - Report
on Consumption" and an analysis of 1985 TVA load data from 11 plants. During development
of the 1985 NAPAP point source inventory, 1985 hourly'generation data were obtained for a total
of 59 TVA coal-fired units. These data were assumed to represent the most accurate temporal
information available for the 11 plants.
Canadian Temporal Factors
Canadian area source emission estimates were developed using methodologies similar to
those used for area sources in the United States. The estimation of area source emissions in
Canada was performed directly by Environment Canada and provided to the provincial
environmental agencies for their review. Point source emissions and facility data for the 1985
NAPAP emissions inventory were collected in a cooperative effort between Environment Canada
and the provincial governments.
Natural Source Temporal Factors
Data on alkaline paniculate emissions from natural sources in the United States and
Canada were obtained from NAPAP's task Group II, EPA/AREAL, and Environment Canada.
A summary of natural alkaline particulate categories included in-the 1985 NAPAP inventory is
provided in Table 2-4 in of the original document. Temporal allocation factors were developed
individually for each category at varying levels of specificity. Canada temporal information for
natural sources were supplied by Environment Canada and is included among the Canadian area
' source temporal factors, which are described earlier.
A-42
-------
PROCESSING OF TEMPORAL ALLOCATION FACTORS
Temporal allocation if annual emissions data is accomplished during processing of the
point and area source data base by the Flexible regional emission data System (FREDS), in
brief, FREDS is an integrated data processing system designed to generate a resolved emission
inventory based on the application of temporal, spatial and species allocation factors to annual
¦emissions data. Resolution of annual emissions is accomplished by the execution of a series of
"allocation modules" which produce gridded, speciated and temporally allocated output files. The
majority of the FREDS software-is written in SAS, and relies heavily on the use of SAS macros
for flexibility in processing different file structures.
Point and area source emissions are temporally resolved during execution of the Temporal
Allocation Module (TAM) of FREDS. Different versions of TAM have been developed to
process point and area source data'on anthropogenic and natural emissions in the United States
nd Canada. However, the overall methodology used to apportion these data is similar. The
generalized TAM structure accepts four main input files. The emissions data file, which is
already been files processed by preceding FREDS modules, is imported by TAM along with a
Control Options File (COF). The COF passes file types information to TAM to permit file-
specific execution of the module., the temporal Allocation Factor File (TAFF) is a separate file
containing the seasonal, daily, and hourly multipliers used for temporal allocation. Adjustments
of local temporally allocated emissions to Greenwich Mean Time, including an offset for
Daylight savings Time where appropriate, is accomplished by applying factors from the Time
Zone file (TZF).
The TAFF provides allocation factors for up to 12 "typical days' (a weekday, Saturday
and Sunday for each season). These days are commonly referred to as "temporal scenarios".
Individual records in the file are distinguished by one or more identifiers. For emission sources
in the United States these identifiers include source category, state, plant, and point. For sources
in Canada a special linking variable is created. TAM uses these identifiers to link temporal
profiles with appropriate emission records. For U.S. sources, profiles are applied to emission
records by starting at the most specific level and differing in steps to the most general level. As
an exception, seasonal factors for point sources in the United States are based primarily on
seasonal throughput percentages supplied with the emissions records; seasonal factors form the
TAFF are used secondarily.
In the absence of specific temporal allocation factors for point sources in the unite States,
TAM provides daily and hourly emissions allocation based on operating schedule data from
emission records (i.e. day per week and hours per day of process operation, (seasonal throughput
percentages on emission records are automatically used when they are present.) As a final step,
both point and area source TAM programs assign uniform default allocation profiles if temporal
factor data are missing or incomplete.
A-43
-------
Gardner, L. et. al. Procedures for the Preparation of Emission Inventories for Carbon Monoxide
and Precursors of Ozone, Volume II: Emission Inventory Requirements for Photochemical Air
Quality Simulation Models. EPA-45Q/4-91-014. Systems Applications International under EPA
Contract No. 68-DO-0124. U.S. Environmental Protection Agency. Research Triangle Park, NC.
May 1991.
BACKGROUND
This document is the second volume of a two-volume series designed to provide
assistance in preparing and maintaining emission inventories. The first volume of this series
describes procedures for preparing inventories of VOC, CO, and NOx on annual or seasonal basis
at the county level. Volume II provides technical assistance for developing detailed emission
inventories of VOC, CO, and NOx for use in photochemical air quality simulation models at the
cell level of a grid system with special emphasis on fulfilling the input requirements of the Urban
Airshed Model (UAM).
PHOTOCHEMICAL AIR QUALITY SIMULATION MODELS
Photochemical air quality simulation models attempt to simulate, at the cell level of a grid
system, the photochemical reactions that occur over an urban region during each hour of the day
or days for which the model is being applied. Photochemical simulation models provide detailed
spatial and temporal information on concentrations of both ozone and precursor pollutants.
In order for photochemical simulation models to accurately predict temporal and spatial
variations in modeled ozone and CO concentrations, the emission inventories input of these
models must be more detailed than an annual or seasonal inventory generated at the county level.
Specifically, additional resolution of emissions is required, namely spatial, temporal, and
chemical. This summary is geared towards the temporal resolution of emissions.
OVERVIEW OF THE UAM EMISSION PREPROCESSOR SYSTEM
The UAM Emission Preprocessor System (EPS) consists of six programs which are
executed sequentially to generate the emission input files for the UAM. The PREPNT program
prepares the annual average or seasonal point source inventory for subsequent chemical speciation
by the CENTEMS module. PREPNT also assigns temporal distribution profiles based on the
operating information contained in the annual or seasonal inventory. One output produced by
PREPNT is a gridded point source inventory in Model Emissions Record format (MERF) which
undergo subsequent processing in the CENTEMS module.
CENTEMS, the central program of the UAM EPS, adjusts annual average daily emissions
to account for monthly variations and assign them to the hours of the modeling episode based
on temporal distribution profiles. In addition to the MERF emissions files from PREPNT,
CENTEMS requires a diurnal variation factors files, containing hourly profiles for the diurnal
codes contained in the MERF files, which are used to allocate the daily emissions to the fours
of the modeling episode. CENTEMS also requires a weekday factor file for adjusting emissions
based on the day of week by the weekday code in the MERF record.
A 44
-------
The PREGRD program reformats an annual or seasonal area source emission inventory
and prepares it for gridding. Emissions are separated into area sources and on-road motor vehicle
sources. PREGRD requires several input files including a file containing motor vehicle
adjustment factors used to adjust annual or seasonal average mobile source emissions for episodic
conditions and the annual or seasonal area source inventory, containing county-level emission
estimates by source category.
The GRDEMS program allocates the pre-processed county-level emissions from PREGRD
to the modeling region grid cells and assigns temporal distribution profiles by source category.
GRDEMS requires several input files including the county-level MERF emissions files generated
by PREGRD.
TEMPORAL RESOLUTION OF EMISSIONS
Several approaches can be employed to provide the temporal detail needed for the
photochemical simulation models. The most accurate approach is to determine the emissions (or
activity levels) for specific sources for each hour of a typical day during the modeling period.
A second approach is to develop typical hourly patterns of activity levels for each source
category, and apply those to the seasonal or annual inventory to estimate hourly emissions.
In developing temporal resolution of emissions, emissions must be adjusted to reflect
typical levels for the particular non-attainment season (ozone or CO). Similarly, emissions are
adjusted to represent the day of the week on which polluting activities are at a maximum,
normally a weekday.
Temporal Allocation of Point Source Emissions
For most urban regions, a basic annual or seasonal point source emission inventory will
already exist which contains most or all the information required for photochemical modeling.
One notable exception may be the lack of sufficient operating schedule information to accurately
estimate hourly emission rates.
Existing point source inventories more likely contain annual or typical ozone season day
estimates of emissions and a general description of the operating schedule. However, for
photochemical modeling purposes, this information may need to be augmented to represent day-
specific emission estimates for each hour of the modeling episode. Several approaches for this
augmentation are available. Ideally, each facility would be contacted to obtain hourly operating
records for the modeling episode, or if this information is unavailable, representative operating
schedules for a typical ozone season day. Other approaches include contacting local agencies,
extrapolating from information contained in the existing inventory, and using engineering
judgment.
For many sources, daily operation is confined to one or two workshifts; thus, hourly
operation during working hours would be determined by dividing the daily operating rate by 8
or 16. Above-ground fixed-roof petroleum product storage tanks present a unique situation that
A-45
-------
should be handled outside of EPS because breathing loss emissions appear to be a function of
time of day rather than operation,
UAM EPS's Temporal Allocation of Point Source Emissions
The UAM EPS supports incorporation of episode-specific hourly emissions estimates for
individual point sources into the modeling inventory. Day-specific MERF records allow the user
to specify the hourly fractions used to diurnally allocate total daily emissions. However, Version
1.0 of the UAM EPS does not provide software to generate day-specific MERF records;
accordingly, this information must be incorporated into the modeling inventory outside of EPS
using supplemental software or by manual editing of the point source standard MERF file
produced by PREPNT.
The UAM EPS can perform temporal adjustments automatically based on the operating
data contained in the basic inventory. The PREPNT program converts seasonal fractions of
throughput to a monthly variation profile (assuming uniform variation throughout the season) and
assigns default day-of-week and diurnal activity profiles based on the number of days per week
and the number of hours per day in operation.
The PREPNT program automatically applies one temporal adjustment based on the
number of weeks per year in operation for each source. The PREPNT program assumes that any
source not operating for the entire year will be in operation during the modeling episode.
The UAM EPS assigns a flat operating profiles (i.e., equal seasonal fractions of annual
throughput, and 52 weeks/year, 7 days/week, and 24 hours/day) to each source in the inventory
with missing temporal variation data. A flat operating profile is also assigned to those sources
with invalid operating data.
Tables 5-4 and 5-5 show the default weekly and diurnal activity profile codes provided
with the UAM EPS. The values in these tables represent relative levels of activity by day of
week and hour of day, respectively. These temporal variation codes are assigned in PREPNT
for point sources and GRDEMS for area and mobile sources. The weekly profiles listed in Table
5-4 are used in conjunction with monthly activity fractions to compute representative 'episodic
emissions levels from the annual or' seasonal average emissions contained in the existing
inventory. This calculation can be mathematically represented as
Em = (Em) x (Mf x 12 monthsjyr) x [(7 daysfwk total)/DF]
where E^ denotes episodic daily emissions, ElVA denotes annual average daily emissions, MF is
the fraction of annual activity occurring in the episode month, and DF is a day of week
adjustment factor obtained by dividing the value in Table 5-4 for the day of week of the episode
by the total for the day.
A-46
-------
Temporal Allocation of Area Source Emissions
Emissions modelers can often use an existing county-level area source emission inventory
as the basis for the modeling inventory. Several approaches can be employed to develop hourly
emissions resolution from the basic area source inventory. The approaches involve the use of
assumed diurnal and daily patterns of activity.
If the county-level inventory contains annual emission estimates, the first step is to
estimate the seasonal components of activity for each area source. Chapter 6 of Volume I
discusses seasonal adjustment in detail, for many sources, activity is fairly constant from season
to season. Table 6-11 lists recommended seasonal adjustment factors that can be used if local
activity distribution data is unavailable.
After the seasonal adjustments are estimated, the weekly variation is determined. Some
area source activities are fairly constant from day to day, making it a simple matter to estimate
daily activities, many area sources, on the other hand, are generally more active on weekdays.
In these cases, the seasonal activity should be distributed to only those days on which the source
is active.
After the daily activity level has been determined, the next step is to estimate hourly
emissions, this is generally accomplished by applying a 24-hour operating pattern to the daily
activity level.
Table 6-13 lists some approaches for incorporating temporal resolution for several area
source categories into the detailed emissions inventory. These can be used for temporal
distribution in the absence of more specific data. However, local working hours and seasonal
activity may differ from those suggested in Table 6-13. The most general default option is to
assume complete temporal uniformity.
UAM EPS's Temporal Allocation of Area Source Emissions
The UAM EPS program GRDEMS assigns temporal distribution codes and monthly
activity factors by source category code contained in a cross-reference file. The temporal
variation codes contained in this file correspond to those shown in Tables 5-4 and 5-5. In
addition, the program requires that seasonal adjustment factors be defined as monthly fractions
of annual activity. The cross-reference files provided with the UAM EPS contains temporal
distribution profiles by NAPAP source category which reflect national average or default data.
Accordingly, the default temporal profile assignments in this file should be reviewed for local
applicability.
Seasonal adjustment factors in Table 6-11 tan be converted to monthly fractions in the
following manner: for each month of the ozone season, the monthly fraction of annual activity
for a particular source category will be the seasonal adjustment factor for that category divided
by 12. for the other months, the monthly fraction can be assumed to be 1/9 of the remaining
activity or
A-47
-------
[1 - (3 x monthly fraction for an ozone season month)]
9
If the inventory contains seasonal emission estimates, no additional seasonal adjustment should
be applied, A flat monthly variation profile for all source categories should be used (i.e., each
month factor would be 1/12 or 0.083.
Hourly operating information can be incorporated directly into the modeling inventory
using day-specific MERF.
Temporal Allocation of Mobile Source Emissions
An existing annual or seasonal area source emission inventory generally contains adequate
estimates of emissions for all sources except on-road motor vehicles. The UAM will be applied
for specific episode days. Mobile source emissions must be computed specifically for those days
or adjusted to reflect conditions on those days.
In general, one of two methods can be employed to develop the on-road portion of the
modeling inventory:
1. compiling an episode-specific on-road vehicle emission inventory using the
methods described in Procedures for Emission Inventory Preparation, Volume TV:
Mobile Sources (hereafter referred to as Volume TV); or
2. adjusting an existing annual or seasonal inventory to reflect episodic conditions.
Temporal adjustment of the mobile source inventory into monthly, daily and hourly
specific totals is not significantly different from the treatment of other area source categories.
Procedures recommended for temporal adjustment of area sources also apply to mobile sources.
As a special consideration for weekend inventories, diurnal variations in weekend driving activity
generally differ from weekday patterns which typically exhibit a double-peaked profile, with the
most activity occurring during the morning and afternoon commute hours. If hourly vehicular
speeds and vehicle miles travelled (VMT) distributions are available from local Metropolitan
Planning organizations, these should be utilized in estimating hourly mobile source emissions.
UAM EPS's Temporal Allocation of Mobile Source Emissions
The UAM EPS daily and hourly variation codes are shown in Table 5-4 and 5-5
respectively. In Table 5-4, the default assignment for daily variation of mobile emissions is code
#23. In Table 5-5, the default assignment for diurnal variation of all on-road source categories
is code #48 for weekends and code #50 for weekdays. However, in any UAM application, these
values should be checked with locally-specific data to determine their applicability. In addition,
the UAM EPS assumes a flat profile for monthly variation of on-road mobile emissions. This
is not characteristic of any particular region, but is an indication of the wide variance of monthly
factors depending on location.
A-48
-------
Hanna, S.R. and PJ. Drivas. "Modeling VOC Emissions and Air Concentrations from the Exxon
Valdez Oil Spill," Journal of Air Waste Management Association, Vol. 43. March 1993.
The purpose of this article is to model both atmospheric emissions and atmospheric
concentrations of VOC's from the March 1989 Exxon Valdez oil spill. Emissions were estimated
using a multicomponent evaporative emissions model developed for the purposes of this study.
Concentrations were estimated using the Atmospheric Turbulence and Dispersion Laboratory
(ATDL) dispersion model and the Offshore and Coastal Dispersion (OCD) model. Fifteen
specific compounds were considered: "benzene, ethylbenzene, toluene, o-xylene, m&p-xylenes,
n-hexane, n-heptane, n-octane, n-nonane, n-decane, n-undecane, n-dodecane, n-tridecane, n-
tetradecane, and n-pentadecane. Emissions of the most volatile compound, toluene, peaked at
about 20,000 kg/hr, or about 5 g/rrf-hr at about 8 hours after the spill. Maximum predicted
hourly concentrations did not exceed 10 ppmv for any compound. The highest reported
concentration over the spill was for toluene which was modeled at 8.24 ppmv the first hour of
the spill.
Emissions estimates were based on evaporation rates from a liquid spill, which are
controlled by one of two processes: mass transfer from the surface of the spiE or diffusion
through the liquid layer. The authors based the mass transfer rate on Raoult's law for a liquid-
gas equilibrium. The emission rate in terms of mass is expressed as:
dm/dt = -(kgAp/nx) mt
where is the mass of compound i remaining, t is time, kg is a mass transfer coefficient, A is
the spill area, pj is the pure component vapor pressure, and nT is the total gmoles of oil
remaining. In the study, this equation was integrated to allow hour-by-hour calculations, and
parameters were assumed to be relatively constant on an hourly basis, so that the average hourly
evaporative emissions could be calculated. The result is:
my= mij-i exP[-(kgAp/nT)j(tj-tj.,)]
where j indicates the number of hours since the spill. An exponential decay was assumed for the
molar evaporation rate of the oil to estimate number of moles remaining. The authors used kg
values based on the work of Mackay and Matsugu cited in the article.
The other process controlling evaporation is diffusion. The authors assume a simplified
mass-thickness relationship based on diffusion for a semiinfmite solid. This results in the
diffusion equation being expressed as:
dm/dt = -0.632(A/V)m1(D/t)w
where m, is the mass of compound i remaining, t is time, D, is the molecular diffusivity of liquid
i, V is spill volume, and A is spill area. Similar to the mass-transfer approach, the equation was
integrated to allow hour-by-hour calculations, and average hourly emission rates can be
calculated. The result is:
A-49
-------
m}~ m^, exp[-1.264(A/V)(\'Dl)(\'tj - VtH)]
where j indicates the number of hours since the spill. The authors used Dj values based on
molecular diffusivity, but the diffusivities were adjusted,to account for increased diffusion due
to wave motion. The effective diffusivites were based on experimental data.
In order to determine whether diffusion or mass transfer is dominant, the authors
compared the exponential terms in the two hourly calculations above for each hour and for each
compound. Whichever term is smaller will indicate which process is limiting the evaporation
rate. It is expected that high-volatility compounds will be limited by diffusion, and heavier
compounds will be limited by mass transfer.
In applying the emissions estimation methodology to the Exxon Valdez case, compositions
of compounds of interest in the initial oil had to be determined. Because no measured data were
available for spilled oil, an average reported literature value for Prudhoe Bay crude oil was used.
These values are reported in the paper. The area of the spill was determined from maps of
observed spill area, and from a NOAA spill model. Emission rates were then calculated for each
hour during the first two weeks of the spill for the 15 compounds of interest.
In the interest of space, the paper only reports hourly emissions calculated for the three
most volatile compounds - toluene, benzene, and hexane. These emissions are reported for the
first 15 hours of the spill. Rates of over 20,000 kg/hr are reported for toluene, about 10,000
kg/hr for benzene, and about 4,000 kg/hr for hexane. Generally, evaporation rates of these
compounds peaked at 9 hours, and were entirely evaporated in about the first 12 hours of the
spill. The paper notes that these emissions were diffusion limited. The paper also notes that the
less-volatile compounds evaporated at a much slower and more constant rate over the entire two
weeks, and that these emissions were mass-transfer limited. Although the paper reports
quantitative results for only three compounds, it does reference a more detailed report where all
hourly quantitative results can be found.
Next, the paper briefly discusses the effect of the spill temperature being higher than the
ambient temperature. This was the case for the Valdez spill, but was not accounted for in the
model. It is noted that if evaporation rates at the tanker temperature of 40 °C are compared to
rates at the ambient temperature of 3 °C, the maximum evaporation rates would be very similar,
but would occur about two hours sooner. This does not affect the peak concentrations, only the
timing of them.
After completing the emissions estimation section, the authors use their emissions
estimates to model peak ambient concentrations of the compounds. Before modeling, the authors
discuss the determination of wind fields necessary for model input. The authors analyzed
National Weather Service (NWS) wind data from several stations and determined which ones
were relevant to the spill area, considering proximity to the spill, terrain, and local effects. The
best stations were temporary stations measuring wind direction near the spill. However, these
were not installed until the last two days of the study, so the authors analyzed the relationship
between the temporary and permanent stations to select the permanent stations which are most
representative of wind direction near the spill. They determined that wind vectors over the spill
A-50
-------
location could best be interpolated using two stations, Potato Point and Middleton Island, together
with the observed movement of the spill. The authors present these measured and interpolated
hourly wind speeds and directions in Figures 5 and 6.
For modeling, two dispersion models were used. For concentrations over the spill, the
ATDL area source diffusion model was used. For receptors at the edge of the spill and over the
land, the OCD model was used. The ATDL model is a simple formula used to calculate
concentrations over an area source. The ATDL model has been validated in many field studies.
The OCD model is a steady-state Gaussian model developed to simulate plume dispersion and
transport from offshore point, area, and line sources to receptors on land or water. The authors
use it to calculate concentrations at the nearest shoreline to the oil spill, but do not report the
OCD results in the paper.
For concentration results, the paper presents a table showing maximum hourly average
concentrations over the spill, together with the hour that they occurred, for all 15 compounds.
In addition, it presents a table detailing hourly concentrations for all 15 compounds at selected
time periods during the study. The authors note that the maxima for the more volatile
compounds all occurred during the first three hours, with the magnitude of the maxima depending
on wind speed. Toluene had the highest modeled concentration at 8.24 ppmv, occuring in the
first hour. They also note that for the less-volatile compounds, (tetradecane and pentadecane)
the maxima did not occur until the second day (hour 29), when the wind speed reached a low
point. The maxima for tetradecane and pentadecane were 0.00628 ppmv and 0.00186 ppmv
respectively. Finally, they note that all lighter compounds (through nonane) are predicted to
completely evaporate by the end of the first day, while concentrations of the heaviest compound
studied, pentadecane, remained fairly constant at a low level throughout the study.
Finally, after summarizing the modeling results, the authors make recommendations for
improving the accuracy of the calculations. These include: improving ambient VOC monitoring,
improving knowledge of the composition of the spilled oil over time, improving wind
measurements and other atmospheric data close to the spill, using mesoscale wind models, doing
a sensitivity analysis, comparing results with other studies, and rerunning models with improved
data when available.
A-51
-------
Heisler, S.L. et. al. Interim Emissions Inventory for Regional Air Quality Studies. EA-607Q.
ENSR Consulting and Engineering for Electric Power Research Institute. Palo Alto, CA,
November 1988.
SUMMARY
An emissions inventory for 1982 was developed for the contiguous United States and part
of southeastern Canada. Inventoried materials include total emitting particulate matter (TEP),
alkaline TEP, S02, S04, NO, N02» NH3, and hydrocarbons in nine reactivity classes. Emission
rates, geographic locations, and stack parameters are provided for major TEP, S02, NOx. and HC
emitters. Emission rates for all other sources are aggregated into 1/2° longitude by 1/3° latitude
elements of a grid system. Emission rates were developed for several averaging times: the entire
year, each season, weekdays and weekends within each season, and each of the eight 3-hour
periods during weekdays and weekends in each season.
This summary only discusses the sections of this document that addressed temporal
allocation issues. Therefore the remainder of this summary discusses:
• Derivations of seasonal, weekday/weekend and' diurnal emission rates
Data uncertainties of these temporal profiles
Resolution of temporal allocations uncertainties
SEASONAL, WEEKDAY/WEEKEND, AND DIURNAL EMISSION RATES
Seasonal, weekday/weekend, and diurnal average emission rates were calculated from
annual average emission rates and temporal variations in source operating rates. It was. assumed
that the proportionality between emission and operating rate was constant throughout the year,
so that the average emission rate in units of mass per unit time during a particular time period
was equal to the product of the annual average emission rate and the ratio of the average
operating rate during the time period to the average operating rate for the year.
Operating schedules (days per week and hours per day) and seasonal operating levels
(percentages of annual production) were reported for individual point sources, and these data
were used for the calculations. Seasonal variations in United States area source residential,
commercial, and institutional fuel use were calculated from 1982 heating degree day data for each
county. Flight schedules at three major airports during 1982 were used to calculate temporal
variations in emissions from commercial aircraft landings and takeoffs. Temporal changes in
emission rates for other area source categories were calculated from characteristic patterns in the
literature, when possible. When patterns were not available in the literature, variations were
based on common sense.
The average emission rate during a season, ER^ (g/s), was calculated from the annual
average emission rate, ER, (g/s), by:
A-52
-------
ER, = 0.04 PS ER,
where:
PS = percentage of annual operations that occurred during a season
The factor 0.04 converts PS from a percentage to a fraction and accounts for the difference in
length between a season and a year.
Weekday/weekend and diurnal average rates were calculated by the use of temporal
profiles. Weekday/weekend profiles specified the ratio of daily average operating rates to
seasonal averages for weekdays and weekends. Distinctions were not made among the five
weekdays or between Saturday and Sunday. Diurnal profiles specified the ratio of average
operating rates during each of the eight 3-hour periods during a day to daily average rates.
The average emission-rate during a particular 8-hour period on a particular day of the
week during a particular season, ERh is calculated by:
ERb - PH PD ER,
where:
PD = the appropriate weekday/weekend profile value
PH = the appropriate diurnal profile value
1024 emission rates can be calculated from each emission source for 16 pollutants,'4 seasons,
2 days (weekday and weekend day), and 8 diumal periods (8 diurnal periods of 3 hours each).
Point source weekday/weekend profiles were based on reported operating days per week.
A single value was reported for the entire year for each source, and therefore a single
weekday/weekend profile was used for all four seasons. The profile was determined as follows:
* For five or fewer days per week, or when no value was reported, operations were
assumed to be uniform from Monday through Friday, and the profile values were 1.40 for
weekdays and zero for weekends.
For more than five days per week, operations were assumed to occur uniformly during
the week, and the profile values were all equal to one.
Operating hours per day determined point source diurnal profiles for all sources except fossil
fueled electric utility boilers as follows:
~ For eight or fewer operating hours per day, or when the value was missing, operations
were assumed to occur uniformly between 0900 and 1700 hours local time.
For nine to sixteen hours per day, operations were assumed uniform between 0800 and
2400 hours local time.
A-53
-------
For more than sixteen hours per day, operations were assumed uniform during the entire
day.
Diurnal variations for fossil fueled electric utility boiler operations were based on hourly SO-
emission data collected during development of the Sulfate Regional Experiment (SURE)
emissions inventory. These data were collected from 300 utility plants in the northeastern United
States for Januaiy, July, and October 1978. Average diurnal profiles were calculated for each
plant during each of the three months by dividing the average emission rate for each hour of the
day (i.e., the average of emission rates between 0000 and 0100 hours, 0100-0200 hours, etc. for
all days during the month) by the average emission rate for the entire month. The profiles for
all the plants were then averaged arithmetically for each month. January and July data were
assumed to represent winter and summer, and October was assumed to represent both spring and
fall.
Discussion is also presented on temporal allocations for area sources. Temporal
allocations assumptions are presented for each of the area source categories.
DATA UNCERTAINTIES
Temporal Ratios
Annual emissions were multiplied by temporal profiles to calculate seasonal,
weekday/weekend, and diurnal average emission rates. Uncertainties in these profiles need to
be estimated for both point and area sources. Point source seasonal emission rates were
calculated from reported percentages of annual operations that occurred during each season.
Uncertainties in these values should be evaluated as discussed previously for emission source
data. Weekday/weekend point emission rates were calculated using reported days of operation
per week as an indicator for weekend operations. The profiles that were used are correct only
if sources operate at the same rate each day they are in operation. This requirement is probably
not satisfied by many sources, particularly sources whose operating rates are driven by customer
demand. For example, generating rates at electrical utility plants can be affected by electrical
demands throughout a utility's service area or in the service areas of other utilities to whom
power is sold. Diurnal operating patterns can also vary from day to day because of similar
factors.
Variations in control device efficiency and fuel quality can also cause temporal changes
in emission rate independently from changes in operating rates. These variations introduce
additional uncertainties in calculated emission rates.
Facility operators and industrial representatives should be consulted to evaluate the day-to-
day viability of operations. While it is probably not feasible to contact operators of each major
emission source, appropriate representatives of important industries should be contacted to
determine if these variations are consistent within their industries. If they are consistent, a subset
of facilities could be contacted to determine this variability. Establishing these variations may
be difficult if daily or diurnal operating records are not maintained. However, some sources are
A-54
-------
equipped with continuous emission monitors. Data from these monitors could be used for this
activity,
Seasonal, daily, and diurnal emission rates for area sources in individual counties and grid
elements probably also deviate from values in the inventory. For example, residential fuel use
is affected by daily temperature, although the profiles used for calculating weekday and weekend
values produced constant emission rates during a season.
The most feasible approach for estimating the range of deviations from area source
temporal profiles would probably be to develop more accurate profiles for some counties. This
effort might involve contracting fuel distributors, small industrial and commercial facilities, state
and local transportation agencies, forest services, farmers, and construction contractors to obtain
the required data. It will probably not be feasible to obtain data for all sources in the county.
However, subsets could be sampled, and appropriate statistical surveying techniques could be
used for interpreting the results.
TEMPORAL RESOLUTION
Point source seasonal average emission rates were calculated from annual averages using
percentages of annual operations that occurred during each season for individual emission
sources. This approach neglected seasonal differences in fuel sulfur and ash content for
combustion sources and seasonal differences in control device efficiency. Weekday and weekend
averages were calculated from seasonal averages using reported number of days of operation per
week. Weekday and weekend rates were set equal to the seasonal average for sources that
operated more than five days per week. This approach neglects effects of possible reductions in
operating rates on weekends for sources that operate every day. Three-hour average, non-utility
point source emission rates were calculated from the weekday and weekend averages using
reported hours of operation per day. Daily emissions were assumed to occur uniformly between
0900 and 1700 hours local time, 0800 and 2400 hours, or during the entire day for sources that
operated less than nine, between nine and 16, emission rates as a function of season for utilities
were derived from hourly emission rate data acquired for developing the emissions inventory for
the Sulfate Regional Experiment.
.Seasonal, weekday/weekend, and diurnal average emission rates for area source categories
were calculated using temporal profiles, which were the ratios of the average operating rate
during each subperiod to the average during, the next longer period. These profiles were derived
from: (1) variations in operating rates recommended for use by the U.S. EPA, (2) variations in
operating rates reported in the literature (3) variations in surrogates for operating rates, such as
seasonal heating degree days, which was used as a surrogate for residential fuel use, and (4)
"common knowledge", such as setting the emission rate from orchard heaters to zero during the
summer. Emission rates were assumed to be constant when none of these methods could be used
for deriving a profile.
The profiles should be improved for source categories that are important emitters. In
particular, profiles for motor vehicles, which are a major NOx source, were derived from
measurements of traffic volumes on only five roadways. Data for other roadways should be
A-55
-------
obtained from state transportation agencies to provide better basis for calculating temporal
variations. Wind erosion dominates TEP emissions; however, temporal variations were not
calculated for this source. Seasonal meteorological data should be used for calculating seasonal
average emission rates for this category. Since these emissions axe episodic in nature and are
a nonlinear function of wind speed and rainfall, joint frequency distributions of meteorological
data should be used for these calculations. Modelers should also consider calculating emission
rates for this source category using meteorological data specific to the time period being modeled
in conjunction with the solid data acquired during this project. Seasonal variations in NH3
emissions from land surfaces should also be calculated using relationships that account for
variations in emission rate with temperature.
A-56
-------
Hilst, G, et. al. Time-Variable Air Pollutant Emission Strategies for Individual Power Plants.
EA-418. TRC - The Research Corporation of New England for Electric Power Research
Institute. Palo Alto, CA. April 1977.
SUMMARY
This report presents the concept of varying pollutant emissions from power plants during
times where sufficient atmospheric dispersion would allow for adequate waste disposal while
complying with applicable air quality standards. As of the date this document was prepared, the
practical application of this concept had been well established. This concept is called time-
variable emission control strategies. The intent of the study was to develop emission control
strategies which meet all applicable emission standards and to minimize emission control costs.
As such, an assessment was made of the following alternative control strategies: (1) pre-
combustion fuel treatment for sulfur removal; (2) fuel switching; and (3) post-combustion flue
gas desulfurization. Data from a two-year case history of dispersion in Col strip, Montana was
used in this analysis.
The report discusses in detail, physical modeling analysis as well as both linear and non-
linear cost of control analysis to determine the optimum number of operation modes needed in
order for power plants to comply with emission standards while minimizing control costs. The
physical model used is a single isolated power plant which emits a single pollutant through a tall
chimney. The basic scenario requires the plant to be operated in such a manner so that the
ground-level concentration does not exceed the annual standard and exceeds a 3-hour and a 24-
hour standard only once per year.
Inputs into the time-variable emissions model included dispersion over time, the maximum
emission rate which the plant could achieve, and the frequency for which the plant can switch
from one emission mode to another. The model assumed (from a realistic standpoint) a 500
megawatt (MWe) plant which burns 1 percent sulfur coal with an maximum emission rate of
1000 grams/sec and a minimum emission switching time of 3 hours. Emission mode refers to
varying the emission rate through the process of either switching from one fuel to another or
going from one pollution control device to multiple pollution control devices.
For S02, the analysis determined the emission rate needed to achieve the applicable
emission standard from: (1) an emission rate of 1000 grams/sec (with a sulfur content of 1
percent); and (2) an emission rate of 3000 grams/sec (with a sulfur content of 3 percent). Using
time-variable emission controls, the results of the analysis showed that the emission standards
for each of the scenarios could be met with minimal environmental impacts by using intermittent
control systems. Specifically, for a maximum emission rate of 1000 grams/sec, the emission rate
would need to be reduced by 18% and for a 3000 grams/sec maximum emission rate, a 56
percent drop would be needed to meet the standards. It should be noted that for a single
continuous emission rate of 1000 grams/sec, 90 percent of the S02 would need to be removed
and for a 3000 grams/sec emission rate, a 96 percent removal rate of S02 using controls would
be necessary to comply with the standard.
The discussion contained in the paragraph above applies to achieving the 3-hour, 24-hour
A-57
-------
and annual standard during the first year of operation.
Multi-mode time-variable emission control scenarios for power plants includes using the
following control techniques in various combinations to achieve the appropriate emission
reductions: (1) pre-con-,bastion fuel selection or desulfurization (which can be used in
combination with fuel switching); (2) load reduction; and (3) post-combustion flue gas
desulfurization. The results of the study indicated that the optimum number of emission modes
needed depends on plant size and fuel sulfur content and the largest number of modes needed
would be five 'Or six.
A single multiple scrubber system (FGD) can handle up to 250 MWe of effluent SO, and
power plants built with greater than 150 MWe capacity generally have two or more.FGD units
operating in parallel. Such systems have been designed to. operate in time-variable mode in
response to boiler load variations.
Finally, this report does not contain any information regarding the development of hourly
emission rates, but only discusses the optimum emission modes necessary to comply with
applicable air quality standards. However, some of the assumptions used in the modeling
analysis may be helpful in the estimation of temporal allocation factors for power plants.
A-58
-------
Hodgson, A.T. et. al. "Emissions of Volatile Organic Compounds from New Carpets Measured
in a Large-Scale Environmental Chamber," Air & Waste, Vol. 43, March 1993.
The purpose of this study was to quantify the emissions of VOC's released by several
types of new carpets. Of particular interest was the compound 4-phenylcyclohexene (4-PCH),
which is associated with odor and possibly health complaints. Four carpets were tested; all had
a primary backing of polypropylene, and the other characteristics differed as follows: Carpet 1
had nylon fibers and a polypropylene secondary backing with a styrene-butadiene rubber (SBR)
latex adhesive; Carpet 2 had nylon fibers and a polyurethane secondary backing with an
unspecified adhesive; Carpet 3 had nylon fibers and a PVC secondary backing with a an
unspecified adhesive; Carpet 4 had a combination of 75% olefin and 25% nylon fibers and a
polypropylene secondary backing with a SBR latex adhesive. These carpets were selected for
study by the Consumer Product Safety Commission (CPSC) as representative of home, school,
and office carpets.
First, a small (4-liter) chamber was used to initially-screen each carpet sample for VOC
emissions. The small-chamber samples were taken at room temperature in an atmosphere of dry
N2. The ventilation rate of the chamber was 6.3 per hour, and the loading ratio (carpet
area/chamber volume) was 2.65 m2/m3. Small-chamber samples were collected at 1, 3, and 6
hours after establishing gas flow.
Second, the main portion of the carpet testing was done in an environmental chamber set
up to simulate building conditions. The environmental chamber was 20 nr (3.65 m by 2.4 m by
2.23 m). The chamber was maintained at an operating temperature range of 23 °C ± 1 °C and
a relative humidity range of 50% ± 5%. The ventilation rate of the chamber was held at 1.0 ±
0.1 per hour, and the loading ratio was 0.44 m2/m3. Fans were positioned to generate an average
air velocity above the floor of 9 cm/s, typical of building air. The environmental chamber
operated within the specified ranges throughout the experiment. Background concentrations of
the target compounds in the environmental chamber were determined during the 3 days prior to
the beginning of the experiment. Next, replicate chamber samples were conducted over a week-
long (168-hour) period. An inlet air sample was also collected. During the first day, sampling
was conducted at 1, 3, 6, and 12 hours. Subsequent samples were collected at 24 hour intervals.
Third, a field study was conducted in an occupied townhouse with a volume of about 400
m3 and a carpeted area of 93 nr. 12 and 24 hour samples were taken, and ventilation rates were
measured with an SF6 tracer. The carpet used was similar to Carpet 1.
Regarding sample collection and analysis, individual VOC's and total VOC's (TVOC)
were collected on multisorbent samplers packed with Tenax-TA, Ambersorb XE-340, and
activated charcoal, in series. Separate formaldehyde and acetaldehyde samples were collected
on Cj8 Sep-Fak cartridges impregnated with a 2,4-dinitorpheny[hydrazine solution. Samples were
then thermally desorbed and separated with a gas chromatograph system. A mass spectrometer
scanning a range of m/z 33-250 was used as the detector. In addition, an unseparated sample
was taken for measuring TVOC with a flame ionization detector. The separate formaldehyde and
acetaldehyde samples were analyzed using a liquid chromatograph and a diode-array UV detector.
A-5 9
-------
The screening measurements from the small chamber detected numerous compounds.
Many of these were identified based on comparison to standards, while others were tentatively
identified based on their spectra. Others were identified only by class, and some were not
identified at all. The 34 most abundant compounds found in the carpets were listed in the paper
(a reference for the complete report is also included). The screening measurements found that
Carpets 1 and 4, the two with the SBR latex adhesive, emitted primarily styrene and 4-PCH, and
also emitted 4-ethenylcyclohexene and alkyl benzenes. Carpet 3, the one with the PVC backing,
emitted mainly vinyl acetate, acetic acid, 2,2,4-trimethylpentane, 1,2-propanediol, and
2-ethyl-l-hexanol. Caxpet 2, with the polyurethane backing, emitted hexamethylcyclotrisiioxane,
1 butanol, dipropyiene glycol, methyl ethers, and 2,6-di-?err-butyl-4-methylphenol (BHT). Based
on the results of the preliminary screening, five to seven VOC's were selected for each carpet
for more detailed quantitative analysis in the environmental chamber.
In the environmental chamber measurements for Carpet 1, an SBR carpet, two separate
carpet samples were tested. The second sample was held in storage for 21 days longer than the
first, its storage bag was opened inside the chamber, and it was sampled in more detail during
the first six hours of measurement. The second sample had a higher initial styrene concentration,
but. based on measurement of the air inside the storage bag, the authors conclude that the
additional styrene is not due to the release of the air from the bag. They attribute it to analytical
uncertainty. For Carpet 1, the authors plot the concentration of styrene, 4-PCH, and 4-
ethenylcyclohexane over time. Curves are fit to the concentration data. An exponential decay
fits the data from 1 to 12 hours, but a power curve fits better from 24 to 168 hours. For Carpet
4, the other SBR carpet, initial emissions of styrene were much greater than Carpet 1, but other
results were similar.
For Carpet 2, with the polyurethane backing, initial concentrations of the target
compounds were relatively low. BHT was the highest, at 14 ppb, followed by 1-butanol, at 9
ppb. For Carpet 3, with the PVC backing, the authors present the time-dependent concentrations
of the five most important compounds — formaldehyde, vinyl acetate, 2,2,4-trimethylpentane, 1,2-
propanediol, and 2-ethyl-l-hexanol. The initial concentrations were 46, 290, 21, 120, and 8 ppb,
respectively. As with Carpet 1, thedecay was found to be exponential for the first 12 hours, and
found to fit power curves for 24 to 168 hours.
The authors note that, in general, the chamber concentrations decreased rapidly during the
first 12 hours, as described by the exponential curves used to fit the data. Decay coefficients' of
these curves were related to compound volatility; the most volatile compounds decayed most
rapidly. In addition to the time decay, the authors calculated "quasi steady-state specific
emission rates" for each compound at times of 24 hours and 168 hours. These were calculated
according to the following equation:
Quasi Steady-State Rate = Va(C-C0)/A
where V is chamber volume, a is average ventilation rate, C is average measured concentration,
C0 is average background concentration, and A is the carpet area. At 24 hours, the highest
emission rates were for vinyl acetate and 1,2-propanediol from Carpet 3. Other high rates
included styrene from Carpet 4, 4-PCH from Carpets 1 and 4, and BHT from Carpet 2. Carpet 3
A-60
-------
had the highest TVOC emission rate. They also compared ratios of the 168-hour quasi steady-
state emission rate to the 24-hour quasi steady-state emission rate and found that,the rates
decreased by over 60% for all compounds except 4-PCH, BHT, and the dipropylene glycol
methyl ethers. These last three compounds are less volatile, and their emissions are expected to
remain high for a longer time.
Also, the authors calculated total (integrated) specific mass emissions for the first 12
hours, and for the entire 168 hours. Again, Carpet 3 had the highest emissions of individual
compounds (vinyl acetate and 1,2-propanediol) and TVOC. They noted that the ratio of total
emissions for the two time periods were similar to the ratio of the quasi steady-state emissions
rates for the two time periods. They also noted that total mass emission of 4-PCH from the two
SBR carpets were very similar.
After reporting the results of the environmental chamber study, the authors compared
these concentrations with the small-chamber concentrations from the initial screening. Operation
rates were selected so that the ventilation rate/material loading ratio was the same for each
chamber. However, the difference in chamber geometry led to greater sorptive losses of several
compounds to the walls of the small chamber. They felt that this raised questions about the use
of small chambers to measure VOC emissions from carpets.
For the field study, the sample carpet was installed in the townhouse, and the doors and
windows were left open for several days. For this reason, the measured ventilation rate was high
and initial concentrations were low. When the ventilation rate decreased, concentrations
subsequently increased. The authors report initial 4-PCH emissions rates higher for the carpet
in the townhouse than for Carpets 1 or 4 in the environmental chamber, but emissions from these
three carpets eventually level off at similar values. Styrene emissions were similar for the
townhouse carpet and chamber Carpet 1, and were considerably lower than the 4-PCH emissions.
In the discussion section, the authors hypothesize about which specific ingredients or
components of the various carpets led to the emissions of the compounds. They note that in
Carpets 1 and 4, the 4-PCH, styrene, and several other compounds came from the SBR adhesive
backing. Other compounds were left from the textile finishing. In Carpet 3 they report that the
vinyl acetate and 2-ethyl-l-hexanol probably came from the secondary backing, while other
compounds may have come from carpet manufacture or from the unspecified adhesive. Most
compounds emitted from Carpet 2 were also probably residual from carpet manufacture.
Finally, the authors briefly discuss health effects. They focus on eight major compounds
found during the experiment: styrene, 4-PCH, formaldehyde, vinyl acetate, 2,2,4-trimethylpentane,
1,2-propanediol, 2-ethyl-l-hexanol, and BHT. Of these, formaldehyde is the best understood, and
is known to be an irritant at levels of 70 ppb. The maximum measured concentration was only
half this value, but may, when considered with other sources, lead to irritancy. Other possible
irritants may include vinyl acetate, 4-PCH, and BHT, but their irritancy levels are not well
quantified. In addition, the odor from 4-PCH, styrene, or vinyl acetate may produce a response
in some people. 4-PCH odor from SBR carpets would be the most persistent, and may linger
for several months. It is noted that TVOC emissions from carpet axe low relative to other TVOC
sources found in buildings, but if control of VOC from carpets is desired, it could be
A-61
-------
accomplished by airing out carpets for 12-24 hours prior to installation. This would result in
substantially less emissions for all except the slow-decaying compounds like 4-PCH.
6
A-62
-------
Inventory of Organic Emissions from Fossil Fuel Combustion for Power Generation. F, A-1394.
GCA Corporation for Electric Power Research Institute. Palo Alto, CA. April 1980.
This report presents a summary of organic emissions data from fossil-fuel-fired electric
power generation facilities. It examines the effects of combustion conditions, type of control
device, sample collection, sample preservation, and sample analysis on the reliability of the
reported data.
After going through the report the only material I could find that might be useful was the
conversion factors used for generating emissions for combustion sources. A variety of factors
can affect the operation of the combustion source and hence the composition of combustion
products emitted. Such factors include size, fuel feed rate, fuel homogeneity, boiler size, and
load during sampling. The fact that information on these factors is seldom included in literature
sources presents a major difficulty in attempting to compare organic emissions data. Authors use
emissions units including pg/Btu, ng/m3 stack gas, pg/kg coal, ng/g particulate, and ng/J. In
order to normalize all data to 10~5 ng/J fuel input, operating conditions were assumed to be that
of known conditions for similar boilers. The conversion 'proces is summarized below.
Conversion Factors for Emission Tables
To Convert To ng/J Multiply By
ug/Btu
9.48 x 10'4 Btu/J x 1000 ng/ug
ng/m3
9.48 x 10*4 Btu/J x 3 mVmin x a Btu/min
pg/kg
3.3 x 10'2 kg/J x 1000 ng/ug
ng/g
9.48 x 10'4 Btu/J x 1 m3/min x 1 g/m3 x J Btu/min
A-63
-------
Klemm, H.A, and R.J. Brcnnan. Emission Inventory for the SURE Region. EA-1913. GCA
Corporation for Electric Power Research Institute. Palo Alto, CA. April 1981.
BACKGROUND
The Sulfate Regional Experiment (SURE) is a major air pollution monitoring and
modeling study sponsored by the Electric Power Research Institute (EPRI). The study included
the eastern United States and parts of southern Canada. The primary purpose of the SURE
program was to define the mechanisms that link emissions of sulfur oxides (SOx), total .emitted
particles (TEP), nitrogen oxides (NOJ. and hydrocarbons (HC) to ambient concentrations of
sulfur dioxide (S02) and sulfates or to what is sometimes referred to as the "sulfate particulate
complex" (SPC). The report summarized here describes the phase of the SURE program
involved with the development of a detailed emissions inventory for the SURE region.
PREPARATION OF SEASONAL EMISSIONS INVENTORIES
Under Task 1 of the SURE program, inventories of all man-made emission sources were
prepared for each season of the year. In addition, the variations in emission rates between
weekdays and weekends and throughout the day were determined. Variations during the day.
were developed for the eight 3-hour periods in a day. Emissions estimates were prepared for five
sources: (1) electric utility; (2) industrial; (3) commercial; (4) residential; and (5) surface
transportation sources.
Point sources within major sources in the electric utility, industrial and commercial
sources were modeled as point sources. Major sources were defined as those plants whose total
emissions exceeded any of the following: 10,000 tons/year for SOx; 7,000 tons/year for
particulates; 3,000 tons/year for NOx; or 5,000 tons/year for HC. Emissions from the smaller
point sources as well as emissions from residential sources* and surface transportation were
aggregated into grid cells.
The task of developing the seasonal emissions inventories was divided into four subtasks:
(1) develop emission factors; (2).obtain existing emissions inventories; (3) review,.modify,
improve, and correct existing emissions inventories; and (4) calculate seasonal emissions.
Develop Emission Factors
Emission factors for particulates, SOs, NO,, and total HC were generally available.
Emission factors for specific species of these pollutant classes were not always readily available.
Literature searches and extensive contact with researchers were used to develop the required
emission factors. When sufficient test data were not available to develop emission factors for
specific sources, estimates were made based on chemistry, thermodynamics, and engineering
judgment.
A-64
-------
Obtain Existing Emissions Inventories
United States data were obtained from EPA's National Emissions Data System (NEDS).
In addition, Regional EPA Offices and State and local pollution control agencies were contacted
to update the NEDS point source inventories. Canadian government agencies were contacted to
obtain data for point sources in the provinces of Quebec and Ontario. A copy of the Ontario
point source inventory was available. Emissions data for the year 1976 aggregated by industry
and province were also obtained.
Review, Modify, Improve, and Correct Existing Emissions Inventories
This was the most important phase in preparing the seasonal emissions inventories. The
approach involved (1) the development of the basic U.S. point source emissions inventory;
(2) modifications, improvements, and corrections to major U.S. point sources; (3) development
of the Canadian point source inventory; and (4) development of the U.S. and Canadian area
source emission inventories.
In addition to the point source emissions inventory, an area source emissions inventory
was developed for the SURE. The area source emissions inventory was composed of two
components: the small point sources obtained from the basic United States and Ontario emissions
inventories, and the area sources which were not included in the point source inventory (i.e.,
residential and transportation sources).
Calculate Seasonal Emissions
The final step in developing the seasonal emissions inventories was (1) calculating the
emission rates for various component species, and (2) calculating seasonal, weekly and diurnal
variations in the emission rates. Point sources and area sources were handled differently because
different types of information were available for the two source types.
Point Sources
Emission rates for all electric utility and industrial point sources were adjusted on the
basis of source type and operating schedules. Annual emissions for particulates, SO,, NOx, and
HC were adjusted to seasonal rates and broken down into their component species. Annual
species emissions were calculated by determining what portion of the total emissions were
attributable to a particular operation.
. Much of the data necessary to develop seasonal emissions were available in the NEDS
database. Seasonal emission rates were determined by multiplying the seasonal fraction of annual
throughput on the NEDS files by the annual emissions. Average hourly seasonally adjusted
emission rates were calculated from
A-65
-------
_ . . , ' emissions(tons!season)
Emissions(tons/hour) ~
13 x hours/day x days/week
where hours/day and days/week are the normal operation of the source obtained from NEDS file.
Factors were also developed for each of the eight 3-hour periods throughout the day, as were
weekend and weekday factors. The number of days per week of operation was used to estimate
the weekday/weekend variations and the number of hours per day. of operation was used to
develop the diurnal variation. Design flow rates were similarly adjusted to reflect the actual
operation during any particular season from
_ , fT> - r-i \ Seasonal Throughput weeks/year
Seasonal Flow = (Design Flaw) x s-i- x —
Annual Design Throughput 13
Area Sources
For commercial and residential area sources, the temporal variation in emissions were
based on the known seasonal variations in degree days. For transportation-related emissions, the
variation was primarily a function of monthly fuel sales and the daily vehicle miles travelled
(VMT).
CORRELATION BETWEEN EMISSIONS AND EXTERNAL FACTORS
The purpose of this part of the emissions inventory program was to develop relationships
describing the influence of external parameters, such as meteorological conditions, on emission
rates by source category. The categories investigated included utilities, residential sources, and
transportation sources.
Development of Power Plant Correlations
The objective was to develop correlations between boiler emissions (or load) and
definable factors that influence the load. An initial analysis was-performed for the New England
area on a monthly basis and then for all months combined. A similar analysis was performed
for the State of Georgia.
The parameters investigated as having potential influence on plant average and peak loads
included (1) average daily wind speed, temperature, sky cover, and dewpoint; (2) 12:00 p.m. and
3:00 p.m. temperatures; (3) cooling and heating degree days; and (4) a weekend/weekday factor.
The analysis of the meteorological and load data was performed using stepwise multiple
regression techniques on six data sets resulting in the following equations:
A-66
-------
1. New England - Annual
Avg = 11,084 x (WE{WD) + 2,699 x (HDD) + 1,688 x (7)-4,023
Peak = 14,116 x (WE/WD) + 3,186 x (HDD) + 2,133 x (I)-2,094
2. New England - January
No significant fit was found
3. New England - July
Avg = 14,913 x (WE/WD) + 3,345 x (7) - 1,617 x (7^ + 802 x (DP) - 11,617
Peak = 16,845 x (WEI WD) + 1,721 x (T) + 9,334
4. Georgia - Annual
Avg = 2,379 x (CDD) - 416 x (T12) + 5,051 x (WE/WD) + 44,019
Peak = 2,615 x (CDD) - 393 x (TJ + 3,609 x (WE/flKD) +54,161
5. Georgia - January
Avg = -1,175 x (DP) + 2,316 x (7) - 837 x (T3) + 43,286
= -1,368 x (DP) + 3,238 x (7) - 1,367 x (F3) + 52,565
6. Georgia - July
Avg = 942 x (jy + 1,077 x (SC) + 17,623
Pea* = 1,127 x (T3) + 1,014 x (SQ + 23,800
where:
Avg
Peak
Average hourly 24-hour load (10 Btu/hr)
Peak load in a day (106 Btu/hr)
WE/WD
1 for weekday, 0 for weekend
HDD
CDD
Heating degree days fC), Minimum = 0
Cooling degree days (°C), Minimum = 0
-------
T = Average daily temperature (°C)
DP = Dewpoint (°C)
T3 = 3:00 p.m. temperature (°C)
T,2 = 12:00 p.m. temperature (°C)
SC = Sky cover (eights)
Development of Variations in Emissions for Residential Sources
Because the development of NEDS residential home, heating fuel use is based primarily
on degree day information, the same type of information was used to estimate variations in fuel
use on a diurnal and seasonal basis. To develop the seasonal and diurnal use patterns, a
representative meteorological site was chosen in each State and station summary data were
obtained for 1977.
An analysis was performed on monthly variation in degree days as a function of latitude
in order to determine whether the selection of one meteorological site in a State could give a
realistic picture of the fraction of degree days occurring in each month. The variation between
sites was considered reasonable and on this basis, it was determined that this method was
adequate to define the monthly, and thus seasonal, variation for a State.
Monthly averages of the 3-hour meteorological records were obtained for each of the
stations to determine the diurnal space heating fuel use degree day pattern. The resulting eight
values were proportional to the variation in diurnal heating for a selected station for a month.
Months were then averaged to obtain seasonally adjusted' diurnal factors for each State.
Seasonal factors were determined in a similar manner except that monthly average degree
day data were used. Monthly factors were averaged to obtain the seasonal variation in fuel use.
Development of Variations in Emissions for Transportation Sources
Monthly urban and rural VMT data were available for each State for 1977. Because the
data were broken out on a monthly basis, the task of developing seasonal variations in emissions
via travel patterns was straightforward. Monthly VMT were divided by annual VMT to obtain
monthly factors and these factors were subsequently averaged by season.
To determine the daily variation in VMT, a nationally averaged diumal traffic pattern was
developed and used for all States in all seasons because good data of this type were not available
on a State or seasonal basis.
For off-highway fuels, the assumption was made that the emissions occur between 6 a.m.
and 6 p.m., and for railroads the emissions are evenly distributed throughout the day. The
seasonal factor used for off-highway fuels was the same as that used for VMT.
A-68
-------
Lebowitz, L. and A.S. Ackerman. Flexible Regional Emissions Data System (FREDS)
Documentation for the 1980 NAPAP Emissions Inventory, EPA-60Q/7-87-025a. Alliance
Technologies Corporation under EPA Contract No, 68-02-3997. U.S. Environmental Protection
Agency, Research Triangle Park, NC. November 1987.
SUMMARY
The Flexible Regional Emissions Data System (FREDS) is a software system developed
in the early 1980's that's designed to process emissions data for input to regional acid deposition
and oxidant models. This earlier version of FREDS consisted of five modules which reformatted
data and applied allocation factors to the annual emission data and was used for the 1980 NAPAP
Emissions Inventory.
FREDS was also used for the 1985 NAPAP Emission Inventory and consisted of two
additional modules for a total of seven modules. For this project, the only module of interest is
the Temporal Allocation Module (TAM) which is essentially the same module in the 1985
FREDS NAPAP application. Also, the peripheral software that's used to assure data quality and
maintain the allocation files is essentially the same in the 1985 FREDS NAPAP application.
Therefore, the reader is referred to the summary of the 1985 FREDS NAPAP application by Lysa
Modica for the TAM summary.
A-69
-------
Lipfert, F.W. and R.E. Wyzga. "On the Spatial and Temporal Variability of Aerosol Acidity and
Sulfate Concentration. Air & Waste. Volume 43. No. 4, April 1993,
SUMMARY '
It has been proposed that acid aerosols be considered as a criteria air pollutant. Recent
research has shown high correlations between sulfate concentrations and acidity on both a
temporal basis at a local monitoring site and between certain sites in the same region. In this
study, relationships between S04= and H+ were examined based on extant data in locations with
high acidity, such as Southwestern Pennsylvania and with high population density, such as, New
York City. Both spatial variability in these relationships and temporal tracking at specific
locations were, examined.
Spatial Variability
Spatial variability in S04=, FT, and their ratio increases as the distance between monitoring
sites increases. For a one month sampling period, a high correlation (R = 0.95) was found in H+
and SO/ between two rural sites approximately 35 km apart. Spatial variability in H+ and SO/
was also determined to be minimal in a small town.
On a larger spatial scale, a high degree of similarity in mean values and correlations was
found between sulfate and acidity in White Plains, Albany, and Buffalo, New York over a 20
month period. All of the peak S04 periods and most of the peak acidity periods were seen
nearly simultaneously in all three locations.
On the Eastern United States, considerably more variability was found. The variation in
H7S0/ was approximately a factor of twelve due to spatial differences in atmospheric
neutralizing capacities. The highest degree of neutralization was seen in Newark, New Jersey
which was the most densely populated city sampled.
Temporal Variability
Temporal variability decreases as averaging time increases. Although temporal
correlations'between SO/ and H* are often quite high (R > 0.9), data from summer sampling
campaigns indicate that SO/ is not always a reliable predictor of H= for individual events at a
given site. For example, at two Pennsylvania sites, the molar ratio of H7SO/ varied from about
0.5 to 1.5 in the 18 to 29 ug/m3 SO/ range. One reason for this variability is that neutralization
increases as an air mass ages. Other data from Southwestern Pennsylvania indicated larger day-
night differences for H+ than for SO/.
Hourly measurements of H2S04 vary extensively; the ratio of HLSO+ZHT ranged from 0.18
to 0.72 for H+ > 3 ug/m3. Therefore, it is difficult to predict a detailed sulfate composition on
the basis of HP concentrations.'
A-70
-------
Long-term temporal variability indicates that a typical acidity monitoring record consists
of long periods of near-zero levels with occasional spikes of varying intensity and duration.
Therefore, it can be concluded that an annual average may have less relevance to health
responses than the statistics of the spikes.
Seasonal median S04= at several sites in New Jersey dropped by more than a factor of two
from 1988 to 1989 while W only decreased by about 25 percent Possibly, higher ambient
temperatures in 1988 increased photochemical activity but also released more ammonia from
natural sources.
CONCLUSIONS
The acidity and composition of sulfate aerosol are highly variable in both time and space.
However, for a given time period, the relationship between FT and S04= tends to be much more
consistent at a specific site or region than it is among sites, especially when the sites involve
large cities. Since a high correlation between H+ and S04= does not ensure reliable predictions
of acidity values based on measured sulfate, conclusions about the effects of H+ should not be
inferred from studies in which only S04 was measured.
A-71
-------
Modica, L.G. and D.R. Dulleba. The J 985 NAPAP Emissions Inventory: Development of Spatial
Allocation Factors. EPA-60O/7-89-0l0b. Alliance Technologies Corporation for U.S.
Environmental Protection Agency. April 1990.
The objectives of this document are to discuss the software used to generate spatial
allocation factors applicable to the 1985 NAPAP emissions inventory. The document contains
program descriptions and source codes, for programs executed for the 1985 spatial factor effort,
with input/output formats included. The allocation factor development process is briefly
described while documentation of the development is discussed in:
EPA-600/7-S9-012a;
EPA-600/7-88-024a; • .
EPA-600/7-85-035; and
EPA-600/7-89-010a,
Quality control procedures and resulting problems noted are also included in this document.
The flexible regional emissions data system (FREDS) is a software system which produces
an emissions inventory suitable for input to the regional acid deposition and oxidant models.
FREDS extracts emission data, pertinent modeling parameter and source identification
information from NEDS records or preprocessed data files, and applies appropriate temporal,
spatial, and pollutant species allocation factors to derive a gridded, speciated, and temporally
resolved emissions file. The spatial allocation module (SAM) accepts information from the
spatial allocation factor preprocessor (SAFP) output. SAFP contains SCC, state, county, column,
row, and a spatial surrogate value assigned to each grid cell. Spatial allocation is accomplished
by matching area source emission records to those in the spatial allocation factor file (SAFF) and
multiplying county-level area source emissions by their corresponding spatial fractions.
Spatial allocation methodology is based on two main sources' of data: U.S. Department
of Commerce, Bureau of the Census, Census of Population and Housing, 1980, and land use data
from the 1972-1973 Landsat satellite maps. Software programs were created to process the
Census and Landsat data. The programs provided spatial fractions which were combined to form
a data set containing state, county, column, row, and 14 spatial surrogates on each record.
Spatial allocation factors were developed using EPA's system, the IBM 3090. These factors were
matched with area source categories in the SAFP file. For the 1985 NAPAP allocation,
population, housing, urban land, agriculture land, composite forest, and land area category
surrogates were used. The SAFP output file is compatible with SAM of FREDS. The surrogate
assignments are based on activity level surrogates used for calculating area source emissions and
on previous assignments.
Several problems were noted with the results of the allocation factors. These problems
were that information on eleven counties was missing, four counties had missing grid cells,
several incorrect county code assignments resulted in nonmatches between land use and census
data and therefore incorrect county totals for the spatial surrogates, incorrect location data for two
Virginia counties, and several algorithms concerning column and row calculations were incorrect.
A-72
-------
This document discusses the computer computation of the spatial allocations for the 1985
NAPAP. Very little input data is given in the documentation. The majority of discussion
concerns the input and resulting output description. The computer code used for several of the
programs are included in the appendix section. Terry Wilson briefly looked at this document and
commented that reviewing the computer codes would not be beneficial. An analysis of the raw
data and the use of the data would reveal a lot. If the WAM was satisfied with their analysis,
then we can not argue with the results. Reference documents listed above may provide the raw
data which we seek.
A^73
-------
Ohio Trends Analyses Graphs of the Cirteria Pollutants (except PM10). Memo (with disks) from
Gary L. Engles, State of Ohio Environmental Protection Agency to Phil Marsosudiro, TRC
Environmental Corporation. May 14, 1993.
SUMMARY
This preliminary draft data presents Ohio ozone data for the 1972-1992 period. Trends
are presented and derived from this data to highlight ozone exceedances, and other indicators of
ozone air degradation. Several major Ohio metropolitan areas are presented in this data. No new
temporal allocation data is presented other than the day of the week the ozone exceedances
occurred.
A-74
-------
Pechan, E.H. An Air Emissions Analysis of Energy Projections for the Annual Report to
Congress. DOE/EIA-0102/16. U.S. Department of Energy, Energy Information Administration.
September 6, 1978.
SUMMARY
This report discusses the major factors involved with changes in air emissions from
energy related sources at the airshed (i.e., in various air quality control areas), regional and
national levels. This document provides only global projections of total air emission levels in
these areas for various pollutants as a result of changes in fuel use and economic growth. No
information at the facility level is given. The only type of allocation factor discussed is a factor
associated with predicting fuel use from a regional to an airshed level.
A-75
-------
Ryan, R. et. al. User's Manual for the Hourly Emissions Database for the Acid-Modes Field
Study. Final Report. Alliance Technologies Corporation for U.S. Environmental Protection
Agency under Contract No. 68-09-0M3. Research Triangle Park, NC. October 1990.
INTRODUCTION
The purpose of the manual summarized below is to document the development and use
of the Hourly Emissions Database for the Acid-Modes Field Study. This inventory provides •
real-time hourly emissions data for large sources of sulfur dioxide and nitrogen oxides in the
eastern United States and Canada. The real-time data are to be used in conjunction with the
1985 National Acid Precipitation Assessment Program (NAPAP) Emissions Inventory as the
emissions input for the verification runs of two acidic deposition computer simulation models.
METHODOLOGY AND CALCULATIONS
The purpose of this section is to explain the calculations and underlying assumptions
used to estimate the hourly emissions, hourly stack gas flowrates, and hourly temperatures. A
number of variations of the basic methodology were necessary due to the variety of formats
that were used by the participating companies.
Heat Input Estimates
All estimates of emissions were derived by taking the product of an emission factor and
some indicator of the activity level. Because the goal of the project was to provide actual
hourly emissions estimates, rather than disaggregated "typical" estimates, a readily available
measure of hourly activity level was needed. For electric utilities, the megawatt (MW)
generation on either a gross or net basis is usually accurately measured and recorded. These
megawatt-hours can be converted to total heat input to the boiler for each hour using heat rate
data supplied by the companies.
Each hour's heat rate is either interpolated from a table of megawatts versus heat rate
or calculated from a polynomial function of megawatts. For those cases where a table is
used, no extrapolation beyond the upper MW limit of the table is performed but extrapolation
is performed beyond the lower MW limit.
Emissions Estimates
Emission factors in units of pounds of pollutant per million British thermal units
(BTUs) of heat input were obtained in a variety of ways. Possibly the most accurate data
were from continuous emissions monitors (CEMs). These data were usually supplied with
hourly resolution, but daily or yearly averages were used for some boilers. Site-specific
emission factors obtained from a recent stack test were supplied for some boilers, either in
units of pounds per million BTU or pounds per ton of coal burned. In lieu of CEM data or a
site-specific emission factor, standard AP-42 emission factors were used.
A-76
-------
The AP-42 factors provide emissions estimates in units of pounds per ton of coal
burned, pounds per 1,000 gallons of oil burned, or pounds per million cubic feet of gas
burned. The factors for SO, and S04 also depend on the sulfur content of the coal or oil.
Fuel analysis results were therefore requested for each source in order to convert to a pounds
per million BTU basis. The frequency of these analyses varied from daily to weekly to
monthly. Some companies also felt that historical average values would be sufficiently
accurate, and reported just those averages. In some cases daily or weekly coal analyses
reflected "as received" values, rather than "as burned." A monthly average of these "as
received" values was used if the company indicated that their mode of operation precluded a
more precise assignment of the day or week that a particular fuel was burned.
Stack Parameter Estimates
Stack gas flows and temperatures were estimated for each hour from that hour's
megawatt load, and from the fixed design data provided at the beginning of the project by
each company. The fixed design data included estimates of flow and temperature at both full
load and half load. Linear interpolation was used to estimate the hourly values recorded in
the database. Data for half load flow or temperature were not provided by a number of
companies. Missing values were estimated from the average ratios of the half load to full
load values for those companies that did supply such data. On average, the flow at half load
was 57.2 percent of full load flow, and half load temperature was 95.4 percent of full load
temperature, expressed in degrees Rankine.
No extrapolation of temperature above full load or below half load was performed in
determining the hourly stack temperatures. The temperature versus megawatt curve therefore
"flattened out" at either end. Stack gas flow rate was also limited to being less than or equal
to the reported flow at full load, but extrapolation below half load was permitted. Because
the data provided by some companies would have led to the extrapolation of negative flow
rates at low loads, a minimum flow rate equal to 10 percent of the foil load flow was. used
unless the megawatt load was zero.
PROBLEM DATA
The database was comprised of fixed design and hourly operating data from 31 major
utilities, with a combined total of 382 units. CEM data were submitted by nine utilities,
either hourly or as a daily average. Daily averages were applied uniformly to the entire
24-hour period. For missing or unreadable data, averages were taken of values directly
preceding and following the point in question.
The database did not contain entries for all days for some boilers. The modelers
therefore used the 1985 typical values where those missing days were to be modeled. This
situation occurred only when data were not submitted for large blocks of time (i.e., at least
one month) before the Modeler's Emissions Files were created. For some plants, data were
often missing for several hours up to several days. Various methods were used to estimate
the missing data for these periods. The following discusses some of the most prevalent
problems.
A-77
-------
Hourly generation data for Associated Electric Corporation's Thomas Hill Unit 3 boiler
were not submitted for February 1990. The downtime periods for the unit were known,
however. A typical weekly pattern for this unit was therefore copied from the January
1990 period, and the known downtimes were then superimposed onto the pattern.
Hard copy data for several Detroit Edison units were impossible to read for periods of
from one to eight hours. An average of the preceding and following hours' data was
substituted in these cases.
The Tennessee Valley Authority data for June through September 1988 were based on
hourly generation data. Beginning in October 1988, only monthly total generation and
the outage periods were available for each TVA unit. A standard operating pattern was
used to allocate the monthly generation totals .to the operable time periods.
Daily total net plant generation was submitted for Illinois Power, along with a listing of
major outage time for each of the 3 boilers. Hourly generation was calculated for each
boiler using allocation factors from the State of Ohio, assuming that all operable boilers
would carry an equally distributed load:
Hourly Generation = Daily Total x Hourly Factor
No. of Boilers Operating
The monthly data submittal by Kentucky Utilities included monthly total net MW
generation in addition to a list of operating periods for each boiler. The monthly totals
were allocated to hourly net MW generations, utilizing allocation factors from the state
of Ohio:
Hourly Generation = Hourly Factor x Daily Factor x Monthly Total.
ZHourly Factors XDaily Factors
Boiler down-time was accounted for in the hourly factors by inserting a "zero" when
the boiler was not operating. Round-off error caused by inserted zeros resulted in a
slight difference (< 1 percent) between the submitted monthly total and that calculated
by summing hourly generations. To correct for this, the net hourly MW was adjusted
by a normalizing factor resulting in a calculated monthly total identical to that
submitted.
A-78
-------
Scire, J,S. and J, Chang. "Analysis of Historical Ozone Episodes in the SCCCAMP Region and
Comparison with SCCCAMP 1985 Field Study Data." Journal of Applied Meteorology.
American Meteorological Society. Volume 30. May 1991.
SUMMARY
This report presents the results of a comprehensive air quality and meteorological
monitoring project conducted in the Santa Barbara Channel and adjacent areas from Point Sal to
Point Dume during a five-week period in September-October 1985. The primary objective of
the 1985 SCCCAMP was to assemble a detailed database for use in the evaluation, refinement,
and application of photochemical dispersion models for estimating the impacts of offshore oil and
gas activities in the Santa Barbara Channel and surrounding areas on air quality. The 19S5
SCCCAMP data was also compared against a six-year historical ozone and meteorological
database to assess its representativeness.
This report documents the seasonal and diurnal behavior of ozone concentrations and
meteorological conditions in the SCCCAMP region during the study period. Statistical analysis
were performed to evaluate the diurnal and seasonal behaviors of ozone in this study area. The
diurnal periods of analysis were hourly and the seasonal periods of analysis were the fall, spring
and summer. No temporal allocation approaches or techniques were used since most of the
diurnal data was from air monitoring stations in the SCCCACMP region which recorded hourly
ozone concentrations.
A-79
-------
Sellars, F.M. et. at. National Acid Precipitation Assessment Program Emission Inventory
Allocation Factors. EPA-600/7-85-035. GCA Corporation under EPA Contract No. 68-02-3698.
U.S. Environmental Protection Agency. Research Triangle Park, NC. September 1985.
SUMMARY
This summary contains information taken from Sections 1 and 2 of the report. In order
to develop an Eulerian model input tape (for applications in an. atmospheric model) allocation
factors were developed to apportion the National Acid Precipitation Assessment Program
(NAPAP) annual emission inventory for point and area sources into gridded, hourly emission
totals with emissions of NOx split into NO and NO, components. Components of VOCs were
resolved into 10 photochemical reactivity classes. The NAPAP inventory is maintained by EPA's
emission inventory system (EIS). The EIS is not formatted to allow the inventory data to be used
directly by the Eulerian acid deposition model. Therefore, the data must be converted into a
compatible file format. This is accomplished through the use of a preprocessor called the
Regional Model Data handling System (RMDHS).
The RMHDS can apportion the annual emissions inventory data from the NAPAP
inventory into hourly totals for a typical weekday for any season. A program within RMHDS
called TPSPLIT searches the temporal factor file and assigns an appropriate temporal distribution
depending on the specific area source category or plant or process under investigation.
Functional multipliers are applied to the annual emissions totals to convert the emissions to
quarterly emissions for a particular season with subsequent conversion to a daily total for a
typical weekday (Monday through Friday). Daily totals are then apportioned to hourly rates by
dividing total daily emission by one of 24 hourly fractions representing an entire diurnal pattern
or by the number of hours representing a standard work day.
Temporal allocation factors had been developed for area and point sources in the.
Northeast Corridor Regional Modeling Project (NECRMP) EPA-450-/4-82-013q (see reference
2 contained on pg. 24 of text). Also, many of the temporal allocation factors developed for
NAPAP were derived, from the NECRMP study. As such, some area source categories are
sometimes broken down into several, more refined categories. The document provides the
example of partitioning onsite incineration into three subcategories: (1) industrial, (2)
commercial/institutional; and (3) residential onsite incineration (see Table 1 - pg. 5). Also,
temporal allocation factors for area sources are taken directly from the temporal variation of the
activity or phenomenon causing the emission to occur. The activity or phenomenon causing the
variation could be the activity itself (e.g., railroad traffic variations causing variations in railroad
emissions) or some factor such as temperature which cause variations in emissions from
residential heaters. For area sources, activities which cause such variations in emissions to occur
may not vary much from state to state. Therefore, the temporal allocation factors may be
specified on a countywide, statewide, or regionwide level.
Brief descriptions of the development of temporal allocation factors for the various
source categories represented in the NAPAP inventory are contained in the following paragraphs,
as previously mentioned methods used in the NECRMP study were used to develop the temporal
allocation factors for many of the source categories discussed below.
A-80
-------
Area Source Fuel Combustion
This source category was divided into residential, commercial/institutional, and industrial
subcategories. The combustion of all fuels within any of these subcategories were assumed to
be represented by the same temporal pattern. Temporal allocation factors for each of these
subcategories are explained below,
residential fuel combustion - temporal allocation factors were developed from the
methodolgoy used in the Sulfate Regional Experiment (SURE) as follows. Seasonal
patterns were developed from average monthly temperatures and monthly averages of 3-
hour meteorological records were used to develop season-specific, hourly patterns for each
State in the Eastern U.S. Seasonal temporal factors were derived from the average
monthly temperatures obtained from state, regional, and national monthly and seasonal
heating degree days weighted by population. Hourly variations in residential fuel use
were derived from data contained in climatic data from the National Oceanic and
Atmospheric Administration (NO A A). Both the seasonal and hourly data reflect 1980
measurement data.
• commercial!institutional fuel use - all subregional daily and hourly patterns used were
derived from the Procedures for the Preparation of Emission Inventories for Volatile
Organic Compounds; Volume II-Emission Inventory Requirements for Photochemical Air
Quality Simulation Models. EPA-450/4-79-018. September 1979.
~ industrial fuel use - all seasonal, daily and hourly patterns were developed from EPA-
450/4-79-018 and from emissions information contained in the Emission Inventories for
Urban Airshed Model Application in the Philadelphia AQCR, March 1981, (see reference
7 on pg. 25 of the text)
\
Onsite Incineration/Open Burning
This source category was divided into residential, commercial/institutional, and industrial
subcategories. The temporal factors were derived from existing inventories (i.e., the SURE
Region, the St. Louis AQCR, Tulsa Oklahoma, State of New Jersey, and the State of New York,
New York Metropolitan Region; the references for these are contained on pages 24 and 25 of the
text).. Specifically, these inventories assumed a uniform seasonal distribution for both onsite
incineration and open burning occurring between the hours of 5 a.m. to 8 p.m., 5 to 6 days per
week. This text notes that a refinement to these assumptions needs to be made since temporal
variations of each of these subcategories exists.
Highway Vehicles
The category for highway vehicles was divided into eight subcategories. These
subcategories reflect activities on rural and urban roads for light duty, medium duty and heavy
duty gasoline powered vehicles and heavy duty diesel powered vehicles. For all subcategories,
uniform seasonal distributions were used for all States and separate hourly patterns for gas and
diesel vehicles were developed from regional studies (see references #s 7 and 13 contained on
A-81
-------
page 25 of the text). Seasonal variations of emissions form these vehicles occur more as a non-
linear function of temperature than as a linear function of activity level. The text noted that the
RMDHS does not permit adjusting temporal allocation factors on a county level and a better data
handling system for this purpose needs to be developed,
Off-Highway Vehicles
For off-highway vehicles, hourly patterns were developed separately for gasoline and
diesel powered off-highway vehicles for five subcategories: (1) agricultural equipment; (2)
construction equipment; (3) industrial equipment; (4) lawn and garden equipment; and (5)
motorcycles. Developing these hourly patterns was first done by developing a single composite
hourly pattern for all five subcategories using data from Table 6-6 of EPA-450/4-79-018 and
from an EPA study entitled, Regional Air Pollution Study; Off-Highway Mobile Source Emission
Inventory, EPA-600-4-77-041 (the reference is contained on pg. 25 of the text). Then, using
information obtained from Table MF-24 of Highway Statistics, 1980 and a combination of
information from area source emission inventories for Delaware, New Hampshire, Ohio and
Virginia (references for all of these sources are contained on pg, 25 of the text), percentages of
gasoline and diesel usage were apportioned for each of the five subcategories. In so doing,
nationwide daily variations for both gasoline and diesel powered off-highway vehicles for each
of these five subcategories were developed as a composite of daily patterns (weighted by
' emissions strength).
Railroads
Temporal allocation factors for railroads were developed from factors contained in EPA-
450/4-79-018 for States not contained in the NECRMP study. For States contained in the
NECRMP study area, information provided by Conrail and the Philadelphia AQCR were used
to develop these factors.
Aircraft
Commercial aircraft were assumed to operate 7 days per week, 24 hours per day with 90
percent of such aircraft operating between 6 a.m. and midnight. Data obtained from Seasonally
Adjusted Traffic and Capacity, for 1980 (see reference 22 contained on pg 26 of the text) were
used to compute the temporal allocation factors. Hourly trends for military aircraft were
developed from information obtained from the NECRMP study. For civil aircraft, temporal
factors were developed nationwide. Other references used for developing temporal factors for
aircraft include EPA-450/4-79-018, data from the Civil Aeronautics Board and several area source
emission inventories (see reference numbers 3, 6, 7, 11 and 22 starting on pg. 24 of the text).
Vessels
Information contained in EPA-450/4-79-018 was used to develop temporal allocation
factors for both diesel oil and residual oil vessels. In general, seasonal allocation factors are
based on the number of months in which the mean temperature exceeds 45°F. State-specific
temperature data was obtained using State, Regional, and- National Monthly and Annual Average
A-82
-------
Temperatures Weighted by Area (see reference 23 contained on pg. 26 of the text). Daily
allocation factors were developed from EPA-450/4-79-018 for gasoline vessels known as pleasure
craft and these factors are used nationwide.
Solvents Purchase
This area source category is a composite of the following categories: (1) small
industrial/commercial degreasing; (2) dry cleaning; (3) architectural surface coating; (4) autobody
refinishine; (5) small industrial surface coating; (6) graphics art; (6) commercial/consumer solvent
use; (7) cutback asphalt; and (8) pesticides. A composite temporal allocation factor was produced
for all of these categories by calculating a weighted average hourly pattern. Pesticides were
ignored during the weighting process since these sources only represent 3 percent of the total
annual VOC emissions. Annual VOC emissions were used in the weighting process and this data
was taken form Table 8 of NECRMP - Volume XVII, EPA-450/4-82-013r (see reference 24
contained on pg. 26 of text).
Gasoline Marketed
Hourly temporal allocation factors used in the NAPAP study were developed from the
NECRMP study using information contained in EPA-450/4-79-018 for stage I gasoline
evaporation, stage II gasoline evaporation storage tank breathing and gasoline loading/transit
operations. The daily and hourly patterns obtained were weighted on the basis of category
VOCs listed in EPA-450/4-82-013r (see reference 24 contained on pg. 26 of the text). The
composite temporal allocation factor developed for this category was used for all states.
Agricultural, Structural, and Forest Fires
Within this category, managed burning and agricultural field burning were assumed to
occur 7 days per week, between 5 a.m. to 8 p.m and no burning occurs during the oxidant
season. The same field/slash burn patterns used in the NECRMP study were also used in
NAPAP. Both structural and forest fixes occur randomly, 7 days per week, 24 hours per day.
However, such fires are represented uniformly since RMDHS cannot handle random distributions.
About, 90 percent of forest fire activity is estimated to occur during the summer or fall and 10
percent occurs evenly between the winter and spring seasons of the year.
Manure Field Application
Due to a lack of study data, manure field applications for beef, dairy cows, hogs and pigs,
broilers, and other chickens are assumed to be uniformly distributed. The same assumption is-
also used for beef cattle feed lots.
Anhydrous NH3 Fertilizer
Hourly averages of monthly N02 emissions are contained in Natural and Fertilizer-
Induced Emissions of Nitrous Oxide from Soils, 1980 (see reference 25 contained on pg. 26 of
the text). It should be noted that this information does not represent the true temporal variability
A-83
-------
in emissions of XH, from fertilizer applications and the text recommends that a more accurate
representation be located.
Minor Point Sources
This category represents ground level sources emitting less than 300 tons per year (TPY)
each of N02 and VOC. The schedule assumes 52 weeks per year, 5 days per week and 8 hours
per day of operation beginning at 7 a.m.
Power Plants
Temporal allocation factors for power plants were developed using hourly power plant
fuel use data collected during the SURE study (see reference 4 contained on page 24 of the text).
A-84
-------
Solomon, p.A. et. al. "Spatial and Temporal Distribution of Atmospheric Nitric Acid and
Particulate Nitrate Concentrations in the Los Angeles Area." Environment, Science and
Technology» Vol. 26, No. 8, 1992.
(This summary contains applicable segments directly lifted from the original paper. This should
be remembered when referencing the given information).
Atmospheric oxides of nitrogen, including nitric acid vapor, are major contributors to the
dry deposition flux of strong acids to the earth's surface in Southern California. Nitric acid
vapors also may react with ammonia to produce visibility-reducing fine aerosol nitrate and with
sea salt or soil dust particles to produce coarse particle nitrates. This paper describes the results
of a 1-year field experiment in which HN03 plus fine and coarse particle nitrates were measured
throughout the Los Angeles area. Measurements made in the urban area will be compared to
observations at a remote background site on an offshore island and to data taken at a high
elevation receptor site in the mountains downwind of the metropolitan area. Spatial and temporal
trends in pollutant concentrations will be discussed in light of the transport patterns and
atmospheric processes that govern the HNtyaerosol nitrate system.
During the calendar year 1986, a monitoring network designed to measure gas-phase
HN03 and atmospheric particulate matter was operated at nine sampling sites located throughout
the South Coast Air Basin (SOCAB), which surrounds the Los Angeles area. All sites except
Tan bark Flats and San Nicholas Island were co-located with present South Coast Air Quality
management District (SCAQMD) continuous air-monitoring stations. The Tanbark Flats site was
located in the mountains north of San Dimas, in the Angeles national Forest, at an elevation of
approximately 870 m. This site was chosen to determine the concentration of acidic pollutants
present in the national forests to the north of Los Angeles, and because air quality modeling
calculations suggest that nitric acid concentrations may be different at higher elevations than are
observed near the ground within the urban area. The ninth site was located at the meteorological
station on San Nicholas Island (SNI), approximately 140 km southwest of the Los Angeles
coastline. This remote, off-shoe location was chosen to determine background pollutant levels
present in the marine environment upwind of Los Angeles.
RESULTS AND DISCUSSION
Nitric acid and aerosol nitrate concentration patterns in the Los Angeles area have been
examined previously by means of short-term photochemical air quality models that are capable
of explaining the observed concentrations patterns on a cause and effect basis. The findings of
the present long-term monitoring study are, consistent with these prior air quality model
predictions. These modeling results indicate that the highest N02 concentrations accumulate near
the coast in the western portion of the air basin overnight and during the early morning hours.
As the day proceeds, NO and N02 typically are advected eastward across the air basin; N02 is
oxidized to form nitric acid, and high nitric acid concentrations are predicted to occur in the
middle portion of the airbasin (e.g., at Bur bank and Upland in the present study). As this nitric
acid-laden air mass passes over the Chino dairy area (just to the west of Rubidoux), very large
amounts of ammonia are injected into the atmosphere from livestock waste decomposition and
A-85
-------
from other agricultural activities. Ammonia measurements made at the same time as the present
study show a 1986 annual average NH3 concentration of 30 pg m"3 downwind of Chino at
Rubidoux. This is approximately 10 times higher than the NH3 concentrations measured at
upwind sites in the western portion of the air basin during 1986. The available nitric acid reacts
to form large amounts of ammonium nitrate aerosol, resulting in exceptionally high aerosol
nitrate concentrations and.the low HNO, levels measured farther downwind at Rubidoux. The
modeling study of Russel et al. predicts that Rubidoux should have the highest aerosol nitrate
concentrations and the lowest HN03 concentrations of any routine monitoring site in the air basin
during 1982 summertime conditions. The present monitoring study shows this was a year-round
condition at Rubidoux during the year 1986.
Monthly average time series graphs of HN03, fine- and coarse- particle nitrate and total,
inorganic nitrate at Hawthorne, Downtown los Angeles, and Upland are illustrated in Figure 7,
Inorganic nitrate production almost always is in great excess of the amount of HN03 in the
atmosphere, with average fine-particle nitrate concentrations in winter often higher than in the
summer. These results suggest that the pronounced summertime peak in HN03 observed at
inland sites is apparently due to factors governing the partition of inorganic nitrate between the
gas and aerosol phases in addition to the expected increase in production of inorganic nitrate
during the summer as compared to the winter. Based on thermodynamic considerations, it is-
predicted that atmospheric HNO, and NH3 often are in equilibrium with NH,N03 aerosol and that
the equilibrium dissociation constant for NH4N03 often governs the concentration product of
HN03 times NH3 in the atmosphere and the partition of inorganic nitrate between the gas and
aerosol phases. The NH3-HN03-NH4N03 equilibrium condition is very sensitive to temperature,
with greatly increased ambient HN03 concentrations predicted to be in the gas phase at higher
ambient temperature. Aerosol NH4N03 formation also is sensitive to the absolute magnitude of
the concurrently observed NH3 concentrations, and measured NH3 concentrations at most of the
monitoring sites studied here are lowest during summer months. Therefore, it is likely that the
broad summer seasonal HN03 peak observed at inland low elevation monitoring sites (e.g.,
Burbarik, Upland) results from the higher summer temperatures and lower summer NH,
concentrations that shift the NH4No3-HN03-NH3 equilibrium toward the aerosol phase. Nearer
to the coast, seasonal temperature extremes are moderated by the presence of the ocean, NH,
concentrations are lower, and there is little seasonal pattern in the HN03 concentrations.
Coarse-particle nitrate concentrations during 1986 were found to be comparable in many
cases to the fine-particle nitrate concentrations observed. Chemical analysis of samples collected
in the Los Angeles area and elsewhere have shown that the coarse-particle nitrates are largely
composed of the non-volatile reaction products of HN03 with sea salt or soil dust, while the fine-
particle nitrates consist largely of NH4N03 which may dissociate to release HN03 and NH3.!
Coarse-particle nitrate formation has been examined via photochemical modeling calculations in
which nitric acid transport to the surface of the sea salt or soil dust particle is the governing
factor limiting coarse-particle nitrate formation. As seen in Figure 7, coarse particle nitrate
concentrations display approximately the same seasonal variations as HN03 concentrations, with
a flat seasonal distribution near the coast and a summer seasonal peak at inland sites such as
Burbarik and Upland. This is consistent with the pattern expected if coarse-particle nitrate
formation is limited by HN"03 diffusion to an existing coarse-particle surface that acts as an
irreversible sink for HNO,; coarse-particle nitrate formation is driven by the availability of HN03
A-86
-------
in the gas phase. Coarse-nitrate concentrations also may be affected be sedimentation losses of
large particles, particularly in winter months when resultant wind speed are slow and retention
times for air masses in the Los Angeles airshed increase.
The highest HN03 levels observed in this air basin in 1986 occurred at the high elevation
monitoring site at Tanbark Flats, while total inorganic nitrate concentrations at that site are much
lower than at the nearest monitoring stations on the populated valleys below. The most likely
explanation for this increased HN03 concentration at high elevation in the presence of lower
total inorganic nitrate concentrations is that aerosol nitrate formation has been suppressed.
Previous modeling studies have shown that ammonia concentrations should be much higher at
low elevations at night and in the early morning hours than at higher elevations. This is because
NH-, is released principally from the ground-level sources located on the floor of the air basin.
Nitrogen oxides emissions may be produced by both ground-level sources and elevated sources.
Hence, there is the possibility that the ratio of NOx to NH3 may vary with elevation in the
atmosphere. Ambient NH3 measurements made concurrently with the present study show that'
annual average NH3 concentrations at Tanbark Flats are very low (only 0.6pg m"3), compared to
2.1-4.4 ug m"3 at most other urban sites and 30 ug m" at Rubidoux. The high HNO}
concentrations observed at Tanbark Flats when compared to the other sites are consistent with
suppression of N"H4N03 formation due to the near absence of available ammonia in the gas phase.
A-87
-------
Users Guide for the Urban Airshed Model, Volume IV: Users Manual for the Emissions
Preprocessor System 2.0. EPA-450/4-90-007D, - U.S. Environmental Protection Agency.
- Section 2.3 Data Requirements
Five Categories of Input-
Emissions Data
System Default Parameters
Region-specific data
¦ Episode-specific data
Optional data
Emissions Data - ' input in two distinct files: point sources and area sources.
These emissions data must be formatted according to the
AFS work file format or the AMS work file format These
formats are found in Tables B-l and B-2 in Appendix B.
The information in the AFS work file can be divided into
seven data types: inventory description, geographical,
source identification, stack characterization, operating
schedule information, control technology description, and
emissions.
Data items contained in the AMS work file are for area and
mobile sources. The information in the AMS work file can
be divided into six categories: inventory description,
geographical, source identification, control technology
description, and emissions.
EPS 2.0 comes with a set of files containing default inputs
for certain data, which are intended to provide the user with
an initial EPS setup. These files include:
Speciation profiles for chemically allocating the
emissions to the carbon bond species used by the
UAM (derived from data in EPA's AIR Emissions
Speciation Manual);
Default speciation profile code assignments by
Source Classification Code (SCC) and Area Source
Category (ASC);
Default reporting code assignments (process, activity
control, and pod) by SIC/SCC(ASC);
Default reporting code descriptions;
System Default
Parameters
A-88
-------
National average temporal allocation profiles;
Default temporal profile code assignments by
SCC(ASC);
Economic and demographic projection data for
developing projection factors (from the Bureau of
Economic Affairs).
The user must specify the region to be modeled the
/UAMREGN7 packet of the global US ERIN user-input file.
Items the user must supply include
The domain reference .origin (in UTM coordinates),
Thp UTM 7nn?
i liV W iiYi cUl lVi
The grid origin (in meters from the reference
origin), and
Grid cell size- and number of grid cells in the X and
Y directions.
The episode to be modeled must be defined in the
/EPISODE/ packet of the US ERIN global user-input file.
The data in this packet include the number and names of
the Carbon Bond Mechanism species that will be included
in the final UAM-formatted emissions file, as well as the
beginning and ending Julian data and time for the modeling
episode.
In addition to the episode definition, hourly temperatures
for the episode are required to be able to adjust onroad
motor vehicle emissions to levels appropriate to episode
conditions. (Note that the hourly temperature data are input
to the MOBILE model to generate hourly adjustment
factors, and are not input directly to the EPS 2.0 system.)
Optional inputs to the EPS systems include - link-based
motor vehicle emissions data, source-specific temporal and
chemical composition data, and special inventories.
A-89
-------
Specification of Level
of Detail in Emissions
Data - EPS 2.0 allows the user to input both source-specific inputs.
and default inputs without the need for intensive data
manipulation and processing. For example, if detailed
specification data axe available for only a single source in
the modeling domain, the EPS allows the data to be used
without requiring that all sources have data at the same
level of detail
In general, one or more of the following fields is used to
identify an emissions source;: HPS state/county code,
subregion code, SCC(ASC), SIC, and plant and stack
identification codes. If the input data are to be applied to
a specific source, then all fields that are applicable are
specified. However, for an overall application of the input,
the specific fields are either left blank or filled with zeros.
A FTPS code of 00000 can be used to indicate data for the
entire modeling region. Table 2-4 (EPS User's Guide Vol'.
IV; Part A) lists the identification fields necessary for
varying levels of detail
US ERIN, The EPS 2.0
User Input File - Each of the core EPS 2.0 modules accesses the user input
file. US ERIN. The USERIN file consists of a number of
sections, referred to as "packets". Each packet begins with
an identifier keyword enclosed in brackets (actually they're
foreword slashes 7') and ends with the keyword "/END/".
The packets, which may be specified in any order within
the USERIN file, contain input parameters that are
inventory specific; output options, default parameters,
region description, modeling episode description, etc. Each
core model in the EPS 2.0 system requires a packet in
USERIN defining the input parameters specific to the
module.
In addition to the module-specific packets, USERIN
contains three packets that describe the' modeling episode
and domain.
/EPISODE/ This packets defines the time period of the
episode and the species list used for the
speciation of emissions into the Carbon
Bond Mechanism species. For consistency
the UAM, the contents of the packet are
identical to the CONTROL packet used by
all of the UAM preprocessor; accordingly,
many of the parameters included in
/EPISODE/ are not actually required in the
A-90
-------
EPS 2.0 modules. Example attached,
Exhibit 2-1.
/UAMREGN/ This packet defines the modeling region and
grid resolution. The contents of this packet
are identical to the REGION packet used by
all of the UAM preprocessors. The
parameters defining the number of cells in
the modeling grid, cell size, and modeling
domain are found in this packet. Example
attached, Exhibit 2-2.
/COUNTY/ This packet contains a list of the counties in
the region of interest. Data from any county
other than those in the county list will not be
processed. Each county is identified by the
state/county FIPS code. The first twenty
columns of each county line are optional and
can be used to specify the name of each
county. Each line also contains fields for
specifying design values for ozone, carbon
monoxide, and PM-10. Example attached.
Exhibit 2-3.
A-91
-------
Veldt, C. Emissions of SOx, NOx, VOC and CO From East European Countries. Air and Waste
Management Association, 1991.
This article describes the first attempt to do an emission inventory for eastern Europe.
All of Europe was included in the inventory however, eastern European emissions were estimated
due to lack of sufficient information. The focus of the paper is the construction of a database
to be used for model calculations of long-term ozone levels. Much of the information contained
in the paper is of little value to our project. However, there is a section dealing with temporal
distribution of emissions. This section will be summarized more thoroughly.
Statistical data is required to assess time patterns of emissions. This data was collected
for source types like utilities and road transport but not from industry and solvent losses. The
estimation of time dependencies, month, day and hour factors can be used. They are defined by
XFm - 12, 2Fa - 7, and XFh = 24. An emission at any hour can be derived from the
corresponding mean annual emission E with E/8760 x Fm x Fd x Fh. It was decided to use a
uniform time pattern for the whole model area, with climatic differences being expressed by
different fuel consumption rates, for the projects first phase. Some generalizations and
assumptions were made about stationary combustion sources; 40% of fuel consumption for area
sources is used by utilities, 2/3 of industrial fuel consumption is used continuously while the rest
is used during work days, and for utilities F^ = 0.9 from April to September and Fm = 1.1 during
the remainder of the months, Fh = 1.1 between 0700 and 1900 h and Fh = 0,9 during the
remaining hours. Petroleum refineries and chemical production processes were assumed to emit
continuously. Solvent emissions were assumed to originate for 2/3 from the'discontinuously
operating industry and for 1/3 from non-industrial applications and for the most part occurring
during daytime. Below is a table giving the temporal allocation factors;
Month Factor
Day Factor
Hour Factor
April-
Sept.
Other
Mon -
Fri
Sa,Su
700-
1900
Other
Stat, Comb, Source
Point Source
0.9
1.1
1.06
0.85
1.1
0.9
Area source
- Util + Ind
0.96
1.04
1.08
0.8
1,24
0.76
- Non Ind
0.45
1.55
1.0
1.0
1.5
0.5
- all Sources
0.8
1.2
1.06
0,85
1.3
0.7
Petroleum Refin
1.0
1.0
1.0
1.0
1.0
1.0
Chemical Process
1.0
1.0
1.0
1.0
1.0
1.0
Transportation
1.0
1.0
1.0
1.0
1.8
0.2
Solvent Evaporation
1.0
1.0
1.26
0,35
1.9
0.1
A-92
-------
Wilkerson, G, SF6 Tracer Studies, The Lake Michigan Ozone Study 1991 Simmer Field
Program. North American Weather Consultants, Salt Lake City, Utah, for Lake Michigan Air
Directors Consortium, December 1991.
SUMMARY
Two tracer studies were performed during the summer of 1991 in the Lake Michigan area.
The first type of tracer experiment involved the use of a ground based release of sulfur
hexafluoride (SF6) at a shoreline location at least one hour after the onset of a well-defined lake
breeze. The second type of tracer experiment consisted of an inland release point very much to
the west of the lake front breeze.
Only two tracer experiments were conducted (both shoreline releases) due to the limited
number of true lake breeze situations. Meteorological conditions were much more favorable
during the second tracer experiment as a convergence zone developed along the lake breeze front.
Purpose of Experiments
The primary purpose for these studies was to provide dispersion and transport data to
assist in validation of a mcsoscale meteorological model This model will be used in forecasting
complex transport and diffusion processes occurring over Lake Michigan and near its western
shorelines.
Data Analysis
The following four data sets were created during the tracer studies:
SF6 data measured from the aircraft,
SF6 data measured from the mobile van,
SFs data measured from the array of syringe samplers, and
SF6 release data obtained from the mass flow meter used at the SF6 release site.
During field measurements, aircraft and mobile van data were collected each second,
Since much of the SFh plume structure would have disappeared if a ten second average had been
applied, averages were not applied to these data. Aircraft data were subjected to a screening
program which checked for periods of constant values, values outside of a preset range, and
spikes. Mobile van data were inspected visually for outliers.
Tracer Experiment Summaries
The tracer studies consisted of the following four main components: the SF6 release
system, an aircraft measuring platform, a sampling van, and an array of stationary programmable
syringe samplers. SF6 (in a gaseous state) was released from cylinders resulting in relatively
simple release techniques and accurate measurements of release rates.
A-93
-------
Continuous SF6 analyzers were used in the aircraft and van during both experiments.
Using a continuous SF6 analyzer in an aircraft allowed for the real-time observation of actual
plume trajectory and dimensions, as well as, quantitative concentrations within the SF6 plume.
The SF6 analyzer used had a fast response (time constant < 1 second) and could measure SF5
concentrations as low as 10 parts-per-trillion (ppt). Also, multiple, stationary, programmable
syringe samplers were used at selected sampling sites.
Both the sampling aircraft and van were able to locate the SF6 plume in both tracer
experiments. However, due to the wind direction on the first experiment and the lack of
transport time to reach the sampling network on the second experiment, the stationary syringe
samplers did not detect any SFfi. The dimensions associated with each experiment are given
below.
DIMENSIONS OF EXPERIMENTS
First Tracer Experiment
Second Tracer Experiment
Sixteen plume transects were conducted.
Seventeen plume transects were conducted.
East-west transects were conducted at
approximately 10 and 30 kilometers
downwind from the SF6 release site.
Multiple east-west transects were conducted
at approximately 15 and 50 kilometers
downwind from the SF6 release site.
Transects were conducted over 14 altitude
levels, ranging from approximately 400 to
1650 meters msl.
Transects were conducted over 11 altitude
levels, ranging from approximately 400 to
1650 meters msl.
Plume widths ranged from 2 kilometers,
close to the release site, to nearly 20
kilometers far out from the release site.
Plume widths ranged from 5 kilometers
above the marine air mass to over 20 *
kilometers within the marine air mass.
Conclusions
The two tracer studies were successfully performed and provided valuable information
which can be used to stress the models that will be used for the Lake Michigan area. Transport
and dispersion characteristics noted during the two experiments will provide a foundation on
which to build a working understanding of the complex flow patterns in the area.
Flow patterns observed from the shoreline SF6 release indicated that the plume appeared
to mix through the marine layer and up into the synoptic, flow. In addition, pollutant
measurements indicated an area of increased 03 concentration near and over Lake Michigan.
Since weather conditions under which the tracer experiments were conducted were not
typical, transport and diffusion processes that would occur under "classic" lake breeze conditions
may not have been accurately described. In addition, since the inland SF6 release was not
performed, transport and dispersion of a source outside of the marine air mass was not examined.
A-94
-------
TRC
Environmental Solutions through Technology
TRC Environmental Corporation
100 Europa Drive, Suite 150
Chapel Hill, NC 275 U
-a (919) 968-9900 Fax [919) 968-7557
MEMORANDUM
TO: Charles Mann
FROM: Ritchie Buschow and Theresa Kemmer Moody
DATE: . May 27, 1993
SUBJECT: Results of Telephone Interviews conducted for the Identification of Allocation
Methodologies and Available Data
EPA Contract No. 68-D9-0173, WA No. 3/314, Task 2
TRC Reference Number 1-637-314-3
1.0 INTRODUCTION
The purpose of this memorandum is to present the results of information received as a
result of telephone contacts made under Task 2 of the above referenced Work Assignment, In
general, Task 2 required TRC to contact (by telephone) technical staff members of selected State
agencies, universities, and other government or private research organizations in order to identify
and obtain information pertaining to the development of temporal allocation factors. To date,
TRC has conducted interviews with most of the individuals identified on the approved contacts
-list in addition to other referral contacts. TRC will continue pursuing those individuals for which
follow-up calls are required. All initial contacts have been completed.
2.0 RESULTS
Copies of project contact reports for all individuals interviewed by telephone are contained
as attachments to this memorandum. Highlights of informadon regarding the development of
temporal allocation factors contained in these reports are given below organized by the groups
contacted. TRC will provide follow-up coverage to obtain as much as possible of the information
described below.
Results of Discussions with EPA
The results of the Southern Oxidants Study (SOS) include 7 weeks of measurement data
for ozone and related pollutants at 5 minute intervals. Some mobile source data from 36
separate measurement locations are included as a subset in the database. Most of the
database is not presently available to the public since it is currently undergoing a quality
assurance inspection. Approximately 2 gigabits of data are available for release to the .
public. The estimated size of the entire database is 6 gigabits. EPA has contacted Carlos
Offices in Califomio, Colorado, Connecticut, Illinois, Louisiana, Massachusetts, New Jersey, New York, North Carolina, Pennsylvania, Texcs,
Washington, Washington, D.C., and Puerto Rico
r ^ 3
Printed on Receded Pap«r
-------
Cardelina, Coordinator of the SOS at the Georgia Institute of Technology, and has
'requested a copy of the SOS database. (See report dated 5/26/93; interview with Mike
Rogers, Georgia Institute of Technology), TRC has pursued the Georgia Department of
Environmental Protection to obtain a copy of the stationary source database from the SOS
and the Atlanta Regional Commission for a copy of the SOS project summary report.
TRC expects receipt of this database around June 4, 1993. In addition, the Georgia
Department of Transportation was contacted and they will send a copy of the database
pertaining to the 6-week study of hourly emissions from mobile sources. TRC expects
receipt of this mobile source data on diskette June 1, 1993.
The EPA documentation pertaining to the development of temporal allocation factors from
data contained in the St. Louis Regional Air Pollutant Study (RAPS) is probably outdated
and will not be of much use under this Work Assignment. [See report dated 5/19/93;
interview with Chuck Masser, EPA Atmospheric Energy and Engineering Research
Laboratory (AEERL)j •
The EPA Emissions Inventory Branch (EIB) stated that hourly emissions data for VOC
storage tanks will be available from the TANKS program around June 4, 1993. Robin
Baker Jones of the Midwest Research Institute (MRI) in Cary, NC has been contacted for
obtaining this information. It should be noted, however, that the TANKS program may
not provide an accurate hourly emission rate since the equations used in the program were
not designed to give hourly data. TRC will follow-up by contacting MRI and EIB and
requesting a copy of these data, when available. (See report dated 5/17/93; interview with
Ann Pope, EPA EIB)
The results of EPA's 1990 Atlanta Ozone Precursor Study contain a huge database on
quantified non-methane organic compounds on a continuous hourly basis (suspected to
have been developed from ambient monitoring). TRC will contact Chuck Lewis of
AREAL to obtain a copy of the report pertaining to this study. (See report dated 5/20/93;
interview with Chuck Lewis, AREAL)
Results of Discussions with State!Local Regulatory Agencies
The South Coast Air Quality Management District (SCAQMD) uses diurnal codes from
their airshed filing system. For each facility, the emission rate and daily operating hours
are given. An estimate of daily emissions is sometimes determined. SCAQMD
determines the emissions on an hourly basis by the use of a code from the standard Urban
Airshed Model emissions processing system (UAMEPS), which they receive from the
California Air Resources Board (CARB). The operating schedule for each facility is
available through the Emissions Information System (EIS). TRC will follow-up by
contacting the EIS in order to obtain electronic copies of the database. (See report dated
^ 5/17/93; interview with Julia Lester, SCAQMD.)
2
A-96
-------
The Virginia Department of Environmental Management (DEM) is planning to obtain
temporal operational data for major point sources from major industry representatives and
is also working with EPA on defaults to the Urban Airshed Model (UAM). TRC will
make follow-up contacts- with the Virginia DEM to obtain additional information
regarding further progress on the temporal allocation work, with the Virginia DOT to
obtain any mobile source vehicles miles-traveled (VMT) information which they may
have available, and with the Washington D.C. Council of Governments for mobile source
information. (See report dated 5/20/93; interview with Kirit Chaudhari/Harry Quin,
Virginia DEM.)
* The Ohio State agency has information pertaining to operating schedules (i.e., point-by-
point and plant-by-plant) and gives seasonal throughput information on an hours/day,
days/week and weeks/year basis. This information is in a database format and TRC will
obtain a copy of these data. (See report dated 4/23/93; interview with Bill Juris, Ohio
State Agency.)
The emissions inventory system used by the Illinois Environmental Protection Agency
(IEPA) contains much information pertaining to throughput and scheduling data for 1990
through 1992. TRC has contacted Dave Afflemeier of the IEPA to determine what
information they may have available. TRC will need to obtain this information through
EPA. (See report dated 5/20/93; interview with Dave Afflemeier, IEPA.)
The' Texas Air Control Board (TACB) maintains an electronic database system which
contains information on operating schedules (along with facility contact information for
obtaining operating schedules), enforcement, emissions inventories, inspections, emission
permits, and New Source Review permits. This database can be imported into a dBASE®
format for data manipulation. TRC will obtain a copy of this database. (See report dated
5/21/93; interview with Roger Laprelle, Texas Air Control Board.)
The Wisconsin Department of Natural Resources has conducted a study of source-specific
hourly emissions from over 200 companies in the U.S. Facilities contained in the study
include power plants and paper mills. The raw data are available and include hourly
emissions of air pollutants. The data are in ASCII format and can be imported into a
SAS® format for data manipulation. TRC will obtain a copy of this database, (See
report dated 5/20/93; interview with Jens Lass, Wisconsin Department of Natural
Resources.)
Results of Discussions with Other Government or Private Organizations
* Many of the major metropolitan planning organizations (MPOs) throughout the U.S. are
doing work on mobile source emission estimates (many on an hourly basis) based on
percentages of VMT over various traffic segments and links. The Denver Regional
Council of Governments and the Baltimore Metropolitan Council have sent TRC daily and
hourly mobile source activity data. The Delaware Valley Regional Planning Commission
3
A-97
-------
has also indicated that they have an extensive vehicle activity database. TRC is
continuing to contact other major MPOs in order to determine what information they have
which may be of use in developing temporal allocation factors for mobile sources. [See
report dated 5/10/93; interview with Robert Newhouser, Southeast Michigan Council of
Governments. Also see the following additional reports: (1) report dated 5/19/93,
interview with Jeff May, Denver Regional Council of Governments; and (2) report dated
5/21/93; interview with Matt DeRouville, Baltimore Metropolitan Council.]
The Coordinating Research Council (CRC) (an automobile manufacturing consortium) has
purportedly released a report in May 1993 regarding the recommendation of temporal
profiles for mobile sources in the Los Angeles emissions inventory. The CRC has also
developed a report on diurnal profiles of vehicle evaporative emissions. TRC will follow-
up by obtaining copies of these reports. (See report dated 4/26/93; interview with Alan
Dunker, General Motors - Auto Oil Program). ,
The Lake Michigan Air Directors Consortium (LADCO) is currently involved with the
Lake Michigan Ozone Study (LMOS). The LMOS includes the development or
improvement of temporal allocation profiles. The LMOS uses scheduling information
from point sources (i.e., hours/day and days/week of operation). The LMOS contains
much information for various point and area sources and LADCO has a great deal of
information on diurnal profiles for mobile sources. TRC is on the distribution list for the
LMOS electronic database, which is expected to be released around June 8, 1993. (See
report dated 4/23/93; interview with Mike Koreber and Mark Jantzen, LADCO).
Radian Corporation can supply the following information from the Geocoded Emissions
Modeling and Projections (GEMAP) system: (1) look up files; (2) point and area source
data: (3) weekday/diurnal data; (4) the data dictionary; and (5) user's manual. TRC will
draft a purchase order requisition to obtain this information. (See report dated 5/17/93;
interview with Ron Dickson, Radian Corporation, Sacramento, California).
~ The Energy Information Administration (EIA) produces energy consumption and fuel
production reports with monthly statistics. The Department of Energy electronic
publication bulletin board system also contains some of the weekly production statistics.
TRC will obtain copies of these data.
Additional Information Sources for Data Collection
The North Carolina National Pollution Discharge Elimination System (NPDES) office and
the Washington D.C. Office of Water to determine the availability of industrial
wastewater flow data that may be used as a surrogate indicator of potential emissions
from wastewater treatment plants.
4
A-98
-------
The TRC Corporation subsidiary, Raymond Keys and Associates, to . determine the
availability of mobile source activity data.
The Federal Bureau of Labor Statistics and the State of North Carolina Employment
Securities Commission to obtain additional information related to data collected on form
790 dealing with employment hours and wages and any shift data. .Such information may
prove to be useful as a surrogate indicator.
The graduate school programs through the School of Business, Economics, etc. at Duke
University, UNC-Chapel Hill, and NC State University to obtain any potential surrogate
indicator data useful for estimating temporal allocation factors from various industries.
Other federal organizations (e.g., the Federal Drug Administration) which may have any
surrogate indicator data useful for estimating temporal allocation factors from various
industries.
Discussions Concerning Temporal Allocation Factors Improvements
Personnel interviewed by telephone were also asked whether they could perceive of any
ways in which temporal allocation factors could be improved. Several responses were given and
are summarized below.
The Ohio State Agency suggested that a sensitivity analysis of air quality models be
performed in order to determine how modeling responds to a change in a temporal
allocation factor. (See report dated 4/23/93; interview with Bill Juris, Ohio State
Agency).
• The Wisconsin Department of Natural Resources (DNR) stated that problems with the
classification of the ozone season may be apparent since the season does not coincide
with quarterly data used by the State agency. DNR urges EPA to amend the guidance
to allow the classification of the ozone season to be location specific. (See report dated
4/26/93; interview with Ralph Peterson, Wisconsin Department of Natural Resources).
* LADCO suggested that day specific data, as opposed to average data, are needed to obtain
the best results from modeling efforts. (See report dated 4/23/93; interview with Mike
Koreber and Mark Jantzen, LADCO).
The General Motors Auto Oil Program had two suggestions for improving temporal
allocation factors for mobile sources; (1) improving diurnal behavior of evaporative
emissions and resting losses with respect to ambient temperatures; and (2) improving the
magnitude, diurnal behavior and utilization profiles for off-road vehicle emissions. (See
report dated 4/26/93; interview with Alan Dunker, General Motors - Auto Oil Program).
5
A-99
-------
PROJECT CONTACT REPORT
Date of Contact:
Name of Contact:
Representing:
TRC Staff Member:
4/23/93
Thomas Carr (313) 872-4311
Motor Vehicle Manufactures Association of the US
Douglas P. Bamdt
Summary of Discussion:
Mr. Thomas Hanna, the Association President, was the initial point of contact. However, access
to Mr. Hanna was denied and the telephone call was instead referred to Mr. Thomas Carr, the
Association Vice President. The Association has conducted no documented work on allocation
of temporal factors for the motor vehicle manufacturing industry. However, the need for
temporal factors has been recognized during certain discussions. Mr. Carr suggested that the
automotive companies themselves may be doing work on this topic.
Mr. Can- stated that no reports are known to exist on the topic, and that the Association has no
specific operations data for the industry, since it is often proprietary. The only information the
Association has is very general and only for a few locations.
Mr. Carr agreed to pass our information request to Mr. Larry Slimak (another Association
employee) who is assigned to a committee which discusses needs of the Association. If the
Association decides to make a statement or help with the topic, Mr. Carr said the they would
contact either TRC or USEPA. Mr. Carr did not commit a date to respond.
A-100
-------
PROJECT CONTACT REPORT
Date of Contact:
Name of Contact:
Representing:
4/23/93
James Conner (919) 549-4800
Automotive Products Emissions Committee
Douglas P. Barndt
TRC Staff Member:
Summary of Discussion:
Mr, James Conner said the Automotive Products Emissions Committee has not been active during
the past 10 years and has done no work on allocation of temporal factors. (Mr. Conner
acknowledged his unfamiliaxity with the topic.) Furthermore, he is unaware of any reports or
other work on allocation of temporal factors or operations data for the automotive products
industry.
Mr. Conner suggested the car companies themselves be contacted for operations information.
A-101
-------
PROJECT CONTACT REPORT
Date of Contact;
Name of Contact:
Representing:
TRC Staff Member:
4/23/93
Paul Martina (202)682-8562
American Petroleum Institute
Douglas P. Bamdt
Summary of Discussion:
Mr. Paul Martino is with the Environmental & Health Section. He is unaware of any past or
current American Petroleum Institute (API) efforts concerning allocation of temporal factors for
the petroleum industry. Mr. Martino is also unaware of any operational data available to develop
temporal factors.
Three possible leads were suggested:
(1) Contact the API Refining Department - Jim Williams (202-682-8155)
(2) Contact NTIS for DOE studies on petroleum industry operational information, done
between 1980-'85. Document series was tided: ECIRs - Environmental Characterization
Information Reports
(3) Contact the API Statistics Department for possible operations data
Mr. Martino agreed that allocation of temporal factors is needed for the petroleum industry,
especially refineries.
A-102
-------
PROJECT CONTACT REPORT
Date of Contact;
Name of Contact;
Representing:
TRC Staff Member:
4/23/93
Bill Juris
Ohio State Agency
Sanjay Saraf
Summary of Discussion
According to Mr. Bill Juris, Ohio State Agency is not involved in any type of project for
developing temporal allocation factors. They use the results from the earlier study (conducted
around 1985) on development of allocation factors.
Even though Mr. Juris is not directly involved in any study for temporal allocation factors, he
reported that Pacific Environmental Services in Connecticut was doing some work pertaining to
allocation factors. Mr. Juris did not give name of any contact person.
Mr. Juris reported that he did not have any reports or documents that could assist us in evaluating
the current situation on temporal allocation factors.
Interestingly, Mr. Juris reported that he has information pertaining to operating schedules. This
data is point-by-point and plant-by-plant and pertains to seasonal throughput, hours/day,
days/week, weeks/year, of operation. Mr. Juris did not feel it necessary for TRC project
members to come to his office to collect this information. This data is in, a database format and
Mr. Juris agreed to send this data on disc to Theresa Moody.
Mr. Juris recommended that TRC should try a "Sensitivity Analysis" of the model. He suggested
that it would be interesting, for example, to determine how the model responds to a change in
the temporal allocation factor.
Mr. Juris was not aware of any other person in office involved in developing allocation factors.
A-103
-------
PROJECT CONTACT REPORT
Date of Contact:
Name of Contact(s):
Representing;
TRC Staff Member:
4/23/93
Mike Koreber and Mark Jantzen
Lake Michigan Air Directors Consortium (LADCO)
Sanjay Saraf
Summary of Discussion
According to Mr. Mike Koreber, LADCO is currently involved in the Lake Michigan Ozone
Study (LMOS). As part of this project, LADCO is developing an emission inventory model. An
important constituent of this emission inventory model is developing or improving temporal
allocation profiles. Mr. Koreber reported that, for point sources, they are using • information
pertaining to hrs/day, days/week, etc., of operation. For area source, LADCO is using the same
methodology as is being used by California.
The report pertaining to LMOS is expected to be complete by April 26, 1993. Mr. Koreber
agreed to send a copy or portions of the report pertaining to allocation factors to Theresa Moody.
Mark Jantzen was contacted on 04/29/93 for additional information.
Mark Jantzen reported that as part of the LMOS study he has vast amounts of data (about 50MB)
for various point and area sources. Mark said that if TRC provides him with a disc, he will send
us this data. He reported that LADCO has vast amounts of data pertaining to diurnal profiles for
mobile sources. Furthermore, Mark pointed out that as part of the LMOS study, Wisconsin
conducted a survey and collected data (throughput data hrs/day, days/week, etc.) from various
major point sources. He recommended contacting Jen Laas (Wisconsin Department of Natural
Resources) for additional information.
Mark suggested that in order to get best results from the model, it is essential to have day
specific emissions inventory data rather than average data. Mark felt that TRC should prepare
a questionnaire for the various State agencies requesting information because he feels that the
state agencies have a lot of data that may be very useful. He reported that the State inventory
community go through several stages of data processing to get the emissions inventory results.
Hence, he feels that one of these intermediate stages may actually pertain to allocation profiles.
A-104
-------
PROJECT CONTACT REPORT
TRC Staff Member:
Date of Contact;
Name of Contact:
Representing:
4/26/93
Alan Dunker (313) 986-1625
General Motors - Auto Oil Program
Douglas P. Barndt
Summary of Discussion:
Mr, Alan Dunker is very involved in estimating emissions from mobile sources for inventory and
modeling. While he (and General Motors) has not been involved in work specifically on temporal
allocation of emission factors, some of his work has included examining different temporal
profiles of mobile sources.
Mr. Dunker discussed a variety of projects and reports which include temporal allocation
concerns.
(1) He's been involved in a project which examines the Los Angeles emission inventory, and
recommends different temporal profiles for mobile sources. The report on this study is
due for release in May 1993, and will be available from the Coordinating Research
Council (CRC) (an automobile manufacturing consortium).
(2) The CRC, in conjunction with Radian Corporation, has developed a report on diurnal
(time) profiles of vehicle evaporation emissions. The CRC point of contact is Tim
Bellian at (404) 396-3400.
(3) Another CRC Air Quality Report (title unknown) dated January 1993 discusses diurnal
(4) Mr. Dunker reported that many states are doing work involving temporal allocation while
estimating mobil sources for SIP inventories (e.g., VMP). Southern California air districts
are developing time profiles for hot and cold vehicle starts. The Lake Michigan Ozone
Study was mentioned.
Mr. Dunker had two recommendations for improving temporal allocation factors for mobile
sources.
(1) Diumal behavior of evaporative emissions from vehicles with respect to ambient
temperatures needs improvement Diurnal resting losses also need improvement.
(2) The magnitude, diurnal behavior, and utilization profiles for off-road emissions needs
further improvement
profiles.
A-105
-------
PROJECT CONTACT REPORT
Date of Contact: A/26193
Name of Contact: Sherry Edwards (202) 659-0060
Representing: Synthetic Organic Chemical Manufactures Association (SOCMA)
TRC Staff Member: Douglas P, Barndt
Summary of Discussion:
TRC identified Ms. Sherry Edwards as the point of contact for government air emission issues
at the SOCMA. The SOCMA has no information regarding allocation of emissions factors for
the synthetic organic chemical manufacturing industry. Ms. Edwards further explained that
SOCMA is a small trade organization with a limited staff representing specialty chemical
manufactures.
Ms. Edwards remembered some operating/schedule procedures information was submitted to
USEPA - OAQPS for developing the SOCMI - Batch Processing Control Technology Guideline
document
Partially because Ms. Edwards was-very unfamiliar with the temporal allocation subject, she had
no ideas for future needs for improvement or use of temporal allocation factors. Thus, she
suggested that TRC submit a FAX to her with the project description and requested information,
since the SOCMA Environmental Quality Committee will be meeting in three weeks and could
possibly offer some help.
NOTE: TRC faxed some project information and temporal allocation questions per Ms.
Edward's request.
A-106
-------
PROJECT CONTACT REPORT
Date of Contact:
Name of Contact:
Representing:
TRC Staff Member:
4/26/93
Karil Kochenderser (202) 862-0500
American Textile Manufactures Institute (ATMI)
Douglas P. Barndt
Summary of Discussion:
TRC identified Ms. Karil Kochenderser as the point of contact for air emission issues at the
ATML The ATMI has not done much air emissions work, and has no information regarding
allocation of emissions, factors for the textile manufacturing industry, Ms, Kochenderser is
unaware of any industry work on allocation factors. However, she said the North Carolina
Department of Economics, Health, and Natural Resources may have some information pertinent
to the State Air Toxics Program.
Ms. Kochenderser was unsure if the ATMI has written documentation on industry
operating/schedule procedures. She recommended that TRC submit a FAX to the her with the
project description and requested information, since the ATMI will be conducting a business
meeting on May 10,1993 and TRC's information request can be discussed among the 70+ person
staff. Ms. Kochenderser recognizes that temporal allocation of emission factors for the textile
industry is important; however, she noted that the ATMI has had staff cutbacks and is
concentrating on a variety of other environmental issues (NPDES - water, stormwater, and
hazardous waste).
NOTE: TRC faxed some project information and temporal allocation questions per Ms.
Kochenderser's request. She suspects that the ATMI library may have some operations data or
other information which may be useful.
A-107
-------
PROJECT CONTACT REPORT
Date of Contact:
Name of Contact:
Representing:
TRC Staff Member:
4/26/93
Marc Deslariers
Environment Canada
Sanjay Saraf
Summary of Discussion
Mr. Marc Deslariers reported that he is not aware of any new work being done by Environment
Canada for developing temporal allocation factors. According to Mr. Deslariers, they still use
the factors developed as part of the 1985 inventory. The emissions data for the 1990 inventory
is currently being processed. Information from various area sources is being collected, analyzed,
and compiled.
Mr. Deslariers is not aware of any other organization or agency currently developing temporal
allocation factors. He does not think that any one at Environment Canada is involved in any
study pertaining to allocation factors.
Mr. Deslariers does not have any documents or reports on temporal allocation factors. However,
he has operating data for various area and point sources. This data is in form of large printouts
from a database program.
Mr. Deslariers reported that he has operating schedules data pertaining to the 1985 inventory for
various area and point sources. Environment Canada does not have complete data for the 1990
inventory. Mr. Deslariers suggested contacting Eva Voldner (416-739-4467), who is involved
in air emissions modelling, for additional information.
Mr. Deslariers admitted that he was not very familiar with factors that impact development of
allocation factors. However, he expressed a strong need for improved guidelines for collecting
data. He felt that improved data collection would help the modelling community immensely.
A-108
-------
PROJECT CONTACT REPORT
Date of Contact:
Name of Contact:
Representing:
TRC Staff Member:
4/26/93
Ralph Petterson
Wisconsin Department of Natural Resources
Sanjay Saraf
Summary of Discussion
Mr. Ralph Petterson reported that the Wisconsin Department of Natural Resources (WDNR) is
currently not involved in a project for developing temporal allocation factors. According to Mr.
Petterson, WDNR has adopted USEPA guidelines and hence uses USEPA-approved equations
to account for emissions for a typical ozone season day.
Mr. Petterson was not aware of any work being done by any agency or association in Wisconsin
for developing temporal allocation factors. However, he mentioned that the Lake Michigan Air
Directors Consortium (LADCO) may be studying allocation factors as part of the Lake Michigan
Ozone Study (LMOS).
Mr, Petterson had no suggestions for any reports or documents pertaining to temporal allocation
factors. However, he is writing a technical memorandum explaining how WDNR uses EPA-
approved equations in their emissions calculations. This memorandum is expected to be
completed by April 30, 1993 and he agreed to send a copy of this memorandum to Theresa
Moody as soon as possible.
According to Mr. Petterson, WDNR does not have data pertaining to operating schedules for
point-by-point or plant-by-plant permitted facilities. Instead, WDNR conducted a survey and
collected quarterly percentage data to use with standard equations listed.in EPA guidelines.
Mr. Petterson did not have any suggestions directly impacting temporal allocation factors. He
feels that the modelling community at WDNR faces a problem because of the ozone season
classification. The ozone season does not coincide with the WDNR quarter which makes it
difficult to apply their quarterly data to the ozone season calculations. He recommended that
EPA should modify the guidance document so that ozone season classification can be location
specific.
Mr. Petterson recommended contacting Mr. Jens Laas (WDNR), to obtain additional information
pertaining to survey data.
A-109
-------
PROJECT CONTACT REPORT
Date of Contact:
Name of Contact:
Representing:
TRC Staff Member:
4/27/93
Gene Parschan (313) 872-4311
American Automobile Manufactures Association
Douglas P. Barndt
Summary of Discussion:
Mr. Gene Parschan contacted TRC based on his discussions with the Association Vice President,
Mr. Thomas Carr, whom TRC contacted on 4/23/93. Mr. Parschan noted that the Association
has recently changed names from the Motor Vehicle Manufactures Association of the U.S. to the
American Automobile Manufactures Association (AAMA).
Mr. Parschan has been helping automobile manufactures in southern Michigan prepare emission
estimates for the State inventory and modeling effort, which uses the Urban Airshed Model. He
is not very familiar with temporal allocation factors, and stated the Association has not done any
specific work on this topic.
Mr. Parschan said the AAMA could possibly help with operational data for automobile
production. He remarked that the weekly publication entitled Automotive Industries publishes
weekly automobile production statistics. This is apparently the most specific source of available
production statistics. Mr. Parschan commented that if more specific operations/schedule
procedures is needed, such as daily production statistics, the AAMA could act as a liaison
between EPA and the industry (GM, Ford, Chrysler).
Mr. Parschan called again to discuss EPA's interest in temporal allocation factors for mobile
sources. Another AAMA employee, Marcel Halberstadt, has done a great deal of mobile
emissions inventory work with the Michigan emissions study. However, Mr. Parschan thought
that much of the methodology for estimating mobile sources was from the State. He suggested
the Mr. Halberstadt may have helpful information.
A-110
-------
PROJECT CONTACT REPORT
Date of Contact;
Name of Contact:
Representing:
TRC Staff Member:
4/30/93
Dick Forbes
Illinois Environmental Protection Agency (IEPA)
San jay Saraf
Summary of Discussion:
At the beginning of our conversation, Mr. Dick Forbes admitted that his knowledge on temporal
allocation factors is limited. He has not been directly involved in any project pertaining to
allocation factors. He reported that IEPA currently is not conducting any study for developing
temporal allocation factors.
Mr. Forbes was not aware of any industry or agency conducting any study pertaining to temporal
allocation factors. He reported that Commonwealth Edison has been keeping hourly throughput
data which IEPA may use to develop temporal profiles in future.
Mr. Forbes did not have any reports or documents pertaining to temporal allocation factors. He
was unable to give any citations for any reports pertaining to allocation factors.
He felt that the Emissions Inventory System at IEPA may have vast amounts of throughput data
from various sources, which may be helpful in developing temporal profiles. Mr. Forbes
suggested contacting Dave Affelmeier (219-782-7048) for data for additional information on the
type of throughput data that IEPA may have.
Since Mr. Forbes has not used temporal allocation profiles extensively in Ms work at IEPA, he
had no other suggestions for improvement of temporal allocation factors.
A-lll
-------
PROJECT CONTACT REPORT
Date of Contact:
Name of Contact:
Representing:
TRC Staff Member:
5/10/93
Larry Cupitt
U.S. EPA AREAL (Integrated Air Cancer Project)
Ritchie Buschow
Summary of Discussion:
Larry Cupitt is not involved in any current work involving the development of temporal
allocation factors. He suggested calling Chuck Lewis of AREAL (919-541-3154) to obtain
information pertaining to the development of VOC datasets to be used in receptor modeling
applications. Larry stated that temporal allocation factors could be developed through the use
of these models.
A-112
-------
PROJECT CONTACT REPORT
Date of Contact:
Name of Contact:
Representing:
TRC Staff Member:
5/10/93
Bill Gill
Texas Air Control Board (TACB), Austin, TX
Ritchie Buschow
Summary of Discussion:
Mr. Bill Gill was not currently involved with the development of temporal allocation factors but
that the modeling group with TACB had been developing factors from inventory data for use in
their Urban Airshed Model. The contact person for obtaining such information is Mr. Dick Karp
at 512-908-1462. Bill also stated that TACB's database has operating schedules for various
sources itemized by SCC code but printing the information in this form involves a detailed
algorithm and would be time consuming. However, a facility screen data retrieval could be done
in less time and this would give a printout of all facility information and the operating schedules.
To obtain this printout, the contact person at TACB is Roger Laprelle at 512-908-1529. Bill also
stated that this database could be pulled into a dBASE format for data manipulation purposes.
Finally, Bill stated that any throughput data contained in the database is confidential but could
possibly be released to another regulatory agency or EPA.
A-113
-------
PROJECT CONTACT REPORT
TRC Staff Member:
Date of Contact:
Name of Contact:
Representing:
5/10/93
Marcel Halberstadt
American Automobile Manufacturers Association
Ritchie Busehow
Summary of Discussion:
Mr. Marcel Halberstadt was contacted as a result of a recent conversation with Mr, Gene
Parschan of the same representing organization given above. Marcel stated that one of his staff
had been working with the Southeast Michigan Council of Governments (SMCOG) on developing
mobile source emission factors and any information regarding the development of temporal
allocation factors for mobile sources could be obtained through SMCOG. The person to contact
at SMCOG is Mr. Robert Newhouser at 313-961-4266.
A-114
-------
PROJECT CONTACT REPORT
Representing:
TRC Staff Member:
Date of Contact:
Name of Contact:
5/10/93
Robert Newhouser
. Southeast Michigan Council of Governments (SMCOG)
Ritchie Buschow
Summary of Discussion:
Mr. Robert Newhouser stated that SMCOG has been working on hourly de facto VOC emission
factors for mobile sources. He will be sending us a copy of their 1990 base emissions inventory
for mobile sources (it will be rather large, approximately 400 pages) which will explain how their
emissions estimation program works. In the meantime, the following paragraph explains, in some
detail, the type of mobile source emission estimation work which SMCOG is currently doing.
Traffic segments are broken down into individual links which are then broken down into various
hours of the day and travel in two directions. Daily traffic count data is obtained and converted
to hourly data. The hourly data is converted to a percent vehicle miles traveled (VMT) which
is then applied in a speed equation to determine VOC concentrations. The estimated VOC
concentrations can then be separated by pollutant and vehicle type. Running emission factors
from the MOBILES model are used and a composite emission factor consisting of both cold-start
and hot-start is used. Both trip-start and hot-soaks factors are not used since the running factors
from the MOBILES model are used. Also, refueling loss emissions arc not treated as a function
of VMT but only as a function of fuel sold (i.e., where the fuel is dispensed). Although there
are 80 possible emission scenarios for which emissions can be estimated, they use only 26
distinct scenarios for which the emissions factors are used to estimate hourly emissions from
mobile sources. Robert stated that the 1990 base year emission inventory document will explain
each of these scenarios. Robert also explained that this procedure for estimating emissions from
mobile sources has been ongoing for some time.
Robert provided some additional contacts for obtaining emissions information for mobile sources.
These contacts include:
American Association of State Highway Transportation Officials (202-624-5800)
to obtain a listing of State DOTs which have been doing much work on mobile
source emissions estimates. Robert suggested that such States might include
Texas, New York, New Jersey, Pennsylvania, California, Florida and Illinois.
National Association of Regional Councils contact Mark Howard at 202-457-0710;
A-115
-------
Robert also provided a listing of other metropolitan planning organizations and council of
governments which have been leaders in estimating emissions from mobile sources. These
include:
• Denver Regional Council of Governments; contact George Scheurnstuhl (Director)
at 303-455-1000
North Central Texas Council of Governments; contact Mike Morris (they are
located in Dallas, Texas)
• Houston/Galvaston Area Council; contact Allen Clark (Transportation Director)
at 713-627-3200
• Metropolitan Washington Council of Governments; contact Ron Kirby at 202-962-
3200
« San Francisco Metropolitan Transportation Commission; contact Chuck Purvis at
510-464-7731
• Chicago Area Transportation Study; contact Andrew Plummer at 312-793-3456
• Baltimore Metropolitan Council; contact Matt Derouville at 410-333-1750
• Delaware Valley Regional Planning Commission; contact Paul Pezzotta at 215-
592-1800
• Minneapolis/St Paul Metropolitan Council; contact Steve Alderson or Natlio Diaz
at 612-291-6337
• Boston Central Transportation Planning Staff; contact John Quakenbush at 617-
973-7100
• Portland (Oregon) Metropolitan Service District; contact Keith Lawton (Director)
the MPOs in California for the South Coast (contact Ralph Cipriani, Director of
Forecasting) and Sacramento (see Mark Smith for phone numbers for both MPOs)
Finally, Robert also suggested calling the Federal Highway Administration in Washington, D.C.
The contact people are Jim Shroads in the Air Quality Planning branch at 202-366-2074 or Pat
Decorla-Souza at 202-366-4076,
A-116
-------
PROJECT CONTACT REPORT
Date of Contact:
Name of Contact(s):
Representing:
TRC Staff Member:
5/11-5/12/93
Rich Rehm, Roger CawkweU, and Glenn Reed; John Langstaff
Pacific Environmental Services; SAI
Phil Marsosudiro
Summary of Discussion:
An earlier project contact report by Sanjay Saraf (04/23/93) stated that Bill Juris of the Ohio EPA
thought that a PES-Connecticut office was doing some work with temporal allocation factors.
I made phone calls to several contacts at PES, Sanjay at the Chicago office [he had been out of
town earlier while I was talking to PES], David Ocamb, and John Langstaff at SAI. The
following bullets summarize what I think is really going on:
John Langstaff at SAI initiated a temporal allocation project with AEERL, but the
project never came to fruition. This project (the same project that David Ocamb
mentioned to Theresa Moody) was to be completed under subcontract to PES.
This may be the project that Bill Juris was thinking of, although John Langstaff
said that the project had "nothing to do with Connecticut."
PES doesn't have a Connecticut office.
The work that Bill Juris mentioned to Sanjay may have had nothing to do with
Phil win check with Connecticut to see if they're doing anything with temporal allocations.
PES.
Followup:
A-117
-------
PROJECT CONTACT REPORT
TRC Staff Member;
Date of Contact;
Name of Contact(s):
Representing;
5/12/93
Kieman Wholean, A1 Leston
Connecticut Department of Environmental Protection
Phil Marsosudiro
Summary of Discussion:
Spurred by rumors that Connecticut was conducting temporal allocation work, I called the
Connecticut DEP. Mr. Kiernan Wholean indicated that "Connecticut has increasing interest in
temporal allocation of ozone emissions, but that little work has been done thus far." Mr.
Wholean mentioned that the DEP recently requested activity data from utilities and gas stations
in Connecticut, but that data had not yet been received. Mr. Wholean then connected me with
A1 Leston, in the DEP's monitoring branch.
Mr. A1 Leston has conducted preliminary data analyses of ozone concentrations in Connecticut
from 1980-1992. Although his work is still "in progress," Mr. Leston indicated these preliminary
findings:
Connecticut ozone concentrations peak on the weekends. Analyses of 1980-1985
data show "highs" on Saturday, Sunday, and Monday. Tuesday-Friday
concentrations are significantly lower. Analyses of 1986-1992 data show highs
on Saturday, only. Sunday-Friday concentrations are significantly lower. Mr.
Leston chose 1985 as a cutoff date for two reasons: tolls were taken off the
highway in 1986, and the DEP changed its monitoring instrumentation.
Mr. Leston has "called around" to see if anyone else is doing temporal allocation
work. Thus far, he has found little activity. (I gave him Chuck Mann's telephone
number.) He has spoken with Gary Engler at Ohio EPA, who has done similar
ozone monitoring analyses. Interestingly, Mr. Engler s results were nearly the
opposite of Mr. Leston's. Mr. Engler's data showed lower concentrations during
the weekends, and higher concentrations during weekdays. Mr. Engler's data also
indicated that concentrations increased on a daily basis from Monday through
Friday, dropping off sharply on Saturday.
An article titled "Weekend Smog May Be Worse" was sent by Mr. Engler to Mr.
Leston. The article discusses smog concentrations in the South Coast Air Quality
Management District The article cites work done by James Birakos, formerly of
the SCAQMD, and now a private consultant. Mr. Leston forwarded copies of
materials sent to him by Mr. Engler.
Followup:
Phil will call Gary Engler at Ohio EPA to see what other work they're doing.
A-118
-------
PROJECT CONTACT REPORT
Date of Contact:
Name of Contact:
Representing:
TRC Staff Member:
5/13/93
Ron Meyers
U.S. EPA EIB
Ritchie Buschow
Summary of Discussion:
Mr. Ron Meyers did not have any information regarding the development of temporal allocation
factors. He stated that the only emission factor development work of which he was aware
involved daily averages. He is currently involved on a project pertaining to the investigation of
the uncertainty of emission factors. He also stated that' Ms. Ann Pope of EIB is currently
involved with estimating daily VOC emissions from tanks during the three-month ozone season.
A-119
-------
PROJECT CONTACT REPORT
Date of Contact:
Name of Contact(s):
Representing:
TRC Staff Member:
5/13/93
Gary Engler
Ohio EPA
Phil Marsosudiro
Summary of Discussion:
At the suggestion of A1 Leston at the Connecticut Department of Environmental Protection, I
called Gary Engler at Ohio EPA.
Mr. Engler has examined 1983-1992 AIRS data for Ohio to correlate the frequency of ozone
exceedances to the day of the week. His analysis indicated the following:
The ten-year data demonstrate a strikingly uniform increase in the likelihood of
an ozone exceedance beginning on Monday and continuing through Friday. The
likelihood of an exceedance drops off sharply on the weekend.
* The "clean-looking" trend is exhibited when ten years of data are examined in
aggregate. Daily data from individual years do not always exhibit the same trend.
1983 and 1988 had far more exceedances than any of the other years between
1983 and 1992. The 1983 data do not demonstrate the same trends as the ten-year
data, but the 1988 data generally do.
Mr. Engler plans to do the following additional work in this area:
* Look at Ohio metro-area (Cincinnati, Cleveland, Columbus, etc.) monitoring data
to see if any patterns exist at the metro level.
« Include the 1983-1992 trend data in a pollutant information center (PIC) pamphlet.
Copies of Mr. Engler's data were sent to TRC by Mr. Leston (CT DEP), yesterday, and have
been placed in the project files.
Mr. Engler indicated that Ohio EPA is probably not doing any other temporal allocation work
at the moment. This study was done almost "on whim" after Mr. Engler's Air Division Chief,
Bob Hudenbose (sp?) received a newsletter clipping that mentioned temporal ozone patterns in
Southern California. (Mr. Leston also gave a copy of this clipping to TRC.)
A-120
-------
Mr. Engler does not know of any other group working on temporal allocations, but suggested the
following;
* Talk to staff at OAQPS AIRS group.
Post a query on the AMTIC bulletin board,
Followup:
The project team may wish to post a query on the AMTIC bulletin board (or on other bulletin
boards) requesting information on "who is doing what?" with temporal allocations.
1 1
-------
PROJECT CONTACT REPORT
Date of Contact;
Name of Contact(s):
Representing:
TRC Staff Member:
5/13/93
Mr. Kieman Wholean
Connecticut Department of Environmental Protection
Phil Marsosudiro
Summary of Discussion:
At the end of our conversation, yesterday, Mr. Wholean agreed to check around the CT DEP
office to see if any other temporal allocations work was being done. He called again this
morning, stating that CT DEP is doing the following;
David Wackter is working on regulations that will affect sources that may have
, exceedance problems on a seasonal basis. Regulations will incorporate trade-offs
and operating practice changes from season to season that allow the source to be
"in compliance" (?) on an annual basis.
Mr. Wholean has been "asked by Mr. Wackter to examine emissions by day of
week." Mr. Leston's monitoring data study (discussed in a 5/12/93 contact report)
provided the impetus for Mr. Wholean's work.
Mr. Leston is continuing with his analysis of 1983-1992 CT ozone monitoring
data.
Project team needs to keep in touch (over the summer and beyond) with Connecticut to see what
new work they do. They may be doing the "cutting edge" work, as far as State agencies go.
Followup:
A-122
-------
PROJECT CONTACT REPORT
Date of Contact:
Name of Contact:
Representing:
TRC Staff Member:
5/17/93
Julia Lester (714) 396-3162
South Coast AQMD
Lula Harris
Summary of Discussion:
The SCAQMD uses diurnal codes from the standard airshed filing system. For a particular
facility, SCAQMD is provided the annual emissions rate and in some cases, the hours/day that
the facility is in operation. SCAQMD sends this information to CARB, and a file with records
including the facility, equipment in the facility, and a code from the standard urban airshed model
emissions processing system (UAMEPS) is returned to SCAQMD. Using this information, the
SCAQMD then determines the total emissions per hour from the facility.
Ms. Julia Lester was unaware of additional projects for developing temporal allocation factors.
However, she referred Mr. Jim Fries of the CARB (916-324-7165) for additional projects.
Julia stated that the EPA BBS contains an Urban Airshed Model (UAM) and UAM Emissions
Processing Manual (Volume V). EPA may be contacted for a copy of this information.
Julia stated that activity/operating schedules are available from the SCAQMD via the Emissions
Information System (EIS). Julia also mentioned that this is a wealth of information and that it
is available on diskettes. A request form must be submitted to the SCAQMD. Julia was not sure
if a fee is required to obtain this information.
Julia suggested using available temporal allocation factors as opposed to developing new or
additional ones.
A-123
-------
PROJECT CONTACT REPORT
Date of Contact*. 5/17/93
Name of Contact: Ron Dickson
Representing: Radian Corporation, Sacramento California
TRC Staff Member: Theresa Moody
Mr, Ron Dickson was contacted for additional information on GEMAP. Ron stated that Mr.
Chuck Mann would need to draft a requisition letter indicating the information needed. It
appears that TRC needs the following information from GEMAP:
* logic and code
look up files
* area source data
weekday/diurnal data
* data dictionary
Ron further stated that he was not familiar with a temporally allocated mobile source emission
inventory developed by Radian for the South Coast Air Quality Management District, Chuck
Mann followed with a letter to Ron Dickson requesting the above information on May 24, 1993.
A-124
-------
PROJECT CONTACT REPORT
Date of Contact: 5/17/93
Name of Contact: Ann Pope
Representing: EPA EIB
TRC Staff Member: Theresa Moody
Ms, Ann Pope was contacted about the development of temporal allocation factors for emissions
of VOCs from storage tanks. Ann stated that data pertaining to hourly emissions from storage
tanks may be available using the TANKS program. The contact for obtaining this information
is Robin Baker Jones of the Midwest Research Institute. Ann also stated that she did not think
that the program provided an accurate estimate of emissions since the equations used in the
program are not designed to quantify the emissions on an hourly basis. For example, if a tank
holds 1,000 gallons of a liquid, the program may not accurately account for heating and cooling
effects (i.e., temperature changes) which could change the emission rate by a significant amount.
A-125
-------
PROJECT CONTACT REPORT
Date of Contact;
Name of Contact:
Representing:
TRC Staff Member:
5/18/93
Mike Koreber
Lake Michigan Air Directors Consortium
Craig EUis
Mr. Mike Koreber was previously contacted by San jay Saraf of TRC in April 1993. The LMOS
database requested will be forwarded once the QA/QC work by Radian is completed. Currently,
technical comments are being addressed and the final version of the database is expected in June
1993.
A-126
-------
PROJECT CONTACT REPORT
Date of Contact:
Name of Contact:
Representing:
TRC Staff Member:
5/18/93
Mike Meyer
Georgia Institute of Technology
Craig Ellis
Mr. Mike Meyer was contacted concerning the Southern Oxidants Study. He referred me to Mr.
Mike Rogers of Georgia Tech who is most familiar with the database associated with this study.
Mike Rogers was contacted via voice mail on 5/21/93 and are awaiting a response.
A-127
-------
PROJECT CONTACT REPORT
Date of Contact:
Name of Contact;
Representing:
TRC Staff Member:
5/18/93
Susan (Labor Mgt. Division Assistant) (733-2936)
North Carolina Occupational Safety and Health Administration
(NC OSHA)
Geary McMinn
The NC OSHA neither requests nor keeps any production information relating to hours worked,
time/shifts, production levels, percent seasonal/quarterly runs, or output. This information is not
reported to the Federal OSHA by North Carolina and it is assumed that other states do not report.
The Industrial Commission was also contacted about case-by-case accident information. No
summary data on hours, percent seasonal/quarterly runs, or output information are available.
Only seasonal/yearly accident information by industry is available. All facility information
obtained (e.g., shift worked, type of job, job duties) is considered confidential.
(Shirley - 733-4820)
A-128
-------
PROJECT CONTACT REPORT
TRC Staff Member:
Date of Contact:
Name of Contact:
Representing:
5/19/93
Jeff May
Denver Regional Council of Governments (303) 480-6746
Susan Clayton
Summary of Discussion:
Jeffs office develops estimates of travel during a,m. and p.m. peak hours and off hours on a link
basis. The data is given to the Health Department for hourly VMT allocation. Call Rick Berritt
(303) 692-3123. YMT variation on a weekly basis does not exceed error in estimated values
(3—5%). Jeff's office also develops percent of trips started during 15 minute time intervals and
average length and purpose of trips. He will send percent travel by purpose by hour if this is
helpful.
Susan Clayton - call Rick Berritt to obtain their temporal allocation factors for VMT.
Question - how to identify links to roadtype? Hope there are speeds assigned to links.
FoOowup:
A-129
-------
PROJECT CONTACT REPORT
Date of Contact;
Name of Contact:
Representing:
TRC Staff Member:
5/19/93
Professional Organizations
See below
Geary McMinn
Several professional societies and organizations listed below were contacted regarding production,
hourly operating, time/shifts, percent seasonal, and output information for industry groups. None
of the organizations contacted were able to provide information or refer to any other organization
which could provide this information.
1. American Production and Inventory Control Society - George Johnson - (716-475-2098)
2. Association of Productivity Specialists - Phone number changed; no record in same city
3. Council of Logistics Management - George Jackawitch - (708-574-0985)
4. Warehousing Education and Research Center - Thomas Sharp - (708-990-0001)
5. Herbert W. Davis & Company - John Poss - (201-871-1760)
A-130
-------
PROJECT CONTACT REPORT
Date of Contact:
Name of Contact;
Representing:
TRC Staff Member:
5/19/93
Chuck Masser
EPA AEERL
Theresa Moody
Mr. Chuck Masser of EPA's AEERL was contacted. Chuck commented on a Regional Air
Pollutant Study (RAPS) which compiled an hourly inventory of criteria pollutant data in the St.
Louis Missouri area over a period of two years. This data was placed into a data handling
system. Ms. Joan Novach of AREEL is the contact person for obtaining scheduling information.
The database contains information for the following source categories: (1) airports; (2) water; and
(3) residential heating by ambient temperature.
For point sources, emission factors were developed from source test data which gave fuel use per
hour. For power plants, fuel use data, source test data, and AP-42 emission factors were used.
Also, yearly fuel use by hourly work pattern (i.e., 40 hours/week) were given. Chuck
recommended the following documents to obtain additional information pertaining to developing
temporal allocation factors:
1. EPA 600/4-77-044; RAP Study: Criteria and Noncriteria Source Testing Program,
November 1977.
2. EPA 450/3-75-098; RAP Study; Residential and Commercial Area Source
Emissions Inventory Methodology for RAP, September, 1975.
3. EPA 450/3-77-025; Assessment of Railroad Fuel Use and Emissions for RAP,
April 1977.
4. EPA 450/3-77-019; Line and Area Source Emissions from Motor Vehicles in the
RAPS Program, June, 1976.
5. EPA 450/3-75-002; Methodology for Estimating Emissions for Off-Highway
Mobile Sources for the RAPS Program, October, 1974.
6. EPA 600/4-79-004; Emissions Inventory Summarization for RAPS, January 1979.
7. EPA 450/3-76-035; Methodology for the Determination of Line Sources, February,
8. RAPS Point Source Methodology and Inventories; EPA 450/3-74-054, October,
1974; EPA 600/4-77-014, March 1977; EPA 600/4-78-042, July 1978.
9. EPA 600/4-77-041; RAP Study Off-Highway Mobile Sources, October 1977.
1975.
A-131
-------
10, EPA 450/3-75-048; Airport Emissions Inventory Methodologies, December 1974.
11. EPA 600/4-79-076; Documentation of the RAPS, December 1979.
It should be noted that TRC does not think that any of these documents will be useful since
much of the information is outdated.
A-132
-------
PROJECT CONTACT REPORT
Date of Contact:
Name of C on tact (s):
Representing:
TRC Staff Member:
5/19/93
Maureen Mull ady
Chicago Area Transportation Study (312-793-3456)
Walid Ramadan
Summary of Discussion:
The purpose of the phone call was to talk to Mr, Andrew Plummer. He was out and I was
referred to Ms. Maureen Mullady. She does not know of any data on or related to hourly VMT
at her agency. She recommends contacting Illinois DOT for such information. The phone
number is (708) 705-4079.
Followup:
* Contact Illinois DOT
* Contact Andrew Plummer
A-133
-------
PROJECT CONTACT REPORT
Date of Contact:
Name of Contact:
Representing:
TRC Staff Member:
5/20/93
John Bvun
Federal Highway Administration (202-366-5451)
Susan Clayton
Summary of Discussion:
Jim Shroads, Chief of Noise and Air Quality Branch (202-366-2074), asked me to call John, a
modeler. John uses MOBILES and UAM models. He was not certain about data collected from
the States and referred me to the Travel Monitoring Division, Edward Kashuba at
(202) 366-0175. I spoke with Paul Svercl at that number since Ed was not in. He said they
collect annual VMT from the States. They also have hourly traffic counts for 3,000 U.S.
locations. He suggested the Metropolitan Planning Organizations would be the best source for
the data we need. He also mentioned the NPTS household survey.
Walid - call the Highway Information Office, who publishes the NPTS survey to obtain most
recent publication.
Followup:
A-134
-------
PROJECT CONTACT REPORT
TRC Staff Member;
Date of Contact;
Name of Contacts):
Representing;
5/20/93
Jim Smith
Kenan Flagler Business School, UNC-Chapel Hill
Phil Marsosudiro
Summary of Discussion:
David Winkler spoke, earlier, with Mr. Callan Swensen (sp?) at the Bureau of Economic
Analysis' Regional Studies Group. Mr, Swensen indicated that they publish quarterly and annual
State economic data, and annual county data—Professor Jim Smith at UNC-CH is the local
contact for the Chapel Hill User's Group.
I called Jim Smith to ask if he thought the BEA data would be useful to the temporal allocation
project. Dr. Smith felt that the BEA data, done on a quarterly or annual basis, was just not
refined enough for our needs. He suggested that we contact the electric power industry, which
collects more refined data. Dr. Smith said that the Fed uses electric power as a surrogate for
activity. Electric power is a "generally adequate surrogate" except where there are changes in
performance due to efficiency-improvements and the like.
The level of detail and coverage for electric power use is not always consistent. For example,
the Tennessee Valley Authority (TVA) keeps very good data on select large users—TV A can tell
us the hourly power consumption of users such as primary aluminum smelters in the TVA, but
not for all other users.
Dr. Smith suggested that we contact Mr. Louis Buck, the VJ3. of Finance at the Carolina Electric
Corp. (919-872-0800), to find out what electric power data are available.
Phil will call Louis Buck at Carolina Electric Corp. [note; I tried to reach him on the 20th, but
he will be out of the office until the 25th.]
Folio wup:
A-135
-------
PROJECT CONTACT REPORT
TRC Staff Member:
Date of Contact:
Name of Contact:
Representing:
5/20/93
Julie Tsao (703-274-6445)
The Defense Logistics Agency (Apparel Tech. Centers Director)
Geary McMinn
The Defense Logistics Agency (DLA) has several types of technology centers for studying
manufacturing. These centers include apparel, gears, and food manufacturing units. Each unit
may have several test facilities at a variety of locations. These facilities perform test runs such
as machinery layout; ergonomic studies; and production orders to test manufacturing methods.
Each facility determines run times and schedules. The data available would not be representative
of manufacturing industries since the facilities arc not actual production operations.
The following apparel were identified.
Georgia Tech/Southem Tech. (Bill Cameron - Mgr.) 404-528-3176
Fashion Institute - N.Y. (Hank Seeselling - Mgr.) 212-760-7410
Clemson University Apparel Center (Ed Hill - Mgr.) 803-646-8454
A-136
-------
PROJECT CONTACT REPORT
Date of Contact:
Name of Contact;
Representing:
TRC Staff Member:
5/20/93
Dave Affelmeier (217) 782-2829
Illinois Environmental Protection Agency
Sheila Thomas
According to Mr. Affelmeier, the Emissions Inventory System at DEPA has vast amounts of
throughput and scheduling data for 1990 - 1992 stored in the state database and in AIRS. He
suggested the best way to obtain this information would be to contact the U.S. EPA and felt that
Mr. Chuck Mann would be familiar with this approach. He was not familiar with any other
studies conducted at IEPA for developing temporal allocation factors.
A-137
-------
PROJECT CONTACT REPORT
Date of Contact:
Name of Contact:
Representing:
TRC Staff Member:
5/20/93
Jens Laas (608) 266-7718
Wisconsin Dept. of Natural Resources
Sheila Thomas
Jens Laas conducted a study looking at point source - source specific hourly emissions from over
200 companies in the U.S. Facilities that were studied included power plants and paper mills.
Mr. Lass was interested in studying the impact of hourly fluctuations from these on the ozone
layer. The raw data from this study is available in electronic format and contains various types
of information including hourly emissions for pollutants. This file is written in ASCII text and
can be read into SAS for data manipulation. The results of this study are included in a report
that Mr. Laas coauthored with Mr. Ron Dickeson from Radian Corp.- Sacramento. Mr. Lass
was very willing to provide TRC with any information we might need and suggested we follow
up this call during the week of the May 24.
A-138
-------
PROJECT CONTACT REPORT
Date of Contact:
Name of Contact:
Representing:
TRC Staff Member:
5/20/93
Chuck Lewis (919) 541-3154
USEPA A REAL
Sheila Thomas
Mr, Chuck Lewis has coauthored a paper to be presented during the 1993 AWMA Meeting in
Denver. This paper, titled Receptor Modeling of Volatile Hydrocarbons Measured in the 1990
Atlanta Precursor Study presents the results of a Chemical Mass Balance analyses aimed at
quantifying the contributions of various sources to the measured ambient concentrations of total
During the summer of 1990 the U.S. EPA conducted a two-month, two-site air quality monitoring
study in the Atlanta metropolitan area referred to as the "1990 Atlanta Ozone Precursor Study."
The magnitude of the resulting NMOC (non-methane organic compounds) database, with its
continuous hourly time resolution, large number of quantified species, extended duration and
multiple site characteristics, exceeds any previous study of this type.
Mr. Lewis offered to send TRC a copy of this paper and agreed to answer any questions that we
may have after we have read the paper.
NMOC.
A-139
-------
PROJECT CONTACT REPORT
Date of Contact;
Name of Contact:
Representing:
TRC Staff Member:
5/20/93
Statistical Department (202)682-8155
Administrative Dept (202)682-8510
American Petroleum Institute
Sheila Thomas
Ms. Karen Smith and Ms. Theresa Burke (202) 682-8513 from the statistics department were
contacted. Both are unaware of any in-house information on operational data. They indicated
that much of this information is proprietary in nature and will therefore have to be obtained from
the facility directly. They suggested the following four other points of contact.
• Nelson Refinery Operating Index
• National Petroleum Refinery (202) 457-8513
• Oil and gas Journal (800) 345-4618
Department of Energy
A-140
-------
PROJECT CONTACT REPORT
Date of Contact:
Name of Contact(s):
Representing:
5/20/93
Lawrence Fulco
Bureau of Labor Statistics (202) 606-5604
TRC Staff Member: David Winkler
Summary of Discussion:
* Fulco referred me to either of these:
Thomas Plewis (606-6400)
Marty Ziegler (606-6500)
Followup (responsibility: Winkler)
• Contact Plewis or Ziegler
-------
PROJECT CONTACT REPORT
Date of Contact:
Name of Contact(s):
Representing:
TRC Staff Member;
5/20/93
Thomas Plewis
Bureau of Labor Statistics (202) 606-5604
David Winkler
Summary of Discussion:
* Plewis not in -1 left message on voice mail
* I asked for another contact, was referred to John Stinson (202/606-6373)
Followup (responsibility; Winkler)
* Contact Stinson
A-142
-------
PROJECT CONTACT REPORT
Date of Contact:
Name of Contact(s):
Representing:
TRC Staff Member:
5/20/93
Call an Swenson
Bureau of Economic Analysis (202) 254-6630
David Winkler
Summary of Discussion:
BEA tracks earnings by industry, quarterly by States
Productivity measured by Bureau of Labor Statistics (BLS) - I could call
Lawrence Fulco, BLS (202/606-5604)
BEA has:
state-level revenues quarterly and annually
county-level data annually
Format - paper and electronic
Earnings are calculated from Wage and Salary Data at BLS
He'll send me info on diskette
Followup (responsibility: Winkler)
Receive/review diskette information
Contact Lawrence Fulco
A-143
-------
PROJECT CONTACT REPORT
Date of Contact;
Name of Contacts):
Representing;
TRC Staff Member:
5/20/93
Kirit Chaudhari/Hany Quin
Virginia Department of Environmental Management
David Zimmerman
Summary of Discussion:
Contacted Kirit to enquire whether Virginia has collected any additional data on point, area or
mobile temporal factors. Hairy, a member of Kirit's staff, called me back. Virginia is currently
working with the EPA defaults for UAM, but is planning to develop more specific factors. In
fact, Kirit's group is meeting in June with major industry representatives, including VEPCO, to
discuss a cooperative agreement to obtain temporal operating data for major point sources.
Kirit was not aware of any special effort underway for area or mobile sources, although mobile
sources are not his area of expertise. He indicated that VDOT does have time-of-day VMT, but
he had no specific knowledge of its aggregation or format. He will ask another staff member
with mobile source expertise what data are available and follow up with me,
Followup:
Hairy is interested in whether NC has better temporal data because there is a sliver of NC
in the Hampton Roads modeling domain. Follow-up with Harry after we have this
information from NC.
• Contact in late June and thereafter to check on their progress and relate our own progress.
Contact VDOT directly if no word is received (contact?)
Check with DC COG (Council of Governments, especially if mobile sources are an issue
(contact?)
A-144
-------
PROJECT CONTACT REPORT
Date of Contact*
Name of Contact(s):
Representing:
TRC Staff Member:
5/21/93
John Stinson
Bureau of Labor Statistics (202) 606-6373
David Winkler
Summary of Discussion:
BLS deals with employment hours and wages (mostly)
data collected via Form 790, collected at the State level by Employment
Federal Reserve or Commerce Dept may do physical output
Form 790:
reported at 3- or 4-digit SIC level
done by state, aggregated to national level
provides no. employees, no. production workers
samples all SICs (i.e., all facilities are not surveyed every single
month)
hours and earnings reported for non-farm workers
collected monthly
Some State-level data are available, but national data have more detail
He is sending me March "Employment and Earnings" (in mail today, should be
received early next week)
March and June figures give annual averages
Receive, review, file "Employment and Earnings"
Contact State Securities Commission to determine what are collected, how often
Securities Commissions
Followup (responsibility: Winkler)
A-145
-------
PROJECT CONTACT REPORT
Date of Contact: 5/21/93
Name of Contact(s): Matt DeRouville
Representing: Baltimore Metropolitan Council (410) 333-1750
TRC Staff Member: Walid Ramadan
Summary of Discussion:
Mr. Matt DeRouville will send us a draft report prepared for the 1990 emission inventory and
will fax tables on hourly allocation of VMT. Hourly allocation of YMT was based on hourly
trip volumetric functional class and categorized into rural and urban areas.
Mobile model is run by the Maryland Department of Environment.
We provided him. with our Federal Express number for him to send report.
Followup:
If draft report was not obtained by Tuesday, someone should call Matt again.
A-146
-------
PROJECT CONTACT REPORT
Date of Contact:
Name of Contact:
Representing:
TRC Staff Member:
5/21/93
Roger Laprelle (512) 908-1529
Texas Air Control Board
Sheila Thomas
Mr. Roger Laprelle offered to send a copy of the TACB database in electronic format. He said
that the database contains information of operating schedules (also facility contact information
for operating schedules), enforcement, emissions inventory, inspections, emissions permit and
New Source Review Permits. He added that this database could be pulled into a dBASE foimat
for data manipulation. He hoped to pull together this information by next week. TRC will be
required to make a payment depending on the level of effort required for this task.
A-147
-------
PROJECT CONTACT REPORT
Date of Contact: 5/21/93
Name of Contact: Eva Voldner (416) 739-4467
Representing: Environment Canada
TRC Staff Member: Sheila Thomas
Eva Voldner was not aware of any work being done by Environment Canada for developing
temporal allocation factors. She suggested TRC contact Francois Lavelle at (819) 994-4073 who
is responsible for emissions inventory work.
A-148
-------
PROJECT CONTACT REPORT
Date of Contact: 5/25/93
Name of Contact; Energy Information Administration (202) 254-539.
Representing:
TRC Staff Member: Susan Clayton
Summary of Discussion:
The Manufacturing Energy Consumption report is a triennial survey by SICs 20 and 39. 1988
is the most recent data available. Another report, Petroleum Supply Annual, reports monthly
production values for the year. Both reports are available from the U.S. Government Printing
Office.
Followup:
Susan Clayton to obtain reports.
A-149
-------
PROJECT CONTACT REPORT
Date of Contact:
Name of Contact:
Representing:
TRC Staff Member:
5/25/93
Paulette Yoimg
EIA Coal & Uranium Division (202) 254-5389
Susan Clayton
Summary of Discussion:
The Quarterly Coal Report lists weekly coal production information. Paulette will send the
January 1992 report. She also gave me contacts in the following divisions:
1. Uranium and Alternative Fuels: Luther Smith (202) 254-5565
2. Electricity: Larry Prete (202) 254-5671
3. Oil and Natural Gas: Jimmy Petersen (202) 586-6401
A-150
-------
PROJECT CONTACT REPORT
Date of Contact: 5/25/93
Name of Contact: Steve Aidenson
Representing: Minneapolis/St. Paul Metropolitan Council (212) 201-6337
TRC Staff Member: ' Walid Ramadan
Summary of Discussion:
Steve Aidenson does not have any information on VMT. He thinks it is usually done through
home interviews. He suggested we contact the Minnesota Department of Transportation, Office
of Data Analysis, Transportation Data Section and speak to Mr. Gary Graves (612) 296-1671.
Followup:
Contact Gary Graves of the Minnesota Department of Transportation. He is the Director of the
Transportation Data Section in the office of Data Analysis. Telephone number is (612) 296-1671.
A-151
-------
PROJECT CONTACT REPORT
Date of Contact:
Name of Contact:
Representing:
TRC Staff Member:
5/25/93
Chuck Durvis
Sail Francisco
(510) 404-7731
Walid Ramadan
Metropolitan Transportation Commission
Summary of Discussion:
He did some simulation, but "nothing real." Simulation for estimating VMT by time period was
based on travel surveys and information on trip purpose by time of day. Travel surveys and
model simulated average trip lengths are used to simulate VMT by time period.
A-152
-------
PROJECT CONTACT REPORT
Date of Contact:
Name of Contact:
Representing:
TRC Staff Member:
5/25/93
Chuck Mann
EPA/AEERL
Theresa Moody
Summary of Discussion:
Chuck called to tell me that he had finally reached Carlos Card el in a with Georgia Tech to request
the SOS Atlanta 6-week Summer Intensive Study data. Carlos is still waiting on a subset of the
data. He has only the highway vehicle related data. Also, the GA DOT has traffic data reported
by County with lots of detailed specifics, etc. and traffic counters, etc.; Chuck does not want to
get into this much mobile source detail with the DOT's for each State. He asked me to followup
with Ron Methier (404-363-7016) for the Stationary Oxidant Study (6-week summer intensive)
and also asked me to all the GA DOT, Darrell El well (404-986-1361) to request the highway
mobile source database (SOS).
Followup:
According to Craig Ellis, the highway database is not ready yet.
A-153
-------
PROJECT CONTACT REPORT
Date of Contact:
Name of Contact:
Representing:
TRC Staff Member:
5/25/93
Ron Methier
Georgia Department of Environment Protection
Theresa Moody
Summary of Discussion:
Referred to Mike Fogle; Ron Methier is gone for the day; he did not return my call from
9:30 a.m. this morning. Receptionist left message for Mike Fogle to call me. He's unavailable
now (12:15 p.m.). Need the stationary source database from the 1992 SOS (Atlanta 6-week
intensive) program.
Talked to Mike Fogle on 5/27/93. He will find the database contact person and have them call
me ASAP. Fogle recognizes that the database exists and thinks we can probably have it here by
June 4, 1993.
A-154
-------
PROJECT CONTACT REPORT
Date of Contact: 5/25/93
Name of Contact; Patty Schioup
Representing: Atlanta Regional Commissions
TRC Staff Member: Theresa Moody
Summary of Discussion:
Is there a summary report for the SOS 1992 Atlanta summer intensive. Out til 6/1/93; called
receptionist, Trena Jackson 364-2530 also worked on the project. Left voice mail message for
Trena to call me back and get copy of SOS project summary report, if available.
A-155
-------
PROJECT CONTACT REPORT
Date of Contact;
Name of Contact;
Representing:
TRC Staff Member:
5/26/93
Mike Rogers, Professor of Atmospheric Science at Georgia Tech
Georgia Tech and Southern Oxidant Study (SOS)
Craig Ellis
Mike Rogers role with the Southern Oxidant Study has been with field measurements and data
gathering. He indicated that SOS is a database comprised of 7 weeks of measurement data of
ozone and related pollutants at 5 minute intervals. These measurements were taken in the Atlanta
metropolitan area during July and August of 1992 and does not include mobile source data.
The SOS database is presently undergoing QA and is not completely available for public use and
distribution. However, the portion of the database that has been QA and is available for public
use is approximately 2 gigabits in size. The estimated final size of this database will be 6
gigabits.
Mobile source measurements are presently ongoing and are expected to be ready for input into
this database by October 1993. Carlos Cadelina has oversight for the development of the SOS
database and should be contacted for further details.
A-156
-------
PROJECT CONTACT REPORT
Date of Contact:
Name of Contact:
Representing:
TRC Staff Member:
5/26/93
Tim Christian
State of Georgia Department of Transportation
Theresa Moody-
Summary of Discussion:
Original call was placed to Darrell Elwell. He was out today, Tim Christian works for Darrell.
I requested the electronic database generated from the mobile source study of the Southern
Oxidant Study (SOS) which occurred in 1992 over a 6 week period last summer in the Atlanta
area. Thirty-six locations monitored. Hourly data collected, Tim said he would send the
database this week on 1 floppy diskette. I asked if there was a summary report for the entire
SOS effort. He has not ever seen one, but referred me to Patty Schroup (404-364-2599) at the
Atlanta Regional Commission.
A-157
-------
TRC
Environmental Solutions through Technology
TRC Environmental Corporation
100 Europe Drive, Suite 150
Chapel Hill, NC 275 U
v (919) 968-9900 Fox (919) 968-7557
TO: Chuck Mann, AEERJU^
FROM: Theresa Kemmer Moody and David Winkler, TRC
DATE: June 9, 1993
SUBJECT: Improvement of Temporal Allocation Factor Files Project
The purpose of this memorandum is to summarize available information thought to be pertinent
and useful for the Temporal Allocation project. In summary, TRC has identified the five major
potential sources of new information which may provide pertinent data. The sources of
information are listed below. Data analyses and reviews are not complete on all of these sources
of information.
Business and Labor Statistics data (BLS): These data provide broad coverage
of economic statistics that can be used for surrogates activity (physical output and
production labor hours are two promising data items). These data are often
available on a monthly basis, which would allow generation of documented default
seasonal allocation factors that could be easily updated in the future.
Southern Oxidant Study (SOS): The study contains both mobile and stationary
point source data. We axe reviewing these data to determine how they could be
used to improve TAFFs.
Lake Michigan Ozone Study database i.LMOS): As in the case of the SOS
data, we are reviewing these data to determine their usefulness.
California Air Resources Board (CARB) "Hot Spots" pooled source test
reports: These data may be of help in targeting key SCCs for improvement
Some operating data axe included, but do not represent a large number of
observations.
Texas Air Control Board (TACB) stationary source operating schedule
database: From telephone conversations with TACB, it appears that operating
data suitable for use in developing TAFFs may exist in this database. The
database was received on 9 June and is currently under review.
The AIRS Facility Subsystem (AFS) was searched to determine and rank the top 100 source
classification codes (SCCs) by pollutant (criteria only) on a national basis reporting for the 1990
baseyear inventory and State implementation plans (SIP). All the sulfur oxides (SO.) emissions
are reported by the States as sulfur dioxide (S02). A majority of the nitrogen oxides (NOx) were
reported as nitrogen dioxide(NO,), with only six SCCs reporting as nitrous oxide (NO).
Offices in California, Colorado, Connecticut Illinois, Louisiana, Massachusetts, New Jersey, New York, North Carolina, Pennsylvania, Texas,
Washington, Washinqton, D.C., and Puerto Rico
A-158
Printed on Recycled Paper
-------
For each pollutant, TRC calculated the contributing percentage of the top 100 SCCs ranked to
the total national emissions reported for the same base year period. TTiese AFS reports are
attached to this memorandum. For each pollutant, the top 100 SCCs (indicated on the individual
pollutant AFS reports) account for between 62 and 96 percent of total reported national
emissions. It is important to note here that the top 100 SCCs emitting the individual criteria
pollutants reported in AFS are not always the same categories. We are not suggesting that all
of the top 100 SCCs should be reviewed under this project, but do believe that this subset of the
AIRS database is a manageable list of potential review subjects. Table 1 summarizes the impact
of the top 100 SCCs on national emissions of the studied pollutants.
TABLE 1. FRACTION OF NATIONAL EMISSIONS
REPRESENTED IN SCC STUDY
Pollutant Fraction of National Total from Top 100
PM
PM-10
NO;
voc
SO,
CO
78%
75%
95%
62%
96%
94%
Also attached for your review is a copy of an example telephone survey questionnaire that we
would use (after your approval) to gather information from model plants or facilities. This
approach would be an effort to acquire information necessary for updating or generated temporal
allocation factor files for specific, high-impact source categories.
I have enclosed a diskette copy of the LMOS database and supporting documentation that we
received late yesterday for your files.
A-159
-------
Questionnaire for Determining Relevant Data for Use as Surrogate Indicators of Temporally
Allocated Pollutant Emissions
Facility Name ,
w ¦ ¦ _ '
Facility Location
Contact Person (phone no.)
Facility SIG'SCC(s) (note: interviewer should determine and record this information after the
interview in the space provided below)
1. Type of facility
2. Type(s) of facility process(es)
3. . Operating Schedule (be as specific as possible; i.e., number of and length of work shifts
throughout various times of the year; also, what shifts occur during holidays; if shifts vary
during various times of the year, obtain this information)
4. Relative production/consumption data [i.e., no CBI data, only percentages of a given
product produced during various times of the year broken down as much as possible (e.g.,
by day versus by month or by season);]
5. Relative data on process throughput, fuel throughput, process stack gas flow rate (again,
ask only for percentages and not specific CBI information)
A-160
-------
see
CO
sec
CO
30100504
563985
30100599
10512
30300913
502206
10100902
10424
30700104
340323
30301201
10362
30300101
270661
10100901
10355
ao«O0201
269290
10201101
10234
30300813
191231
30501604
10004
30100509
130298
39000689
9938
30300826
120214
30113299
9250
20200202
114427
30103101
8910
30300914
102195
10100604
8801
30700106
96915
10200202
8772
30100503
96391
10100226
8708
30400301
95244
30300107
8643
10200902
88865
20200301
8045
10100202
75660
10300903
7721
30300103
68903
10200704
7149
30300904
56003
30600106
7123
10200601
50019
10100404
7119
30103503
4 8414
30100104¦
6928
10200901
45129
10100903
6851
30100500
41837
20100202
6021
10100601
40824
30390003
5812
30300105
39590
20200203
5469
30700110
37617
30501099
• 5261
30300908
35422
10100101
5040
10201402
- 34668
30115801
4930
10200602
34595
390 00201
4849
39000699
32807
10200205
4706
30300102
32408
10200603
4603
30300315
29583
30400403
4559
30199999
28471
30100308
4534
10100201
26885
30100305
4498
30100603
25172
30103599
4273
39000702
23658
30500706
4250
30300812
23441
10200701
4242
30400315
22796
_______
10100222
21196
100 racs 4,
520,964
30600104
21068
10200204
20717
All rsca 4,
821,913
10101301
19990
20400401
19524
Pet of Total
94%
30100501
19420
20200201
17183
50100101
16850
10100301
16152
10100212
16010
30101901
15704
30600301
15207
30402004
15127
30300907
14624
10200903
14193
50300105
13539
10200401
13491
30400701
13233
30100103
12857
30300822
12723
30103102
12580
30100506
12479
10100203
12460
10100401
12392
10200907
12165
31000203
11808
10100602
11644
20100201
11033
10100204
10542
-------
see
S02
sec
S02
10100202
5272421
30400402
16254
10100203
1489760
10100501
15730
10100201
1367097
30300825
15131
10100212
998859
30601401
14352
10100301
429799
30103201
13750
10100222
413945
30601201
12889
10200401
310879
30600504
12738
10100401
291584
30301002
12649
10200202
197047
39000789
12576
30500201
188059
30402001
12429
10200204
175909
30100507
12420
31000201
142434
10200206
12299
10100204
129644
30600904
11736
10100303
117937 -
30501402
11708
10100226
98904
10200707
11521
10100404
93458
30600005
11435
10100302
87451
30609904
11084
30700104
86235
10100217
10644
10100225
86222
30300504
10502
30102308
82330
10200402
10499
30S00104
79809
10200602
. 10253
31000205
73023
10200104
10029
10200219
63045
30501511
10023
30500706
58692
31000299
9919
10200799
54649
30300813
9898
30301001
53456
30300502
9887
30S00401
52874
3070010S
9721
30300503
51503
10200902
9585
39000201
50887
30699999
9227
10200205
50774
30609902
8956
10200701
494S8
30300315
8872
10200203
45626.
10200901
8704
10200301
44790
30102316
8691
30300101
43573
39000889
8556
10200201
40948
39000203
8 360
10201402
40360
30103204
39405
100 recs
14,040,316
30199999
37338
30600106
34714
&11 rscs
14,573,313
30300512
34233
30102301
33Q68
Pet of Total 96%
30300306
32862
10100221 '
321-79
10100101
30018
10100801
29837
10300209
29589
10200226
28229
30103202
28099
10300207
27396
10100223
27038
10200601
24671
20100102
23711
10200501
23227
30102306
21685
30700110
20441
10300401
19586
30190099
19535
39000699
19152
30600103
18463
10100601
16835
30300599
16767
30609903
16747
30400403
16360 '
20200202
16332
30501604
16311
-------
see
NO 2
see
N02
10100202
2071479
10300603
11363
10100203
747445
30600102
10977
10100201
679324
30300933
10695
20200202
603036
30500201
10425
10200601
446933
10200502
10173
10100212
394391
30199999
10045
10100222
392735
10300602
9881
10100601
324805
103 0 0209
9808
10100301
279573
30700110
9296
30500206
246955
30600103
8935
10100401
• 169652
50100101
8562
10200602
154110
30501404
8458
10200202
124840
20200102
8416
10100226
122676
10100217
8367
10200401
115517
10200903
8224
30600104
111114
30501212
7987
20200201
91847
10200226
7889
30500606
80995
10100701
7838
31000203
73715
30102308
7790
10100303
73040
50100102
7564
10100604
68832
10201201
7387
10200201
'63485
30101302
7376
10100404
63257
20100101
7155
10100302
62259
30700704
7148
10200204
61430
50100201
7063
10200902
49805
10200301
6963
30500706
47883
10200224
6 878
10100225
46849
30119701
6808
20100202
40694
31000201
6750
10100602
39837
10300502
6565
10200701
39734
30190003
6528
20100201
38961
•20200204
€119
30600201
38643
10100801
5836
10100223
38180
31000414
5802
30600401
32227
10201301
5731
10200603
29963
;
10100101
29157
100 recs
8,921,900
30501402
28739
10200203
28377
All recs
9,340,910
30302312
26804
10200901
25959
Pet of Total 95%
30100306
25699
39000201
25407
10100204
23732
30700104
22308
30600106
22080
30501604
21741
30501403
• 21676
10100221
20995
30700106
20731
20200203
20435
10200799
19326
10200501
19133
20100102
18455
10200205
16876
10200219
15322
10201402
14916 .
10300601
14218
20200401
13871
10200704
13568
39000699
13320
30700105
13211
10300401
12923
30101301
12178
39000S89
11665
-------
see
PM10
see
PM10
30302312
24994
30903099
2249
10100202
22333
30501014
2136
30100603
18356
30101801
2099
30300999
15336
30504099
2091
30200611
15203
31401101
2067
50100101
12737
30588801
1933
10100217
10905
30300307
1871
30500201
10838
30502007
1845
10200401
10724
30501009
1838
30700104
9997
30501403
1820
30500716
9654
30300101
1805
30502003
9033
30501602
1780
10100301
8896
30301301
1778
30302302
8581
30900205
1698
30600201
8550
30200805
1694
30500606
8283
30200789
1687
10100212
7S41 ¦ .
31299999
1677
10100204,
7330
30200608
1666
30501604
6904
30300306
1596
30400104
S829
10100303
1584
30400399
6811
10101202
1579
10100222
64 64
30200786
1575
30501021
6313
10101101
1539
10100201
5362
3Q502004
1447
30200734
5141
10200204
1437
30200508
5047
30600104
1426
30200784
4822
30200606
1426
30502002
4768
' 30703098
1378
30500302
4639
10100404
1378
30400899
4557
30539999
1322
30400217
4363
30302304
1321
30500617
4110
30300512
1318
30501010
4010
10200901
1316
30502006
3927
30504021
1305
30500613
3857
30800199
1279
30502001
3822
30501011
3767
100 recs
448,471
10100401
3630
10200902
3122
All raca
592,718
30299998
3048
50100103
2908
Pet of Total
15%
30300304
2893
10100203
2881
30501039
2873
30501610
2837
30700105
2813
30500303
2771
10200202
2752
30502011
2733
30300302
2716
30501024
2696
30200607
2687
30700704
2686
30501099
2678
30501402
2657
30501008
2636
30500706
260$
30199999
2515
30700106
2457
10100226
. 2450
30500205
2389
30501004
2296
30501007
2291
30200782
2275
30300825
2251
-------
see
voc
see
VOC
30199999
69007
40500599
8473
20200202
61647
30101401
8196
40200901
57279
30800799
8126
49099999
56774
40400250
8093
40200101
49236
40500201
8078
40201901
49205
39999994
7823
40500511
39803
30700704
7812
30600503
34964
40201606
7694
30688801
29672
30600701
7681
30201003
25 882
30300302
7681
40201301
25295
40301097
7601
40299996
24643
40100399
7595
10200902
23745
40202201
7591
30100503
22957
40200410
7330
40600131
22646
40301102
7164
40500501
21354
40299995
7013
33088801
20014
40400111
6872
40301012
19955
30101802
6769
30700104
19744
30600401
6694
40200401
19560
30125405
6619
30600801
19386
30180001
651S
30188801
18945
40100225
6231
30100504
18927
40301199
5969
40200110
16826
30103102
-5951
30101812
16468
40201101
5913
30300306
15602
30600803
5806
40200801
15365
30119705
5733
30102601
15325
30400110
5587
30106003
15111
39000699
5586
39993999
14141
30300813
5537
30102S99
13949
40500401
5457
40202501
13528
40600240
5444
40301099
13394
40201726
5389
30106099
13248
31000299
5311
30600201
13014
40100202
5259
30102501
12371
31000203
12342
100 racs
1,467,500
10100901
11972
30699999
11775
All racs
2,330,741
10200602
11661
40600141
11238
Pet of Total 62%
30600805
11136
40301008
10891
10200901
10883
30101807
10814
49099.998
10761
30119701
10654
30600504
10635
50100701
10388
10100601
10063
40500311
9735
40188898
9559
30101899
9509
40200701
9459
10100202
9428
30600602
9397
40500301
9208
40200501
9203
30600104
9088
40200810
9000
30102505
8828
40201399
8719
30199998
8638
40500101
8548
10200601
8492
-------
see
PB
see
PB
30301001
95
50300599
0
30400217
75
50300506
0
30300823
72
50300501
0
30301015
69
50300205
0
30400402
47
50300202
0
30400301
47
50300201
0
30301005
27
503Q0108
0
30301002
. 15
50300106
0
30301022
14
50300105
0
30103599
13
50300104
0
30300503
12
50300103
0
30300801
11
50300102
0
30301011
10
50300101
0
30301009
10
50290010
0
20200102
10
50290006
0
30400525
8
50200602
0
30400405
8
50200601
0
30500606
7
50200507
0
30300504
7
50200506
0
30103507
7
50200505
0
30400499
6
50200302
0
30301020
6
50200301
0
30301014
6
50200105
0
30300904
6
50200104
0
30300502
6
50200103
0
30501024
5
50200102
0
30400509
5
50200101
0
30400406
5
50190006
0
30301099
5
50190005
0
10100212
5.
50100701
0
30400401
4
50100601
0
30400224
4
50100516
0
30301012
4
50100515
0
30588801
3
50100510
0
30400403
3
50100508
0
30103552
3
10100203
3
100 reca
667
30501404
2
30400704
2
All reca
667
30400701
2
30400599
2
Pet of Total
100%
30301025
2
30301018
2
30301016
2
10200401
2
10100202
2
30500699
1
30500619
1
30500609
1
304S9999
1
30400506
1
30400303
1
30301019
1
30300914
1
10200202
1
50390010
0
50390006
0
50390005
0
50300899
0
50300830
0
50300810
0
50300701
0
50300603
0
50300602
0
50300601
0
-------
see
FT
see
PT
10200902
346251
30200603
5904
10100101
223796
30504099
5838
10100202
112861
30500617
5608
30101301
B4993
30502004
5375
30700104
43566
30501090
5304
30501604
38042
10200903
5048
30200611
37752
10100601
5000
30502003
35404
30200511
4993
30400104
32698
30200804
4937
30502002
32652
30504033
4743
30502006
30218
30200506
4735
10100301
263 83
30299998
4722
30500201
25483
30300825
4697
30500S06
23441
30501905
45 64
30502001
21541
10200701
4540
10100201
20725
30388801
4503
30600201
20095
10200301
4441
10100222
19705
30504021
4375
10100401
19576
30501039
4274
10200901
19529
30200601
4263
10200401
19055
30501001
4251
30200508
19014
30200805
4107
10200202
15215
30200782
4015
30700106
14887'
30500613
4000
10100302
13527
30300307
3038
10100212
13482
30100603
3832
10100203
12949
30501024
3755
30199999
12931
30501402
3673
30300304
12794
30501031
3582
30502007
11961
10200205
3519
30200784
11256
30299999
3507
10100226
10784
30500614
3457
10200204
10745
30200504
3411
30500301
9916
30501012
3409
10200601
9675 ¦
30200606
3355
30500706
9633
30500302
9559
100 recs
1,789,111
30300001
9505
~
30700105
9487
All recs
2,278,359
30200605
9193
30501011
9142
Pet of Total 78%
39999999
8492
30700899
8324
30501010
8016
30200802
7859
30501099
7838
10100225
7634
30400399
7328
30301301
7217
30700799
7137
30300302
7124
30501503
7106
30200505
6790
30502011
6688
30600104
6656
30200608
6444
30300201
6352
30200604
6319
30900202
6288
30599999
6247
30700704
6233
30500303
6042
39000699
6030
10201101
5971
30300999
5960
-------
ACTION PLAN FOR THE DEVELOPMENT AND
IMPROVEMENT OF
TEMPORAL ALLOCATION FACTOR FILE(S)
TECHNICAL MEMORANDUM
EPA Contract No, 68-D9-0173
Work Assignment 3/314
Prepared for:
Charles 0. Mann
U.S. Environmental Protection Agency
Air and Energy Engineering Research Laboratory
Research Triangle Park, NC 27711
Prepared by:
TRC ENVIRONMENTAL CORPORATION
100 Europa Drive, Suite 150
Chapel Hill, NC 27514
July 7, 1993
CH-93-69
-------
TABLE OF CONTENTS
Section Page
List of Tables v
1.0 INTRODUCTION 1- i
2.0 DATA SOURCES............ 2-1
2.1 BUSINESS AND LABOR STATISTICS DATA 2-2
2.1.1 Mu lti-industry Publications 2-3
2.1.2 Beverages and Breweries 2-5
2.1.3 Paper, Paperboard, and Wood Pulp 2-6
2.1.4 Metals Industry 2-6
2.1.5 Plasties Industry 2-6
2.1.6 Printing Industry...... 2-7
2.1.7 Textile and Fibers Industry 2-7
2.1.8 Fossil Fuels and Power Generation 2-8
2.1.8.1 Electric Power 2-8
2.1.8.2 Coal... 2-9
2.1.8.3 Petroleum and Natural Gas 2-9
2.2 CALIFORNIA AIR RESOURCES BOARD DATA 2-10
¦ 2.2.1 Identification of Data Availability 2-10
2.2.2 Description of Data 2-11
2.2.3 Approach for Incorporating Data into a Temporal Allocation Factor
File... .....2-11
2.3 TEXAS AIR CONTROL BOARD DATA .2-12
2.3.1 Identification of Data Availability 2-13
2.3.2 Description of Data........ 2-13
2.3.3 Approach for Incorporating Data into a Temporal Allocation Factor
File .......2-14
2.4 SOUTHERN OXIDANT STUDY DATA 2-14
2.4.1 Data Description 2-15
2.4.2 Approach for Incorporating Southern Oxidant Study Data into a
Temporal Allocation Factor File 2-15
2.5 LAKE MICHIGAN OZONE STUDY DATA 2-16
„ 2.5,1 Description of Data.. 2-16
2.5.2 Approach for Incorporating Data into a Temporal Allocation Factor
File ..2-16
2.6 CONTINUOUS EMISSIONS MONITORING DATA 2-17
2.6.1 Identification of Data Availability 2-20
2.6.2 Approach for Incorporating Data into a Temporal Allocation Factor
File 2-20
2.7 WASTEWATER DATA 2-21
A-169
-------
TABLE OF CONTENTS (Continued)
Section Page
2.7.3 Approach for Incorporating Data into a Temporal Allocation Factor
File 2-22
2.8 OTHER DATA SOURCES 2-22
2.8.1 Waste-to-Energy Source Data 2-22
2.8.1.1 Identification of Data Availability 2-23
2.8.1.2 Description of Data 2-23
2.8.1.3 Approach for Incorporating Data in a Temporal Allocation
Factor File 2-23
2.8.2 Acid-Modes Field Study 2-24
2.8.3 Urban Airshed Model Emissions Preprocessor System Data ..... 2-25
2.8.3.1 Identification of Data Availability 2-25
2.8.3.2 Description of Data 2-25
2.8.3.3 Approach for Incorporation of Data into a Temporal
Allocation Factor File 2-26
3.0 SOURCE CATEGORY PRIORITIZATION 3-1
3.1 METHODOLOGY 3-1
3.1.1 Source Category Ranking Based on Criteria Pollutant Emissions . . 3-1
3.1.2 HAPs-Related Source Category Prioritization 3-3
3.2 SUMMARY AND COMPARISON WITH EARLIER PRIORITIZATION
EFFORTS 3-3
3.3 OTHER ISSUES ..... 3-14
4.0 PLAN OF ACTION FOR TEMPORAL ALLOCATION FACTOR FILE
DEVELOPMENT 4-1
4.1 ORGANIZATION AND RESOURCE REQUIREMENTS 4-1
4.1.1 Data Collection 4-1
4.1.2 Database Design and Management 4-2
4.1.3 Technical Support 4-2
4.2 TEMPORAL ALLOCATION FACTOR FILE CONSTRUCTION 4-3
4.2.1 Tier 1 4-4
4.2.2 Tier 2 4-4
4.2.3 Tier 3 . . .. 4-5
4.3 APPLICATION OF TEMPORAL ALLOCATION FACTOR FILE 4-5
5.0 BIBLIOGRAPHY 5-1
APPENDIX A. STATISTICAL AND ECONOMIC INDICATOR REFERENCES ..... A-1
APPENDIX B. CARS REPORT REVIEW SUMMARY . B-l
A-170
-------
TABLE OF CONTENTS (Continued)
Section Page
APPENDIX C. ACID-MODES FIELD STUDY DATABASE FORMAT C-l
APPENDIX D. EPS DATABASE FORMAT D-l
iv
A-171
-------
LIST OF TABLES
Number Page
2-1 Commodities Inventoried in the Commodity Year Book 2-4
2-2 CEM Data Summary , , , . 2-18
3-1 Rank-ordered SCCs . 3-4
3-1 High Priority Source Categories 3-10
4-1 Characteristics of Intermediary Files 4-4
CH-93-69
V
A-172
-------
1.0 INTRODUCTION
Emission inventories traditionally have been developed to produce estimates of emissions
for annual or daily time periods. In order to be used as input to photochemical and other
atmospheric simulation models, emission estimates on an hourly time basis axe usually required.
Ideally, emissions for specific hourly time periods would be measured or calculated directly at
the emissions source. However, this approach is normally impractical due to technical and
resource constraints. As an alternative, hourly emission estimates can be obtained using surrogate
temporal allocation factors from "temporal profiles" assigned to specific emissions source
categories. Estimates of hourly emissions may then be calculated by applying the appropriate
temporal allocation factors to available annual, seasonal, or daily emission values. This approach
has been followed in a number of previous air pollution studies, including the National Acid
Precipitation Assessment Program (NAPAP), the Northeast Corridor Regional Modeling Project
(NECRMP), and others. Since the performance of atmospheric simulation models is strongly
dependent upon the availability of accurate, temporally resolved emissions values, it is important
that suitable methodologies and databases be available to personnel responsible for developing
the emissions estimates needed for model inputs.
TRC has been directed by the U.S. Environmental Protection Agency (EPA) Air and
Energy Engineering Research Laboratory (AEERL) to evaluate the quality and completeness of
data and methodologies presently being used for temporal allocation of emissions factors. TRC
has collected data to improve existing temporal allocation factor (TAF) file(s). The TAF file(s)
will be used by the emissions model processing systems that calculate temporally resolved
emission estimates for model input. The model processor systems presently in use or expected
to be used in the future include the Emissions Preprocessor System (EPS) for the Urban Airshed
Model (UAM), the Geocoded Emissions Modeling and Projections (GEMAP) system and the
Flexible Regional Emissions Data System (FREDS).
The purposes of this memorandum are: to summarize the information contained in the
various data sources identified during the course of the work assignment and discuss the
usefulness of the data for improving or developing TAF file(s); to discuss the methodology used
to prioritize source categories or Source Classification Codes (SCCs) for incorporation into the
CH-93-69 1-1
A-173
-------
TAF file(s) and present the rank-ordered SCCs; and to propose a plan of action for developing
the TAF file(s).
Section 2 of this memorandum identifies and describes each of the data sources from
which applicable data may be extracted for improving or developing TAF flle(s), including the
units in which the data are given. Section 2 also discusses the approach for incorporating the
data into the TAF file(s). Section 3 discusses the methodology used to prioritize the source
categories for TAF file(s) development.
Section 4 proposes the methodology for incorporating the data into the TAF flle(s) and
defines the specific format of the TAF file(s). Appendices A through D provide additional
information and supporting documentation for Sections 2 through 4 of this memorandum. The
data elements to be used to generate TAF file(s) vary among the data sources. The general
approach used will be to use time series data from the identified data sources that are either
direct measures of or appropriate surrogates to the activities generating emissions. These time
series data will be normalized on a temporal basis to provide the operating profiles required to
develop the final TAF flle(s).
CH-93-69
1-2
A-174
-------
2.0 DATA SOURCES
Descriptions of the data sources identified as providing sufficient information to support
TAF file(s) development are included in this section. The source is identified through an
intensive search of the literature and contacts with many EPA, State and local air pollution
control agencies and other federal agencies. Each data source has been described in the context
of applying these data towards the development of temporal profiles, using the information
gathered to date. Where preliminary information has permitted, the descriptions include potential
applications and methods of creating temporal profiles. However, such analyses are not yet
possible for those data sources where complete information or the databases themselves have not
been obtained. TRC recognizes the applicability and/or limitation of each of these data sources
and will apply the information where appropriate in the development and improvement of the
TAF file(s).
Two distinct types of data are presented. First, business and labor statistics were
identified as potential surrogate-type indicators applicable at a major category level where
process-specific data are absent or incomplete. Second, specific category or plant data were
identified through a variety of sources including state-of-the-art emissions inventories, continuous
emissions monitoring and industrial and federal reports and databases. The data sources
discussed in this memorandum include the following:
Business and Labor Statistics (BLS) data
• Department of Energy (DOE) data pertaining to production/consumption from various
energy industries
• California Air Resources Board (CARB) AB-2588 "Hot Spots" pooled source test reports
• Texas Air Control Board (TACB) stationary source operating schedule database
Southern Oxidant Study (SOS) database
• Lake Michigan Ozone Study (LMOS) database
• Continuous emissions monitoring data
• Wastewater data from publicly owned treatment works (POTWs)
• Operating schedule/parameter data from resource recovery facilities
• Urban Airshed Model, Emission Preprocessor System, Version 2.0 TAF file(s)
CH-93-69
2-1
A-175
-------
The information presented in this section for each data source includes the following.
1. publication/database identification
a. name
b. intended purpose of report/database
c. year/frequency of update
d. responsible agency or organization
2. description of pertinent data
a. definition of data
b. units (e.g., tons, hours, dollars)
c. data collection efforts
3. availability of data (format, cost, delivery, local availability vs. ordering)
4. approach for incorporating data into a TAF file(s)
Item 4 above varies considerably among the data sources listed. The nature of some of
the data sources and the availability of actual data samples for review, allow a prescriptive,
quantitative algorithm. However, other data sources (e.g., the long list of Bureau of Labor
Statistics/Department of Energy data sources) were not predisposed toward a concise, prescriptive
description of the data incorporation approach.
2.1 BUSINESS AND LABOR STATISTICS DATA
Industrial activity and output are often monitored by trade associations, private
organizations, and governmental agencies. The types of statistical information compiled by these
groups include number of employees, labor hours worked, sales, production, capacity, energy
consumption, peak demand, and other operating and economic indicators, such as production rate
per employee. The statistics are published on varying temporal resolutions: seasonally, monthly,
and weekly. The information often is specific to Standard Industrial Classification (SIC) codes.
Labor and economic statistics can be used to develop default TAF file(s) for ozone
atmospheric modeling purposes. The data may be supplemented by industry survey or continuous
emissions monitoring (CEM) data for further temporal resolution to an hourly basis. Information
sources and frequency of publication can be documented for easy retrieval during future TAF
file(s) updates.
A comprehensive index entitled Statistical Reference Index, 1991 Annual is published
annually by the Southwest Research Institute and contains statistical reference sources. Arranged
CH*93h59
2-2
A-176
-------
by subject and names, the index references document abstracts for more detailed information.
Available data compiled at a temporal resolution finer than annual was of interest. If no
information is available for significant industrial sources, TRC will investigate the identified
annual data further to determine if agencies collect information on a more resolved temporal basis
if required.
The general methodology suggested for developing these data into TAF file(s) is
calculating fractional proportions over the temporal basis of the data. For example, monthly
proportions can be developed by dividing each monthly indicator by the yearly total. The basic
assumption is that the operating or economic statistics are surrogate indicators of industrial
processes releasing pollutants. For example, the number of hours worked by employees or the
industry's production rate are directly related to that industry's potential emissions during that
time frame.
The two publications discussed in Section 2.1.1 provide data that cover a wide range of
industrial source categories. These data are compiled on a level suitable for SIC code or third-
level SCC assignment These data will be used to develop seasonal and monthly temporal
profiles. Industrial process-specific data are discussed in Sections 2.1.2 through 2.1.8. These
data will be used to refine the profiles to a higher level SCC classification or finer temporal
basis. Appendix A presents an example listing of all the statistical information obtained by TRC.
2.1.1 Multi-industry Publications
The Commodity Research Bureau (CRB) in New York City publishes the Commodity
Year Book, which represents a dependable and readily available source of temporal data for 97
commodities. The commodities inventoried in the publication are listed in Table 2-1. Monthly
statistics such as production, stocks, consumption, shipments, exports, and imports are reported
for several years. For example, U.S. production of ethyl alcohol and spirits is presented on a
monthly basis in units of millions of tax gallons. The commodity data can be easily assigned
to a six-digit SCC. Seasonal and monthly TAF file(s) could be developed from this data source.
CH-93-69
2-3
A-177
-------
TABLE 2-1. COMMODITIES INVENTORIED IN THE COMMODITY YEAR
BOOK
Alcohol
Aluminum
Antimony
Apples
Barley
Bauxite
Bismuth
Broilers
Butter
Cadmium
Castor Beans
Cattle and Calves
Cement
Cheese
Chromite
Coal
Cobalt
Cocoa
Coconut Oil and Copra
Coffee
Coke
Copper
Com
Corn Oil
Cottonseed aid Products
CRB Futures Index
Currencies
Eggs
Electric Power
Fertilizers (Nitrogen, Phosphate and Potash)
Flaxseed and Linseed Oil
Gas
Gasoline
Gold
Grain Sorghum
Hay
Heating Oil
Hides and Leather
Hogs
Honey
Interest Rates
Iron and Steel <
Lard
Lead
Lumber and Plywood
Magnesium
Manganese
Meats
Mercury
Milk
Molasses
Molybdenum
Nickel
Oats
Olive Oil
Onions
Oranges and Orange Juice
Palm Oil
Paper
Peanuts and Peanut Oil
Pepper
Petroleum
Plastics
Platinum-Group Metals
Pork Bellies
Potatoes
Rapeseed
Rayon and Other Synthetic Fibers
Rice
Rubber
Rye
Salt
Sheep'and Lambs
Silver
Soybean Oil
Soybeans
Stock Index Futures
Sugar
Sulfur
Sunflowerseed and Oil
Tall Oil
Tallow and Greases
Tea
Tin
Titanium
Tobacco
Tung Oil
Tungsten
Turkeys
Uranium
Vanadium
Wheat and Flour
Wool
Zinc
CH-93-69
2-4
A-178
-------
The data are copyrighted and permission for use may be needed from the publishers.
The U.S. Department of Labor, Bureau of Labor Statistics, publishes a monthly document
entitled Employment and Earnings and an annual publication entitled Employment, Hours, and
Earnings, United States. The publications are available for a minimal subscription fee. The
reported data are collected in cooperation with State employment security agencies under the
Cunrent Employment Statistics (CES) Program. Reported statistics include number of employees
(grouped by totals, production and non-supervisory categories), average weekly labor hours
worked, average weekly overtime hours worked, and average hourly and weekly earnings. The
average weekly or overtime labor hours would be of most interest in the development of TAF
file(s) since hours worked correspond to the variations in plant production. The data are reported
on a monthly basis over a series of years. Statistics for mining, construction, manufacturing,
transportation and public utilities, wholesale trade, retail trade, finance, insurance and real estate,
services, and government facilities are generally available to a four-digit SIC resolution.
Seasonal and monthly profiles could be developed from these data. The data may be used
without permission from the publishers. Additionally, the data are available in electronic format
at a modest subscription fee,
2.1.2 Beverages and Breweries
Two publications covering the Beverage Industry have been identified: Beverage Industry
Annual Manual and Brewers Almanac. The Beverage Industry Annual Manual provides
information about production of soft drinks, beer, bottled water, wine, powdered drinks, and
distilled spirits, and information about vending, fountain/food' services, packaging, and
ingredient/sweetener trends. Statistics are reported on an annual production basis in units of
cases, barrels, etc. Several table references indicate that monthly values may be available.
The Brewers Almanac reports monthly statistics for malt beverages, with various tables
on wine, packaged and draught beer, and distilled spirits. Statistics include production (packaged
and draught), shipments, exports, imports, inventories, per capita consumption and
employment/hours/earnings. Production of beer bottles and beer cans is also reported. The data
may be used to develop seasonal and monthly profiles. The data may be easily assigned to broad
SCC descriptions. The copyright status of the data is not known at this time.
CH-93-69
2-5
A-179
-------
2.1.3 Paper, Paperboard, and Wood Pulp
A monthly publication of the American Paper Institute entitled Paper, Paperboard and
Wood Pulp presents seasonally adjusted and unadjusted statistics for two month periods.
Aggregate industrial production and shipment figures are reported in units of tons. Imports,
exports, consumption, inventories, raw materials, employment, and output per man-hour are also
reported. The statistics are also compiled by product type, such as newsprint, tissue paper,
containerboard, and boxboard, and production process groupings are listed, such as unbleached
kraft, solid bleached, and semichemical. These data may be easily assigned to six-digit SCC
categories. The copyright status of the data is not known at this time.
2.1.4 Metals Industry
The "Economic Notes" section of the 33 Metal Producing journal presents monthly and
quarterly statistical data for steel manufacturing. The text cites quarterly shipment values and
new orders for the specific month. Monthly percent-change statistics for steel orders are listed.
A table for steel consumption by end use sector, steel production, and steel output per worker
hours lists values for the current two months. The data, as presented in the journal, could only
be assigned to broad SCC categories (1- or 2-digit resolution). At this time, neither the coverage
of other sectors in the metals industry, nor the copyright status of the material, has been
determined.
2.1.5 Plastics Industry
The Modern Plastics journal was identified as a source of resin and plastics production
data. The journal contains a section entitled "Barometer-Quarterly Update." Reported statistics
vary with each monthly publication. Resin statistics including production, sales, and capacity
utilization may be readily available on a monthly basis from the Society of the Plastics Industry,
Committee on Resin Statistics. Plastic-part production index statistics were identified on a
quarterly basis. Quarterly import and export statistics by industry sector in units of millions of
pounds are also published. The ease of SCC assignment will depend on the statistic and data
CH-93-69 2-6
A-180
-------
source utilized. For example, the import and export data are resolved to industry sector and
plastic type (e.g., polystyrene, polypropylene, etc.). The copyright status of the material was not
identified but will probably depend on the original data source.
2.1.6 Printing Industry
The Graphic Arts Monthly journal publishes a monthly data table of printing industry
statistics. The two issues obtained present different statistics, for two quarters or two months.
Production index statistics, by printing sector, are presented in one table. A productivity index,
paper shipments (thousands of tons), newsprint consumption (thousands of tons), and new orders
and shipments indices are presented in the second table. The data could be assigned only to
broad SCC categories. Data that are not copyrighted can be obtained from the original data
source.
2.1.7 Textile and Fibers Industry
The Fiber Organon journal is a monthly publication of the Fiber Economics Bureau, Inc.
Statistics are reported for two month periods of the current and previous year, and quarterly for
a series of years. Manufactured fiber shipments (measured in millions of pounds) for artificial
(cellulosic) yam and rayon staple, synthetic (noncellulosic) nylon, polyester, olefin, and
staple/tow/fiberfill nylon, acrylic/modacrylic, polyester, and olefin are reported for two month
periods of the current and previous years. Quarterly manufactured fiber data (measured in
millions of pounds) by product type are reported. Statistics include production, shipments
(domestic and export), and ending stock. The journal also reports capacity information from two
surveys compiled each year in May and November. The data are reported at a level of detail
sufficient for assigning six-digit SCCs. Permission for use of the data will be needed from the
Fiber Economics Bureau, Inc.
CH-93-69
2-7
A-181
-------
2.1.8 Fossil Fuels and Power Generation
The U.S. Department of Energy (DOE) publishes an extensive collection of fossil fuel and
energy data through the Energy Information Administration (EIA), the independent statistical and
analytical agency within the DOE. Located in Washington, DC, EIA publishes a large number
of periodic and one-time reports related to fossil fuels, power generation, and other energy related
information. Many of the reports are used or cited in various emission inventory and trends
guidelines, methods, and documents published by the EPA. Subscriptions to EIA reports are
available by mail for a nominal fee (less than $100 per year). Certain groups (including federal,
State, and local governments) are eligible for complimentary subscriptions. EIA also publishes
a bimonthly bulletin of new publications, and maintains an electronic publishing system (EJPUB)
which allows the public to electronically access selected data from many of the available EIA
reports for no fee.
The EIA collects data directly from energy producers, consumers, and other sources
through regular reporting and occasional special surveys. They also use indirect estimation
methods for some types of data. The specific data available and temporal resolution of data
found in EIA reports vary. While many reports are published once or annually, others are
published monthly or weekly. The following sections describe reports reviewed which have
potentially useful data, and which have a temporal resolution of tetter than annual.
2.1.8.1 Electric Power
Electric Power Monthly provides monthly data on power generated at U.S. power plants
(classified by energy source and geographic location), as well as data on consumption and stocks
of fossil fuels at power plants. Also included are sales of electricity in kilowatthours by sector
(industrial, commercial, residential, and other) and various electricity cost information, SCC-
speciflc power consumption data are not contained in this publication although fuel consumption
. is provided by fuel type. Data from this publication are also available through EPUB.
CH-93-69
2-8
A-182
-------
2.1.8.2 Coal
The Quarterly Coal Report is an extensive quarterly report on the coal industry. It
reports quarterly coal and coke production, imports/exports, consumption, and distribution. The
most detailed, data arc the coal receipts data and the coal consumption data, which report tons
of coal received and consumed by end-use sector (electric utilities, coke plants, other industrial,
residential and commercial). These data are further broken down into tons of coal received and
consumed by each two-digit SIC group. These data could be used to derive quarterly fractions
of coal consumption for these broad source categories, and can be linked to industrial activity.
Unfortunately, data with a more finely detailed source categorization are not available in this
report. Data from this publication axe also available through EPUB.
Weekly Coal Production is a short weekly report which summarizes coal production, in
tons, for several U.S. coal producing States. Data, from this publication are also available through
EPUB.
2.1.8.3 Petroleum and Natural Gas
Petroleum Supply Monthly provides monthly data on the supply and disposition of crude
oil and petroleum products, including production of crude oil, processing of natural gas, and
refinery operations. These data include, for each product, U.S. production, imports and exports,
and disposition (i.e., as inputs to refineries or as product supplied). Products reported include
crude oil, liquified petroleum gases, motor fuels, coke, kerosene, unfinished oils, naphthas, waxes,
and many others. Weekly Petroleum Status Report contains much of the same information as the
monthly report, but is published weekly. The detailed breakdown of disposition of petroleum
products may allow identification of product-specific source categories (e.g., asphalt and road oil
as a surrogate for paving operations). Data from both of these publications are also available
through EPUB.
Petroleum Marketing Monthly contains the volumes sold (and the prices) of numerous
petroleum products reported, on a monthly basis. Sales to end users are reported separately from
sales to resellers, but a more detailed source categorization is not provided. This publication is
also available on EPUB.
The Oil and Gas Journal is a weekly publication which contains a section on petroleum
industry statistics. This section contains weekly reports on imports, product stocks, crude
OH-53-69
2-9
A-183
-------
production, and refinery operations. The refinery report, which is supplied by the American
Petroleum Institute, may be useful as a surrogate for refinery operations, and could be used to
allocate refinery emissions on a weekly basis. The crude oil production report may also be a
useful surrogate for crude production operations if emissions from such operations are of concern.
2.2 CALIFORNIA ADR RESOURCES BOARD DATA
In accordance with the California Air Toxic "Hot Spot" Inventory and Assessment Act
of 1987, many types of industrial sources were required to submit air toxic emissions data to
local air pollution control agencies in the State. Collectively, the program is called the Assembly
Bill (AB)-2588 program. These facilities were required to undergo source testing in order to
determine the quantities of toxic air pollutants emitted from their operations. The AB-2588
regulations allowed similar facilities or industries to perform pooled source testing. These test
results would then be applied to all sources or facilities of the same type.
2.2.1 Identification of Data Availability
TRC reviewed available California Air Resource Board (CARB) pooled source test reports
and summarized the information on source/processes tested, useful surrogate data (i.e., plant
activity or throughput data), emissions data (if given on an hourly basis), testing period, and
control equipment (including control equipment efficiencies where available). The extracted
information was used to determine the applicability of these data to the development of profiles
for the various source categories represented. For many of the reports, SIC/SCC codes were not
given, thus, TRC made a determination of applicable SCC codes for each of those source
processes. A list of the AB-2588 reports which TRC reviewed is provided in Appendix B. The
listing also includes the reference numbers assigned to each report for accountability and tracking
purposes.
CH-93-69
2-10
A-184
-------
2.2.2 Description of Data
The data contained in the CARB reports include emissions of various pollutants
determined from actual source test measurements taken at each facility during a certain time of
the year while the facility was operating at a normal capacity. For many of the reports, these
emission rates were given in pounds per hour. In addition to the measured emission rates, many
of the CARB reports also contain various process operating parameters which may prove useful
as surrogate indicators when developing profiles. Operating information includes operating
schedules (hours per day, days per week, weeks per year), product throughput (tons or pounds
per year and tons or pounds per hour), and fuel throughput (e.g., pounds per hour) or stack gas
flow rates (e.g., dry standard cubic feet per minute). Some reports gave relative monthly activity
percentages for the various processes tested. It should be noted that the percent operating
capacity at the time of the test is most likely typical of the capacity at which- the plant operates
during that period (e.g., month or season) of the year when the test was performed. For many
facilides, such operating parameters may change considerably during various times of the year.
Finally, in many of the reports, the type of control equipment used for each of the processes
tested is also listed; however, only a few reports specified actual control efficiencies. In some
instances, the control equipment may have been reported, but the report format made it difficult
to determine the type and percent efficiency of the equipment.
2.2.3 Approach for Incorporating Data into a Temporal Allocation Factor File
Information contained in the CARB reports which may prove useful in developing TAF
file(s) is given in Appendix B. The SCC code applicable to each process is also given and is
listed in the same order as the process(es) listing. In general, the CARB source test data may
be used to provide the necessary surrogate indicator information for priority SCCs for which
other information sources are insufficient (so-called "gap fillers").
Since the data contained in the CARB reports are only representative of plant operating
conditions during that period of the year in which the emission tests were performed, these data
may be used for comparison with hourly emissions estimated from other surrogate data
representative of that time period. For example, emission rates (in pounds per hour) which were
CH-93-69
2-11
A-185
-------
measured at a utility boiler in the month of June could only be compared to hourly emission rates
estimated from other sources using surrogate data representing plant operating conditions during
the summer months (or for June in particular) for a utility boiler. Such a comparison would be
most appropriate for those sources for which plant size, scheduling, and/or production data are
comparable in size and/or quantity to the surrogate indicator data developed for the TAP file(s).
Much of the operating parameter data contained in the CARB reports which would be
useful surrogate indicators of pollutant emissions are already specified on an hourly basis. In
many of the reports, operating schedules were also given. However, specific hours of the day
of operation are not given. Therefore, for sources with schedules which have less than 24 hours
of operation {e.g., operating seven or eight hours per day), applicable hours of operation could
be obtained by contacting a representative or the source to improve the accuracy of the profile.
Several CARB reports provided monthly activity percentages for processes tested at the
facility. These numbers represent the amount of production activity during a particular month.
From these reports, it might be possible to provide a fairly accurate estimate of the activity for
other months of the year. These data could yield an accurate indicator of hourly emissions since
the hourly activity information given in the reports is more of an indicator of the actual
production activity for that particular month of the year.
It should also be noted that for several reports, source testing was conducted during
different times of the year, and the activity data from those months in which the tests were
performed could also be compared and proportioned.
2.3 TEXAS AIR CONTROL BOARD DATA
The Texas Air Control Board (TACB) emissions inventory database, which is maintained
by the Technical Services Division of the TACB, includes an abundance of information
pertaining to source activity for individual businesses and their respective process(es). The
purpose of the TACB database is to track all relevant source information for emissions
inventories, enforcement, permitting, etc. The TACB database is updated on a continual basis
whenever relevant source information is received pertaining to permitting new sources, renewing
operating permits, or inspection results.
CH-93-69 2-12
A-186
-------
2.3.1 Identification of Data Availability
Data from the TACB database may be downloaded into various user formats for a fee. One
of the databases received by TRC was formatted for dBASE®. In addition, TRC obtained a
sample of the 1992 TACB Nonattainment Emissions Inventory Questionnaire to review the
information requested. This report gives information for a single business and includes SIC and
SCC codes for each individual process as well as individual process descriptions. Conversion of
the reports to the data file formats needed for TAF file(s) development will require some
programming. TRC has requested the following information, based on conversations with the
TACB. It should be noted that the fee for accessing this information may be between $1000 and
$1500.
« Company name and SIC code(s)
Contact person/phone number
Business description
• Site operating schedule
Seasonal operating percentages
• Facility identification number (FIN) information consisting of the following for each
individual facility processes):
facility name
operating schedule
seasonal operating percentages
SCC code and facility (SIC) description
speciated emissions data (pounds per day)
2.3.2 Description of Data
Two databases were identified from the TACB. The first database is a listing of facilities,
addresses, points of contact, type of business, operating schedules, and seasonal operating
percentages. This database does not contain either SIC or SCC codes, although SCC-level
process descriptions are available.
The second database contains both SIC and SCC codes for each respective business. The
plant specific process information includes the facility operating status, individual process
CH-93-69
2-13
A-187
-------
descriptions and their respective operating schedules, and seasonal operating percentages.
Emission estimates in pounds per day are also provided for each process. Emission estimates
for volatile organic compounds (VOCs) are speciated. Throughput data are considered
confidential and therefore cannot be accessed. Data are maintained for over 6,000 plant sites
within the State of Texas. Site operating schedules are given for each process in hours per day,
days per week, and weeks per year. Seasonal operating percentages are provided as a percentage
for each of the four seasons of the year. Data are collected, through the permitting, SIP
inventory, and enforcement processes.
2.3.3 Approach for Incorporating Data into a Temporal Allocation Factor File
For sources which operate on less than a 24-hour basis, exact operating times may be
estimated.; alternatively, TRC may need to contact a source representative of a given process SCC
to determine the exact time of day in which the process operation would be performed. For each
SCC given, facility (i.e., process-level) descriptions are also given. The TACB data element
"facility" descriptions are synonymous with the Aerometric Information Retrieval System (AIRS)
data element "process."
The hourly operating schedule values determined by the procedure above for each SCC
contained in the database will then need to be proportioned by the percent of operation for each
of the seasons. This will provide a reasonable computation of the hours of operation during a
given season of the year.
The TACB data are comprehensive and may contain surrogate information for a wide
variety of SCCs. The emission estimates are speciated, and this information may be useful for
prioritizing sources emitting hazardous air pollutants. (HAPs).
2.4 SOUTHERN OXIDANT STUDY DATA
The Southern Oxidant Study (SOS) is a point source inventory for the 1992 Atlanta,
Georgia six-week intensive ozone season, July 15 to August 31. The Georgia Department of
Natural Resources, Air Protection Branch conducted a survey of local facilities for the inventory.
Responses from 57 facilities were received. In addition to summary statistics used to update
CH-93-69
2-14
A-188
-------
AIRS/AIRS Facility Subsystem (AFS) data, day-specific emissions and production data for the
six-week intensive ozone season were collected.
)
2.4.1 Data Description
Day-specific data were collected to update 1990 data in the AIRS/AJFS database. Data
collected include the following: process code and description, normal operating schedule
(including normal start and end time for processes), stack height, type of raw materials, specific
gravity or density, percent by weight VOC, and throughput and emissions. All data not involving
time measure were reported in units chosen by those surveyed. The general survey data
(i.e., non-day-specific) were entered into the State implementation plan (SIP) Air Pollutant
Inventory Management System (SAMS), where the units were standardized. For each process,
the following data were recorded for each day in the six-week intensive ozone season; actual
operating schedule (hours per day, start time and end time); input or production rate; and VOC
emissions. These data were not converted into electronic format. This is not an annual survey
and no further surveys of this type have been scheduled.
2.4.2 Approach for Incorporating Southern Oxidant Study Data into a Temporal
Allocation Factor File
The SOS data set will provide insight into diurnal production and emission cycles for the
summer months, and supplement the AIRS/AFS data. The data can be used to describe daily
operations at the type of facilities and processes found in the survey. Although the survey only
addresses six weeks of the year, the daily production cycles may be used to support estimates
of other season's activities for SCCs lacking diurnal data. The general data have been obtained
and efforts to obtain copies of the 2,000 to 2,300 pages of day-specific, process-specific
information are underway.
CH-93-69
2-15
A-189
-------
2.5 LAKE MICHIGAN OZONE STUDY DATA
Data gathered by the Wisconsin Department of Natural Resources (DNR) to improve
emission rates for carbon monoxide (CO), nitrogen oxides (NOJ, and VOCs for the Lake
Michigan Ozone Study (LMOS) were requested from the Lake Michigan Air Directors
Consortium (LAD CO). Due to the current draft form of the data, personnel at LAD CO were
unable to release these data to TRC, but referred TRC to the Wisconsin DNR. TRC received
data from the Wisconsin DNR, along with related "foundation files," also in draft form.
2.5.1 Description of Data
The LMOS data include emission rates calculated from production data received by the
Wisconsin DNR for approximately 200 of the largest facilities in the 21-county LMOS region
of Wisconsin. Production data, including operating rates, fuel use, and solvent use, were
collected from facilities on a process-by-process basis from June 10 through August 24, 1991,
The emissions estimates were calculated on a process level basis and were identified by SCCs
in order to obtain the most accurate estimates possible. Before emission calculations were
performed for a facility, quality assurance procedures were conducted on the data to identify and
correct any missing, suspect, or erroneous data. Naming conventions for emission sources and
stacks were analyzed, as well as a number of other validation and range checking procedures.
There are approximately 65,000 daily emission estimates included in the LMOS database,
representing 560 unique Source Classification Codes. The size of the available database was
roughly 15 megabytes. The data in the LMOS database should be supplemented by the annual
emissions inventory for these facilities.
2.5.2 Approach for Incorporating Data into a Temporal Allocation Factor File
Even in the current draft form, the LMOS data represent a reliable, documented source
of information for the development of TAF file(s). The hourly emission estimates will be
associated to SCCs found in the LMOS "foundation files." These "foundation files" relate
process-level information with source and process identification codes. The correlation of the
CH-KW9
2-16
A-190
-------
SCCs with hourly emission estimates will be accomplished by match-merging the LMOS data
file with the "foundation file" on the device and process identification codes. Algorithms will
then be developed to construct temporal allocation factor distributions from the hourly emission
estimates related to the various processes. Additional data, available in the annual inventory files,
may supplement these hourly estimates to create temporal profiles at seasonal and weekly levels.
The annual inventory will be requested from Wisconsin DNR.
2.6 CONTINUOUS EMISSIONS MONITORING DATA
Hourly emissions data obtained from continuous emissions monitors (CEMs) are a
potential source of data for developing TAF file(s) at several levels of spatial and temporal
resolution. CEMs are installed at facilities to monitor pollutant emissions on a continuous basis
for demonstrating compliance with permit conditions and/or State or federal regulations. The
continuous measurements are ideal data for temporal profiles. Several pollutants may be
monitored with CEM, however, the most commonly monitored pollutants arc sulfur dioxide
(SO2), NOj, and CO. The pollutant emissions of greatest interest for developing TAF file(s) are
S02, NO,, and VOCs.
CEM data are reported at the facility level and may be used to develop TAF file(s) at
several levels of spatial or geographic resolution. Depending on the modeling application, TAF
flle(s) may be developed at the facility, SCC/county, SCC/State, SCC/region, or SCC/national
levels. The degree of aggregation can be defined by the modeling needs.
CEM data will be compiled on an hourly basis, and, therefore, temporal profiles may be
determined diurnally, by day of the week, or by season of the year. For this project, the CEM
data will be aggregated to develop temporal profiles for each hour of the day for a typical
weekday and weekend day in each season.
The data which will be used to develop TAF file(s) are the facility-specific CEM data for
each source. The data may be recorded at various levels of temporal resolution, depending on
the pollutant and regulatory requirements. In some cases, the data may be as small as five-
minute averages or as large as three-hour averages. As indicated in Table 2-2, the CEM data are
maintained at the individual facilities in most States because there are no regulatory requirements
that the data be submitted to the State.
CH-93-69
2-17
A-191
-------
TABLE 2-2. CEM DATA SUMMARY
to
l
t—L
oo
* STATE
FACILITY
NUMBER OF
POLLUTANTS
DATA ACCESSED AT:
TYPES *
FACILITIES
MONITORED
STATE
FACILITY
ALABAMA
A, C, G, I, P, PP-C, PP-FF
200
NOx, S02, TRS, VOCs, CO
X
ARIZONA
C, CE, CU, I, L, R, PP-C
600
NOx, S02, CO, 02
X
ARKANSAS
PP, P, S
15
NOx, S02, TRS, C02
X
CALIFORNIA
B, C, CE, F, I, P, PP, R, SM
1,000+
NOx, S02, TRS, THC, CO
X
COLORADO
CE, PP, R
35
NOx, S02
X
CONNECTICUT
CG, I, UB
25
NOx,S02, CO, C02.02
X
DELAWARE
C, I, PL. PP,R,S
NOx, S02, CO, HC, 02
X
DISTRICT OF COLUMBIA
B, U
<19
NOx, S02
X
FLORIDA
C, CE, CG, F, I, L, PP, UB
50
NOx, S02, TRS
X
X
GEORGIA
C, P, UB
45
NOx.S02.TRS
X(5/45)
X(40/45)
IDAHO
Gas Lines, P
12
NOx, S02, TRS
X
ILLINOIS
B.C, L. S, U
X
INDIANA
B, I, S
45
NOx, S02, CO, HS
X
IOWA
B, C
40
X
X
KANSAS
CE, PP, R
<30
NOx, S02, H2S, CO
X
KENTUCKY
C, R, U
22
NOx, S02, H2S
X
LOUISIANA
B, R, UB
NOx, H2S, CO, HC
X
MAINE
I, P, PP-W
25
NOx, S02, TRS, C02, CO
X
MARYLAND
B. CE, I, PP
18
NOx, S02, HC1, CO
X
MASSACHUSETTS
B, I, PP, TU
25
NOx, S02
X
MICHIGAN
B,C, CE, I,P, U
100
NOx, S02, TRS
X
MINNESOTA
F,I,M, PP, R, SM, U
100
NOx, S02, TRS
X
MISSISSIPPI
B, CE, I
14
NOx, S02
X
MISSOURI
C, CO, PP
<41
NOx, S02
X
MONTANA
P, PP, SM
8
NOx, S02, TRS
X
KEY TO FACiUTV XYPESt
A
ASPHALTPLANTS
CE
CEMENT KILNS
G
GENERATORS
pp
POWER PLANTS
s
STEEL MILLS
VOC
VOCREMOVAL
AU
AUTO MANUFACTURERS
CG
COOENERATION
1
INCINERATORS
PP-C
COAL FIRED PP
SM
SMELTERS
w
WOOD PROCESSING
B
BOILERS
CO
CONCRETE
L
LIMEKILN
PP-KF
FOSSIL FUEL FIRED PP
SVV
SEWAGE TREATMENT
ID
INDUSTRIAL BOILERS
CU
COWER SMELTER
M
MINING
PP-O
OIL FIRED PP
T
TRONA PLANTS
UB
urnrrYBoiLER
F
FERTILIZER
P
PUIJMPAPER
I'PAV
WOOD FIRED PP
TU
TURBINES
C
CHEMICAL PLANTS
Ft)
POOD PROCESSING
PL
pij\snc
K
KUI'INERIES
u
UTILITIES
-------
1
I
!
TABLE 2-2. CEM DATA SUMMARY (continued)
to
>—*
v©
wmsmmmsismm
FACILITY
TYPES *
NUMBER OF
FACILITIES
POLLUTANTS
MONITORED
DATA ACCESSED AT; _J
STATE
FACILITY
NEBRASKA
CE, F, PP
8
NOx
X
NEVADA
PP-C, U
8
NOx, S02
X
NEW HAMPSHIRE
P, PP-C, PP-O, PP-W, VOC
NOx, S02, TRS, VOC
X
NEW JERSEY
PP, RR(I)
5
NOx, S02, CO
X
NEW MEXICO
CU, PP, R, W
25+
NOx, S02, CO, 02
X
NEW YORK
CG, I, IB, SW
300
NOx, S02, C02, CO, 02
X
NORTH CAROLINA
C, CG,1,P, PP
NOx, S02, TRS, 02
X
NORTH DAKOTA
B, C, I'P-C
24
NOx, S02
X
OHIO
AU, B,I
X
OKLAHOMA
CE, F, PP-C
X
OREGON
B, CG, G, I, P, S, TU, U
100
NOx, S02, TRS, CO
X
X
PENNSYLVANIA
C, CG,I,UB
NOx, S02
X
RHODE ISLAND
CG
. 2
NOx, CO
X
SOUTH CAROLINA
G. P
10
NOx, S02, TRS
X
SOUTH DAKOTA
NONE
0
TENNESSEE
B, C, P, PP-C
30
NOx, S02, TRS
X
TEXAS
CE, CU, P, PP, R
100+
NOx, S02, TRS, H2S, CO
X
UTAH
I, PP, R, S
70
NOx, S02, H2S, C02,02
X
VERMONT
I, UB
5
NOx.THC, C0.02
X
VIRGINIA
C, B.I.P, U
42
NOx, S02, CO, VOC
X
WASHINGTON
B, CE, I, P, R
120
NOx, S02, TRS
X
WEST VIRGINIA
B, C, I, R, PP, U
NOx, S02, CO, 02
X
WISONSIN
IB, P, U
60
NOx, S02, TRS
X
WYOMING
CtCE,PP, R,T
27
NOx, S02, TRS, H2S, 02
X
: » KiiY TO facility types;
A
ASPHALT PLANTS
CE
CEMENT KILNS
G
OENERATOKS
PP
POWER PLANTS
S
STEEL MILLS
VOC
VOCREMOVAL
AU
AUTO MANUFACTURERS
CG
COG EN ER AT ION
I
INCINERATORS
PP-C
COAL FIRED PP
SM
SMELTERS
W
WOOD PROCESSING
D
BOILERS
CO
CONCRETE
L
LIMEKILN
PP-FF
FOSSIL FUEL FIRED PP
SW
SEWAGE TREATMENT
IB
INDUSTRIAL BOILERS
CU
COPPER SMELTER
M
MINING)
PP-O
OIL FIRED PP
T
TRONA PLANTS
UB
UTILITY BOILER
F
FERTILIZER
P
PULPAPAPER
PP-W
WOOD FIRED PP
TU
TURBINES
C
CHEMICAL PLANTS
FO
FOOD PROCESSING
PL
PLASTIC
R
REFINERIES
U
UTILITIES
-------
Generally, CEM data arc reported in a standard unit of measurement, parts per million
(ppm). In some cases, exhaust flow rate data are also necessary in order to calculate hourly
emissions from the concentration data.
2.6.1 Identification of Data Availability
State environmental agencies for the 48 contiguous States and the District of Columbia
were contacted to determine the availability of hourly CEM data. (Note: Alaska and Hawaii were
not included in this effort since they are not included in current modeling domain). The results
of this effort are summarized in Table 2-2. For all but the following States, CEM data will have
to be obtained from the individual facilities: Connecticut, Georgia, Iowa, Kentucky, Nebraska,
New Hampshire, New Jersey, New Mexico, Ohio, and Oregon.
CEM data which are available from the States will be copied electronically to a computer
file as a first choice. In cases where an electronic file is not available, a hard copy of the data
will be obtained. CEM data which are kept at individual facilities will have to be sought out on
a facility by facility basis. The format of the data collected will be the same as that collected
from the States.
2.6.2 Approach for Incorporating Data into a Temporal Allocation Factor File
Once all of the available CEM data have been obtained, a statistical evaluation of the data
will be perfonned to determine the variability of the emissions for each source category at
various levels of spatial resolution (i.e., county, State, region, national). Where the CEM data
are averaged over time periods of less than one hour, emissions will be aggregated to an hourly
average and evaluated on an hourly basis. Statistical parameters to be evaluated include the
mean, median, mode and standard deviation. The variability of these statistical parameters for
each source category will be evaluated diumally, for weekdays and weekends. CEM emissions
data will also be evaluated for seasonal variations. Based on the results of this analysis, the most
representative parameter will be used to develop profiles.
CH-93-69
2-20
A-194
-------
2.7 WASTEWATER DATA
In searching for appropriate surrogates for actual industrial activity levels, TRC has
determined that wastewater data may be a useful, if nontraditional, surrogate. The basic premise
of using wastewater data to allocate air emissions is the assumption that air emission increases
will correspond to wastewater release increases. It is believed that industrial facilities filed
temporal wastewater discharge profiles to municipal or State water regulating agencies; therefore,
TRC investigated the availability and usefulness of these data for generating activity profiles
(probably SIC-specific) based on wastewater discharge that may be used to allocate air emissions.
This section presents the results of this investigation.
2.7.1 Identification of Data Availability
TRC has made preliminary contacts with the State of North Carolina, Division of
Environmental Management and the Regional EPA Office of Water Permits and Compliance.
The N.C. Compliance and Enforcement Section indicated that wastewater flow and pollutant
loading data could be obtained from their database. These data can be assigned to industry SIC
codes. Industrial processes generating wastewater may also be described. These data are available
on a daily temporal resolution for many facilities; other facilities report monthly values at a
minimum.
National wastewater discharge data can be accessed through the EPA Regional Office of
Water Permits and Compliance in Atlanta, Georgia. Only minor dischargers as defined in the
Clean Water Act are tracked at the national level, representing approximately 10 percent of all
discharges. Available wastewater parameters include flow, total suspended solids (TSS), pH, and
biological oxygen demand (BOD) (though less common than the others). Additionally, industry
SIC code and a process description field are available. The data are compiled as monthly
averages.
CH-93-69
2-21
A-195
-------
2.7.2 Description of Data
Industrial wastewater discharge data are maintained under the National Pollution
Discharge Elimination System (NPDES). The NPDES program is managed at the State level.
Individual States maintain databases with daily or monthly values of wastewater flow rates and
pollutant concentrations. Monthly average values of wastewater flow and pollutant concentrations
are maintained at the national level in the Permit Compliance System. The temporal frequency
of a facility's wastewater monitoring data depends on the facility's permit requirements. The
permit requirements are based on the type of wastes discharged, location of discharge (i.e., to a
receiving stream or POTW), and the uniformity or variability of discharge flow. Wastewater
parameters may be reported monthly or daily and may represent average or maximum values.
2.7J Approach for Incorporating Data into a Temporal Allocation Factor File
The use of industrial wastewater discharge data for the development of TAF file(s) could
be used under the assumption that wastewater flow and pollutant loads are good indicators of a
facility's emissions over time. Times of increased wastewater flow or higher concentration of
pollutants in the wastewater may be assumed to indicate simultaneously increased air pollutant
emissions. At least one year of wastewater flow or pollutant loading data would be obtained.
Temporal allocation factors could be developed to the daily level at best and the monthly level
at least. The wastewater flow or pollutant parameters would be added over the one year period
and proportions developed for monthly and daily temporal factors.
2.8 OTHER DATA SOURCES
2.8.1 Waste-to-Energy Source Data
TRC has identified the 1991 Resource Recovery Yearbook, Directory and Guide as a
source of waste-to-energy data. On a need basis (as determined by the steering committee), these
data will be used to characterize emissions from solid waste disposal facilities that generate
energy (included in SIC 4900 through 4999, and SCC 5-01-001-01 through 5-03-900-10). The
GH-93-69 2-22
A-196
-------
report, published by the Governmental Advisory Associates, Inc. (GAA), New York City, is the
fifth edition of previous reports published in 1982, 1984, 1986 and 1988.
The Resource Recovery Yearbook provides statistical data on the number and types of
waste recovery facilities located throughout the United States. Information is provided for
conceptually planned facilities, advance planned/existing facilities, and facilities which have been
shut down on either a temporary or permanent basis.
2.8.1.1 Identification of Data Availability
The directory includes facility-specific information that may be useful as surrogate
indicators for developing TAF file(s) (i.e., operating schedules, average operating throughput, and
other operating parameter data). The data can be obtained in dBASE® format from the GAA
upon request.
2.8.1.2 Description of Data
The report provides both operating schedule and throughput data for specific waste
recovery facilities located throughout the entire United States. SCC codes are not contained in
the report, but an example SCC code applicable to this type of source would be 5-02-001-01.
Operating scheduling data are given in hours per week, hours per shift, shifts per day and
days of operation per year. Average operating throughput is given in tons per day. Other
operating parameters include the amount of ash residue generated daily (tons per day) and steam
output in pounds per hour, generated from the combustion of the solid waste. The GAA collects
information from the waste-to-energy recovery industry on an ongoing basis.
2.8.1.3 Approach for Incorporating Data in a Temporal Allocation Factor File
The facility-specific data can be used as gap fillers for developing TAF file(s) specific
to the waste recoveiy industry. Generally, the operating scheduling data can be broken down into
various periods of the day since the number of shifts of operation for a given day are provided
Although the scheduling data do not delineate information by season, the assumption is given that
these operating schedules are the same throughout the entire year. By multiplying the hours per
CH-93-6? 2-23
A-197
-------
shift by shifts per day the total number of hours per day can be determined. An assessment can
be made as to which hours of the day the data would be assigned to facilities which operate with
fewer than 3 shifts per day. To make a more accurate assessment of the time of day in which
the shifts operate (for facilities which operate with less than 3 shifts per day), several of these
facilities could be contacted to determine at what hours of the day their shifts operate.
The same type of procedure for assessing the actual operating hours per day mentioned
above can be used if process throughput data (e.g., pounds per hour steam produced) are used
for developing the TAF file(s).
2,8.2 Acid-Modes Field Study
Under Contract Number 68-02-4274, Work Assignment 30, EPA's Air and Energy
Engineering Research Laboratory (AEERL) instructed Alliance Technologies Corporation
(currently TRC) to estimate hourly emissions of S02 and NOx from coal-burning electric power
utilities in the Eastern United States. The data were to be used in conjunction with the 1985
National Acid Precipitation Assessment Program (NAPAP) as the emissions input for the
verification runs of two acidic deposition computer simulation models. The data collection
efforts were sponsored by the EPA, the Ontario Ministry of the Environment (OME), the
Atmospheric Environmental Service (AES) of Environment Canada, the Electric Power Research
Institute (EPRI), and the Florida Electric Power Coordinating Group (FCG).
During the project, megawatt load data were collected by mail survey from 124 U.S. and
Canadian power plants. The temporal resolution varied by facility, but in some cases, hourly
load data were submitted. For facilities not submitting hourly data, hourly estimates were made
using the facility-provided weekly or monthly data.
The data files for the project were stored at EPA's National Computing Center (NCC).
In the three years since the project was completed, these data have been archived and
warehoused.. TRC was able to locate an electronic copy of the data maintained by Mr. Ron Ryan
of the U.S. EPA, and has had the data uploaded by NCC to allow their use in this project These
data are in S AS® format data files and are readable and usable.
These data could be used to replace or augment the current allocation profile for electric
utility emission sources. The database includes actual hourly profiles, and the documentation
CH-93-69
2-24
A-198
-------
identifies the sources that submitted actual hourly data. The current profile for electric utilities
is based on similar work performed under NAPAP, but is based on a sample of power plants in
only the Northeast United States. The Acid-Modes Field Study database contains data from a
larger number of sources in a wider variety of areas. The hourly megawatt load data could be
used to develop an improved profile of a typical operating day for specific utility SCCs. These
operating profiles could then replace existing profiles for a few select utility SCCs.
Appendix H includes the database structure for the databases restored to the NCC.
2,8.3 Urban Airshed Model Emissions Preprocessor System Data
2.83.1 Identification of Data Availability
The Urban Airshed Model (UAM) has been designated as the "recommended"
(i.e., prefened) model for "photochemical pollutant modeling applications involving entire urban
areas" by EPA's Office of Air Quality Planning and Standards (OAQPS). The UAM simulates
the hour-by-hour photochemistry occurring for each grid cell in the modeling domain;
consequently the input emissions data must contain the same level of resolution. To
accommodate this level of resolution in the input data, a system of computer programs has been
designed to perform the intensive data manipulations necessary to adapt a county-level annual
or seasonal emission inventory for photochemical modeling use; the UAM Emissions
Preprocessor System (EPS), Version 2.0. In the EPS, SCCs are cross-referenced to a month, day
of the week, and diumal profile code which determines the temporal profiles applied to emissions
being processed for input into the UAM. These monthly, weekly, and diurnal hourly temporal
profiles are available from the EPS. The EPS file format is presented in Appendix I.
2.83.2 Description of Data
In the EPS, TAF file(s) axe cross-referenced to point, area, and mobile source processes
by profiles codes. These TAF file(s) exist for an array of different temporal scenarios consisting
of typical monthly distributions, day to week distributions, and hourly distributions for typical
weekday and weekend scenarios. The TAF file(s) were compiled from operating parameters
found in the NAPAP emissions inventory files, as well as data resulting from field studies
conducted by the CARB conducted approximately ten years ago.
CH-93-69
2-25
A-199
-------
2.83.3 Approach for Incorporation of Data into a Temporal Allocation Factor File
The TAF file(s) will be matched with the corresponding SCCs by using the EPS defined
profile codes. These SCC/TAF file(s) combinations will be incorporated into the final TAF
file(s) developed by TRC. EPS data for high-priority source categories could be substituted into
the TRC TAF file(s). The data can be incorporated into the TAF file(s) developed by TRC
through a method which will prioritize the EPS TAF file(s) along with other data sources. Due
to the dated nature of the EPS TAF file(s), their priority in relationship to other sources being
used should be comparably low. These data will only serve as defaults in the absence of more
recent data.
CH-93-W
2-26
A-200
-------
3.0 SOURCE CATEGORY PRIORITIZATION
Emissions source categories must be prioritized to ensure proper attention to the source
categories that are major national contributors to air pollution. This section outlines the
methodology used to develop a preliminary prioritized list of source categories that will be used
to focus future efforts in developing the TAF f!le(s).
The prioritized source categories presented in this section reflect only data available from
AIRS/AFS. No area or mobile source categories were included because those data are not
currently publicly available from the AIRS Area and Mobile Sources Subsystem (AMS). The
source category prioritization focuses on ozone precursors (VOC, NOx, and CO) because these
pollutants will be targeted in pending ozone/CO SIPs modeling efforts. If requested by the
WAM, future source category prioritization efforts in this work assignment will focus upon the
remaining criteria pollutants [lead (Pb), S02, PM-10, and total particulate (TP)]. Because AFS
data were the sole source of emissions estimates in this study, the source categories themselves
are generally referred to in this section by AFS source classification codes (SCCs).
Source categories were not prioritized by hazardous air pollutant (HAPs) emissions due
to insufficient HAPs emissions data in AIRS. However, an approach for prioritizing source
categories by HAPs emissions is presented based on the source categories listed in Title in of
the Clean Air Act Amendments (CAAA). Additionally, source categories contained in Title HI
are compared with the prioritized list of criteria pollutant source categories derived from AFS to
verify completeness.
3i methodology
3.1.1 Source Category Ranking Based on Criteria Pollutant Emissions
The SCCs contained in AFS were prioritized in the following sequence.
1. Emissions of VOC, N0„ and CO by SCC for the current AFS database (including 1990
base year and SIP data) were retrieved from AFS using the AFS 650 report, (Emission
by SCC Report). This report contains the following information:
CH-93-69
3-1
A-201
-------
A national listing of source categories sorted by emissions of each pollutant
* Total estimated emissions in tons per year by SCC
* Number of records (a record in AFS is considered one emissions generation
process) for each SCC contributing to the total estimated emissions
2. An initial ranking of source categories from the AFS 650 report was performed to identify
the top 100 SCCs emitting each of the criteria pollutants. This list was presented in the
memorandum dated June 9, 1993 from Theresa Moody and David Winkler of TRC
Environmental Corporation to Chuck Mann, AEERL. The rank-ordered scale ranged from
1 to 100 with 1 being the highest total emissions and 100 being the lowest total
emissions.
3. Emissions-per-record for each criteria pollutant were calculated for the top 100 SCCs.
These SCCs were ranked in descending order. Emissions per record was selected as the
criterion that determines rank for the following benefits:
* Simplifies identifying true outliers (i.e., processes emitting relatively small
amounts per process may have significant total emissions by virtue of their
numbers)
Provides a clearer picture of emissions by process
Focuses attention on the higher emission processes that are most likely to be
included in a point source inventory
4. At the request of the WAM, these top 100 SCCs were then scrutinized for plausibility of
the reported emissions. As a result of this quality control measure, some SCCs were
discarded for each of the ozone precursor pollutant SCC lists. The SCCs discarded were
assumed to contain faulty data (probably representing misassigned SCCs).
5. At the request of the WAM, SCCs representing poorly characterized sources (such as
miscellaneous industrial fugitive losses) were also omitted.
6. The remaining SCCs were re-ranked from highest to lowest emissions per record.
7. An overall rank order value (an average rank order value across all the ozone precursors
pollutants) was then calculated for each ozone precursor pollutant SCC (VOC, NO, and
CO) based on the following expression:
£ Rankpou«^
Overall Rank Order =
Number of pollutants
CH-93-69
3-2
A-202
-------
If a category was not represented: among all three ozone precursors pollutants, it was
assigned a rank order of 100 to provide consistency in the calculations and to simulate
the numerical averaging impact of the lowest ranked SCC. Table 3-1 provides details of
the overall rank order of the source categories,
3.1.2 HAPs-Related Source Category Prioritization
Insufficient HAPs emissions data in AFS at the current time make quantified prioritization
of SCCs for HAPs difficult As SCCs and accompanying emissions data for HAPs emissions
become available in the future, however, SCC prioritization based on HAP emissions may be
advisable. The Draft Scoping Document for the MACT Database indicates that there will be
ongoing HAP SCC development to encompass source categories as defined in the CAAA, Title
in and the Federal Register, Volume 57, No. 186, September 24, 1992.
EPA's National Air Data Branch (NADB) is presently assigning and developing additional
HAP source categories in conjunction with the maximum achievable control technology (MACT)
standards development. The ongoing and future HAP SCC development will encompass these
source categories. Therefore, the initial priorities for the HAP SCCs are somewhat defined by
these documents and the CAAA Title HI. However, as HAP emission data become available,
these initial priorities should be re-examined.
3.2 SUMMARY AND COMPARISON WITH EARLIER PRIORITIZATION EFFORTS
Table 3-2 lists the source category descriptions included in Table 3-1. The prioritized list
includes nine industry groups and 1.9 major source categories. The major source categories as
represented in Table 3-2 include:
Electric utility, industrial, and commercial/institutional external combustion boilers
* Industrial internal combustion engines
* Chemical manufacturing
* Primary and secondary metal production
CH-93-69
3-3
A-203
-------
TABLE 3 — 1. RANK-ORDERED SCCs
RANK
see
CO
N02
voc
Average
Rank
1
30700104
24
56
27
36
2
30600201
27
42
46
38
3
30100504
9
100
8
39
4
30103102
15
100
4
40
5
10100901
18
100
2
40
6
10200901
46
44
32
41
7
30300813
11
100'
12
41
8 '
30100603
31
100
6
46
9
10201402
23
19
100
47
10
10100301
40
3
100
48
11
10100202
63
10
72
48
12
30700704 .
100
41
14
52
13
10100201
51
6
100
52
14
10100222
49
8
100
52
15
10100226 .
48
9
100
52
16
10100203
60
1
100
54
• 17
30600401
100
28
38
55
18
30700110
28
40 .
100
56
19
10100101
54
18
100
57
20
10200902
59
62
52
58
21
10100212
64
12
100
59
22
10100601
74
35
• 76
62
23
10100404
70
21
100
64
24
30700106
39
55
100
65
25
10100604
71
26
100
66
26
30500706
73
25
100
66
27
30300826
1
100
100
67
28
30102501
100
100
1
67
29
10100204
65
36
100
67
30
30300913
2
100
100
67
31
10100303
100
2
100
67
32
30102505
100
100
3
68
33
30300914
3
100
100
68
34
10100302
100
4
100
68
35
30300812
4
100
100
68
36
30400315
5
100
100
68
37
10100223
100
5
100
68
38
30201003
100
100
5
68
39
30100501
6
100
100
69
40
30300101
7
100
100
69
41
10100225
100
7
100 .
69
42
30400110
100
100
7 .
69
(Continued)
3-4
A-204
-------
TABLE 3-1. RANK— ORDERED SCCs (Continued)
RANK
see
CO
N02
voc
Average
Rank
43
30300102
' 8
100
100
69
44
30106003
100
100
9
70
45
10200202
80
29
100
70
46
30103503
10
100
100
70
47
30125405
100
100
10
70
48
50100701
100
100
11
70
49
10200226
100
11
,100
70
50
30301201
12
100
100
71
51
30300306
100
100
13
71
52
10100221
100
13
100
71
53
30100503
13
100
100
71
.54
30501604
68
45
100
71
55
102C0301
100
14
100
71
56
50100101
56
58
100
71
57
30300103
14
100
100
71
58'
30500206
100
. 15
100
72
59
10100401
82
33
100
72
60
30600503
100
100
15
72
61
30100509
16
100
100
72
62
10200201
100
16
100
72
63
30600801
100
100
16
72
64
10200704
69
48
100
72
65
10200203
100
17
100
72
66
30103101
17
100
100
72
67
30119705
100
100
17
72
68
40600240
100
100
18
. 73
69
10100602
72
47 ¦
100
73
70
30300105
19
100
100
73
71
30600504 .
' 100
100
19
73
72
30115801
20
100
100
73
73
10100217
100
20
100
73
74
30600803
100
100
20
73
75
30300302
100
100
21
74
76
30101901
21
100
100
74
77
30119701
100
72
50
74
78
10101301
22
100
100
74
79
50100201
100
. 22
100
74
80
30600805
100
100
23
74
81
30501403
100
23
100
74
82
30100306
100
24
100
75
83
40100225
100
100
24
75
84
30100103
25
100
100
75
(Continued)
3-5
A-205
-------
TABLE 3-L RANK-ORDERED SCCs (Continued)
RANK
see
CO
N02
voc
Average
Rank
85
30600602
100
100
25
75
86
20200203
76
49
100
75
87
30300907
26
100
100
75
88
40500501
100
100
26
75
89
30500606
100
27
100
76
90
40201606
100
100
28
76
91
30300908
29
100
100
76
92
30102601
100
100
29
76
93
40201901
100
100
30
77
94
10200204
. 78
52
100
77
95 .
30100305
30
100
100
77
96
10200219
100
30
100
77
97
30106099
100
100
31
77
98
30101301
100
31
100
77
99
10100801
100
32
100
77
100
30300107
32
100
100
77
101
39000702
33
100
100
78
102
40600131
100
100
33
78
103
40600141
100
• 100
34
78
104
20400401
34
100
100
78
105
' 30501212
100
' 34
100
78
106
30501099
35
100
100
78
107
40201726
100
100
•35
78
108
4050051-1
100
100
36
79
109
20200202
85
67
84
79
110
31000203
88 .
68
80
79
111
30101807
100
100
36
79
112
30400301
36
100
100
79
113
20200201
83
53
100
79
114
30300904
37
100
100
79
115
30302312
100
37 •
100
79
116
30688801
100
100
37
79
117
40500599
100
100
37
79
118
30100599
38
100
100
79
119
30501404
100
38
100
79
120
10200224
100
39
100
80
121
40100202
100
100
39
80
122
40301008
100
100
40
80
123
40201301
100
100
41
80
124
30100506
41
100
100
80
125
40301012
100
100
42
81
126
30100104
42
100
100
81
(Continued)
3-6
A-206
-------
TABLE 3 — 1. RANK-ORDERED SCCs (Continued)
RANK
see
CO
N02
VOC
Average
Rank
127
30101812
100
100
43
81
128
30101302
100
43
100
81
129
10200907
43
100
100
81
130
30402004
44
100
100
81
131
40201399
100
100
44
81
132
40500201
100
100
45
82
133
30300315
45
100
100
82
134
20100201
81
64
100
82
135
30501402 .
100
46
100
82
136
10200205
•84
63
100
82
137
• 40200401
100 '
100
47
82
138
. 30600301
47
100
100
82
. 139
40200901
100
100
48
83
140
30102699
100
100
49
83
141
10200799
100
50
100
83
142
30400701
50
100
100
83
143
30180001
100
100
51
84
144
50100102
100
51
100
84
¦ 145
20100202
86
65
100
84
146
10100902
52
100
100
84
147
30100308.
53
100
100
84
148
10100701
100
54
100
85
149
40400250
100
100
54
85
150
50300105
55
100
100
85
151
30101802
100
100
55
85
152
30600104
92
78
86
85
153
30800799
100
100
56
85
154
40200101
100
100
57
86
155
10100903
57
100
100
86
156
30300933
100
57
100
86
157
-10200701
91
66
100
86
158
30400403
58
100
100
86
159
40202201
100
100
58
86
160
40201101
100
100
59 .
86
161
20200204
100
59
100
86
162
10200602
89
84
87
87
163
40200810
100
100
60
87
164
20200401
100
60
100
87
165
40200701
100
100
61
87
166
10201301
100
61
100
87
167
30103599
61
100
100
87
168
40500311.
100
100
62
87
(Continued)
3-7
A^2G7
-------
TABLE 3-1. RANK-ORDERED SCCs (Continued)
RANK
err
CO
N02
voc
Average
Rank
169
10300903
62
100
100
87
170
10200903
77
86
100
88
171
30600701
100
100
63
88
172
40500101
100
100
64
88
173
10200601
79
100
85
88
174
40200501
100
100
65
88
175
10201101
66
100
100
89
176
40500301
100
100
66
89
177
30101401
100
100
67
89'
178
30113299
67
100
100
89
179
31000299
100
100
68
89
180
40200410
100
100
69
- 90
181
10300601
100
69
100
90
182
40200801
100
100
70
90
183
10300502
100
70
100
90
184
10200502
100
71
100
90
185
30101899
100
100
71
90
186
10200401
93
79
100
91
187
30600106
87
85
100
91
188
10201201
100
73
100
91
189
40200110
100
100
73
91
190
30700105
100
74
100
91
191
40500401
100
100
74
91
192
40202501
100
100
75
92
193
20200301
75
100
100
92
194
20100102
100
75
100
92
195
10300209
100
76
100
92
196
40400111
100
100
77
92
197 .
31000201
100
77
100
92
198
40100399
100
100
78
93
199
40301102
100
100
79
93
200
30600102
100
80
100
93
201
40301097
100
100
81
94
202
30600103
100
81
100
94
203
10200501
. 100
82
100
94
204
40301199
100
100 .
82
94
205
30190003
100
83
100
94
206
40301099
100
100
83
94
207
10300603
100
87
100
96
(Continued)
3-8
A-208
-------
TABLE 3-1.
RANK-
•ORDERED SCCs (Continued)
Average
RANK
SCC CO
N02
voc
Rank
208
20200102
100
88
100
96
209
-10300401
100
89
100
96
210
10200603
90
100
100
97
211
20100101
100
90
100
97
• 212
10300602
100
94
100
98
3-9
A-209
-------
TABLE 3-2. ITIGIT PRIORITY SOURCE CATEGORIES
General Description
Specific Description
EXTERNAL COMBUSTION BOILERS-ELECTRIC GENERATION
EXTERNAL COMBUSTION BOILERS-
EXTERNAL COMBUSTION BOILERS-
EXTERNAL COMBUSTION BOILERS-
EXTERNAL COMBUSTION BOILERS-
EXTERNAL COMBUSTION BOILERS-
EXTERNAL COMBUSTION BOILERS-
EXTERNAL COMBUSTION BOILERS-
EXTERNAL COMBUSTION BOILERS-
EXTERNAL COMBUSTION BOILERS-
-ELEC GENERATION:
-ELEC GENERATION:
-ELEC GENERATION:
-ELEC GENERATION:
-EUEC GENERATION:
-ELEC GENERATION:
-ELEC GENERATION:
-ELEC GENERATION:
- ELEC GENERATION:
ANTHRACITE COAL
BIT/SUBBIT COAL
LIGNITE COAL
RESIDUAL OIL
NATURAL GAS
PROCESS GAS
COKE
WOOD/BARK WASTE
LIQUID WASTE
EXTERNAL COMBUSTION BOILERS-INDUSTRIAL
LO
EXTERNAL COMBUSTION BOILERS-
EXTERNAL COMBUSTION BOILERS-
EXTERNAL COMBUSTION BOILERS-
EXTERNAL COMBUSTION BOILERS-
EXTERNAL COMBUSTION BOILERS-
EXTERNAL COMBUSTION BOILERS-
EXTERNAL COMBUSTION BOILERS-
EXTERNAL COMBUSTION BOILERS-
EXTERNAL COMBUSTION BOILERS-
EXTERNAL COMBUSTION BOILERS-
EXTERNAL COMBUSTION BOILERS-
- INDUSTRIALBIT/SUBBIT COAL
- INDUSTRIAL;LIGNITE COAL
- INDUSTRIAL;RESIDUAL OIL
-INDUSTRIAL-DISTILLATE OIL
- INDUSTRI ALNATURAL GAS
- INDUSTRIAL PROCESS GAS
7 INDUSTRIAL WOOD/DARK WASTE
- INDUSTRIALBAGASSE
-INDUSTRIAL-SOLID WASTE
-INDUSTRIALLIQUID WASTE
- INDUSTRI ALCO BOILER
EXTERNAL COMBUST BOILERS-COMMCL/INSTITNL
EXTERNAL COMBUST BOILERS - COMMCL/INSTITNL BIT/SUBBIT COAL
EXTERNAL COMBUST BO ILERS-COMMCI7INSTITNL RESIDUAL OIL
EXTERNAL COMBUST BOILERS-COMMCL/INSTITNLDISTILLATE OIL
EXTERNAL COMBUST BOILERS-COMMCL/INSTITNLNATURAL GAS
EXTERNAL COMBUST BOILERS-COMMCI7INSTITNLWOOD/BARK WASTE
(Continued)
>
I
to
t—»
o
-------
i
i
TABLE 3-2. HIGH PRIORITY SOURCE CATEGORIES rContinue
I
to
-------
TABLE 3-2. HIGH PRIORITY SOURCE CATEGORIES (Continued)
General Description Specific Description
FOOD AND AGRICULTURE:ALFALFA DEHYDRATION
FOOD AND AGRICULTURE
PRIMARY METAL PRODUCTION PRIMARY METAL PRODUCTION: ALUMINUM ORE
PRIMARY METAL PRODUCTION:BY-PRODUCT COKE MANUFACTURING
PRIMARY METAL PRODUCTION:IRON PRODUCTION
PRIMARY METAL PRODUCTION:STEEL PRODUCTION
PRIMARY METAL PRODUCTION:TITANIUM PROCESSING
PRIMARY METAL PRODUCriON:TACONITE IRON ORE PROCESSING
SECONDARY METAL PRODUCTION SECONDARY METAL PRODUCTION:SECONDARY ALUMINUM PRODUCTION
SECONDARY METAL PRODUCTION:GRAY IRON FOUNDRIES
SECONDARY METAL PRODUCTION:SECONDARY LEAD PRODUCTION
SECONDARY METAL PRODUCTION:STEEL FOUNDRIES
SECONDARY METAL PRODUCTION:FURNACE ELECTRODE MANUFACTURE
MINERAL PRODUCTS:ASPHALTIC CONCRETE
MINERAL PRODUCTS:CEMENTMANUFACTURTNGiDRY PROCESS
MINERAL PRODUCTS:CEMENT MANUFACTURING:WET PROCESS
MINERAL PRODUCTS:SURFACE MINING OPERATIONS
MINERAL PRODUCTS:FIBERGLASS MANUFACTURING
MINERAL PRODUCTS:GLASS MANUFACTURING
MINERAL PRODUCTS:LIME MANUFACTURE
PETROLEUM INDUSTRY PETROLEUM INDUSTRY:PROCESS HEATERS
PETROLEUM INDUSTRY:FLUID CATALYTIC CRACKING UNITS
PETROLEUM INDUSTRY:BLOWDOWN SYSTEMS
PETROLEUM INDUSTRY:FUGITIVE EMISSIONS
PETROLEUM INDUSTRY:VACUUM DISTILLATE COLUMN CONDENSORS
¦ PETROLEUM INDUSTRY:COOLING TOWERS
(Continued)
MINERAL PRODUCTS
LO
I
~—*
to
>
I
-------
TABLE 3-2. HIGH PRIORITY SOURCE CATEGORIES (Continued^
General Description Specific Description
PULP AND PAPER AND WOOD PRODUCTS PULP AND PAPER AND WOOD PRODUCTS:SULFATE PULPING
PULP AND PAPER AND WOOD PRODUCTS.PLYWOOD/PARTICLEBOARD OP
OIL AND GAS PRODUCTION OIL AND GAS PRODUCTIONS ATURAL GAS PRODUCTION
ORGANIC SOLVENT EVAPORATION ORGANIC SOLVENT EVAPORATION:OPEN-TOP VAPOR DEGREASING
ORGANIC SOLVENT EVAPORATION:COLD SOLVENT CLEANING/STRIPPING
SURFACE COATING OPERATIONS:SURFACE COATING APPLICATION
SURFACE COATING OPERATIONS:COATING OVEN
SURFACE COATING OPERATIONS:THINNING SOLVENTS
SURFACE COATING OPERATIONS: FABRIC COATING
SURFACE COATING OPERATIONS:PAPER COATING
SURFACE COATING OPERATIONS:SURFACE COATING OF AUTOS/TRUCKS
SURFACE COATING OPERATIONS:METAL CAN COATING
SURFACE COATING OPERATIONS: WOOD FURNITURE SURFACE COATING
SURFACE COATING OPERATIONSsURFACE COATING OF PLASTIC PARTS
SURFACE COATING OPERATIONS:SURFACE COATING OF MISC METAL PARTS
PETROLEUM PRODUCT STORAGE:FIXED ROOF TANKS
PETROLEUM PRODUCT STORAGE:FLAT ROOF TANKS
BULK TERMINALS/PLANTS-PETROLEUM STORAGE TANKS BULK TE RMIN ALS/PLANrS—PETROIJE UM STOR TANKS:FLTNG ROOFTANKS
PRINTIN Q/PUBLISHING PRINTING/PUBLISHING:DRYERS
PRINTING/PUBLISHING:PRINTING
TRANSFORATION AND MARKETING OF PETROLEUM PRODUCTS TRANSFORATION & MARKETING OF PETRO PRODUCTS:TANKCARS/TRUCKS
TRANSFORATION & MARKETING OF PETRO PRODUCTS-MARINE VESSELS
(Continued)
SURFACE COATING OPERATIONS
SURFACE COATING OPERATIONS
OJ
i—»
w PETROLEUM PRODUCT STORAGE
>
I
K>
U>
-------
TABLE 3-2. HIGH PRIORITY SOURCE CATEGORIES (Continued^
General Description Specific Description
SOLID WASTE DISPOSAL-GOVERNMENT SOLID WASTE DISPOSAL-GO VERNMENT:MUNICIPAL INCINERATION
SOLID WASTE DISPOSAL-GOVERNMENT:OPEN DUMP BURNING
SOLID WASTE DISPOSAL- GOVERNMENT:SEWAGE TREATMENT
SOLID WASTE DISPOSAL-INDUSTRIALINCINERATION
LO
*
>
i
to
£
-------
• Mineral products
• Surface coating operations
To check for consistency with previous EPA efforts, TRC compared the source categories
assigned high priority under this methodology to high-priority source categories defined in the
Federal Register notice dated September 24, 1992. The Federal Register notice lists 17 industry
groups and 171 source categories that are targeted for Maximum Achievable Control Technology
(MACT) standard development in order to meet requirements of Title HI of the Clean Air Act
Amendments of 1990. All categories identified in this prioritization effort were found to also be
included in the Federal Register list of source categories. Therefore, it was concluded that the
results of this prioritization effort were consistent with earlier efforts for prioritizing source
categories.
3.3 OTHER ISSUES
Other issues associated with the SCC prioritization include assessing the dynamics of the
AIRS Facility Subsystem (AFS) database, and how variations in AFS data would affect subsequent
prioritization efforts. There are also concerns about how to assess the impact of AIRS Area and
Mobile Source (AMS) data and Hazardous Air Pollutant (HAP) data on future prioritization
efforts.
The AFS database is a dynamic database with new SIP data being regularly uploaded by
state and local air pollution control agencies. This attribute's affect on the current SCC
prioritization approach outlined in this memorandum cannot be quantified. Significant changes in
AFS emissions data may occur as the result of the regular data updates. The rank order for SCC
emissions identified under this work assignment is representative of the AFS database as of June
1993; some change in the ranking of SCC emissions data are likely to occur as new data are
uploaded to AFS. Therefore, it may be advisable to determine the long-term (one year minimum)
impact of the dynamic data.
3-15
A-215
-------
The impacts associated with AMS and HAP categories are not fully known at this time.
However, depending upon the computer model being used, the significance of these impacts
could vary. It is likely that the UAM would be less impacted by the omission of these SCCs,
but the impact on a human exposure model could be more pronounced. It may be advisable to
extend an effort in the future to include these source categories. Although HAPs data may not
be available for several years, area and mobile source data should be available within a year.
CH-93-G9
3-16
A-216
-------
4.0 PLAN OF ACTION FOR TEMPORAL ALLOCATION FACTOR FILE
DEVELOPMENT
This section presents the TRC anticipated technical approach to generating the final TAF
file(s). The plan presented will be implemented as Task 4 of the work assignment. The primary
product from this effort will be the final temporal allocation factor file: a final report
documenting the efforts of the work assignment will also be delivered under Task 4.
4.1 ORGANIZATION AND RESOURCE REQUIREMENTS
The following three major technical tasks will be undertaken during Task 4. Resources
required are as identified in the final Work Plan dated March 17, 1993.
• Data Collection
Database Design and Management
• Technical Support
The teams for these tasks will be interdependent through Task 4. For example, the Data
Collection team will rely on the Database Management team to recommend formats for electronic
data collection that will facilitate data entry. The Database Management team will rely on the
Technical Support team to help develop the methodology for combining data to develop a useful
database. The methodology development will rely on the Data Collection team's providing
accurate descriptions of data provided by each data source. The approach of each task is
described below.
4.1.1 Data Collection
Although some data have been collected since the onset of the project, additional effort
is required. TRC has identified staff to concentrate immediately on collecting data identified in
Section 2 of the memorandum. Issues that will be addressed under the Data Collection task
include identifying the best available format for each data source, obtaining the data (including
CH-93-W
4-1
A-217
-------
issuing purchase orders or telephone requests), implementing phone surveys to collect data from
high priority SCCs, and (when data are collected on paper) inputting data.
Much of the work for this task has been initiated under Tasks 1, 2, and 3 of the Work
Assignment. The SOS outstanding data will be obtained July 7, 1993. The remaining data will
be collected and in electronic format by My 31, 1993.
4.1.2 Database Design and Management
The Database Design and Management team will be responsible for the design of the TAF
flle(s) and aU computer programming efforts required to manipulate the collected data into the
final format. Actual construction of the final TAF file(s) will be the responsibility of the
Database Design and Management team. This team, in conjunction with the Technical Support
team, will be responsible for developing methodologies for selecting and combining data sources.
The Database Design and Management team will operate under procedures developed by
TRC's Software Engineering Group to ensure that the TAF file(s) is developed in a reliable,
consistent, documented manner. An internal TRC Approach Study Document is currently under
development that will explicitly identify the major requirements of the project. This document
will also include specific algorithms for incorporating data from each selected data source into
the TAF file(s). Once this document is complete, the identified software development approach
will be implemented.
4.1.3 Technical Support
The Technical Support team will have three main responsibilities: (1) source category
prioritization; (2) data source prioritization; and (3) methodology development. As discussed in
Section 3, TRC has developed a preliminary prioritized source category list based on AIRS-
reported emissions. During the data collection phase of the project, the TRC Technical Support
team will refine and adjust this list to reflect data availability for highly ranked source categories.
Prioritization may also change as more information is collected and analyzed. For the categories
that are represented in more than one data source, the Technical Support team will be responsible
for determining if and how the data should be combined. For example, if two data sources
CH-93-69
4-2
A-218
-------
include coal-fired industrial boilers operating profiles, the Technical Support team will determine
if the profiles should be combined or maintained separately. If the recommendation is to
combine the profiles, the Technical Support team will recommend a suitable methodology (e.g.,
arithmetic average, average weighted by emissions levels).
4.2 TEMPORAL ALLOCATION FACTOR FILE CONSTRUCTION
TRC is currently developing a detailed approach for constructing the final TAF file(s).
The data sources identified under this Work Assignment primarily target point source emissions.
The traffic patterns that drive mobile source emissions are highly characteristic of each specific
municipality and are not easily generalized as required for this project. Procedures for the
Preparation of Emission Inventories for Precursors of Ozone (Volume I) (EPA-450/4-88-021)
provides adequate temporal allocation factors for area sources; further area source TAF
development would yield little improvement.
A three-tiered approach is expected to be used to develop the final TAF file(s). Tier 1
will provide a baseline TAF file, covering all source categories, but includes data only at a
monthly level. Tier 1 data are useful for constructing seasonal allocation profiles. The low cost
of and easy access to these data allow immediate- improvement of the TAF file by providing
documented, routinely updated seasonal data by which all point source emissions can be
allocated. Within the season, however, a flat distribution (seven days per week, 24 hours per
day) would be assumed. Tier 1 data are expected to provide an acceptable temporal profile for
many sources, but will not contain the detail desired for large emissions sources.
Tier 2 will provide increased temporal resolution for a large number of categories. Data
contained in large databases will be used to provide temporal profiles for many sources,
improving the temporal resolution from that provided by Tier 1 data.
Tier 3 will be used to provide operating profiles for high-priority source categories. These
source categories will be identified by reviewing the results of the prioritization effort described
in Section 3 of this memorandum. High-priority source categories will be candidates for
individual attention, and data sources such as the CARB reports, wastewater studies, or the waste-
to-energy reports will be reviewed to determine the presence of sufficient data to improve these
CH-93-69
4-3
A-219
-------
profiles. If data are inadequate to characterize the. sources, telephone surveys may be used to
collect data to support the studies.
The Tier 1, 2, and 3 files will be used as intermediary work files, from which a single,
final deliverable TAF file will be developed. Table 4-1 summarizes the three levels.
TABLE 4-1. CHARACTERISTICS OF INTERMEDIARY FILES
Relative Number of Source
Temporal Resolution of
Tier
Categories
Data
1
High (all categories)
Monthly
2
Moderate to High
Hourly
' 3
Small
Hourly
All TAF records will include a data source indicator field to provide traceability and
documentation of the final product. TRC is placing a high priority on creating a TAF file design
that will allow future users to identify in-use data sources and readily update the file with future
data.
4.2.1 Tier 1
Predominantly using monthly labor and energy consumption data described in Section 2.1,
TRC will construct monthly operating profiles for all source categories. These profiles will
provide a baseline TAF file(s) and allow quarterly allocations for all sources based on routinely
updated, inexpensive, publicly available data. Tier 1 data will cover a broad range of data
sources, but will only provide seasonal allocation factors (based on monthly data). Tier 1 data
will be used only for those sources not included in Tiers 2 or 3.
4.2.2 Tier 2
TRC will create TAF file(s) based on the information retrieved from data sources such
as TACB, SOS, and LMOS (Sections 2.3, 2.4, and 2.5). The Database Design and Management
CH-93-69
4-4
A-220
-------
and Technical Assistance teams are currently developing methodologies for combining the data
for source categories common to two or more of the data sources. Tier 2 files will be used to
provide hourly profiles for a fairly large number of source categories. Tier 2 data will be used
for emissions sources not included in Tier 3.
4.23 Tier 3
For high priority source categories, TRC will develop operating profiles based on CARB
data, CEM reports, wastewater data, and other data sources (Sections 2,2, 2,6, 2,7, and 2,8),
TRC will conduct telephone surveys of no more than nine facilities to collect data for high-
priority source categories from which little alternative data exist. The data in Tier 3 files will
represent focused efforts toward improving the default operating profiles for a small number of
important emission sources. As with Tier 2 data, methodologies are currently under development
for combining the data for source categories common to two or more Tier 3 data sources. Tier
3 data should be the most detailed and desirable data available to the final TAF file(s), but will
also be the most resource-intensive, since Tier 3 temporal profiles will be constructed individually
for each included source category.
4.3 APPLICATION OF TEMPORAL ALLOCATION FACTOR FILE
The TAF file(s) developed under this project will be in a computer-neutral format
(probably flat ASCII) that will be accessible by photochemical models currently recommended
by EPA [Urban Airshed Model's Emission Preprocessor System (UAM-EPS), the GEMAP
system, and the Regional Oxidant Model (ROM)]. Temporal profiles may supersede the temporal
profiles found in the models or their preprocessors, although some revisions to the model's
temporal processors may be required. Any necessary revisions of the model processors will not
be made under this project. The most extensive revision would be required of GEMAP, for
which TAF file(s) activity values would convert to GEMAP format.
CH-93-W
4-5
A-221
-------
The TAF file(s) database structure is currently under design, but expected key fields to
retrieve an allocation profile include:
* AIRS source classification code (SCC)
« season (winter, spring, summer, fall)
« weekday or weekend
The TAF file(s) will include hourly allocation factors for hours 00 through 24 that may be
multiplied by the annual emissions estimates to yield hourly emissions estimates.
GH-93-69
4-6
A-222
-------
5.0 BIBLIOGRAPHY
Bureau of Labor Statistics, Employment and Earnings. U.S. Department of Labor, Washington,
DC, August 1992,
Bureau of Labor Statistics. Employment and Earnings. U.S. Department of Labor. Washington, DC.
March 1993.
Commodity Research Bureau. 1991 CRB Commodity Year Book. New York, NY. 1991
Energy Information Administration. Electric Power Monthly. DOE/EIA-0226(93/Ol), U.S.
Department of Energy. Washington, DC. January 1993.
Energy Information Administration. Petroleum Marketing Monthly. DOE/EIA-0380(93/03) U.S.
Department of Energy. Washington, DC. March 1993
Energy Information Administration. Petroleum Supply Monthly. TDOE/EIA-0109(93/02) U.S.
Department of Energy. Washington, DC. February 1993
Energy Information Administration. Quarterly Coal Report, October-December 1992. DOE/EIA-
Q121(92/4Q). U.S. Department of Energy. Washington, DC.
Energy Information Administration. Weekly Coal Production. Production for Week Ended: April 24,
1993. DOE/EIA-0218(93-17) U.S. Department of Energy. Washington, DC.
Energy Information Administration. Weekly Petroleum Status Report. DOE/EIA-0208(93-21). U.S.
Department of Energy, Washington, DC. May 21,1993.
Federal Register. Volume 57, No. 186, September 24, 1992.
Fiber Economics Bureau, Inc. Fiber Organon. Volume 62, No. 5. May 1991
Graphic Arts Monthly, Cahners Publishing. September 1988.
Graphic Arts Monthly. Cahners Publishing. October 1988.
National Air Pollutant Emissions Estimates, 1940-1989. EPA-450/4-91 -004 (NTIS PB91-168559).
Office of Air Quality Planning and Standards. U.S. EPA. Research Triangle Park, NC. March 1991.
Oil and Gas Journal Vol. 87, No. 35, PennWell Publishing Company. Tulsa, OK, August 28,
1989.
Paper, Paperboard & Wood Pulp. Vol. 69. No, 1. January 1991
A-223
-------
Textile World. Maclean Hunter Publishing. January 1991.
The Brewing Industry in the United States. Brewers Almanac 1990. Beer Institute.
-------
APPENDIX A. STATISTICAL AND ECONOMIC INDICATOR REFERENCES
A-l
A-225
-------
STATISTICAL AND ECONOMIC INDICATOR REFERENCE DOCUMENTS
PUBLICATION
FREQUENCY
PUBLISHER
FOUND
COMMENTS
PUBLICATIONS COVERING A RANGE OF INDUSTRIES
Industry Week
Weekly
Penton Publishing
Y
Pennsylvania Business Survey
Monthly
Pennsylvania State University
N
Manufacturing hours and employment data
Chemical and Engineering
News
Weekly
American Chemical Society
Y
U.S. Chemical Industry
Statistical Handbook
Annual
Chemical Manufacturers Association
N
Production, price. Financial, employment, health,
and safety data
Business Statistics
Quarterly
New York Slate, Department of
Economic Development
N
Activity, income, and Finance data
Business Week
Weekly
McGraw-Hill
Y
Labor Situation
Monthly
Connecticut, State of
N
Employment and earnings data. Economic
indicators
CRB Commodity Year Book
Annual
Commodity Research Bureau
Y
Production: 19xx Capital
Spending Survey
Annual
Gardner Publishing
N
Production, operating rate, and purchasing data
and projections
KEDI Economic Survey
Monthly
Kent Economic and Development
Institute, Inc.
N
Modelled quarterly income, employment, and
production data and forecasts
Economic Report
Bimonthly
Manufacturers Hanover Corp.
N
Financial and income data
Chemical Engineering
Monthly
McGraw-Hill
Y
BEVERAGES and BREWERIES
Brewing Industry in the U.S.:
Brewers Almanac
Annual
Beer Institute
Y
Beverage Industry
Annual
Edgell Communications
Y
Beverage Industry
Monthly
Edgell Communications
Y
(continued)
-------
STATISTICAL AND ECONOMIC INDICATOR REFERENCE DOCUMENTS (Continued)
PUBLICATION
FREQUENCY
PUBLISHER
FOUND
COMMENTS
PAPER* PAPERBOARD, AND WOOD PULP
19xx Statistics of Paper,
Paperboard, and Wood Pulp
Annual
American Paper Institute
Y
Paper, Paperboard, and Wood
Pulp Capacity
Annual
American Paper Institute
N
Production and employment data
Paper, Paperboard, and Wood
Pulp
Monthly
American Paper Institute
N
Production and employment data
Paperboard Packaging
Monthly
Edgell Communications
Y*
Pulp & Paper 19xx Factbook
Monthly
Freeman, Miller, Publications
Y
Pulp & Paper
Monthly
Freeman, Miller, Publications
Y
METALS INDUSTRY
Foundry Management and
Technology
Monthly
Penton Publishing
N
Production and operation data
Metal Producing
Monthly
Penton Publishing
Y
Annual Statistical Report
Annual
American Iron and Steel Institute
N
Production and employment data
Iron Ore Report
Monthly
American Iron One Association
N
Operation data
Aluminum Statistical Review
Annual
Aluminum Association
N
Production and market data
PLASTICS INDUSTRY
Modem Plastics
Monthly
McGraw-Hill
N
Capacity, sales, and operating data
PRINTING INDUSTRY
Newsprint Statistics
Monthly
American Newspaper Publishers
Association
N
Consumption, production, inventory, and
operating data
Graphic Arts Monthly
Monthly
Cahners Publishing Co.
Y*
(continued)
-------
STATISTICAL AND ECONOMIC INDICATOR REFERENCE DOCUMENTS (Continued)
PUBLICATION
FREQUENCY
PUBLISHER
FOUND
COMMENTS
TEXTILE AND FIBERS INDUSTRY
Textile World
Monthly
Maclean Hunter Publishing
Y
Fiber Organon
Monthly
Fiber Economics Bureau
N
Production and capacity data
Fairchild's Textile and Apparel
Financial Directory
Annual
Fairchild Fashion and Merchandising
Group
N
Production, capacity, employment, and capacity
data
Fairchild Fact File
Series
Fairchild Fashion and Merchandising
Group
N
Production, capacity, employment, and capacity
data
I9xx Sales of Natur.il Gas
Liquids and Liquified Refinery
Annual
American Petroleum Institute
N
Sales and use dala
Basic Petroleum Data Book
3x/Y car
American Petroleum Institute
Y
Gas Facts
Annual
American Gas Association
N
Sales data
Gas Slats
Monthly
American Gas Associalion
N
Sales data
Gas Slats
Quarterly
American Gas Association
N
Sales data
Public Power
Bimonthly
American Public Power Association
Y
Statistical Yearbook of the
Electric Utility Industry
Annual
Edison Electric Institute
N
Production and capacity data
Oil and Gas Journal
Weekly
PennWell Publishing Co.
Y
NUEXCO 19XX Annual
Review
Annual
NUEXCO
N
Sales data
Electric Light and Power
Monthly
PennWell Publishing
N
Securities and annual operating data
Electricity Supply and Demand
Annual
North American Reliability Council
Y
Steam Electric Market
Analysis
Monthly
National Coal Association
Y
(continued)
-------
STATISTICAL AND ECONOMIC INDICATOR REFERENCE DOCUMENTS (Continued)
PUBLICATION
FREQUENCY
PUBLISHER
FOUND
COMMENTS
Fuel Oil News:The Oil Heat
Industry
Monthly
Hunter Publishing Co.
N
Sales and service data
Coal
Monthly
Maclean Hunter Publishing
Y*
Electrical World
Monthly
McGraw-Hill
Y
MISCELLANEOUS
Footwear Manual
Annual
Footwear Industries of America
N
Production, shipment, and employment data
Chilton's Automotive
Industries
Monthly
Chilton Co.
Y*
Datamation
Quarterly
Cahners Publishing Co.
N
Market, salary, and budget information
Electronic Business
Biweekly
Cahners Publishing Co.
Y
Automotive News
Weekly
Crain Communications
N
Production data
* Obsolete data or discontinued serial found
-------
APPENDIX B. CARB REPORT REVIEW SUMMARY
B-l
-------
TABLE B-l. CARB DATA POTENTIALLY USEFUL IN DEVELOPING TAFFS
Ref. #
Source/
Process/SCC Code(s)
Applicable SCC
Useful Surrogate
Data
Emission Data (Lb/Hr)
Test Period
1
fluidized bed wood combustion,
1-02-004-07
operating
arsenic, beryllium cadmium,
9/87
cooling tower, fugitive dust
schedule for all
hexavalenl chromium.
processes
copper,
(hrs/day,
lead,
days/wk,
manganese
wks/yr);
mercury,
product
nickel,
throughput (units
selenium,
defined only for
zinc.
cooling tower
formaldehyde.
BTU/hr);
acetaldehyde,
stack gas flow
benzene.
(for fluidized bed
naplhalene,
in CFM);
benzo(a)pyrene.
plantwide
dibenzofurans.
relative monthly
dioxins
activity given
chlorine (cooling lower only)
(%)
bromine (colling tower only)
3
cement silo loading system,
weigh hoppers, ready mix truck
loading system - concrete batch
plant
3-05-011-07
3-05-011-08
3-05-011-09
operating
schedule for all
processes
(hrs/day,
days/wk,
wks/yr);
product
throughput for all
processes (cubic
yards/hr)
hourly maximum emissions for all
processesarsenic
compoundscadmium compoundstotal
chromium, hexavalent chromium,
lead compoundsinercury, nickel and
nickel compoundsselenium
compounds
5/23/90
(continued)
-------
TABLE B-l. CARB DATA POTENTIALLY USEFULE IN DEVELOPING TAFFs (Continued)
Ref. #
Source/
Process/SCC Code(s)
Applicable SCC
Useful Surrogate
Data
Emission Data (Lb/Hr)
Test Period
3
truck loading,
conveyor belt transfer station,
crushing,
sorting,
pile forming,
storage,
truck hauling,
gasoline storage - rock aggregate
plant
3-05-025-06
3-05-025-03
3-05-025-10
3-05-025-11
3-05-025-05
3-05-025-07
3-05-025-07
3-05-025-04
3-05-025-01
product
throughput
(TPH); for
gasoline storage
(gal/hr given);
plantwide
relative monthly
activity given
crystalline silica, total PM
5/22/90
5
cotton gin
3-02-004-10
operating
schedule
(hrs/day,
days/wk/, wks/yr
given);
product
throughput
(cotton bales/hr
given)
total PM, chromium, arsenic,
beryllium nickel, lead, mercury,
copper, manganeseselenium, silica,
zinc (note: all emissions determined
to be 0 lbs/hr)
? (but report dated
6/6/91)
6
separating hull, delinting, meal
processing -
cottonseed oil mfg. plant
3-04-494-95
3-02-949-98
3-02-019-12
operating
schedules (for all
processes in
hrs/day,
days/wk.,
wks/yr);
relative monthly
activity given for
each process
tested
cadmium, lead, zinc, nickel,
berylliumchromium VI,
crystalline silica
4/30 - 5/1/90
(continued)
-------
TABLE B-l. CARB DATA POTENTIALLY USEFULE IN DEVELOPING TAFFs (Continued)
Ref. #
Source/
Applicable SCC
Useful Surrogate
Emission Data (Lb/Hr)
Test Period
Process/SCC Code(s)
Data
6
hull processing/pelletizer, oil
3-029-99-48 -
operating
same as above plus arsenic, copper,
same as above
cottonseed process, natural gas
3-02-09-03
schedules (for all
manganesemercury,
boiler - cottonseed oil mfg. plant
1-02-006-02
processes in
selenium;
hrs/day,
benzene and formaldehyde
days/wk„
(from the boiler units)
wks/yr);
relative monthly
activity given for
each process
tested; fuel
throughput for 6
gas boiler units
(cubic feet);
stack gas flow
rale (dscfm)
7
natural gas boiler, shipping/
1-02-006-02
<
planlwide
boiler unit (benzene and
? (report dated
receiving grain, hammennilling
3-02-005-01
operating
formaldehyde);
6/4/90)
grain,
3-02-005-01
schedule
all other proceses tested for
pelletizing grain, shipping/
3-02-008-16
(hrs/day,
emissions of cadmium, chromium,
loadout grain - food products
3-02-005-01
days/wk.,
copper, nickel.
plant
wks./yr.);
manganese zinc, chromium VI,
product
throughput for
boiler unit
(MMcuft./hr.);
plantwide
relative monthly
activity (%)
(continued)
-------
TABLE B-l. CARB DATA POTENTIALLY USEFULE IN DEVELOPING TAFFs (Continued)
Ref. #
Source/
Process/SCC Code(s)
Applicable SCC
Useful Surrogate
Data
Emission Data (Lb/Hr)
Test Period
12
HDS heater - oil refinery
3-10-004-01
avg. heat input
(MMBTU/lir);
avg. fuel use
[millions of
cubic feet per
hour (MCFH)]
PAH, phenols, formaldehyde,
acetaldehyde, benzene, HS, toluene
3/5 and 6/90
12
a variety of crude oil, diesel or
gas fired IC engines/compressors
or steam generators
crude oil steam generators
natural gas steam generators
diesel compressors
3-10-004-13
3-10-004-14
3-10-002-03
heat inputs given
for each source
in MMBTU/lir;
dependent on source tested; PAH,
trace metals, VOCs, aldehydes, HCs
4/2-26/90
14
radiant reheat boiler using
residual fuel oil and natural gas -
utility power plant
natural gas
fuel oil
1-01-006-01
1-01-004-01
product
throughput (rated
capacity in lb/hr
(MW) given;
plant was
operating at 96%
rated capacity
during emissions
tests
formaldehyde, metals, cadmium,
CR/CR*6, PAH, benzene using fuel
oil; formaldehyde using natural gas
3/1-6/90
15A
industrial external combustion
boilers burning sawdust or bark
(dutch oven type)
1-02-009-05
product
throughput
(lb/lir); stack gas
flow rates given
but may be
inacurrate
(dscfm);
benzene, PM, metals, hexavalent
and total chromium,
PAH, phenols, (note: lb/hr not
always given but can be calculated
from data presented)
6/7/93
(continued)
-------
TABLE B-l. CARB DATA POTENTIALLY USEFULE IN DEVELOPING TAFFs (Continued)
Ref. #
Source/
Process/SCC Code(s)
Applicable SCC
Useful Surrogate
Data
Emission Data (Lb/Hr)
Test Period
15B
wood fired
boiler (hog fuel)
1-02-009-03
product
throughput
(lb/hr);
stack gas flow
rate (dscfm);
benzene, PM, metals, hexavalent
and total chromium,
PAH, phenols, crystalline silica,
(note: lb/hr not always given but
can be calculated from data
presented)
7/25-30/90
15C
stoker-type bark fired boiler
1-02-009-03
product
throughput
(lb/hr); slack gas
(low rate (dscfm)
formaldehyde, acetaldehyde,
benzene, PM, metals, hexavalent
and total chromium, PAH and
crystalline silica
(note: lb/hr not always given but
can be calculated from data
presented)
7/13-18/90 (tested
emissions of
metals, PM,
formaldehyde,
acetaldehyde,
benzeneand
crystalline silica;
11/3-4/90
(tested for
hexavalent
chromium, PAH
and particulates)
15D
fiuidized bed wood or bark fired
boiler
1-02-009-02
product
throughput
(lbs/hr); stack
gas flow rate
(dscfm)
same as above (note: lb/hr not
always given but can be calculated
from data presented)
6/19-21/90
15E
wood fired boiler (fuel cell type)
1-02-009-06
product
throughput
(lbs/hr);
stack gas flow
rate (dscfm)
same as 15C except PM (note: lb/hr
not always given but can be
calculated from data presented)
9/26-29/90
(continued)
-------
TABLE B-l. CARB DATA POTENTIALLY USEFULE IN DEVELOPING TAFFs (Continued)
Ref. #
Source/
Process/SCC Code(s)
Applicable SCC
Useful Surrogate
Data
Emission Data (Lb/Hr)
Test Period
15F
wood fired boiler (stoker type)
1-02-009-03
product
throughput
(Ibs/hr); stack
gas flow rate
(dscfm)
same as 15C (except crystalline
silica),
phenols
(note: Ib/hr not always given but
can be calculated from data
presented)
9/6-8/90
15G
wood fired boiler (stoker type)
1-02-009-03
product
throughput
(Ibs/hr);
stack gas flow
rate (dscfm)
formaldehyde,
acct aldehyde,
benzene,
PM/metals,
hexavalent and total
chromium,
PAH,
phenols
9/11-13/90
15H
wood fired boiler (dutch oven
type)
1-02-009-03
product
throughput
values given for
summer and
winter seasons
(lbs/hr); stack
gas flow rate
(dscfm)
same as 15C except crystalline
silica
(note: Ib/hr not always given but
can be calculated from data
presented)
8/15-17/90
151
wood fired boiler (stoker type)
1-02-009-03
product
throughput
(lbs./hr);
stack gas flow
rate (dscfm)
same as 15G plus crystalline silica
8/1-3/90 and 10/16-
18/90
15J
wood fired boiler (air suspension
burner type)
1-02-009-03
same as above
same as 15G except phenols
6/6-9/90
(continued)
-------
TABLE B-l. CARB DATA POTENTIALLY USEFULE IN DEVELOPING TAFFs (Continued)
Ref. #
Source/
Process/SCC Code(s)
Applicable SCC
Useful Surrogate
Data
Emission Data (Lb/Hr)
Test Period
16
bleach plant - bleach vents and
smelt dissolver exhausts
3-07-001-99
slack gas flow
rate (dscfm)
same as above
? (report dated
12/13/90)
16
bleach plant (for bleaching pulp)
alkaline and acid/oxidizer stage
hoods
3-07-001-99
stack gas flow
rate (dscfm)
emissions are not reported in mass
per unit time
8/27-31/90
16
kraft pulp mill recovery furnace
3-07-001-04
stack gas flow
rate (dscfm)
metals,
PAHs, organics,
miscellaneous organics
(results arc not in lbs/hr or any
equivalent mass per unit of time
designation)
6/26/90 and
.8/1-10/90
16
recovery furnace, smelt dissolver,
lime kiln, bleach plant, wood-
fired boiler (stoker type) - paper
plant
3-07-001-04
3-07-001-05
3-07-001-06
3-07-001-09
3-07-001-99
product/heat
throughput
(MMBTU/hr;
Ib/hr dry BLS, lb
sleam/hr. tons/lir
dry solids, tons
lime/hr,);
stack gas flow
rate given for
each process
(acfm or dscfm
or % 02 dscfm)
given in g/hr for
metals, formaldehyde, acetaldehyde,
HCL, chlorine,
methanol, benzene, chloroform,
napthalene, BaP, furans, dioxins
Sept-Oct/90
(continued)
-------
TABLE B-l. CARB DATA POTENTIALLY USEFULE IN DEVELOPING TAFFs (Continued)
Ref. #
Source/
Applicable SCC
Useful Surrogate
Emission Data (Lb/Hr)
Test Period
Process/SCC Code(s)
Data
17
IC engine for natural gas
3-10-004-14
operating
given for both inlet and outlet side
4/11/90
combustion - gas company
schedule (report
tests
states continuous
benzene, toluene,
operation during
xylenes,
test period, i.e.,
propylenenapthalene,
24 hrs);
formaldehyde,
average
acetaldehyde.
horsepower
acrolein.
given;stack gas
ammonia
flow rate (avg.
rate in dscfm
given for both
inlet and outlet
sides)
23
electric utility steam boiler
1-01-006-01
avg. product
natural gas (benzene and
2/1-5/90
1-01-006-2
throughput
formaldehyde); distillate oil
natural gas boiler
1-01-006-04
(lb/hr); fuel
(benzene, formaldehyde, total PAH,
distillate oil boiler
1-01-005-01
throughput
also broken down by species)
1-01-005-04
(natural gas
1-01-005-05
(kscfh), distillate
oil (gal/min)
(continued)
-------
TABLE B-l. CARB DATA POTENTIALLY USEFULE IN DEVELOPING TAFFs (Continued)
w
o
Ref. #
Source/
Process/SCC Codc(s)
Applicable SCC
Useful Surrogate
Data
Emission Data (Lb/Hr)
Test Period
24
natural gas heater (residual oil
used as alternate fuel)
natural gas
residual oil
1-01-006-01
1-01-004-01
1-01-004-04
1-01-004-05
1-01-004-06
product
throughput (nal.
gas - ft3/min;
residual oil -
[gallons per
minute (GPM)];
fuel throughput
(BTU/lb);
avg. stack gas
flow rate
(dscfm)
benzene, formaldehyde, PAH,
metals (for both types of fuels)
4/2-7/90
25
rotary dryer - asphalt facility
3-05-002-01
operating
schedule -
depends on
contract and
weather
conditions;
product
throughput (tph);
avg. stack gas
flow rate (dscfm)
organics, aldehydes, HS, gaseous,
trace elements, PAH
4/5/90, 6/27 and
29/90
27
wastewater plant facility -
screening device barminuter for
the headworks
5-01-007-01
gas effluent
through the
headworks (avg.
dscfm); note: gas
is only product
through the
headworks
inorganic compoundsammonia,
chloride, HS, VOCs, xylene,
toluene, benzene, chlorinated HCs
March, May, July,
Sept. 1990
I
>
I
to
u>
vo
(continued)
-------
TABLE B-l. CARB DATA POTENTIALLY USEFULE IN DEVELOPING TAFFs (Continued)
Ref. #
Source/
Process/SCC Code(s)
Applicable SCC
Useful Surrogate
Data
Emission Data (Lb/Hr)
Test Period
28
chlorine mixing facility - water
quality control plant
5-01-007-01
seasonal chlorine
usage data for
winter and
summer given
(Ibs./mgd)
inorganic compounds
HS, ammonia, chloride,
VOCs, benzene, toluene, xylene,
chlorinated HCs
3/8,20,27/90 and
7/25/909/5/909
29
multiple hearth sewage sludge
incinerator - water quality
treatment plant
5-02-001-01
operating
schedule
(hrs.day);
product
throughput
(lb/day and
scrubber flow in
gpm);
fuel throughput
(dscf/min)
metals, HCL, VOC, PAH, ketone,
aldehydes
3/90 and 9/90
31
wastewater treatment plant
processes - liguid process, solids
process, combustion process,
auxiliary process (16 total
processes tested)
entire processes
5-01-007-01
plantwide %
relative monthly
activity given -
hourly process
rates (units not
given) -
operating
schedule
(hrs/day,
days/wk, wks/yr)
form aldehyde, ammonia, benzene,
toluene, PAHs
April and May,
1991
(continued)
-------
TABLE B-l. CARB DATA POTENTIALLY USEFULE IN DEVELOPING TAFFs (Continued)
Ref. #
Source/
Process/SCC Code(s)
Applicable SCC
Useful Surrogate
Data
Emission Data (Lb/Hr)
Test Period
33 App. E
asphalt mixer/diesel powered
rotary kiln
3-05-002-01
product
throughput
(TPH); slack gas
flow rate
different for
metal emissions
versus PAHs
(dscfm)
metals, PAHs,
benzene,
formaldehyde
Nov. 1990
33 App. F
asphalt mixer/natural gas fired
rotary kiln
3-05-002-01
product
throughput
(tons/test period
in minutes);
stack gas flow
rate for kiln
(dscfm)
trace metals, berylliumchromium,
PAH, formaldehyde, benzene
Sept. 1990
33 App G.
asphalt mixer/rotary kiln
3-05-002-01
product
throughput
[tons/lest period
(72 min.) given];
stack gas flow
rate (dscfm)
trace metals, arsenic,
chromium,
PAH,
benzene,
formaldehyde
Aug and Sept.
1990
34
lime recalcing/charcoal furnace -
wastewater treatment plant
5-01-007-01
flow in millions
of gallons per
day (mgd) given
(plant design
capacity data);
not sure whether
this is inflow or
outflow after
treatment
metals, PAHs, furans, dioxins, given
for both processes
4/29/91 and
5/8/91
(continued)
-------
TABLE B-l. CARB DATA POTENTIALLY USEFULE IN DEVELOPING TAFFs (Continued)
Ref. #
Source/
Process/SCC Code(s)
Applicable SCC
Useful Surrogate
Data
Emission Data (Lb/Hr)
Test Period
37
surface treatment tanks - plating
shop
3-09-010-01
avg. stack gas
flow rate (dscfm)
total chromium, lead, copper, nickel
7/17-18/90
38
6 diesel fired generating units at
the LA airport
2-01-001-02
stack gas flow
rates (dscfm)
given for each
unit tested
speciated PAHs and formaldehyde
Oct. 1990
40
delaquering for aluminum MFG
3-03-001-04
Avg. product
throughput
(lbs.hr)
Avg. stack gas
flow rate (dscfm)
metals, HCL, HF, arsenic, dioxins,
furans
6/26-28/91
41
carbon fabric dryer for
carbonization of rayon
3-01-005-02
product
throughput [feet
per hour (FPH)]
avg. stack gas
flow rate [dry
standard cubic
feet per minute
(dscfm)]
formaldehyde, benzene, toluene,
ethyl benzene, phenols, PAH
10/17 and 26/90
42
waste gas burner, neutralization
pit - chemical mfg. facility
3-01-301-10
3-01-195-06
avg. stack gas
flow rate for
waste gas burner
(dscmh)
emission rates in kg/hr from
waste gas burner - methyl
methacrylate and styrene from both
influent and effluent streams; HCL
(ug/hr) from the neutralization pit
12/7/90
(continued)
-------
APPENDIX C. ACID-MODES FIELD STUDY DATABASE FORMAT
A-243
-------
1
Tbe SAS System
09:46 Tuesday, June 22,1993 1
CONTENTS PROCEDURE
Data Set Name: SASTAFEQ.DATA
Member Type: DATA
Engine: TAPE
Created: 1&1X Friday, Much 1,1991
Last Modified: .
Protection:
Data Set Type;
Label:
Observations:
Variables: 103
Indexes: 0
Observation Length; 818
Deleted Observations: 0
Compressed: NO
Sorted: NO
—Engine/Host Dependent Information
Data Set Page Size: 32760
Physical Name: IVSRADM.UNi iiNFO-SASDAT'
Release Created: 6.06
Created by: IVSMERGE
—Alphabetic List of Variables and Attributes
Variable Type ten Pos Label
3
DAY ID
Char
2
16
79
HRATEl
Num
8
618
80
HRATE2
Num
8
626
81
HRATE3
Num
3
634
82
HRATE4
Num
8
642
83
h rates
Num
8
650
84
H RATES
N'um
8
658
85
HRATE7
N'um
8
666
86
H RATES
Num
8
674
87
HRATES
N'um
8
682
88
HRATEiO
Num
8
690
89
HRATEl 1
N'um
8
698
90
HRATE12
Num
8
706
91
HRATE13
Num
8
714
92
HRATE14
Num
8
722
93
HRaTELS
Num
8
730
94
HRATE16
Num
8
738
95
HRATE17
Num
8
746
96
HRATE18
Num
8
754
97
H RATE 19
Num
8
762
98
HRATE20
Num
S
770
99
HRATE21
Num
8
778
100
HRATE22
Num
8
786
101
HRATE23
Num
8
794
102
HRATE24
Nurn
8
802
103
HRATE2S
Num
8
810
2
MONTH ID
Char
4
12
4
MWLOADl
Num
S
18
5
MWLOAD2
Num
S
26
6
MWLOAD3
Num
8
34
7
MWLOAD4
Nuns
8
42
8
MWLOAD5
Num
8
50
9
MWLOAD6
Num
8
58
10
iMWLOAD7
Num
8
66
11
MWLOAD8
Nurn
8
74
12
MWLOAD9
Nuts
8
' 82
24 HOURLY GEN (MW)
The SAS System
CONTENTS PROCEDURE
09:46 Tuesday, June 22,1993 2
# Variable Type Lett Pos Label
C-2
A-244
-------
13
MWLOAD1Q
Num
8
90
14
MWLOAD11
Num
S
98
15
MWLOAD12
Num
8
106
16
MWLOAD13
Num
8
114
17
MWLOAD14
Num
8
122
18
MWLOAD15
Num
8
130
19
MWLOAD16
Num .
8
138
20
MWLOAD17
Num
S
146
21
MWLOAD18
Num
S
154
22
MWLOAD19
Num
8
162
23
MWLOAD2Q
Num
8
170
24
MWLOAD21
Num
8
178
25
MWLOAD22
Num
8
184
25
MWLOAD23
Num
8
194
27
MWLOAD24
Num
8
202
23
MWLOAD25
Num
8
210
1
NED ID Char 12
0
54
NOXCEMl
Num
8
418
55
NOXCEM2
Num
8
426
56
NOXCEM3
Num
3
434
57
NOXCEM4
Num
8
442
58
NOXCEM5
Num
8
450
59
NOXCEM6
Num
8
458
60
NOXCEM7
Num
8
466
61
NOXCEM8
Num
8
474
62
NOXCEM9
Num
8
482
63
NOXCEMIO
Num
8
490
64
NOXCEMH
Num
8
498
65
NOXCEM12
Num
8
506
66
NOXCEM13
Num
8
514
67
NOXCEM14
Num
8
522
68
NOXCEMlS
Num
8
530
69
XOXCEM16
Num
8
538
70
NOXCEMH
Num
8
546
71
NOXCEMlS
Num
8
554
72
NOXCEM19
Num
8
562
73
NOXCEM20
Num
1
570
74
NOXCEM21
Num
8
57S
75
NOXCEM22
Num
8
586
76
NOXCEM23
Num
8
594
77
NOXCEM24
Num
8
602
78
NOXCEM25
Num
8
610
29
S02CEM1
Num
8
218
30
S02CEM2
Num
8
226
31
S02CEM3
Num
8
234
32
S02CEM4
Num
8
242
33
S02CEM5
Num
8
250
34
S02CEM6
Num
8
258
35
S02CEM7
Num
8
266
36
S02CEM8
Num
8
274
37
S02CEM9
Num
S
282
38
SO2CEM10
Num
8
290
39
S02CEM11
Num
8
298
40
S02CEM12
Num
8
306
The SAS System
0*46 Tuesday, June 22,1993 3
CONTENTS PROCEDURE
# Variable Type Len Pos Label
41 S02CEMX3 Num
42 S02CEM14 Num
43 S02CEM15 Num
44 S02CEM16 Num
8 314
8 322
8 330
8 338
C-3
A-245
-------
45
S02CEM17
Num
S
346
46
S02CEM18
Num
8
354
47
S02CEM19
Num
8
362
4S
SO2CEM20
Num
i
370
49
S02CEM21
Num
8
378
50
S02CEM22
Num
8
386
51
S02CEM23
Num
8
394
52
S02CEM24
Num
8
402
53
S02CEM2S
Num
S
410
The SAS System
09:46 Tuesday, June 22,1993 4
CONTBNTS PROCEDURE
Directory
Lib re f: SASTAPEO
Engine: TAPE
Physical Name: rvSRAJDM.UMi lNFO.SASDAT
Unit: tape
Volume: NORMS
Disposition: SHR
Device: 3400
Blocksize: 32760
# Name Memtype -Indexes
1 DATA DATA
€-4
-------
APPENDIX D. EPS DATABASE FORMAT
D-l
-------
Structure for database: C:\EPS\SCCTPL.DBP
Number of data records: 28667
Date of last update : 06/26/93
Field
Field Name
Type
Width
Dec
Index
1
see
Character
10
N
2
SARAOD
Character
5
N
3
MONTH
Character
4
N
4
WEEK
Character
4
N
5
DAY
Character
4
N
** Total **
28
D-2
A-248
-------
Structure for database; C:\EPS\MONTH.DBF
Number of data records: 27
Date of last update : 06/26/93
Field
Field Name
Type
Width
Dec
Index
1
IMNCD
Character
5
N
2
JAN
Numeric
4
N
3
FEB
Numeric
4
N
4
MAR
Numeric
4
N
5
APR
Numeric
4
N
6
MAY
Numeric
4
N
7
JUN
Numeric
4
N
8
JUL
Numeric
4
N
9
AUG
Numeric
4
N
10
SEP
Numeric
4
N
11
OCT
Numeric
4
N
12
NOV
Numeric
4
N
13
DEC
Numeric
4
N
14
TOT
Numeric
5
N
** Total **
59
D-3
A-249
-------
Structure for database; C:\EPS\WEEK.DBF
Number of data records: 11
Date of last update : 06/26/93
Field
Field Name
Type
Width
Dec
Index
1
IWKCD
Numeric
5
N
2
MON
Numeric
4
N
3
TUB
Numeric
4
N
4
WED
Numeric
4
N
5
THU
Numeric
4
N
6
FRI
Numeric
4
N
7
SAT
Numeric
4
N
8
SUN
Numeric
4
N
9
TOT
Numeric
5
N
** Total **
39
D"4
-------
Structure for database: C:\EPS\DAY.D3F
Number of data records: 40
Date of last update : 06/26/93
Field
Field Name
Type
Width
Dec
Index
1
IDYCD
Numeric
5
N
2
HR0000
Numeric
4
N
3
HR0100
Numeric
4
N
4
HR0200
Numeric
4
N
5
KR0300
Numeric
4
N
6
HR0400
Numeric
4
N
7
HRQ500
Numeric
4
N
8
HR0600
Numeric
4
N
9
HR070Q
Numeric
4
N
10
HR0800
Numeric
4
N
11
HR0900
Numeric
4
N
12
HR1000
Numeric
4
N
13
HR1100
Numeric
4
N
14
HR1200
Numeric
4
N
15
HR1300
Numeric
4
N
16
HR1400
Numeric
4
N
17
HR1500
Numeric
4
N
18
HR1600
Numeric
4
N
19
HR1700
Numeric
4
N
20
HR1800
Numeric
4
N
21
HR1900
Numeric
4
N
22
HR2000
Numeric
4
N
23
HR2100
Numeric
4
N
24
HR22 00
Numeric
4
N
25
HR2300
Numeric
4
N
26
TOT
Numeric
5
N
** Total **
107
-------
(JAM EPS TEMPORAL PROFILES
see
SAROAD
MOM
WK
DAY
42603
1
5
24
43104
1
5
24
42101
1
5
24
42401
1
5
24
11101
1
5
24
81102
1
5
24
2101000000
42603
1
7
24
2101000000
43104
1
7
24
2101000000
42101
1
7
24
2101000000
42401
1
7
24
2101000000
11101
1
7
24
2101000000
81102
1
7
24
2101001000
42603
1
7
24
2101001000
43104
1
7
24
2101001000
42101
1
7
24
2101001000
42401
1
7
24
2101001000
11101
1
7
24
2101001000
- 81102
1
7
24
2101002000
42603
1
7
24
2101002000
43104
1
7
24
UAM EPS MONTHLY PROFILES
CODE
JAN
FEB
. MAR
APR
MAY
JUN
1
83
83
83
83
83
83
2
79
79
82
03
84
84
3
1
1
1
1
1
1
4
178
178
143
107
36
36
5
178
178
143
107
36
36
6
190
238
238
190
0
0
7
333
167
0
0
0
0
8
61
73
98
122
98
73
9
6
6
9
9
10
10
10
32
43
64
85
106
106
11
39
65
65
39
39
130
12
13
27
40
67
133
133
13
0
0
15
74
147
147
14
0
0
167
333
333
0
16
100
100
100
83
83
83
17
76
76
76
79
79
79
18
100
100
100
93
93
93
19
96
96
96
100
100
100
20
76
76
76
76
76
76
21 .
106
106
106
100
100
100
>
KJ
U»
to
JUL
AUG
SEP
OCT
NOV
DEC
TOT
83
83
83
83
83
83
996
85
87
85
84
84
84
1000
1
1
1
1
0
0
10
36
36
36
36
36
143
1001
36
36
36
36
178
178
1178
0
0
0
0
24
119
999
0
0
0
0
167
333
1000
61
61
73
122
98
61
1001
10
10
9
8
7
6
100
106
106
106
106
85
53
998
130
104
104
130
104
52
1001
133
133
133
107
67
13
999
147
. 147
147
147
29
0
1000
0
0
0
0,
0
167
1000
66
66
66
83
83
83
996
83
83
83
93
93
93
993
59
59
59
79
79
79
993
56
56
56
79
79
79
993
103
103
103
76
76
76
993
53
53
53
73
73
73
996
-------
UAM EPS WEEKLY PROFILES
CODE MON TUE WED THR FRI SAT SUN TOT
1
1
1
1
1
1
0
0
5
2
0
0
0
0
0
1
1
2
3
1
1
1
1
1
0
0
5
4
1
1
1
1
1
0
0
5
5
1
1
1
1
1
0
0
5
6
1
1
1
1
1
1
0
6
7
1
1
1
1
1
1
1
7
13
4
4
4
4
4
3
3
26
21
1
1
1
1
1
2
2
9
22
10
10
10
10
10
7
4
61
23
5
5
5
5
5
4
4
33
UAM EPS
DAILY
TEMPORAL
PROFILE
CODE
00
01
02
03
04
05
06
07
08
09
10
11
12
13
14
15
16
17
18
19
20
21
22
23
TOT
6
0
0
0
0
0
0
1
1
1
1
1
1
0
0
0
0
0
0
0
0
0
0
0
0
6
7
0
0
0
0
0
0
0
0
0
1
1
1
1
1
1
1
0
0
0
0
0
0
0
0
7
8
0
0
0
0
0
0
0
0
1
1
1
1
1
1
1
1
0
0
0
0
0
0
0
0
8
9
0
0
0
0
0
0
0
0
1
1
1
1
1
1
1
1
1
0
0
0
0
0
0
0
9
10
0
0
0
0
0
0
0
1
1
1
1
1
1
1
1
1
1
0
0
0
0
0
0
0
10
11
0
0
0
0
0
0
0
1
1
1
1
1
1
1
1
1
1
0
0
0
0
0
0
11
12
0
0
0
0
0
0
1
1
1
1
1
1
1
1
1
1
1
1
0
0
0
0
0
0
12
13
0
0
0
0
0
0
0
0
1
1
1
1
1
1
1
1
1
1
1
1
1
0
0
0
13
14
0
0
0
0
0
0
0
0
1
1
1
1
1
1
1
1
1
1.
1
1
1
1
0
0
14
15
0
0
0
0
0
0
0
0
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
0
15
3 16
0
0
0
0
0
0
0
0
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
16
o 17
0
0
0
0
0
0
0
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
17
18
0
0
0
0
0
0
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
18
19
0
0
0
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
0
19
20
0
0
0
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
0
20
21
0
0
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
0
21
22
0
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
0
22
23
0
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
23
24
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
24
26
0
0
0
0
0
0
0
0
21
21
21
21
21
21
21
21
21
21
21
21
21
21
54
54
402
>
I
to
Ul
u>
-------
APPENDIX B
DATA SOURCE MATRIX
B-i
-------
temporal alloc at
ON DAT
V SOURC
ES FOR
SCCs
:
i
I
J
1
i
i
* 1
FREQUENCY C
FOCCt
JRRENCE IN THESE DAI
A SOURCES *
i
f
r I-
i
1
see
DATA
|
i I i
PRIORITY
GAP
i
,
I | |
RANK
see
CODE
CARB
TACB
sos LMOS
CEM
W-T-d
ACID—M
NAPAF
BLS
E/Ei CRBi PETl
!¦ .
! i !
19
10100101
2x
13x
t 1
10100102
2x
13*
1 1
13
10100201
2x
13x
1 1 !
11
10100202
10X
I 30X
2x
>
66x
1 -1 !
16
10100203
I 7X
2x
X
18x
I 1 1
29
10100204
X I 4X
2x
X
13x
• i i
10100205
I
2x
X
13x
i j |
21
10100212
3X
| 2X
2x
i i i
73
10100217
I
2x
% 13x
i 1 !
52
10100221
I X
2x
X
i i
14
10100222
6X
! 6X
2x
X
i I
37
10100223
,
2X
! ' W
2x
X
:
10100224
i
I 3>| 2*
I X
41
10100225
i '
I \2x I ! X
I 15
10100226
12X
I X2x !
X
i
10
10100301
I18X
! 12x
13x
1
34
10100302
12X
I 12x
i ¦
| 31
10100303
2x
13x
; ! i
10100304
4X
2x
13x
'
i ;
10100306
2x
13x
1 1
59
10100401
I Xl2x
16x
! 1 !
23
10100404
I2x
16x
• ; :
10100405
I 2x
16x
I i I i
10100406
1X
2x
16x
I : !
10100501
14X
26X
2x
16x
I ! ! i
I I ( I
10100504
3X
2x i
16x
I ! ! !
10100505
1X
2x I
16x
I i i i
22
10100601
201X
23X
2x
13x
i ,
l ! i «
69
10100602
110X
19X
2x
13x
i i . i
25
10100604
38X
2x
13x
148
10100701
35X
2x
I
i i ;
10100702
9X
2X
2x
| i
99
10100801
2x
' 1 '
5
1Q10QS01
2X
2x
! ! i
146
10100902
4X
2X
2x
1 i 1
155
10100903
1X
3X
2x
! • i i
10101001
4X|2x
1 1 !
10101002
5>J 2x
i 1 1
10101101
I 2x
\
1 1 !
10101201
7X
7X| 2x
i l i
10101202
12x
1 1 !
78
10101301
2X
I ax
1 j
10101302
2X
I 2x
10200101
Ix
10200104
>4
10200107
xl
62
10200201
2)4
45
10200202
1X
sxj
65
10200203
1X
sxj
94
10200204
2X
48X1
136
10200205
2X
1 sxj
10200206
18M
10200210
n
I
10200212
,
n
1
10200213
z
I 1
10200217
z
I
) ' !
13—Sep-93
B-2
-------
TEMPORAl
.ALLOCAT
ON DAT
\ SOURC
ES FOR
SCCs
I
' i '
I
! 1
FREC
3UENCY OF OCCI
JRRENCE IN THESE DAT.
A SOURCE
s *
1 1
i
| |
; scd
DATA
i
) 1
PRIORITY
GAP
|
1 I
RANK
see
CODE
CARB
tacb
sosi
LMOS
GEM
W-T-E ACID-M
NAPAF
BLS
E/E
CRBl PET!
I I
1 1
96
10200219
9X
I
1 !
10200221
z
I
1
10200222
z
I
I
10200223
z
I
1 1
120
10200224
4X
I
i '
10200225
z
I
i i
49
10200226
z
I
I I
J0200229
6X
I
1 '
55
10200301
3X
i
1 :
10200302
z
I
i |
10200303
z
|
I 1
10200304
z
j
i ! :
1 1
10200306
z
i
III.
10200307
z
! I
1 ! :
186
10200401
26X
I 70Xl
Xi I
1 1 '
10200402
1X
|. 3X
I i
1 1 r
10200403
) X!
I I
i !
10200404
3X
I 10X
I
1 ! i
10200405
z
I
i * i
10200407
X
i
1 1 :
203
10200501
103X
101X
I
i i i
184
10200502
8X
10X
I
I t !
10200503
6X
8X
I
1 i • i
10200504
7X
3X
I
i '
10200505
z
i
! i i i
173
10200601
418X.
62X
i
111.
162
10200602
2X
704X
X
496X]
i
III:!
210
10200603
571X
31SX
i
i
j. : i '
10200604
24X
j
1 i i
157
10200701
193X
X
i
t
1 i !
64
10200704
2X
i
1 i 1
10200707
1X
i
1 ' 1 i
10200710
4X
i
i
1 1 1
141
10200799
56X
X
i
1 1 t
10200802
3X
i
1 1 I
10200804
z
i
1 I ' 1
6
10200901
4X
17X
;
! 1 i
20
10200902
18X
X
15X
1
i i 1
170
10200903
12X
SIX
1
1 ' '
10200904
4X
i i
10200905
m
! 1
10200906
X
10X
25X
¦ i
129
10200907
X
' 1
10201001
z
I
10201002
4X
25X
1
175
10201101
4X
' t
188
10201201
11X
10X
I
10201202
z
|
166
10201301
12X
6>
-1
10201302
11X
4X
|
10201401
z
• 1
9
10201402
7X
3X
1
10201403
X
1
10201404
2>
1
10300101
X
1
10300102
X
1
13-Sep-93
B-3
Page 2
-------
TEMPORA
.ALLOCAT
ON DAT
SOURdES FOR
SCCs I
. 1
|
1
FREQUENCYC
FOCC
JRRENCE
N THESE DAT
A SOURCB
s*
1
r
see
DATA
PRIORITt
GAP
i i
RANK
see
CODE
CARB
TACB
sosi
LMOS
CEM
W—T—B
ACID— W
NAPAF
BLS
E/E| CRBl PET
1 ! i
10300103
X
1 1 i
10300205
z
1 1 i
10300206
1X
1 1 !
10300207
6X
1 1 )
10300208
13X
! 1 i
195
10300209
7X
1 I !
10300211
2X
; ! ¦ !
10300214
X
1 \ '
10300216
z
1 1 1
10300217
z
I
| ;
10300221
z
1 i 1 • i
10300222
z
: |
10300223
z
I
i
; i ¦
10300224
z
)
1
I I I ! :
10300225
I
i xl
, I I I
i 1 : i
10300226
z
• I f
i ; 1
10300305
z
I
I i ! i
10300306
z
I t
III!
10300307
z
|
1 I ' 1 1
10300309
z
I .
III!
209
10300401
X
I
III!
10300402
z
i
| j i ;
10300403
z
I
1 i i i
10300404
z
I
1 1 I ¦ ! ¦ !
10300501
124X
19X
[
! I ;
183
10300502
IIX
3X
I
1 ' ;
10300503
Six
i
i i
10300504
hx
I
i • i
181
10300601'
25X
X
1X
2X
! I
21 a
10300602
38X
X
79X
X
I
t i
207
10300603
84X
SOX
|
' 1 '
10300701
1X
I
! 1 !
10300799
2X
I
1 i . 1
10300901
X
I
1 1 i
10300902
3X
5X
I
1 1 i
16S
10300903
9X
8X
i
! 1 i
10301001
z
! 1 1
10301002
8X
i
1 i 1
103012Q1
5X
X
1
1 I i
. 10301202
z
|
! I S
10301301
z
1 1
10301302
2X
1 1
10301303
z
i
10500102
z
10500105
5X
X
¦
10500106
72X
30X
• i
10500110
z
10500113
z
10500114
z
¦
10500202
z
10500205
X
1
10500206
26X
6X
1
10500209
z
1
10500210
X
i
10500213
z
1 1
10500214
z
1 1
13-Sep-93
B-4
Page 3
-------
TEMPORA
. ALLOCAT
ON DATA SOURG
ES FOR
sees
I '
I
i' I
i
* i
I
|
FREC
HJENCY Q
FOCC
JRRENCE {N THESE DAT
A SOURCES *
i
l
I
see
DATA
I
PRIORmr
GAP
1
RANK
see
CODE
CARB
TACB
sosl
LMOS
CEM
W-T-E
ACID—M
NAPA Ft
BLS
E/E
ORB
PET
211
20100101
18X
3X
16x
1
194
20100102
9SX
4X
16x
!
134
20100201
115X
11X
9x
* i
145
20100202
207X
4X1
Sx
! !
20100702
1X
! 1
20100801
z
1 1
20100802
z
-
20100901
2X
!
16x
i |
20100902
j
16x
I I i
20101001
z
I
. !¦ i I I i
20101010
z
I
t ! i I ! i j
20101020
z
I
i i !ll!;
20101021
z
i
I I . I i ! i !
j
20101030
z
t
I I I I i ! ;
i
20101031
z
I
• I I ¦ ! I !
i
20190099
z
i I I
i 1
20200101
9X
xi
I I i
I i I
208
20200102
87X
8X
I I I
! • i I
20200103
z
I ! I
J i 1
20200104
X
I -1
j |
113
20200201
401X
X
4X
I I
1 !
109
20200202
2954X
X
12X
I I I
! 1 i
86
20200203
28X
I ! I ! I I l
161
20200204
11X
I I i i i ! I
193
20200301
17X
2X
i I i I i ! ;
164
20200401
8X
] sx!
I
Ill's
20200402
9X
I
i
111!
20200403
z
I
I
I 1 !
20200501
1X
I
I
1 1
20200901
1X
]
- I
i !
20200902
1X
I
• i i
1 i !
20201001
IX
I X
i
I t
20201002
z
]
I
20300101
15X
I X
I
1 1
20300102
z
I
1 i
20300201
90X
7X
I
1 1
20300202
11X
3X
|
I
20300203
z
i
1
|
20300204
z
1
1 I
20300301
2X
l
1 ¦ 1
20301001
1X
1
20301002
z
1
20400101
30X
X
1
20400102
z
20400110
z
20400111
z
L. .
20400112
z
20400201
z
20400202
z
20400301
5X
20400302
z
104
20400401
12X
20400402
8X
X
24100252
X
1
28888801
4X
I
23888802
13X
3X
1
1
13—Sep—93
B-5
Page
-------
TEMPORA
. ALLOCAT
ON DAT,
* SOURC
ES FOR
SCCs |
I 1
I
] i
I
1 1
|
1 l
FREQUENCYC
IF OCCURRENCE tN THESE DAT
A SOURCES? *
! i
I
i 1
see
DATA
I
I' !
PRIORITY
GAP
i
i i
RANK
see
CODE
GARB
TAG Bl
sosi LMOS
CEM
W-T-E
ACID—W
NAPAF
BLS
E/E
CRBl PET.
! i i
28888803
6X
i i
i i ;
30100101
3X
X 1
30100102
X !
34
30100103
13X
X - 1
;
126
30100104
29X
X 1
i
30100105
20X
X i 1 i
30100106
12X
x i i !
30100107
1X
! !
301001 OS
i
X ; l :
30100109
2X
!
X I !
30100180
1X
1
1
x 1 1 ;
30100199
15X
i ! i
x : ;
. 95
30100305
f
It 1 U
1 !
82
30100306
21X
|
(
|x
j i
30100307
i
i
i
X
! i
147
30100308
2X
i
1
X
i • ;
30100309
3X
i
1
X
| t
30100399
14X
I
|
X
! ! :
39
30100501
1X
1
I
| 1 ;
30100502
9X
i
1
1
: ¦ s
53
30100503
69X
i
1
1
1 i i
3
30100504
9X
I
1
1 1 !
124
30100506
25X
I
1
! ¦ ! '
30100507
90X
i
i
1
! •!
30100508
34X
i
I ¦
i : ;
81
30100509
67X
i
I
1 i J
118
301CC599
23X
• 1 I
|
1 ; ;
30100601
41X
i X
1 : ;
S
30100603
9X
I
I ;
30100604
1X
[
1
30100605
z
i
1
i i
I
30100699
8X
1
i !
30100799
40X
i
1 :
) 1
30100801
21X
|
1 ' i
30100802
5X
i
1 i
30100803
2X
1
i
30100804
1X
1
1
30100805
8X
[
I i
30100899
87X
1 1 !
30100901
12X
X
1 1
301 COS 02
5X
X
1 ' i
30100910
X
I ;
30100999
109X
X
1
30101011
z
1
30101012
z
1
30101013
z
1
30101014
z
)
30101015
z
t
30101021
z
|
30101022
z
1
30101023
z
1
30101030
1X
l
30101099
1X
1
30101101
2X
|
X
1
130101198
27X
1 X
X
i
1
30101199
13X
1
* ! 1
13-Sep-93
B-6
Page S
-------
TEMPORAL allocat
ON DAT
SOURCt
ES FOR
SCCs I
• 1 1 . 1
I
1 1 1
' 1 f
I
FREC
3UENCYC
>F OCC
JRRENCE
N THESE DAfi
A SOURCE
S *
1 1 1
I
! t )
see
DATA
I
1 1 S
PRIORm
GAP
I
1 1 i
RANK
see
CODE
CARB
TAC8
SOSI LMOS
GEM
W-T-E
ACID —W
NAPAF
BLS
E/El CR8I PET
I
30101202
9X
I
x ' \ ! :
30101203
1X
1 i I
30101204
X I 1 j
30101205
5X
X 1 i (
30101206
X i | !
30101299
12X
X ] 1 1
98
30101301
3X
X
126
30101302
2X
* i ! ; ;
30101303
X 1 1 } I
30101304
7X
X I ! !
30101399
5X
x 1 i 1
177
30101401
63X
X
5X
i ;
30101402
27X
i
( t i
30101403
5X
I !
30101404
30X
! 1!!
•
30101499
80X
I
1 ! • ¦
i
30101501
IX
i
1 | I
i :
30101502
z
I -
1
t :
30101503
22X
I
i
i ! !
30101505
2X
i
! i
30101599
18X
I X
I
1 I !
30101601
10X
i
i
' i
X
I ! •
30101602
9X
1
i
X
i ' ;
30101503
I
i
* ! ! S !
30101699
1
;
x 1 .1.1 1
30101702
1
I
x ! ! 1 !
30101703
i
x i • ! f !
30101704
1
i !
x ! ! i ¦:
30101705
!
I
i
X 1 i
30101706
i
X
! i 1
30101707
X
* ; ^
30101708
X
! i ;
30101799
1
* J 1
XI J
30101801
205X
1
X
X ! ! :
151
30101802
230X
1
X
x ( i i
30101803
25X
x 1 1 i
30101805
34X
1
i 1 i
111
30101807
414X
I
X
X • ! 1 !
30101808
36X
!
1
x 1 I i
30101809
10X
1
1 1 i
30101810
127X
i
!x I I
30101811
333X
t
X
127
30101812
595X
X
X
I !
30101813
43X
X
X
i
30101814
22X
X
X
1
30101815
106
X
X
30101816
79X
X
X
!
30101817
105X
X
30101818
30X
X
30101819
12X
X
|
30101820
24X
X
1
30101821
113X
X
J
X
|
30101822
1X
|
X
*
30101827
3X
X
X
t
30101832
2X
X
i
30101837
5X
6X
1
X
i
13-Sep-33
B-7
Page:
-------
TEMPORA
L ALLOCAT
ON DAT*
\SOURd
ES FOR
SCCs I
I
j
1
FREQUENCY C
>F OCC
[iRRENCE
N THESE DAT
•A SOURCE
s*
1
|
!
see
DATA
|
t
PRlORm
GAP
[
;
RANK
see
CODE
CARB
TACB
SOS
LMOS
CEM
W-T-E ACID—M
NAPAP
BLS
E/E
CRBl pet
!
1
30101838
1X
1
X
1
30101839
1X
!
X 1 i
30101840
13X
1
I 1 i
X 1 I
30101842
1X
!
X
1 •
30101847
69X
I . X
1
X ' I 1 !
30101849
1X
•I '
1
x i Li
30101852
i
X I- 1 1
30101880
9X
I
X
x ! 1 1
30101861
7X
, I
i
X
x ! 1 i
30101863
19X
I
I
X
x 1 Si
30101854
52X
I
i *
l-x
•X' 1
30101865
17X
]
i i-
i i
X |x 1
30101866
16X
I
i ;
i :
X 1
30101870
sx
i 2X
h '
30101871
1X
|
! 1
i x !
30101872
14X
I
i i
!x i
30101880
5X
I
! 1 i
Ix >
30101881
2X
I
1 i
X !
30101882
|
X ¦
!
30101883
18X
]
t
x ! ! ;
30101884
2X
|
x • ! i - !
30101885
4X
I
I
x 1 1
30101890
41X
1
x ! 1 !
30101891
212X
i
x • !
30101892
51X
I xj
I ;
30101893
74X
I- 1 x ! '
,
30101894
52X
1
1 1 V i
1 i x I
)
185
30101899
493X
3Xi
: I l ix i
76
30101901
6X
i
i
1
|
j
30101902
z
i
1
»
30101904
z
i
• i
1
30101905
z
30101906
z
l
1
;
30101907
z
1 1
30102001
z
i
1 I
i I
30102002
z
1
1 i .
30102003
2X
1
i I
30102004
z
! !
30102005
2
i
30102099
z
1
30102101
z
¦ 1
30102102
z
I
30102103
z
30102104
z
30102105
z
301025 06
z
30102107
z
30102108
z
1
30102110
z
• 1
30102111
z
30102112
z
30102113
z
30102114
z
30102120
z
30102199
3X
30102201
13X
X
13—Sap—93
B-8
Page ~e
-------
TEMPORA
_ ALLOCAT
ON DAT
* SOURC
ES FOR
SCCs
|
1 t 1
I
1 1 1
FREQUENCY C!
(FOCC
JRRENCE EN THESE DAT
A SOURCE
S*
1 1 1
• |
k 1 !
see
DATA
j
i : ;
PRIORITY
GAP
|
5 ! 1
RANH
see
CODE
CARB
TACB
SOS LMOS
OEM
W-T-E
ACID-IV
NAPAF
BLS
E/El CRBl PET
!
j i ;
30102301
5X -
|
X
' • ;
30102304
3X
|' "
X
30102306
I
Ix
? ]
30102308
11X
|
1 X
i i
30102310
|
Ix
» .1 *
! 1 «
30102312
|
|x
i !
30102314
|
! X I • ' i
30102316
I
Ix I ! i I
30102318
j
Ix t i i
30102319
1X
|
I
Ix 1 ; . :
30102320
6X
I
I ' i
Ix 1 !
30102321
8X
I,
I I
1X 1 i ;
30102322
2X
I
I I '
i I
Ix i ;
30102399
2X
!
t I
fx I i
30102401
3X
i
I I
1 IX :
30102402
i
t
j
X
30102405
|
j
X i
30102410
|
|
x : ! i
30102414
3X
.
|
x t ; ,
3010241S
1X
i I
i
X ! j :
30102416
|
x ! ! ¦
30102421
I
|
x : J
30102422
| ¦
|
)
x J i !
30102423
1X
I 4XI
]
1
X , j
30102424
5X
I
I
1 1 * i i
30102425
I
1 Ix 1 1.
30102426
|
|
I
x ; ;
30102427
2X
]
1
1
x ; 1 1
30102499
3X
|
!
1
x i 1 ;
30102501
z
]
1
S 1 :
32
30102505
z
I
! i ;
30102599
z
I
! 1 i
92
30102601
61X
1 1 1
30102602
4X
! 1 i
30102608
4X
l 1 !
30102609
127X
! 1 1
30102610
z
i
i : i
30102611
z
i
; 1 !
30102612
1X
i
1 I
30102613
4X
i
i 1 i
30102614
12X
i
! |
30102615
6X
s
1
30102616
z
1
30102617
51X
I
:
30102625
10X
1 ¦
30102630
z
i
140
30102699
282X
| |
30102701
Z
j j
30102704
2X
I
30102705
1X
|
30102706
2X
X
1
!
30102707
Z
1
30102708
z
30102709
6X
30102710
z -
30102711
z
13-Sep-93 Pages
B-9 .
-------
TEMPORA
. ALLOCAT
ON DAT)
* SOURC
ES FOR
sees I - !'
1
i 1 i
I I
1
i
FREQUENCY OF OCC
JRRENCE IN THESE DAI
A SOURCBS*
j
|
1
sea
DATA
I
1
!
PRIORm
GAP
I
|
! 1
RANK
see
CODE
CARB
TACB
sosl LMOS
CEM
W-T-E ACID—W
NAPAri
BLS
e/E
CR81 PETi
j
i .
30102712
2X
I
1
t
i i i
30102713
1X
|
i ;
\ : i
30102714
z.
|
! ' !
30102717
1X
I
! i !
i ! !
30102718
Z
I
1 ! i
30102720
I.
1
1 : !
30102721
z
|
|
1 i !
30102722
2
|
i
)
1 1 i
30102723
Z
I
' [
|
j ; ;
30102724
Z
l
t
1 i : : i
30102725
z.
I
I
! 1 ;
30102727
z
j
1 1 II-: ! 1
30102728
z
i
is I i !
30102729
2
;
I ! i i . i ' 1
30102730
z
I
1 ! ! I i !
30102801
1X
i
1 !'•(•!! :
30102803
z
i I ! i I I I
30102804
z
I I ! i . I • ! ! ! :
30102805
z
| |
i 1 1 ! ' . !
30102806
I
I I
i
II : .
30102807
z
I I
i
[
i : i
30102820
z
|
t
|
1 '
30102821
z
!
I I i !
30102822
z
) «
1
30102823
z
i
i
|1 ! i
30102824
z
!
! ! i ; . 1
30102825
z
; i
~ i
}
i i i •
30102903
z
t
1 -1. ! : i
301Q2S04
z
l
¦ 1
iiit
30102905
z
i
I
ill!
30102906
z
t
1
i
! !
30102907
z
i
!
; i i
30102906
Z
¦ i
!
30102920
Z
i
1
i i ¦ i
30102921
z
I
1 i i
30102922
z
1
1
! I
30102923
2X
|
i i
30102924
z
1
I
1 !
30102925
z
1
i
i i t
30103001
IX
3X
1
i - !
30103002
7X
! 1 !
30103003
5X
1 1 I
30103004
1X
! ! 1
30103020
3X
i
30103021
z
• i
30103022
z
i
30103023
z
i
30103024
2X
t
30103025
z
I
30103099
3X
i
66
30103101
X
1
4
30103102
X
1
30103103
X
j
30103104
X
1
30103105
i
1
X
1
30103180
i
1
X
1
13-Sep—93
B-10
Page 9
-------
TEMPORA
- ALLOCAT
ON DAT
\SOURC
ES FOR
SCCs I
|
1
1 1
|
I
]
1 i
FREQUENCY Cj
FOCC
JRRENCE IN THESE DAT
A SOURCES *
1 1
1 ! l
SCO
DATA
1 i
PRIORITil
GAP
I i :
RAN*
see
CODE
CARB
TACB
SOSj
LMOS
CEM
W-T-E
ACID—V
NAPAF
BLS
E/E
CRBI PET!
1
30103199
1X
|
X i i !
30103201
16X
)
x ! 1 •
30103202
iex
I
X ^ [
30103203
3X
1 Ix ! i
30103204
42X
! i X 1 i |
30103299
25X
|
1 Ix ; ! :
30103301
|
¦ 1 1 * i ¦
30103311
13X
)
1 IX i i
30103312
18X
i
1 Ix ;
30103399
201X
I
1 Ix i i
30103402
ix ¦
i
1 i X
30103403
[
!
I t x !
30103404
!
1
i ! x
30103405
I
i |
i ix
30103406
2X
I
i i
i . ! x
30103410
5X
j
• ! Ix
30103411
3X
1 Ix
30103412
6X
1 1 i • Ix i 1
30103414
23X
1 i i Ix ;
30103415
10X
1
! Ix : i
30103420
l
t
1 Ix : :
30103425
1
1 ix
30103430
i
1
¦| ix :
30103435
i |
! ix :
30103499
107X
!
i i X
30103501
2
]
¦ 1 !
30103502
Z
I
I | i : i
46
30103503
z
|
I ! ¦ . ' :
30103506
z
|
| | : ; i
30103507
2
i
1 ! ; '
30103510
1X
i
I
lll:i
301OSS 15
.z
I '
|
111!.
30103520
z
I
t
1
; s :
30103550
Z
1
i i ¦
30103551
Z
I
1
1 !
30103552
z
1
1
1 ¦ :
30103553
1X
1
i
i
30103554
2X
|
|
167
30103599
11X
i
30103801
3X
1 I
i
30103901
10X
|
! 1
30103902
z
1
1 !
30103903
1X
1
1 i
I i
30104001
2
1
!
30104002
1X
i
i
30104003
IX
1
30104004
1X
1
30104005
Z
• 1
30104006
2X
1
30104007
1X
!
30104008
z
1
30104009
z
1 1
30104010
z
1 1
30104011
z
1 1
•
30104012
z
1 1
30104013
z
1 1
13— Sep—93
B-l 1
Page 10
-------
TEMPORA
_ ALLOCAT
ON DATA SOURC
ES FOR
SCCs
|
1
1 i
J
i i
FREC
2UENCYC
IFOCC
JRRENCE W THESE DAT!
A SOURCes *
1 1
|
i !
SCO
DATA
1
j | i
pRioRrn
GAP
1
¦ I
i
RANM
see
CODE
GARB
tacb|
SOS
LMOS
OEM
W-T-E ACID—W
NAPAP
BLS
E/El CRB
PET.
1
1 ! !
30104101
Z
|
t 1 i
30104102
2
1
t 1
30104103
Z
i
1 ! !
30104104
z
1
I r
1 i
30104199
z
1
! - ! 1
30104201
1
x lx 1 1
30104202
19X
|
x Ix i j
30104203
X |X 1 i
30104204
i |
x 1 x ! !
30104301
! i
i
x ix ;
30104501
26X
| i
1
! i
30105001
z
! 1
[
1 1 :
30106001
X! I
f i ¦ !
30106002
2X
i
1
i t ! 1
44
30106003
Z
j
! ! : : :
30106004
z
|
i 1 ! ;
30106005
z
|
1 i ! ! :
30106006
z
i
t
l 1 1 I
30106007
z
|
' 1 i i
30106008
z
|
1
1
30106009
1X
i t
1
1
30106010
5X
1 ;
1 1 !
30106011
z
i
|
i
30106012
z
1
97
30106099
20X
| |
1
i
30107001
24X
1 • 1
1
j
\
:
30107002
121X
3X
I
i
:
30109101
8X
i !
1 ' '
X 1
30109105
25X
1 ix i
,
30109110
X i I - i
30109151
4X
I
X i } !
30109152
4X
1
x i ! • i
30109153
3X
|
X ! 1
30109154
1X
1
X
I '
30109180
5X
X
1
X
1 i
30109199
14X
1
X
1 ' 1
30110002
z
;
30110003
z
i
1 i
30110004
z
1 i
30110005
2
1
! i
30110080
2X
1
\ ;
1
30110099
10X
i
30111103
z
I
1
30111199
2
i
30111201
1X
j
X
30111202
4X
X
30111299
X
30111301
2'
30111401
X
30112001
X
x
30112002
7X
X
X
30112005
14X
X
X
30112006
2X
X
X
30112007
2X
X
X
!
30112011
22X
X
i
30112012
X
i
13-Sep-93
B-12
Pag# 11
-------
TEMPORAL allocat
ON DAT;
\ SOURC
ES FOR
SCCs I
|
|
j
FREQUENCY OF OCCl
JRRENCE 2N THESE DATA SOURCE
S *
1
see
DATA
1
PRIORITY
GAP
1
RANK
see
CODE
CARS
TACB
SOS
LMOS
OEM
W-T-E
ACID —M
NAPAF
BLS
E/El CRBi PET
I 1
30112013
X | j
30112014
6X
X j 1
"30112017
X 1 I
30112021
34X
X 1 1
30112031
Z
1 i
30112032
z
1 f
30112033
2
1 i
30112034
Z
! !
30112037
z
1 '
30112099
15X
I
' 1 i
30112199
z
I
l
• J
30112401
|
j
X 1 i
30112402
|
t
X i '
30112403
!
i i i ! I x :
30112404
|
I I" II*! !
30112405
i I !
x ;
30112406
I I
X i !
30112407
I I
X j i ;
30112480
1X
I
X 1 i
30112501
4X
I
X 1 i
30112502
4X
I
x ! !
30112504
)
x I i •
30112505
I •
X ! i 1
30112S06
! I
x 1 !
30112509
7X
t I !'
x ! !
30112510
28X
I i
X 1 i
30112511
2X
! I
1 I.
30112512
¦
! I
X i '
30112514
22X
I I
x 1 i
30112515
I
x 1 1
30112520
I
1 i 1
30112521
1X
X .1 !
30112522
1X
X
1
30112524
17X
X
1 ¦
30112525
2X
X
1
30112526
%
1
30112527
X
1
30112528
X
1
30112529
X
1
30112530
%
30112531
X
-
30112532
X
30112533
X
30112534
11X
X
30112535
X
30112540
17X
X
30112541
7X
X
30112542
X
30112543
X
30112544
IX
X
30112545
X
30112546
X
30112547
3X
X
30112550
17X
X
30112551
X
30112552
X 1 1
t3-Sep-93
B-13
-------
TEMPORA
_ ALLOCAT
ON DAT
\SOURC
ES FOR
SCCs
I
j
II I I
J
i i i i
FRE<
2UENCY OF OCC
JRRENCE
N THESE DAT
A SOURCES * i I I I
111!
see
DATA
I I ! 1
PHioRrri
GAP
i 1 <
1 1 1
RAN*
see
CODE
GARB
TACB
SOS
LMOS
CEM
W-T-E
ACID—M
NAPAF
BLS
E/Ei CRBi PET
1 ! !
30112553
x I ! j
30112555
x 1 ! ;
30112599
MX •
x ! 1 !
30112699
x . 1 I ^
30112701
9X
X
i
30112702
2X
X
i
30112703
1X
X
i <
i
30112720
x I j
30112730
X 1 i
30112740
2X
X ! i :
30112780
SX
! 1
30113004
3X
x | ! ;
30113005
I
x 1 !
30113201
12X |
I I i ix : =
30113205
1
!. I' | i x i i
30113210
! 4X I
li i I
X i
30113221
14X
il i i
X i i
30113222
i i
! I*. ¦ i
30113223
2X
I I * ' ;
30113224
5X
1 x | I <
30113227
26X
Ix i 1
178
30113299
244X
x j 1 ;
30113301
40X
x ! I :
30113302
x ! '
30113303
X ' i I !
30113380
-
(X ! 1 i
30113701
X
i
!x ! !
30113710
3X
!
Ix'l | • i
30113799
102X
[
1 ; 1 i
30114001
12X
i
xl 1 ;
30114002
i
X 1 1 !
30114003
X
! !
30114004
X
1 i
30114005
16X
X
i i
30115201
23X
x ! 1 i
30115301
31X
x 1 I i
30115310
X
i
30115311
2X
X ! |
30115312
X I !
30115320
X 1 1
30115321
'x | 1 !
30115322
2X
X • ]
30115380
17X
X
1
30115601
5X
30115602
Z
30115603
z
30115604
z
30115605
z
30115606
z
30115607
z
30115680
z
30115701
2X
30115702
z
1
30115703
z
30115704
8X
30115780
5X
1
13—°Sep—93
B-14
Page 13
-------
TEMPORA
_ ALIOCAT
ON DAT]
\SOURd
ES FOR
SCCs
I
i I
: I •
I
i t i
FREfl
aUENCYOF occ
JRRENCE BM THESE DATlA SOURCES *
I t I
i
i 1
i i 1
see
DATA
f
i i !
PRlORfTl
GAP
|
i i !
RANK
see
CODE
CARB
TAG H
SOS
LMOS
CEM
W-T-E ACID- M NAPAF
BLS
E/Ej CRBi PET!
s ]
30115801
1X
X
72
30115801
1X
30115802
1X
X
> j
30115803
1X
X
30115821
i
X
;
30115822
7X
|X
30115880
I
x 1 I i
30116701
9X
|
X ! i i
30116702
2X
|
X j
30116703
3X
i-
|
X ! !
30116704
8X
! i I
X < -
30116780
5X
! ! I Ix ! r
30116799
85X
I
I
l
I | x : :
30116901
29X
I
i
! i : i
30116902
Z
I
«
I . I . :
30116903
2
I
i
I
!
30116904
Z
I I
!
30116905
3X
I
|
j 1 '
30116906
2X
I
|
5 ;
> i .
30116980
18X
|
1 I '
30117401
40X
|
X « 1 :
30117402
2X
I
x i 1
30117410
I
x ! I :
30117411
I
I
X ! !
30117421
8X
x ,i !
30117480
10X
X . i ! :
30117601
x |x i ! !
30117610
X
X i !
30117611
|
X
x i • !
30117612
I
X
x ¦ ' :
30117613
X
X I t =
30117614
X
x ! ! :-
30117615
X
x ! !
30117616
X
x ' i
30117617
9X
X
X ! 1
30117618
X
x 1 i i
30117630
X
x ! 1
30117631
X
x i i
30117632
X
x 1 )
30117633
i
X
X I- 1
30117634
X
x i ;
30117680
4X
|
X
X
30118101
63X
X
I
I
30118102
z
|
30118103
.
1X
|
30118104
z
1
30118105
z
1
30118106
z
1
30118107
z
1
30118108
z
I
30118109
z
1
30118110
51X
1
30118180
1X
1
30119001
12X
X
1
30119002
X
1
30119003
X
1
13-Sep-93
B-15
Page 14
-------
TEMPORA
- ALLOCAT
ON DAT]
\ SOURC
ES FOR
SCCs
1 ) 1
1 I
FREC
2UENCYC
>F QCcI
JRRENCE
N THESE DAT]
A SOURCd
s *
see
DATA
1 1
PRJORIT*
GAP
j
RANK
see
CODE
CARB
TAC8
SOS
LMOS
GEM
W-T-E ACID—M
NAPAF
BIS
e/e! cnal peh
1 i
30119004
X ' i
1
30119010
X 1
i
30119011
|
x ! 1 I
30119012
|
x ! i :
30119013
3X
(
X ' i 1 '
30119014
1X
i
s
X !
30119080
5X
i
X i
30119501
2
I
!
!
30113502
Z
l
:
30119503
z
i
30119504
z
i i ;
30119505
z
I
i
i
! i '
30119506
z
j )
! ! ! 1
30119580
hx
I
i j » i 5
77
30119701
170X
1
I 1 x ;
'
67
30119705
12X
1
1 Ix i
:
30119706
1
i ]x :
30119707
2X
1
i |x ;
30119708
3X
i
i
I ! x ,
i
30119709
17X
I
1 1X ! 1
30119710
15X
1
j x ' !
30119741
38X
1
1 * ! ;
30119742
SOX
1
ix ! j i
30119743
4X
1
Ix : ! ,
30119744
14X
ix : 1 |
30119745
34X
1 x i ; ;
30119749
S4X
)
1 ix • :
30119799
146X
1
i
i 1 x s !
30120201
sex
|
1
i j_ ;
30120202
z
I
I
t
; , |
30120203
z
! I i
30120204
z
' ! i i
30120205
1X
1 1 •;
30120206
2X
i 1
30120210
z
1 1
30120211
z
1 1
30120280
6X
1 i i
30120501
3X
x ! |
30120502
11X
X 1 t
30120503
2X
X 1 1
30120504
x i i
30120505
X i |
30120506
1 !
30120507
X 1 1
30120508
X 1 !
30120509
1
30120520
1
30120521
3X
X 1
30120522
X 1
30120523
X 1
30120524
X 1
30120525
X 1
30120526
X
30120527
X
30120528
X 1
30120529
X I
13-Sep~93
B-16
Page 1
-------
TEMPORA
. ALLOCAT
ON DAT
Si SOURC
ES FOR
SCCs
|
1 ;
I
1 I
*
FREQUENCY 0
FOCC
JRRENCEC
N THESE DAT
A SOURCE
s *
1 i
! ;
see
DATA
! t
PRioRrn
GAP
RANK
see
CODE
CARS
TACB
sosl
LMOS
CEM
W-T-E
AC1D-M
NAPAFI
BLS
E/E
CRBl PET:
j
ill:
30120530
t
I* I i
30120531
1 x ! ¦ '
30120532
Ix ! ! !
30120540
Ix i ! i
30120541
IX
1 x I \
30120542
XI ! ;
30120543
x 1 i !
30120544
|
_* i L . i
30120545
i
i
\ x i •
30120546
i
1 iv *
|X 1 !
30120547
!
i ix i ¦ j ;
30120548
1 ;
1 Ix i ;
30120549
i . 1 !
! \x • :
130120550
1
i 1 t i x :
30120551
! ! i . !* ! . ¦
30120552
! '1 ! ix i !
30120553
s I i . !x i • •
30120554
i
i x ! - i
30120555
I i x i !
30120580
1
1 Ix ! ; :
30120601
79X
i
j | i I
30120602
7
1 1
¦ 1 S i
30120603
Z
| |
1 1 f i
30120680
26X
1 1
I i
30121001
22X
1 i 1 1 U ! ;
30121002
14X
r
X
30121003
1 1 i tx ! i
30121004
7X
1 ! i ix !
30121005
9X
i I ! ix ! : :
3012100S
10X
i ! ; i x :
30121007
7X
I
1 Ix I i !
30121008
47X
i
i
1
i
X 1 : :
30121009
3X
1
X i ! ;
30121010
14X
1
x i 1 i
30121080
11X
!
x ! 1 i
30121101
1
x j 1 •
30121102
1
X ( ! I
30121103
1
x 1 i i
30121104
1
x I i t
30121121
l
X '1 i !
30121122
1
x i i ;
30121123
i
X
1 ¦ '¦
30121124
x
' !
30121125
X
1 !
30121180
X
1 i
30125001
SOX
X
1 1
30125002
4X
X
1 1
30125003
2X
X
1 1
30125004
3X
X
t 1
30125005
36X
X
X
X 1 1
30125010
X
X
X i 1
30125015
SX
X
X
x l !
30125020
152X
X
X
Ix I 1
30125025
%
X
IX 1 !
30125099
111X
%
X
IX I 1
30125101
113X
X 1 1 1
13~Sep-33
-cc
B-17
Page M
-------
TEMPO HA
. ALLOCAT
ON dat!
\ SOURC
ES FOR
sees
r
I 1 !
j
1 t 1
FREC
JUENCY OF OCC
JRRENCE IN THESE DAT
A SOURCE
s *
i i !
¦!
' i
i 1 i
see
DATA
.1
1 ! :
PBIORm
GAP
1
i 1 :
RANK
see
CODE
GARB
TACB
SOS
LMOS
CEM
W-T-S ACID-M
NAPAF
BIS
E/Ei CRB PET,
. j
| . | |
30125102
i
1
1
x ! :¦ 1
30125103
1
i
x j i i
30125104
5X
j
1
x 1 j 1
30125180
12X
1
x 1 1 !
30125201
22X
>d
1
x ! I I
30125301
23X
1
1
X [ [ ;
30125302
7X
1
1
X ! i i
30125305
1X
1
|
x i i !
30125306
1X
5
1 .
1
|
x ; !
30125315
2X
I
1
j
X
30125316
I
i i
) i
1 Ix ;
30125325
IX
\
i
i i [ ! j x :
30125326
4X
1
1 " 1 i ! j x
30125380
17X
|
i ! i i !x
30125401
7X
1
1
1 ! 1 i ix
47
30125405
47X
1
II I ! ix
30125406
4X
1
1 ! i 1 1 x ' :
30125407
3X
i
1 1 1 i |x ; : :
30125408
1X
i
1 ! ! ! x
30125409
8X
I
1 1 i Ix • i ;
30125410
I
1 I i ix i ! :
30125411
10X
[ • 1 i Ix .
30125412
1 ¦ i i 1 x ! i ;
30125413
1 1 1 ]x ! i »
30125415
sox I I
1 1 ! ix i i i
30125416
]
! 1 i i ¦ ix ; i
30125417
i i 1 1 X 1
30125418
! 1 i i* |
30125420
9X
i 1 ( ix
30125499
40X
t
! | X ! I
30125801
14X
1
x ! ! i
30125802
13X
I
X 1 ! !
30125803
1
X i ! !
30125805
16X
x ] ! !
30125806
1
x ! s
30125807
1
x ! i
30125810
17X
i
x 1 i i
30125815
8X
• 1
X i i ¦ 1
30125816
|
x i ! ;
30125317
2X
1
x ! i
30125880
19X
X 1 ' 1
30125899
83X
X 1 !
30130101
1X
x 1 !
30130102
x 1 ! i
30130103
X 1 i 1
30130104
x i i
30130105
X
i
30130106
X
i
30130107
X
i
30130108
X
i
30130110
X
30130180
X
30130201
10X
X
30130202
X
;
30130203'
14X
X 1 I
30130260
X i 1
13-Sep-93
B-18
Page 17
-------
TEMPORA
ALLOCAT
ON DAT
\sound
ES FOR
3 O 0 •>
t
i
} |
' 1
! j
FREQUENCY 0
FOCCt
JRRENCE H
N THESE DAT.
A SOURCE
5 *
. j
¦ 1' i
see
DATA
i i =
PRIORITY
GAP
|
! ' i
< 1
RANK
see
CODE
GARB
TAC8
sosl
LMDS
CEM
W-T-E ACID—M
NAPAR
8LS
E/E
CAES! PEP
1 ! ;
30130301
x I 1 i
30130302
1
x i 1 ;
30130303
i
* 1 1 i
30130304
1
x 1 ! 1
30130305
1
X j i !
30130380
ex
i
1
1
X j !
30130401
2X
| !x ! 1
30130402
5X
|
x i | i
30130403
|
| '
x 1 ! :
30130404
I I i
t
x j 1
30130405
I I
!
30130480
3X
I
I - ! ! !x ! ;
30130501
5X
I
)
I ' I i U : :
3Q130502
13X
!
! i ! ! : x '
30130503
I : ! i ix ! ;
30130504
1X
' !' ! i ¦ ! X ! :
30130505
1X
I I ! I |x ! !
30130580
3X
• i i i v ' 1
J ; i | 1*
30140101
z
i I
i ! :
30140102
z
| 1
i i s
30140103
2
! I
i | •
143
30180001
682X
! 1
x i j
30181001
33X
1 1
X 1 i I
30182001
2
130X
• 1
X { 1 1
30182002
77X
i ' i 1 - j x ! ;
30182003
13X
1
U 1 1 '
30183001
2
2889X
|x . ! ¦ \
30184001
120X
1
x ¦ 1 1 i
30187001
22X
i
X
1 1 ;
1 t !
30187002
20X
i
t
X
¦1 1 1
30187003
z
I '• :
30187004
2
.1 1 !
30187005
5X
X
i 1
30187006
!
X
\ 1 .!
30187007
3X
I
X
1 1 i
30187008
3X
)
X
30187009
63X
!
X
i ! • i
30187010
23X
1
X
! !
30187097
245X
I
i
1 1 !
30187098
163X
1
• ' - ;
1 ! i
30187501
Z
)
I 1
30187502
z
t
30187597
1SX
1
30187598
19X
j
30188501
36X
X
30188502
z
30188503
33X
X
30188504
z
30188505'
4X
30188599
48X
30188801
866X
2>
30188802
102X
30188803
38X
!
30188804
33X
1
30188805
22X
|
I
30190001
1X
' |
13-Sep-93
B-19
Page 1.
-------
TEMPORA
-ALLOCAT
ON DAT,
\ SOURC
ES FORI
SCCs
1 !
1 1 !
FREQUENCY C
FOCCi
JRRENCE IN THESE DAT
A SOURCE
S *
1 !
. 1 i
I
! ! :
see
DATA
-
( !
pRioRrrr
GAP
I i
RANK
see
CODE
GARB
TACB
SOS
LMOS
CEM
w-T-d ACID-M
NAPAF
BLS
E/E
CflSI PET
|
1 |
30190002
6X
i 1 i
205
30190003
511X
I 1 5
! 1 t
30190004
SOX
i !
30190011
z
! | !
3019(5012
2X
1 ! 1
30190013
59X
1 ! i
30190014
8X
1 ! !
30190099
376X
1 ! i
30199998
301X
i
i ; ,
30199999
1280X
> 16X
30200101
2
1X
!
x 1 !
30200102
4X
!
X ' '
30200103
IX
1
1 i
X ' 1
30200104
2X
! ! i • i i x ;
30200199
! 1
! ' 1 1
X i 1
302C0201
24X
i
! 1 Ix •
X !
30200202
46X
I
f
Ix
!X : :
30200203
!
i
IX
,X • ;
30200299
44X
|
t
X
x j :
30200301
1
t
1
X
ix i 1
30200401
|
,
ix
30200402
9X
)
Ix 1 i
30200403
8X
1 '
i
ix 1
30200404
25X
i
1
\ x ! i
30200410
100X
1
1
IX !
30200499
6X
1
1
t 1 !
30200501
3X
3X
, 1
i
i ! i
30200502
2X
I
1 i :
30200503
1X
1
1
t ! !
30200504
16X
1
)
1 i :
30200505
97X
1
i ) i
30200506
22X
1
1
1 i ;
30200507
16X
1
I
i 1 ;
30200508
24X
|
1 1 !
30200509
1X
1
1 1 ;
30200510
z
1
i i !
30200511
2
1
!
1 i 1
30200512
Z
1
1
1 i 1
30200601
13X
- 1
|
t | ^
30200602
3X
1
t
' 1 1 i
30200603
21X
1
1 1 1
30200604
35X
1 1
i : :
30200605
4ax
1 I i
30200606
38X
' i
1 ! i
30200607
11X
| i
30200608
20X
i
30200609
2X
'
;
30200610
z
30200611
3X
30200701
z
30200702
1X
30200703
1X
X
X
30200704
3X
X
30200705
X
X
i
30200711
X
30200712
X
i
13-Sep—93
B-20
Page 1 $
-------
TEMPORAL allocat
ON DAT
SOURd
ES FOR
SCCs
FREC
3UENCY C
IFOCC
JRRENCE tN THESE DAI
A SOURCE
s *
j
see
DATA
I
1
PRioRrn
<3AP
!
RANK
see
CODE
CARB
TACE
SOS
LMOS
GEM
W-T-S
ACID —W
NAPAF
BLS
E/E
CRBl PE71
1 1 !
30200713
X 1 1
30200714
X 1 ' 1 !
30200721
X
* . ! 1 1
30200722
X
X
• i I
30200723
X
X '
I i
30200724
X
X
1 I
30200730
X
X
1 1
30200731
7X
X
X i 1 !
30200732
3X
X
X 1 1 j
30200733
11X
X
XI |t
30200734
14X
X
v ( 1 !
X 1 ! ;
30200741
6X
I *
X IX j !
30200742
18X
X
x j X i
30200743
11X
x Ix i.x i ;
30200744
5X
x | x | X I :
30200745
5X
X i X IX i ;
30200751
IX
X
Ix 1
30200752
I
• tx
ix 1
30200753
2X
X
IX ! i
30200754
5X
X
!x i !
30200755
X
!x ! 1
30200756
8X
X
!x ! 4
30200760
3X
X
x [X j 1
30200771
21X
X
X 1 ! i
30200772
14X
I
X
X i 1 1
30200773
34X
I 1
X
X ! ! !
3C200774
38X
1
X
x : i !
30200781
)
x !x ! i
30200782
1
x ix 1 i
30200783
)
X ! X
i
302007S4
i
x iX
i
30200785
1
x IX
1
30200786
3X
1
x IX
t
30200787
x |X
30200788
x !x
1
30200789
X
X
x I 1
30200790
1X
X
X
30200791
X
X
30200799
11X
1
30200801
2X
X ! !
30200802
39X
X 1 1
30200803
18X
X I 1
30200804
36X
X
1
30200805
33X
X
t
30200806
18X
X
1
30200815
X
30200816
X
2X
X
30200899
34X
X
X
30200901
56X
X
X
30200902
9X
X
X
X
30200903
19X
X
X
30200904
3X
X
X
30200905
z
30200998
33X
30200999
31X
30201001
Z
13—Sep-93 ^ Page 2!
-------
TEMPOBA
ALLOCATION DAT7
^ SOURC
ES FOR
sees
1 i
1 1
FREC
1UENCY 0
F OCC
JRRENCE CN THESE DAT4
A SOURCB
S *
1 i
1 !
see
DATA
1
I
PRIORm
GAP
| 1
RANK
see
CODE
CARB
TACB
SOS
LMOSl
CEM
W-T-B ACID—M
NAPAF1
BLS
E/E
CRBi PETj
j
i ;
30201002
Z
|
| 1
i i
38
30201003
X
'
30201004
|
X
1 ' t
30201099
-
|
X
| 1
30201103
I
X
1 1
30201104
1X
j
X
! 1
30201105
X
30201106
X
! »
30201199
1X
I
|
X
• i j :
30201201
z
I
i I i
30201202
z
1
i ! !
30201203
z
I
l
I i
1
! ;
30201204
2
!
)
i
30201205
z
I
j
i
30201206
j
1
i
: l :
30201299
1X
|
I
i
i
i ;
30201301
17X
] 23X
i
i
1 X
x ! X i
30201401
4X
I
1
1
i ¦¦ :
30201501
1*
x 1 ! !
30201599
17X
' i *
x .! !
30201601
>4
1
i
X 1 I
30201699
t
i
x ! 1
30201701
2
1
I
1 ; I
1 '
|30201702
2
I
1
i
1 :
130201703
Z
I
i
i
1 ' ¦
30201704
Z
I
I
i
1 ! ¦
30201799
61X
I
I
i
! i '
30201899
5X
X
|
t
x i Si
30201906
z
|
1
i
IX i i
30201907
2X
62X
1
1
x i i
30201908
z
I I
X 1 !
30201909
z
j
! i !
30201912
X
.
i i !
30201916
z
Ix ! !
30201917
3X
Ix ! 1
30201918
ax
lx 1 !
30201919
z
ix 1 i
30201920
z
Ix 1 !
30201999
115X
1
X ! !
30202001
6X
!
X
X 1 1
30202002
4X
I
X
X ! !
30202101
X
x 1
30202102
X
X 1 .
30202105
X
X 1
30202106
X
X 1
30202201
70X
x I
30202601
9X
X 1
30202801
z
1
f
30203001
4X
3X
X
X
X 1
30203099
>
X
X
X 1
30203103
2X
X
1
30203104
X
1
1
302031 OS
4X
X
I
30203106
4X
X
30203107
2X
%
30203108
11X
X
13-Sep-S3
B-22
Page 21
-------
TEMPORAL ALLOCAT
O
z
o
5
\sound
ES FOR
SCCs
. I
1
FREC
3UENCYq
F OCCt
JRRENCE EN THESE DAT
A source
s*
1
1
!
i
1
scd
DATA
1
i
PRionrnl
GAP
1 ! !
RANK
see
CODE
CARB
TACB
SOS
LMOS
CEM
W—T—E;
ACID—M
NAPAF
BLS
E/Ei CRBI
PET1
i 1 i
30203109
3X
X
1 1 1
30203110
X
} ?
> .
30203111
X
• i r
30203201
'
X
X 1 • 1
30203202
24X
X
X 1 1 1
30203299
8X
X
X i j i
30203399
X
j I ;
30203801
7X
6X
X
X 1 i !
30203801
X
! X j !
30203901
z
1 1 i
30204001
IX
i ! :
30288801
43X
I 3»
I I : i :
30288802
10X
I
! i ' !
30288803
3X
I
1
I I i !
30288804
1X
I
j
i i :
30288805
2
s r • ; ;
30290001
2
'
i I ' ;
I I •
30230002
2
I | ' !
30290003
21X
5Xl
i I i ; :
30294912
Z
i I ) i i.
30299998
29X
2X
1
I j ; i i
30299999
38X
f
1
1 ' ' =
30300001
7X
I '
1
I i i X 1 i
30300002
IX
j
" 1
! ! /x i ;
30300003
I
|
l
Ix- !
40
30300101
t *
l
1
lx
X
X
43
30300102
!
Ix
X
IX '
j
57
30300103
1
ix
X
X
i
30300104
1X
X
iX
i
70
30300105
*
iX
i
30300106
X
X
Ix
I
100-
30300107
X
X
IX
i
30300108
1
X
X
ix
i
30300109
1
1
X
X
ix
1
30300110
I '
X
X
ix
1
30300111
I
X
X
IX
1
30300199
z
I
i i !
30300201
48X
I
X
x ! ! i
75
30300302
I
x .1 1 i
30300303
-
X ! ! !
30300304
X 1 ! !
30300305
1X
X ! I 1
51
30300306
x l ' I !
30300307
2X
X 1 i !
30300308
'
x 1 I 1
30300309
2X
1 1 1
30300310
2X
x ! - 1
30300311
X I 1
30300312
34X
X 1 |
30300313
-
11X
X 1 1
30300314
X 1 1
133
30300315
X
x 1
30300316
z
1
30300399
1X
1
30300401
X
1
30300502
5X
I
X
IX
13-Sep—93 jg "2^ Pag® 2*
-------
TEMPO RA
. ALLOCAT
£
D
Z
o
^ SOURC
ES FOR
SCCs
i i
i
I I
FRE«
3UENCY c
FOCCl
JRRENCE iN THESE DATA SOURCE
s *
see
DATA
I
PRioRrn
GAP
I
I
RANH
see
CODE
GARB
TAGS
SOS
LMOS
CEM
W-T-E AGID-M
NAPAF
BIS
E/E
CRB
PET1
|
| 1
30300503
4X
I
X
X i i
30300504
5X
I
X
X 1 I
30300505
I
X
X ! !
30300506
I
X
X 1 i
3030050?
j
X
X i !
30300508
I
X
x 1 i
30300509
|
X
!x ! !
30300510
1X
I
Ix
Ix I !
30300511
3X
f
I
1 X
IX j .!
30300512
i
I
I
tx
Ix i :
30300513
I
t
!x
ix !
30300514
!
I
tx
x :
30300515
/
I
t
lx j ix ; i
30300516
1X
I
!x ! X
30300517
I
I
X 1 Ix i :
30300518
I
i
X
|X :
30300519
I
X
ix i i
30300521
i
X
X 1 I
30300522
'
X
X 1 1
30300523
[
X
X ( i
30300524
1
1
X
X 1 1
30300525
j
X
IX 1 s
30300526
1
X
Ix I
30300527
i
X
IX I i
30300528
)
X
Ix . 1 ¦ i
30300529
!
X
Ix \
30300530
I
! x
ix i
30300531
!
Ix 1 Ix I t
i 30300532
1
ix
lx ] !
30300533
1
IX
Ix i !
30300534
1
Ix
Ix I i
30300535
i
lx
Ix ( 1
30300539
3K
X
1
Ix
ix ! i
30300601
1
X
I !
30300602
1
X
j I i
30300603
' 1
X
1 1
30300604
i
X
1 i
30300605
I
X
1 i
30300606
!
X
! i
30300607
i
(
X
X 1
30300610
t
X
1
30300611
1
X
c
30300613
1
X
30300614
2X
1
X
30300615
1
X
3030C61S
1
X
30300617
|
X
30300699
2X
X
1
X
30300701
X
30300702
X
30300703
X
X
30300704
X
x
30300799
X
30300801
1X
X
X
x
30300802
X
X
x
30300804
X
X
X
13-Sep—93
B-24
Page 22
-------
TEMPORA
ALLOCAT
ON DAT)
^ SOURC
ES FOR ISGCs I
1 !¦ !
r i
i i
FREQUENCY <3
focc
JRRENCE (N THESE DAT
A SOURCE
s *
1 1
¦
I
i
J
)
see
DATA
|
i I
PRioRrn
GAP
I
I
|
1 1
RANK
see
CODE
CARS
TACB
SOS LMOS
CEM
W-T-E ACID-W
NAPAF
BLS
E/E
CRBl PET]
I
1 1
30300805
X
X
X !
30300808
1X
X
X
X !
30300809
7X
X
X
X | 1
30300811
X
X
X 1 1
35
30300812
3X
X
* Ix I • I
7
30300813
X
x ix 1 !
30300814
|
X
X Ix 1 i
30300815'
I
|x
x ix i j-
30300816
I
!x
x |X i s
30300817
i
i
iX
x IX !
30300818
j
|x
x i X i
30300819
i
Ix
x X
3Q3CG820
1
1X 1X 1X ;
30300821
I * - i x -X ;
30300822
1
! x | x ; X
30300823
1
1 x
x |X !
30300824
1
Ix
x X
30300825
1X
i
|x ix Ix i ;
27
30300826
1
1*
x Ix I ;
30300827
I
Ix
x ix I i
30300831
1X
i
i
tx
x Ix i i
30300832
1X
i
Ix
X IX I ;
30300833
1X
I
Ix
x iX 1 !
30300834
1X
i
Ix
x iX 1 1
30300S41
1
!x 1 x 'X i :
30300842
i ,
Ix )x 1X !
30300899
]
X
1 1 1 1 !
30300901
1
i
Ix' ix IX i
114
30300904
28X
1
I
Ix
x 'X j
30300906
IX
i
i
i
|x
x 1X j I
07
30300907
1X
|
1 t
X
X ix I i
91
30300908
2X
1
X
x 1 x i !
30300910
1X
1
X
x ix | i
30300911
1 ¦
•
X
x Ix i . 1
30300912
46X
X
x IX 1 ' i
30
30300913
X
x ix i i
33
30300914
2X
X
X IX I !
30300915
|
X
x IX 1 1
3030091S
X
x IX t 1
30300917
1X
X
X
X ! !
30300918
X
X
X 1 i
30300919
X
X
X l 1
30300920
X
X
X
1
30300921
X
X
X
30300922
3X
X
X
X
30300923
3X
X
X
X
30300924
1X
X
X
X
30300925
X
X
X
30300931
5X
X
X
X
30300932
3X
-
X
X
X 1
156
30300933
15X
X
X
X i
30300934
11X
2>
X
X
x 1
30300935
2X
X
X
x 1
30300936
X
X
X - i
30300998
4X
X
X
x I
30300999
55X
X
X
X
X 1
13-Sap-93
B-25
Page 24
-------
TEMPO RA
.ALLOCAT
ON DAT)
* SOURC
ES FOR
SCCs _j
I
t 1
I 1
FREd
2UENCY OF OCC
JRRENCE IN THESE DAI
A SOURCE
S *
1 i
i i
see
DATA
1 I i
PRIORITM
GAP
RANK
see
CODE
CARB
TACB
SOS
LMOS
OEM
W-T-i ACID—M
NAPAF
BLS
E/EI CRBI PET,
j i !
30301001 I
X
X
ix i I
30301002
4X
X
X
ix I i
30301003
4X
X
X
ix l i
30301004
X
X
X 1 i
30301005
X
X
ix | I
30301006
X
X
IX j :
30301007
1X
X
X
ix t !
30301008
X
X
ix j ;
30301009
X
X
- ix i i
30301010
X
|
X
: X '
30301011
X
i
x I i X i
30301012
2X
X
i
!
i x t 'X j
30301013
!
X
I
|x [ 'X
30301014
I
X
i
ix | X
30301015
j
X
I | ix t X
30301016
i
i
X.
I
Ix I X
30301017
X
i
X
ix ;
30301018
x ! I
X
; X !
30301019
X
- I
X
¦X \
30301020
X
I
X
!X ¦ :
30301021
X
I
X
ix ! i
30301022
]
X
I
X
:X ! !
30301023
I
X
I
Ix
ix i i
30301024
I
X
I
u
-------
TEMPORAL allocat
ON DAT
Si SOURC
ES FOR
SCCs
• I
I
I
FRES
1UENCY C
IF OCCI
JRRENCE EN THESE DAT
A SOURCE
S *
I
1
see
DATA
]
1
PRIORm
GAP
)
1 .
RANK
see
CODE
GARB
TACB
SOS
LMOS
CEM
W-T-E ACID —V
NAPA R
BLS
E/E
CRBl PET]
1
30302401
X
[
30302402
X
1
30302403
X
1
30302404
X
i 1
30302405
X
1 i
30302406
x i r 5
30302407
X
I i I
30302408
X
30302409
X
I ! !
30302410
I
X
: > f
30302411
IX
I
X
i 1 !
30303002
¦
I
X
IX ! j
30303003
I
j l
X
;X I . !
303030C5
[
] j
X
! X 1
30303006
22X
I
i I ix I ix s I
30303007
3X
I
I I Ix
!X i [
30303008
2SX
I
I I
X
1X 1 i
30303009
35X
I
I
X
IX [ |
30303010
3X
I
j
X
IX j !
30303011
1X
I
i
X
ix i • i
30303012
1X
X
x i j
30303014
3X
I
ix
X ! i
30303015
]
x
IX i 1
30303016
!
X
!x ! !
30303099
16X
!
1 i 1
30388801
3X
I
| | 1
30388602
3X
i
! j i
30388803
2
!
! ! I
30388804
Z
I
i ! 1
30388805
z
i
I
1 j i
30390001
z
!
j • i !
30390002
|
1 i I
30390003
8X
3X
I
1 1
30390004
z
I
i i
30390011
z
I
1 1
30390012
z
t
1 1 i
30390013
z
I
1 1 1
30390014
z
I
I i
30390021
z
I
t i i
30390022
z
i
j |
30390023
z
1 i
30390024
z
1 i
30399999
208X
i 1 .
30400101
14X
2X
X
X
30400102
119X
14X
X
X
30400103
25X
31X
X
X
30400104
X
X
30400105
X
X
30400106
X
X
30400107
X
X
30400108
4X
X
X
30400109
3X
3X
X
X
42
30400110
X
X
30400111
X
X
30400112
1X
2X
X
X
1
30400113
1X
X
X
1
13-Sep-93
B-27
Page 26
-------
TEMPO RA
» ALLOCAT
ON DAT
ftSOURd
ES FOR
SCCs I
. I
|
|
I
FREQUENCY
>F OCC
JRRENCE SN THESE DAT
lA SOURCES *
see
DATA
!
PRIORm
GAP
!
RANK
see
CODE
CAR8
TACE
SOS
LMOS
CEM
W-T-E
ACID-N
NAPAF^
BLS
E/E
CRBt PETi
i 1
30400114
3X
X
X
1 I
30400120
1X
X
x I ! i
30400150
3X
X
x I I !
30400199 '
11X
X
X
x ! ) i
30400204
2
2X
X
I I I
3040020?
X!
X
I \ j
30400209
X
! I i
30400209
X
i ) i
30400210
6X
X
! i !
30400211
|
t
X
i ! i
30400212
i
I
X
! ' I
30400214
1X
j
I
X
j j
30400215
|
X
i 1
30400217
ix
30400219
1X
¦3X
!
j
!x
30400220
I
Ix ! I ! !
30400221
1X
I
i
I* I i ! ;
30400223
I
|x | j ! ' !
30400224
7X
I
I* I I ! i
30400230
1X
|
Ix I j | i
30400231
]
Ix
I 1 t
30400232
I
Ix
• ! ?
I | "
30400233
1X
I
ix
1 \ 1
30400234
I
Ix
1 ! I
30400235
I
I x I ! i ;
30400236
I
Ix i ! i
30400237
I
Ix | i ! i
30400238
1X
t
ix • ) i i ,
30400239
8X
i
X
30400299
9X
X
j
i
112
30400301
13X
21X
X
X
1
30400302
1X
X
X
30400303
22X
10X
X
X
1
30400304
10X
2X
X
X
30400305
ax!
X
X
1
30400310
2X
2X
X
X
36
30400315
X
X
1
30400320
4X
10X
X
X
j
30400325
2X
X
X
I
30400330
X
X
30400331
16X
33X
.[
X
X
30400332
2X
X
I
X
X
30400333
5X
2X
]•
X
X
30400340
5X
X
X
30400341
1X
X
X
30400342
1X
X
X
30400350
48X
X
X
30400351
25X
8X
X
X
30400352
3X
X
X
X
30400353
1X
3X
X
X
30400354
1X
X
X
X 1
30400355
X
X
X I
30400356
7X
X
X I
30400357
X
X 1
30400358
X
X 1
30400360
28X
I
X
x 1
J3-Sep-93
B-28
Page 2
-------
TEMPO RA
. ALLOCAT
ON DAT]
^ SOURC
ES FOR
sees
I ¦
I
I
v
|
• I i
FREC
2UENCYC
>F OCcI
JRRENCE
N THESE DAI
A SOURCES *
I
|
J !
see
DATA
|
I
1
PRIORm
GAP
|
!
RANK
see
CODE
CARB
TAC8
SOS
LMOsI
GEM
W-T-E
ACID—M
NAPAfl BIS
e/e
CRBl PET;
I ' i
30400370
1X
2X
X Ix ! !
30400371
1X
4X
x |x i i
30400398
xl
X IX ! I
30400399
67X
XI
X ix i i
30400401
5X
X
X i ;
30400402
9X
X
x i !
158
30400403
9X
X
ix I • l
30400404
IX
|
X
Ix I I
30400405
I
X
ix I
30400406
[
X
Ix !
30400407
12X
i I
X
ix
30400408
9X
I !
X
ix :
30400409
1X
! I
x I :X
30400410
4X
I !
x I . (X :
30400411
3X
I ' ! I*
'X
30400412
! I !x
X
30400413
1X
I I Ix I IX !
30400414
8X
I I Ix [ .:X '
30400499
19X
I- I |x I !x •
30400501
5X
I I Ix
i - ;
30400502
z
I I I
i i
30400503
5X
I I I
I i ;
30400504
z
I I I
l I !
30400505
26X
' I
IX
x • I
30400506
37X
I
I X
X I ! !
30400507
18X
!
I X
X
30400508
4X
I
IX
x ;
30400503
12X
! X
X •
,
30400510
6X
X
x !
30400511
6X
X
x i
30400512
6X
x | x !
30400513
X Ix I
30400521
X
X I i i
30400522
X
X I I !
30400523
X
X ! I ;
30400524
X
X | I i
30400525
X
X I I )
30400526
X
X I i ]
30400527
X
X i i i
3040052S
X
x i :
30400529
X
x I i i
30400530
X
X | | i
30400599
16X
X
x I i i
30400601
2
12X
| j
30400699
60X
i
142
30400701
26X
13X
X
X
X
30400702
X
X
30400703
1X
X
X
X
30400704
15X
2X
X
X
X
30400705
10X
X
X
X
30400706
16X
X
X
X
30400707
4X
X
X
X
30400708
13X
7X
X
X
X
30400709
11X
4X
X
X
X
30400710
X
X
x !
30400711
13X
X
X
X t
13-Sep~S3
B-29
Page 2
-------
TEMPORAL allocat
ON DAT]
\SOURti
ES FOR
SCCs
i
1 1 1
1
1 1 1
FREC
QUENCY 6FOCC
JRRENCE
N THESE DAT
A SOURCES *
1 1 I
I 1 1
see
DATA
i 1 1
PRIORm
GAP
1 i 1
RANK
see
CODE
GARB
TACB
SOS
LMOS
OEM
W-T-6
ACID—K/
NAPAF
BLS
E/Ei CRBl PET
1 1 1
30400712
X
x Ix ! 1
30400713
1X
X
x ix i !
30400714
4X
X
x ix i !
30400715
4X
X
X Ix ! 1
30400716
SX
x Ix ix t 1
30400717
x ix Ix 1 ¦
30400718
x Ix :X j I
30400720
x Ix :X • 1
30400721
2X
x Ix ix ! 1
30400722
X Ix :X ¦ !
30400723
1X
xx X • !
30400724
2X
x |x 'X : ;
30400725
SX
I
i
i
ix [x X :
30400726
1
ix lx .X i
30400730
1X
I
i
1
1 x i x X
30400731
! Xl
!
I x 1x .X
30400799
40X
XI
! i ^
| x j x 'X 1
30400801
X
1
| x j x x
30400802
I
X
X :
30400803
3X1
i
X
x : i
30400805 -
11X
3X
i
x i X !
30400806
j
x t i X : I
30400807
l
x I IX j i
30400809
IX
1
x 1 :X i ;
30400810
I
X I : X 1 !
30400811
i
X ] ! X ! i
30400812
4X
1
x 1 ; x ; !
30400814
l
x ! XI ;
30400818
x 1 iX ! :
3Q4QQ824
X
x i |X i
30400828
2X
X
X I 1
30400834
X
:X ! i
30400838
1 *
!X 1 i
30400840
1 x
ix 1 |
3040084.1
X
ix 1 1
30400842
|x
Ix I
30400843
1
1 x
IX !
30400851
1*
IX !
30400852
1X
ix !
30400853
tx 1 ix ; 1
30400854
x | ix 1 I
30400855
x 1 Ix 1
30400861
X
IX' 1
30400862
X
Ix . 1
30400863
X
X 1
30400864
X
X 1
30400865
X
X 1
30400866
X
X 1
3O40C867
X
X 1
30400868
3X
X
X 1
30400869
X
x !
30400870
X
ix
30400871
X
Ix.
30400872
X
|x
30400873
X
!x
30400874
X
IX
13-S#p-93
B-30
Pag# Z
-------
TEMPORA
L ALLOCAT
ON DAT,
\ sourcIes FOR
SCCs
1
III!
|
1
i
FREQUENCYC
>F OCC
JRRENCE SN THESE DAT1
A SOURCE
s«
i •
I
1
f !
see
DATA
|
1
: !
priority
GAP
|
1
1 i
RANK
see
CODE
CARBl TAC8
SOS
LMOS
CEM
W-T-E ACID —M
NAPAF
BLSl E/El CRBi PET
j
1 [
j]!1
30400875
1 1
x I 1X ; !
30400876
1
x 1 Ix ! 1
30400877
I
1
%
IX ! 1
30400899
3X
1
1
X
IX i :
30400901
2
I
1
i 1 1
30400999
z
|
1
I 1 !
30401001
z
j
1
! i !
30401002
z
j
1
1 i ' J
30401004
z
|
!
\ \ ;
30401005
2
|
i
; ; - s
30401006
z
|
I 1 ! !
; ¦ ¦
30401007
z
|
!
' i i ! i : ¦
30401008
2
|
i ! i i i 1 ' ¦
30401010
z
i
i ; i i 1 : :
.30401011
z
1
| ! 1 ! ! ' :
3040101S
z
! j !• | 1 :
30401016
z
1 'I'll!;1
30401017
z
• i i
1 i i i
30401010
z
i 1
111!
3040101S
z
1
i i i !
30401061
z
i 1
Ill;
30401062
z
1 1 I 1 i ! 1
30401063
z
! i lit;;
30401099
3X
! 1 1
i i
30402001
4X
1 1 1
i ! !
30402002
z
I I
30402003
z
! ¦ 1 111..
130
30402004
IX
1 I
It!!
30402099
29X
1 1
' ! j ; ;
30402201
18X
5X
i
30402210
z
|
i s i
30402211
z
1
- 1 !
30404001
z
s
t 1 -
30404801
z
f
»
I- ! i
30404902
X
I
-
1 1 1
30404999
2X
1
1 1 1
30405001
z
(
¦ ! 1 1
30405099
1X
1
I
I !
30438801
9X
3X
1
1 1
1 i
30488302
4X
X
i
i !
30488803
1TX
i
i
30488804
z
i
1
3048880S
1X
1
30490001
z
30490002
z
30490003
7X
30490004
z
30490011
z
30490012
z
30490013
1X
30490014
z
I
30490021
z
30490022
z
30490023
2X
30490024
z
30490031
z
13-Sep-93
B-31
Page 3
-------
TEMPO RA
_ ALLOCAT
ON DAT
*SOURC
ES FOR
SCCs |
1
!
i
-
FREQUENCYC
Tl
0
o
o
JRRENCE (N THESE DAT
A SOURCES *
! 1
(
! i
! I
see
DATA
1
i \ !
PRIORrTY
GAP
|
i 1 1
RANK
see
CODE
CARB
TACH
SOS
LMOS
CEM
W-T-E ACID —W
NAPAR BLS
E/E! CRBl PETi
1
i ¦
I 1 1
30490032
z
1
! ! i
30490033
2X
1
1 1 :
30490034
z
|
i i !
30499999
92X
|
i !
30600101
14X
1
I 1
30500102
10X
I
i 1
30500103
5X
1
30500104
1X
1
30500105
13X
i
i I i
30500110
2X
i
1
\ i
30500111
2X
j
!
30500112
1X
i
! i i i
30500113
1X
i
| ! . I 1 ! :
30500198
15X
i i ! ! : , '
305001SS
21X
! 1 ! 1 . i :
30500201
95X
I 65X
! i !
! :
30500202
6X
]
|
i
1
30500203
20X
I
1
i
30500204
13X
i
i
i
i i ;
30500205
SOX
I X
1
I !
58
30500206
27X
I 3X
1
i
i '
3050020?
2X
) x!
I
III'.
30500208
I 11X
t
1 1
1 1 ' 1
30500209
Z
!
|
I r " i !
30500211
2
1
•
! : - ;
30500299
8X
1
i
i
} : :
30500301
SX
)
1
i 1 ix ! S i ;
30500302
2X
1
l 1
x ! ! : :
30500303
1X
I
i
X
1 i
30500304
2X
1
I
X
30500307
1X
t
1
X
; ;
30500308
1X
1
1 x
1 1 i
30500309
2X
1
1 X
1 ! •
30500310
l
Ix
i i i
30500311
27X
I
X
1 1 !
30500312
1X
¦ 1
X
! !
30500313
1X
1
X
1 I !
30500314
34X
I
X
S i
30500315
1
X
1
30500316
i
X
I i
30500398
14X
X
1
30500399
7X
X
1 i
30500401
Z
1 1
I 1
30500402
Z
l ;
30500403
z
• 1
30500404
z
1
30500405
1X
:
30500406
z
1
30500499
z
I
30500501
z
l
30500502
4X
30500503
z
1
30500504
1X
1
30500505
8X
i
30500593
1X
i
30500599
2X I
1
13-Sep-93
B-32
Page 31
-------
TEMPORA
_ ALLOCAT
ON DATi
\SOURC
ES FOR
SCCs
1
1 1.1
f
1 1 1
FREQUENCY <2
F OCC
JRRENCE
N THESE DAI
A SOURCes*
I I .
1 1 i
see
DATA
1 1 1
PRiORm
GAP
1 ! i
RANH
see
CODE
CARB
TACB
SOS
LMOS
CEM
W-T-B
ACID-W
NAPAP
BLS
E/EI CRBl PET
I 1 i
89
30500606
62X
X
X
Ix | 1
30500607
16X
X
Ix 1 !
30500608
33X
X
ix I :
30500609
16X .
x
Ix 1 !
30500610
3X
X
ix i !
30500611
2X
X
Ix [ !
30500612
94X
X
IX ! ;
30500613
37X
I.
X
IX i !
30500614
27X
1 Ix '
IX ¦ j !
30500615
29X
i Ix
Ix i
30500616
79X
1 |x
X !
30500617
44X
i
1 Ix
;X ;
30500613
62X
I
1 ix i ; X
30500619
62X
i
! x i ' X i
30500699
23X
!
!
I x | X
26
30500706
21X
|
i
Ix i i X !
30500707
6X
|
j • Ix 1 :x : •
30500708
1X
i
i
! ! x ! 'x
30500709
32X
I
Ix
'X
30500710
2X
I '
1*
!X : ;
30500711
1 ¦
h
Ix !
30500712
6X
|.
Ix
ix 1
30500714
|
1*
IX 1 1
30500715
1X
1
! x
X i 1
30500716
2X
|
1 x
ix
30500717
58X
|
- Ix
1X i i
30500718
33X
|
u
Ix 1 1
30500719
74X
I
i
1 X
ix ! 1
30500799
14X
|
Ix
! X :
30500801
102X
1
Ix
i j i
30500802
5X
. |
X
I ! i
30500803
5X
X
j ;
30500899
46X
X
i I ;
30500901
2X
X
1 I i
30500902
Z
1 1 i
30500903
7X
1 1 1
30500904
2X
1 1 i
30500905
2X
1 ! 1
30500906
z
1 i ;
30500907
z
1 1 i
30500908
13X
1 ! I
30500909
1X
1 ! •!
30500910
1X
i
30500915
3X
1
30500916
z
30500917
z
30500999
25X
30501001
X
X
30501002
X
X
30501003
X
X
30501004
25X
2X
X
X
30501005
X
X 1
30501006
X
X |
30501007
X
X 1
30501008
13X
X
X I
30501009
33X
X
X I
13-Sep-93
B-33
Page 3:
-------
TEMPORA
L ALLOCAT
ON DATA SOURC
ES FOR
SCCs I
I
1
1 1
1 '
1 1
FREC
3UENCYd
IFOCC
JRRENCE IN THESE DAT1
A SOURCE
S *
|
|
see
DATA
1
PRIORITY
GAP
1
RANH
see
CODE
CARB
TAGS
SOS
LMOS
CEM
W-T-E ACID—M
NAPAF
BLS
E/E
GRB? Pt li
j I
30501010
32X
X
x 1 1
30S01011
57X
X
X I '
30501012
3X
1
X
X I :
30501013
X
X 1 !
30501014
15X
X
x ! I
30501015
9X
X
IX 1 1
30501016
6X
X
|x
I
30501017
1X
I
X
ix
i
30501021
! 1 ¦
Ix
30501022
1 I 1
Ix I ;
30501023
13X
- i i i
. IX ! :
30501024
2X
I 1 1
IX i
30501030
! 1 1
ix
3C501031
! 1 i ! ix ;
30501032
! ! ! ! ix
30501033
i 1 II ix s
30501034
! ' 1
X i
30501035
! ill
X !
30501036
! Ill
X i !
30501037
1 III
X ! :
30931038
2X
¦ i I i
X | !
30501039
2X
i I 1
X ) !
30501040
1 1 1
IX 1 !
30501041
l 1 1
!x i !
30501042
i I r i i x i i
30501043
i 1 ] 1 !x i 1
30501044
j 1
1 1 x 1 <
30501045
1 1
1 i X i •
30501046
1 1
! IX I
30501047
! j
! x |
30501048
i |
IX I !
30501049
1 j
IX • i 1
30501050
i
IX i
30501090
|x i :
106
30501099
2X
;
Ix | i
30501101
559X
!'
! !
30501106
12X
1
1 ;
30501107
X
27X
1
I I
30501108
X
12X
i • i
30501109
6X
1 :
30501110
sx
' i 1
30501111
1X
1 i
30501112
11X
1
30501113
3X •
1
30501114
2X
1
30501115
z
1
30501120
4X
1
30501199
86X
2X
!
X
30501201
2X
1
30501202
2
1
30501203
5X
1
30501204
5X
30501205
2X
30501206
3X
30501207
Z
30501208
6X
1
13—Sep—93
B-34
Page 35
-------
TEMPORAL allocat
ON DAT
\ sourcJes for
sees
¦ |
1
1 1
i
I
1
FRE<
3UENCY C
F OCC
JRRENCE IN THESE DAI
A source!
s *
1 j
1 1 i
see
DATA
1 1 i
PRIORmf
GAP
1 1 !
RANK
see
CODE
CARB
TACB
SOS
LMOS
GEM
W-T-0
ACID-*/
NAPAF
BLS
E/El
CRBl
PETt
S 1 1
30501209
4X
) ! !
30501211
4X
i I !
105
30501212
sx
! I i
30601213
2X
1 1 «
1 1 I
30501214
5X
j j 1
30501215
33X
I I !
30501221
z
I
i I 1
30501222
2X
1 l 1
30501223
IX
i
1 . I i
30501224
z
I
i
1
; . ! :
30501299
75X
I
)
I
f
j j
30501301
z
I
]
1
i i :
30501339
z
I
! ! 1 ! 1 i
30501401
Z
I
I ! 1 !
135
30501402
10X,
; 2>5
3X
x ;
f
81
30501403
13X
i
i
3X
i •
1 | X : i
119
30501404
i
j
Ix i j !
30501406
27X
! 2X
I
i
x ! •
30501407
28X
I
I
x i i :
30501408
17X
I
l
X | ! :
30501410
59X
I
1
x I i ;
30501411
i
I
' f
X ! t •
30501412
I
! I :
30501413
5X
i I
1
I
x i i !
30501414
!
1
! ix 1 !
30501415
5X
i
1 1 ' !
x ! i
30501499
5X
:
i i i
! i J
x i !
30501501
3X
l
1
ix
i i !
30501502
IX
j
i
fx
! i !
30501503
28X
|
i
>
Ix
f i 1
30501504
iax
I
i
¦1*
! i i
30501505
1X
I
IX
j i ;
30501506
1X
]
Ix
f >
i ;
30501507
I
tx
t !
30501508
3X
]
Ix
i I :
30501509
1X
i
\
X
I :
30501510
J
!
IX
i ;
S !
30501511
|
jx
i i i
30501512
1X
I
X
! i
30501513
1
X
! |
30501514
9X
i
X
i i
30501515
1X
1 .
X
|
30501516
IX
1
X
30501517
X
!
30501518
X
i I
30501519
X
30501520
7X
X
30501521
3X
X
30501522
X
305015S9
1SX
X
30501501
27X
30501602
26X
30501603
9X
54
30501604
20X
6X
30501605
5X
30501606
2
I
13-Sep—93
B-35
Page 34
-------
TEMPO RA
. ALLOCAT
ON DAT)
\ SOURC
ES FOR
SCCs
I
1 i 1
1 ! I
FREC
5UENCYq
FOCC
JRRENCE G
N THESE DATA SOURCE;
S *
1 1 1
J
1
1 1 1
see
DATA
I
|
i i >•
PRiORITV
GAP
|
|
1 ! !
RANK
see
CODE
CARB
TACB
SOS
LMOSl
CEM
W-T-d ACID —M
NAPAF
BLS
E/El CRBl PET
I
! i
30501607
6X
t !
30501608
5X
I !
30501609
12X
1 i '
30501610
z
t 1 1
30501611
z
i i i
30501612
2
1 1 I-
30501613
12X
! 1 '
30501S14
10X
I ;
30501615
5X
I
1 1
) | !
30501816
19X
I
1 !
i i '
30501617
5X
I
[ i S i 1 :
30501699
30X
| |
1 1 ; ' i i -
30501701
IX
j XJ
[ ! ! I ! ! !
30501702
Z
I |
1 ! ! ) !
30501703
1X
I X
1 1 1 1 !
30501704
! x
1 i i ( 1 ! i
30501705
I
1 1 ! ! !
30501799
2X
I
I j |
1 i S
30501801
5X
I
! 1 1
1 ^ '
30501899
7X
i
i i "1
1 i i
30501901
z
I
1
I
i l <
30501902
10X
j
1 ! 1
i ! !
30501903
4X
i
]
1 1 1 i i
30501904
z
(
!
i I ;
30501905
z
j
30501906
z
i ! ! 1 I- i
30501907
z
I
! 1 i 1 ! i
30501999
z
1
1 1 1 i 1
30502001
227X
1
t ! ! ] 1 ! i
30502002
24X
|
I l i i I
30502003
9X
1
1 1 ! I
30502004
3X
1
1 ! 1
30502005
2X
1 1
1 ! 1
30502006
46X
1
.1 i !
30502007
55X
1
1 1 .
30502008
1X
t
1 1 1
30502009
X
1
i ! i
30502010
3X
I
1
! i 1
30502011
39X
1
1 ! !
30502012
23X
1 ! i
30502013
z
t ! • !
30502014
2X
i
i
1 1
30502015
z
1
30502016
z
! !
30502020
z
i :
30502099
5X
:
30502101
14X
30502102
z
30502103
z
30502104
z
30502105
z
30502106
1X
30502201
z
i
30502299
z
i
30502401
3X
i
30502499
I2X
i
i ¦
i
13-Sep-93
B-36
Page 35
-------
TEMPO RA
- ALLOC AT
ON DAT,
k SOURC
ES FOR
SCCs
1
|
I
1 1 !
1
|
1
I 1
FREQUENCY C
FOCCt
JRRENCE IN THESE DAT
A SOURCES *
1 !
|
i ! i
see
DATA
1
i i ;
PRioRrn
GAP
1
1 i !
RANh
see
CODE
GARB
TACB
SOS
LMOS
CEM
W-T-E ACID- W
NAPAF
BLS
E/Ei CRSi PET1
1 1 1
30502501
14X
X
1 1 1
30502502
2X
I ! ;
30502503
X
19X
1 1 1
30502504
X
5X
1 1 i
30502505
X
2X
' I i )
30502506
X
1X
r j
30502507
X
2X
s 1 !
30502508
4X
X
i i i
30502509
z
i i i
30502510
X
IX
I
I- I
1
|
I j ;
30502511
X
|.
i
! 1 i
30502599
22X
|
1
! f !
30502601
3X
\
1 1
i I
30502699
z
;
1 !
i : :
30503099
z
i
i ¦
i i ,
3G503101
z
i
i
i
fill!
30503102
z ¦
t
1' i ! [ !
30503103
z
j |
i
i i ! i i
30503104
z
I
1 i 1 •[ i i i
30503105
z
i
i i I i i i . i
30503106
z
i
i
i i 1
30503107
z
1
1 1 1 1
30503108
z
i
1 i ! !
30503109
z
!
1 i
r ¦ ; . :
30503110
z
i
i |
III:
30503111
z
i
1 ! • !
30503199
z
1 !
1 1 !
30503201
z
1. !
I 1 !
30503202
z
!
i
) 1 !
30503203
z
i
i
! i 1
30503204'
z
i
I ! 1
i i i
30503205
z
i
30503206
z
i
i 1 1
30503299
z
i •
30503301
1X
i
! 1 !
|30503401
z
i
i
30503402
z
t
1
i
i
30503901
z
i
i
i.
I
30503902
z
i
i
i
!
30504001
z
i
i
1
30504002
5X
i
30504003
z
i
30504010
z
i
30504020
7X
I !'
1
30504021
5X
30504022
z
30504023
z
30504024
3X
30504025
9X
30504030
8X
30504031
4X
30504032
1X
30504033
5X
30504034
70X
30504036
z
30504099
54X
!
13-Sep—93
B-37
Page 36
-------
TEMPO RA
L ALLOCAT
ON DAT
* SOURC
ES FOR
SCCs
I
I I
FRE«
2UENCYC
IF OCCURRENCE
N THESE DAT
A SOURCE
S *
see
DATA
1
1
i
PRIORiT*
GAP
1
1
!
RANK
see
CODE
CARB
TACB
SOS
LMOS
CEM
W-T-E ACID —M
NAPAF
BLS
e/e! crb
PET
! 1
1 !
30509001
z
i
I
30509002
z
1
|
30509101
2
1
1 I !
30510001
z
4X
1 ! i 1 !
30510002
z
2X
1 1 1 1 1
30510003
z
1X
i i j ! . ;
30510004
z
1 ! ! 1
30510005
z
1 i I.I !
30510006
z
!!!!':
30510007
z
IX
' ! ; l
30510101
z
1X
[ | i 1 :
30510102
2X
¦ i I I* i , |
30510103
93X
! { |x I .. !
30510104
4X
I I i i I ;
30510105
26X
I It i
30510106
1X
!
j
1 1 ; :
30510107
z
i t . i
30510108
1X
1 i i i
30510196
z
. 1 ! • i
30510197
4X
1
! 1 !
30510198
7X
]
1 ! •
30510199
38X
1
i j 1
30510201
z
|
' i 1 1
30510202
1X
!x t i 1 i
30510203
20X
1X | ! ! i
30510204
1X
| j ; (
30510205
5X
1 1 i
i
30510206
z
I i !
30510207
z
i I i i i
30510208
6X
! | i i :
30510296
4X
1
i ' 1
30510297
3X
i
i 1 i
30510298
4X
i
30510299
ex
till!
30510301
z
I l-l 1 • !
30510302
x i ill
30510303
45X
x 1 1 1 !
30510304
1X
! 1 1 i
30510305
7X
III!
30510306
z
i ;
30510307
z
;
30510308
z
1 1
i
30510396
z
i
30510397
1X
1
30510398
3X
30510399
6X
30510401
z
30510402
1X
X
30510403
•
10X
X
30510404
1X
30510405
14X
30510406
1X
l
30510407
z
30510408
5X
30510496
1X
I
30510497
3X
1
13-Sep—93
B-38
Page 31
-------
TEMPORA
. ALLOCATION DATA SOURC
ES FOR
sees I
1 I
1
j
1 i
FREQUENCY C
FOCC
JRRENCE THESE DAI
A SOURCES *
1 i
,
J
I
1 t
see
DATA
|
j
i !
prioriti
GAP
i !
RANK
see
CODE
CARB
TACB
SOS
LMOS
GEM
W-T-6
ACID—M
NAPAF
BLS
E/E
CRBl PETi
I i
3C510498
3X
1 !
30510499
IX
i ;
30510501
z
1 !
30510502
X
i 1
30510503
4X
X
1 i
30510504
1X
i I *
30510505
7X
i i i
30510506
z
1 1
1 i
30510507
z
! 1
30510508
3X
1 ! f
30510596
4X
III-.
30510597
2X
j
1 i :
30510598
3X
I
i
I !
30510599¦
13X
I
ill);
30515001
z
i
1 1 ! i ;
30515002
1X
I
II!!-
30515003
7X
i
(
i 5
30515004
3X
I
i
1
1 - !
30515005
9X
i
i
i i ;
30588801
56X
i
1
i
1 ;
30588802
3X
i
lx
I 1 j
30588803
14X
U
1 i ;
30588804
3X
I
!
1 i i
30588805
4X
i ;
30590001
1X
X
i
t lx «
30590002
z
I i i
30590003
27X
X
4X
i i
i !
305S0011
3X
! I
r ¦ ! :
305S0012
I I
! lx ;
30590013
X
! i
1 1 :
30590023
2X
) I
1 ¦
30599999
97X
1 f i
30600101
2X
X
X
1 i
200
30600102
81X
X
X
1 !
202
30600103
36X
X
X
X
I
1S2
30600104
1374X
X
X
I
30600105
159X
X
X
" i
187
30600106
128X
X
X
X
i
30600107
X
X
X i
30600108
X
X
I
30600111
X
X
1
30600199
24X
X
X
!
2
30600201
107X
X
X
X
X
X
30600202
10X
X
X
X
i
133
30600301
42X
X
X
X
1
17
30600401
44X
X
X
X
X
i
i
30600402
13X
X
X
X
I
60
30600503
163X
X
X
X
X
|
71
30600504
27X
X
X
X
!
30600505
186X
X
X
X
i
30600506
21X
X
X
X
1
85
30600602
70X
X
X
X
X 1
30600603
54X
X
X
X
x 1
171
30600701
281X
X
X
X
X
i
30600702
SOX
X
X
X
63
30600801
61X
X
X
X
1
13-Sep-93
B-39
Page 3S
-------
TEMPORA
. ALLOCAT
ON DAT
\SOURd
ES FOR
SCCs
I | j
1 i 1
FREQUENCY 0
FOCCi
JRRENCE tN THESE DAT
A SOURCE
S *
i i i
j
1 i
see
DATA
t 1
PRIORm
GAP
i r 1
RANK
see
CODE
GARB
TACB
SOS
LMOS
CEM
W-T-E
ACID—M
NAPAfi
BLS
E/E! CRBi PET
1 1 1
30600802
7X
*
X
ix 1 :
74
30600803
3X
X
X
!X
!
30600804'
10X
X
X
Ix
j
80
30600305
46X
X
X
X
Ix
|
30600806
1X
X
X
X
(X
j
3060080?
X
X
IX
i
30600811
11X
X
X
Ix
1
30600812
16X
X
X
>x
i
30600813
X
' I
X
¦ x ! ;
30600814
X
I
X
ix l '
30600815
X
I
X I iX 1 i
30600816
X
I
x 1 -X ! :
30600817
2X
]
X
I
x 1 X ¦
30600818
j
X
I
X ; X !
30600819
6X
t
X
I
x l x i ;
30600820
2X
!
X
I
I
X ! -X !
30600821
4X
I
X
|
IX I X ! >
30600822
IX
X
I
I x i X
30600903
44X
X
|
Ill;
30600904
106X
X
I
I i i i
30600905
2X
X
i
I I ! |x !
30600999
34X
X
I
t • ! : i
30601001
X
I
I ! i I 1
30601101
13X
X
X
i
U I ill
30601201
15X
,X
I
lk ! ix i ;
30601401
20X
X
I
1x ! % i
30601402
X
!
i i '
30601599
X
I
i 1 i
30609901
X
i
X
' |X i
30609902
1X
X
I
Ix
: i i
30609903
18X
X
- 1
' i ! i
30609904
11X
X
i ¦ 1 !
30609905
X
Ix
xi if
30610001
139X
X
X
X 1
116
30688801
691X
X
K
x ix
l
30688802
79X
X
X
X 1 I 1
30688803
sax
X
X
X i ! !
30688804
18X
X
X
x ! 1 j
30688805
31X
X
X
*• ! [ i
30699998
SOX
X
! 1 !
30699999
230X
X
i i !
30700101
13X
x Ix .
i
30700102
12X
X
2X
x |X
30700103
9X
x |X
1
30700104
18X
5X
x |X
180
30700105
10X
6X
x |X
24
30700106
11X
X
3X
x |X
30700107
1X
X
x |X
30700108
2X
x |X
30700109
16X
2X
x |X
18
30700110
2X
X
X
x IX
30700199
129X
3X
X
X
30700203
3X
x
X
30700211
X
X
30700212
%
X
30700213
|
X
X
13-Sep—93
B-40
Page 39
-------
TEMPORA
.ALLOCAT
ON DAT.
k SOURC
ES FOR
SCCs I
1
]
I <
1 ' 1
|
1
1
| ; ;
FREQUENCYC
focc(
JRRENCE W THESE DAlt
A SOURCES *
1
i i !
see
DATA
I
I :
PRIORlTi
GAP
1
1 1
RANK
see
CODE
GARB
TACB
SOS
LMOSl
CEM
W-T-S
ACID-M
NAPAP
BLS
E/E! ORB' PET
I
!¦ I I
I : ;
30700214
|
! I I
X
:X J
30700215
3X
I
I
1
X
.'X
30700221
! xi
I
1 I
1 1
X
X
30700222
I
j
| |
X
;x
30700223
I X
I
1 !
X
30700231
I
i I II
X
ix :
30700232
i
I i
1
X
X I
30700233
I
i I
• 1
X
X
30700234
IX
;
j i
|
X
:x
30700299
I 3X
I I
X
¦x
30700301
i
! I
I
X
x
130700302
!
'
X
x
30700303
I
2XI
X
x
30700304
I
: I ! ! ! i ! X
30700401
5X
I : ! ! ! X
1
30700402
3X
XI
x
30700499
10X
! I' i I L I l : .
30700501
18X
X
¦ X ;
30700597
2X
X
X
30700598
1X
r
! i I I*
X
30700599
14X
I
i I
jx
.X
30700701
32X
I
i I
X
X
;x
30700702
15X
!
I I
X
X
X
!
30700703
31X
! 4Xt ! I
X i X
X
12
30700704
: I
i
X | X
X
30700705
i
X
X
X
30700706
! X
X
X
X
30700707
21X
>3
X
X
X
|
30700708
X
X
.X
30700709
1X
i
I i
1*
X
x ;
30700711
;
! I
1 x
X
X
30700712
i
1
Ix
X
X ;
30700713
i
j
Ix
X
;x
30700714
i
i
1
Ix
X
: X
30700715
38X
x !
I
lx
X
IX
30700716
3X
i
|
Ix
X
IX ;
30700758
29X
! 3X
i i
¦ Ix
X
IX |
30700799
43X
! ' x
! I
1 X
X
X
30700801
6X
I
I t
X
X
X 1 f
30700802
3X
!
1 !
X
X
¦X :
30700803
21X
|
1
%
X
!X i
30700804
31X
I
X
X
,X i ;
30700805
16X
I
X
X
:X :
30700806
3X
I
X
X
ix j I
30700807
2X
I
X
X
ix ] I
30700808
18X
I
X
X
Ix | !
30700896
2X
I
X
X
ix ! !
30700897
1X
! 14X
X
X
Ix 1 j
30700898
12X
|
X
X
IX 1 !
30700899
31X
I x
X
X
ix ] ;
30701199
12X
I sx
x 1 ! 1
30701301
3X
I
X
X
ix ! :
3Q701399
5X
I 21X
X
!X : !
30702001
z
I
|
i i i
30702002
z
|
|
t. f !
30702003
z
t
\ j ¦
13-Sep-93
B-41
Page 40
-------
TEMPORAL alloc at
ON DATJ
\SOURCES FOR
SCC s
1
! 1 !
1 ! ;
I
I
1 i
! !
"
FREQUENCY OF OCC
JRRENCE (N THESE DATtA SOURCES *
i ; ;
1
1
I I !
1 i ;
scd
DATA
1 "
! 1 I
1 ; J
PRiQRrnj
GAP
1
1 1 ill]
RANK SCC
CODE
GARB
TACB
sosi LMOS
CEM
• W-T-e ACID —M NAPAR BLSl E/E! CRBi PET
|
|
1 i 1 1 ! i
|30702004
z
1
! i 111.-
1307020S8
4X
1
! 1 ! 1 1 i
130702099
41X •
1
1 ' i j ' !
130703001
17X
1
i 1 i I ! i
130703002
ax
I
1 ' i i i ;
j 30703096
z
1
! 1 i ! ! » •
]30703097
z
i
i ! ' j ! I
!30703098
1X
i
I I I ! ! : i
I 30703099
41X
|
i ; ! ' |
130788801
14X
1 SX 1 1 ! • ! !
!30783802
2X
i
i i ! ; ;
! 1' - ,
!30788803
1X
; i
i 30788804
14X
j
!
130788805
3X
!
i : i : ;
!30788898
4X
i
i
i
130790001
!
2X1
' I ' ' i
130790002
z
i
t
1 • I I . 1 , :
I 30790003
6X
i i
i : ;
130790011 .
z
i
1 ! 1 j ! :
[30790012
z
i
I
! i
: : i
; 307900',3
3X
i
i
i
; i
I ;
I 30790021
z
i
1 < j :
I 30790022
z
1
j
! | ¦ :
130790023
z
i
|
! i I
30799998
6X
!
!
30799999
4X
i 16X
t I i 1 i i i
30800101"
1
i i i X ! -X
30800102
1X
i
i i ; x x
30800103
!
! j : X I X
30800104
IX
(¦
1 i :x S x
30800105
! X
! 1 ! X 1 : X ;
30800106
j
! : ;X ] : X
30800107
i
1 1 i X j ; X :
30800108
i
r ! IX I ' jx i
30800109
1X
1 4X
1 ! ix 1 ix !
30800110
i
! 1 .x ! |x i
30800120
j
i 1 ix I Ix j
30800121
I
! i !x | Ix |
30800122
|
1 iX I IX !
30800123
| 1
i ;X
i X !
30800197
18X
I xl
1 !x
IX !
30800198
4X
1 )
i ix
• ix i :
30800199
21X
i ix ! !x I
30800501
1X
2X
IX 1 IX i !
30800699
28X
15X
IX
f I :
30800701
22X
2X
Ix
! 1 i
30800702
24X
4X
ix
1 1 !
30800703
10X
3X
Ix
1 1
30800704
14X
x
2X
IX
1 1 i
30800705
7X
1
1 1 !
30800720
24X
X
Ix
! ! 1
30800721
1X
1
i i i
30800722
4X
3X
ix I r !¦ 1
30800723
4X
1 X
IX
! I i
30800724
18X
1 8X
IX
1 1 !
153
30800799
287X
X 1 5X
IX
! ! i
13-Sep-93
B-42
Page 41
-------
TEMPORAL ALLOC AT
ON datJ
\SOURC
ES FOR
SCCs
1 I 1
i : ¦
•• 1
i i 1
1 i i
FREQUENCY OF OCCl
JRRENCE IN THESE DATA SOURCES »
1 ¦ 1 !
|
i
I
t 1 i
; | :
sea
DATA
i
i ! ! ' i ! I :
PRIORITY
GAP
1 '
1 1 ! 1 i , :
RANi^ SCO
CODE
GARB
tacbI
SOS
LMOS
GEM
W-T-E ACID —M
NAPAfi BLS
E/E! cab: pet
i
! 1 !i:;.
30800801
z
i
! I i i ¦ !
i 30800802
5X
1
I
! I lli;
130600803
SX
1
! f ! !
130800901
21X
1
1
II i i
!30890001
z
1
l
II!!'
130890002
z
1
1 i i 1 i i
|30890003
48X
! x
I 1 i ! ¦ :
I 30890004
z
]
; . i • i
!30890011
z
i
! ¦ 1 ! ] 1 ! '
130890012
2
!
! ! - j l ¦
=30890013
*5X
ill! ! : j
^ i 30890023
2
i i 1 • ! 1 : j
I 30899999
12X
1 35X1 i ! ! 1 !
i30900198
33X
i 1 ! ! i i ' i
(30900199
8X J
; >d 1 : ! ,
130900201
3SX
I'll i : : !
130900202
66X
!
' : ! > t
• i ! i s
|30900203
5X
i
! ! ! !
130900204
IX
j
1 i ! ! ! : i"
130900205
42X
1
! ' i I i ' :
130900206
2
j
! i ! i 1 ¦
130900207
21X
1
! i ! ; ! > ;
130900208
3X
!
i ! 1 • i ! :
j 30900298
Z
i
! i i !
|30900299
3X
1 |
1 III
|30900301
z
1 i
i | ; :
!30900302
z
;
! ! ; 1 . i
i 30900303
z
i ! ; i : ;
t 30900304
2
I 1
j ; j | ;
130901001
60X
i 3X
! j ; ' ix
130901097
5X
i
1 i J 1 x > !
130901098
13X
l
i | i |x i
130901099
10X
I
1 I 1 Ix ; !
130901101
32X
I 3X
1 1 i 1 x i i i
30901102
52X
i x
1 1 i I x ! i
30901103
15X
j
i i Ix ! ;
30901104
27X
1 i Ix I i
30901199
28X
1
! I 1 Ix !
30901501
11X
1
1 I ! Ix
30901601
i
I Ix :
30901602
1X
1
i -
i • 1 x - i !
30901603
1
' 1
i Ix 1 1
30901604
3X
I
1
1
x 1 ! ' :
30901605
1
¦ 1
1
x 1 1 ;
30901606
1 .
t
1
x 1 i ;
30901607
• I
1
|
X 1 1 |
30902099
12X
I
1
j
1 t 1
30902501
1 2X
t
1
x 1 i :
30903004
9X
|
1
1
1 i 1
30903005
2X
I
i
i
1
' i <
30903006
Z
1
|
j
1 1 1
30903099
28X
I X
i
|
~ ! ;
f ; '
30904001
2X
|
1
|
I 1 i
30904010
1X
|
t
1
! i 1
30904020
3X
1 X
i
|
l I I
30904030
Z
1
l
1
I 1 i
13—Sep—93
B-43
Page 4 Z
-------
TEMPORA
L ALLOCAT
ON DATA SOURdES FOR
SCCs |
1 1 I i . 1 i
I
|
1 It!
i
FREQUENCY 6F OGC
JRRENCE W THESE DATA SOURCES * 1
I
j
1 i
•
see
DATA
1
1 III
PRIORfPr
GAP
¦ 1 1 ' I i
RAfd
see
CODE
CARB
TACB
SOS! LMOS
OEM
W-T-E ACID —M NAPAP BLS E/E
CRBt PET.
]
1 1 5 1
30906001
7X
t >1
! 1 ill
30906099
2X
i
i i ; 1
30988801
37X
i xl
1 . i 1 i i i :
30988802
z
1
1 1 I 1
30988803
1X
j
1 1 1 i ; i
30988804
z
j
I i i I i i
30988805
z
t
1 1 .!! = !¦
30990001
z
1
1 1 i !
30990002
z
i
1 ! 11,'!
30990003
182X
1
! ! i s • ;
130990011
z
I i i
II ! : !
|30990012
z
\ I '
!i :
130390013
13x | : % ¦ ; :
I30990023
17X
; j j i i • ;
]30999997
! isax
! : i ; : :
j 30999998
112X 1
l . 1 :
130999999
'• 191X
"
! ! : -
31000101
1X
!
1 1 ! j* :
31000102
2X
f
1 1 1 !* 1
31000103
; ' \ * i
31000104
2X
I
I ¦ ; i x !
31000105
1X
1
! 1 i ix '
31000199
247X
1
1 i 1 IX | ' !
197
31000201
398X
! x! I i : ;
31000202
152X
1 i ! ! : :
110
31000203
1509X
! ' ; : '
31000204
7X
i i
31000205
217X
ii i .i ; !
j31000206
16X
II i 1 ! i : i
!31000207
53X
III i i i i
131000299
404X
1
1 !
' ! • ! ¦
179)31000299
404X
1 i I '
i !
31000401
2X
I'll i
• ! 1 1 !
31000402
2X
1 n
t
*
1 x 1 1
31000403
47X
T
!
i 1 i i
31000404
473X
1 x
1 I 1 ! • I :
310004OS
20X
l
! . i- 1 ! ; '
31000411
2X
i
1
t j Ix 1 1
31000412
i
1
i J X | ;
31000413
i
\
i ix 1 ! ;
31000414
84X
i
i i'
i
31000415
ax
i
1 1 1
! 1
31088801
396X
1 1 1
! 1
31088802
SOX
1 1 1
1 I
31088803
10X
;
1
i • !
31088804
14X
i
1
I i
31088805
14X
i
| |
31100101
z
I
I 1
31100102
2
1 1
31100103
Z
1
! i
31100199
2
i
1 1 i 1
31100201
2
1
III!
31100202
Z
1
1 1 ! !
31100203
2
1
III!
31100204
2
i
II ! !
31100205
z
!
1
i !
13-Sep-93
B-44
Page 4:
-------
TEMPORAL alloc at
ON DATA SOURC
ES FOR
SCCs I I
1 i 1 i' 1 i
I
1 1 | 1 1 1
FREC
2UENCY (j
F OCC
JRHENCE IW THESE DATA SOURCES • | 1 1 i
I
1 1 II!':
SCO
DATA
1 ! i 1 i 1
pmonrnt
GAP
1 ; | | ;
RANK SCO
CODE
CARS
TAGS
SOS
LMOSI
CEM
W-T-E ACID—M NAPAP, BLS) E/Ei CR8 PET
i ¦ 1 i' . i
i i 1 i ; i
31100206
z
1 till'
31100299
z
1 III!'.
31299999
37X
2X
! ! 1 ; !
31307001
z
1 1 !
!
31307002
z
1 i 1 1 i i
313SC001
z
•! ' 1 | : S :
31390002
z
1 1 I ! ' :
131390003
14X
i ! ill-',
|31399999
313X
i 2X! I ! ! • | !
131400901
I
1 i ' 1X ! X 1
! 31401001
I >0 i ¦ \ \ \ ! :
! 31401002
z
j I ! ! . ; i
131401101
15X
I ! ! I I ! | x
I 31401102
8X
i ! ! ! ( !x
!31499999
4X
x i ! \ i : i ;
i31501001
z
1 ; >
:
i31501002
IX
1 i 1
: j
I31S01003
z
1 1 ! . ! : ! i
•31502001
9X
1 5Xj ! i i I ! x ! :
31502002
1 i ^ i '! ix- :
31502003
i
i 1 ; S 1 x ; i
32099997
! II ' ! ;x : X
32099998
j Xi ! ! . ! 1 X i ; X :
32099993-
5X
1 XS 1 ! | 1 x ! !x ;
133000101
|
1 x j x :
133000102
1
i !x jx
I 33000103
I
i ! I x Ix •
|33000104
5X
1
: I X i *
I330C0105
1
1 ! ! ! X 1 x ¦
133000198
3X
1 , >3 i ! - ix ix ;
|33000199
4X
|
1 l ! I* U : : :
133000201
z
1
I i i 1 i ;
33000202
1 X
j ! j j
33000203
z
1
! 1 III):
133000211
z
]
1
1 : 1 1 ! i :
(33000212
z
|
1 1 1 i t
133000213
1 X
i ' i II:;
33000214
Z '
1
i i list
33000297
IX
1
i i 1
1 i
33000298
z.
1
1 '>¦ \
33000299
14X
1
I ! 1 1 ! i
33000399
z
1
1
1 i i
33CCC4S9
X 1
1
X
i i
33000599
1
X
I !
33066801
2X
9X
1
i i
33085502
z
l
1
1 1
33088803
z
1
1 1
33088804
z
|
1 . 1
33088805
z
1 1
38000101
1X
1 1
38S00101
z
! 1
38500102
1X
I
38500120
z
1
38500201
z
i
38500202
2
38500210
2
1 i
13-Sep-93
B-45
Page 4-
-------
TEMPO RA
-ALLOCAT
ON DAT)
K SOURC
ES FOR
sees i I
I
I ! . i
! '¦
I
! !
FREQUENCY C
>F OCCURRENCE [N THESE DAT
A SOURCES *
' i
i
i
t
|
| :
see
DATA
|
I
I'll
priority
GAP
\
!
I " I : i
RANK
see
CODE
CARB
TACB
SOS , LMOS
GEM
W-T-E ACID-M
NAPAfl
BLS
E/E CR8 PET
1
I
II i
39000189
z
;
I ,
»
I I ; • ;
39000199
2
I
I
[ i ; I
39000201
2X
1
I
! I i
! I '
39000203
7X
i 5X3
i I I ! ; i
39000288
2
i
! Ill ;
39000289
Z
i
I I l i : !
39000299
Z
i
I 111 !
39000389
2
it I i I
139000399
Z
II ! i ! :
39000402
2
I :
! ! j
|39000403
I ; x I ; I ! ! i
|390004 89
! " 3Xj I I : !
| I3SC0C499
I
33Xl I i 1 i
i 139000501
!
42X1 ; ; j j !
I
I
39000502
ix I ; ! I i ) is
39000503
i 2X3 ; ! I :
39000589
: I . ; 13X1 ! ! !
39000598
2X 1 : I i ! 1 : 1
3S0GC599
! 69XJ | i
I ! ;
39000602
: 2Xj j i
I ! = i
39000603
i i sxj j !
i I : ;
39000605
i ; 29X] | !
i I 1 : .
39000689
10X
! 152X1 ' I i
1 • i • i
39000699
34X
i 655X| i i ]
i •
39000701
Z
i | \ i j
I :
101
39000702
z
i i t i i
J
39000788
z
[ j | | j
i :
39000789
z
i • i I i -
i
i
39000797
9X
! X
i
J
39000798
z
I i
1 • 1 :
39000799
! 2X1 | !
i . '
39000801
; xi
I
I
i i
39000889
! • 6X1
|
i
\
39000899
1X
i 13X
|
1 I i
39000989
! X
I
; 1 : :
39000993
: 2ox.
J
i i ;
39001089
! 9XS
i I
39001099
I 32X
|
II-!:
39001299
z
i
I • t
1 1 i
39001389
1X
1
I
i i
| ; t !
39001399
1X
i 10X
I ;
1 - ;
39990001
z
| ;
i ! i r
39990002
z
i : i :
39990003
40X '
I 4X
l
i 1 1
39990004
1X
t
! 1 1
39990011
z
I
I
! 1 i
39990012
z
I
I
1 i 1
39990013
4X
I
I
! ! 1
39990014
2X
I
)
I \ \
33990021
z
I
i
1 I 1
39990022
z
I
I
~ ! 1
39990023
SX
|
I
\
39930024
10X
i
I
•( j <
39999993
158X
i 9X
I I
1 • ¦
39999994
4X
! X
I I
| ; ¦ !
39999995
12X
I 8X
I I
'i i i
13— Sep—93
B-46
Page 4:
-------
TEMPORAL ALLOC AT
ON DATA SOURCES FOR
sees !
i f
(
: : i i
FREQUENCY OF OCCURRENCE IN THESE DAI
A SOURCES *
\
l !
I ;
SCO
DATA
j
! I
1 1 =
PRIORm
GAP
I
1
i |
1 i • i
ranN
see
CODE
CARB
TACB
SOS LMOSI
CEM! W-T-E ACID—M NAPAE BLS! E/E- CRB- PET
|
1 |ii ;
39999996
214X
|
1 1 ! !
39999998
1X
|
j 1
f 1 ;
39999999
68X
i 17X
! 1
i
40100101
I X
1 !
1 : . i =
40100102
2
i
i 1
40100103
f X
1 !
1 - !
40100104
( 2X
! I I I ! i
40100105
1X
1
J
i i i
. i 40100106
Z
!
! : ¦ 1 ! i :
140100198
1X
1
! i -111
140100199
1X
1
1
! I i ! 1
140100201
| 29X
j
II ! ' 1
121 !40100202
j 82X
1 5X3 ! ; j :
140100203
I 22X
! xl » : ! >'
I 40100204
hx
1 2x1 1 ' i !
140100205
31X
! sx
j ! |
140100206
2X
i 3X
! i !
i 40100207
22X
1 • Xl
! i t
i : 1
140100200
Z
1
! ! ! 1 -
j40100209
z
j
1 1
! | i
j40100215
4X
j
! 1
i
140100216
2
I .
1 i
140100217
2X
[
I 1 1 ¦ ! 1
140100221
IX
[
1 1 1 i
140100222 j
1 X]
; | t ; ;
, 1*40100223 f
2X
[
i. 1
i ! |
140100224 j z
1
j 1 i '
83140100225 j
3X
1 5X
| > i ! ;
(40100235 |
2X
|
1 j ! •' ! 1
140100236 j
SX
|
| 1
|40100251 |
3X
I
! i
1 i
40100252
23X
1 5X
1 ,
i I :
40100253
2X
|.
1 i
1 1 '
40100254
Z
I
i
i
1 1 ;
40100255
10X
1 ?x
i
it i
it'll
40100256
! x
i
| : ! ;
40100257
14X
i 2X
:
1 1 ' i
40100258
z
1
i
! 1 : ! :
40100259
z
1
}
1
: i :
40100295
35X
1 2X
j
i i 1
40100296
2X
4X
T
t
! ! : :
40100297
19X
7X
1
i i : '
40100298
7X '
X
1
1 1 i i :
40100299
30X
X
1 1 : i
40100301
3X
2X
| . i :
40100302
16X
3X
! ; ! =
40100303
21X
5X
; i ;
40100304
3X
I . I i
40100305
9X
15X
1 ¦ ¦ I :
40100306
2X
4X
<
! 1 ! :
40100307
9X
X
i i 1
40100308
1X
X
1
1 I
40100309
1X
1 '
i
40100310
SX
1
1 = i !
40100335
13X
i i
I'll
40100336
4X
*
i i
1 i 1
13—Sep—93
B-47
Page 46
-------
TEMPO RAt
. ALLOCAT
ON DAT)
KSOURdES FOR
SCCs
I
i
I
1 ;
FREQUENCY OF OCC
JRRENCE IN THESE DATA SOURCES *
I |
: i
see!
DATA
j j
PRIORITY
GAP
i ' I i . I
RANH
see
CODE
GARB
TACB
soa
IMOS
CEM
W-T-d ACID — M
NAPAR
BLSi
E/E; CRBI Pfc i:
f I 1 s
40100398
55X
48X
1 111 ! :
198
40100399
55X
50X1 | i 1 ) j : i ,
40100401
Z
i y 1
1 1 •
40100499
z
! I i
t i
40188801
35X
5*3
I
1 j ;
40188802
z
(
l
40188803
2
I
i 1
40188804
1X
|
I, .
40188805
2
! ! ! I j
40188898
4X
24x! 1 i ! ! !
40188899
Z
li I 1 I i
40200101
510X
24X1
X
189
401200110
110X
¦ ! 152X1 ! i 1 ; i
40200201
I 25X
5X1
;
X
40200210
3X
18Xi
;
X
40200301
33X
X
5>d
!
i
X
140200310
14X
24X1
X
|40200399
1 X i 1 ! 1 I
137
40200401
24X
3Xj
!
ix
180
40200410
7X
11X1
i*
174
40200501
43X
i xi > ! !
X i :
140200510
10X
1 ioxI ! t -1
x ;
!40200551
1 xi 1 i ' i
i ^
}40200601
25X
4XS
:
.
>„
[ X
40200610
15X
29X1
i
X
155
40200701
44X
: 9X1 :
Ix !
40200706
1X
! X! ; 1
' x
40200707
; 1
ix !
40200710
8X
¦ 14Xj ) :
X |
182
40200801
132X
1 ! ^
1
40200802
Z
1 I !
1 : :
40200803
3X
X3 ! !
1
163
40200810
23X
2Xi
i
1
40200898
2X
|
i
4C200899
z
1
!
II i
139
40200901
36X
14X1
i
Ix i :
40200902
3X
5X1
i
Ix i :
40200903
2X
1 I 1 :
ix j ;
40200904
111 1
ix ! ¦
40200905
1 X i '1
x i i
40200906
I 2X
I
x • i
40200907
i
X j , i i
40200908
i
X 1 : ; :
40200909
2X
i
x 1 ; 1 i
40200910
X
I
X I ¦ j :
40200911
i
x 1 : !
40200912
2X
i
x ! 1 1 I
402C0913
i
x 1 I.I 1
40200914
2X
i
X 1 - ! . j ¦
40200915
I
X 1 ill
40200916
i
x 1 i ! !
40200917
i
X
! I 1
40200918
IX .
4X
i
X
! i
40200919
1
i
X
1 . I
40200920
7X
1SX-
i
X
: 1 1
40200921
11X
3Xj
i
X
i :
13-Sep-93 ti AG Page 47
-------
TEMPORA
.ALLOCAT
on datK sourcJes for
sees III I I I i * !
[ I I
1 1 1 1 ¦' i
FREQUENCY dF OCCURRENCE !N THESE DATlA SOURCES* | 1 !
)
I
1
j ! ; ,
see
DATA
1
i
1
1 ¦ ¦ ;
PRIORITY
GAP
1
1
i
!
RANK
see
CODE
CARB
TACBt
SOS LMOS
CEMi W-T-E ACID-M NAPAP BLSi E/E
CRB> PET
;
!
1 I
;
40200922
25X
] 8»
i
i
lx 1 i 1
40200923
19X
I
1
lx i
40200924
1X
; sx!
i
lx t
I
40200325
lx |
i
40200926
i
1
lx |
!
4020092?
i
l
lx |
40200928 |
i
j
|x L
40200929 |
3X
i t ixl
1
lx i
;
40200930 I
! X
!
IX!
40200931
; 1
I
: X 1
40200998
7X
i 20X1 1 i 1 i x i
40201001
130X
; 26 X3 ! : I ; : ;
40201002
z
i : i ! . ;
1
40201003
IX
! ! i i !
40201004
1X
: II ; i ;
1 160
40201101
3X
X ' 16X1 1 I ! 5! x !
40201103
1X
ill i j • ; < ;
i
40201104
IX
II i 1 1 ! ¦
|
40201105
3X | 1 1 i X 1 ! j
j
40201111
; 1 i i i !x lx !
I
40201112
1X
; 1 1 1 1 lx lx , !
40201113
i
! 1 1 x 1 x 1 ;
40201114
i ]_
' 1 1 X 1 X • ; 1
40201115
i [
: t : X ' X
40201116
i |
! 1 i X 1 X
1
40201199
1X
! 8X1
i 1 1 x | x : i
40201201
z
i
i 1 ! 1 ; ; i
40201210
z
i 1 1 i 1 I 1. ; ' ; !
123
40201301
13X
X ! 6S>3'
; I ! x i x ¦ :
40201303
1X
j
\
| X | X ; i
40201304
12X
1
IX lx i j I
40201305
1X
! 14X
1
1X j X * ¦ :
131
40201399
X 14X
s
¦X
X '
40201401
1X
I X
y
1
]
lx
; ! :
40201402
> 3X
y
i
X 1 j i
40201403
i
x'
1
| 1 > ! <
40201404
2X
X
1
I 1 ! !
40201405
!
1
1 ¦ ! '
40201406
13X
i X *
!
!
i 1 • . i
40201431
1X
! X
X
j
11 ; ; ¦
40201432
|
X
1
1 i i I
40201433
1
*
]
40201434
!
X
1
' '
40201435
2X
i
X
40201436
3X
i
X
1 1
40201437
2X
1
1 1
40201438
I
X
! 1
40201499
1X
i 5X
X
X
1 1.
40201501
sx
X I
X
X ' '
40201502
1
X
1 1
40201503
!
X
X 1 i
40201504
2X
i
X
X I i
40201 SOS
t
X
X 1 t
40201531
i
X
•X i 1
40201599
i
X
X 1 i
40201601
15X
x t
X
i !
13-Ssp-93
B-49
Page 48
-------
TEMPO RA
. ALLOCAT
ON DATA SOURG
ES FOR
SCCs )
I i
1 1 ! ! :
|
| f
j 1 j
FREQUENCY C
F OCCURRENCE IN THESE DAT
A SOURCES * j i
J !
f *
! 1 . : :
scd
DATA
' l '
1 ! ! : :
PFUORm
GAP
j |
II!!-
RANK
see
CODE
CARS
TACB
SOS
LMOSI
CEMl
W-T-E ACID—M
NAPAP
8LS
E/E: CR8
PET
i ;
i i ¦ ;
40201602
7X
X
I :
!x
X
40201603
i x ' x 1
40201604
' ; ' ' ; x ! *
40201605
X
x
X
90
40201606
8X
X
x!
:
:
X
X
40201619
X
2Xl
i
X
X
40201620
X
X
X
40201621
33X
I
X
X
40201622
37X
i :
I :
X
X
40201623
9X
I
i
X
X
40201624
I
*
X
40201625
] 22X
X
X
40201626
| |
-
X
X
. i40201S2T
!
X
X
t40201628
]
•
X
X
j40201629
I
X
X
I 40201S30
X
X
40201631
X
X
40201632
2X
1 i I !*
x ! ;
402016S9
5X
X
X
X"
x i :
40201702
12X
2X
'i
X
X ¦ r
40201703
u
X
40201704
3X
1 i ! ' X j X " :
40201705
!
1X
3X1
x ! X
40201721
XI
>
X
X
140201722
10X
4X5
X
X
I 140201723
7X
3Xj
ix
X
| |40201724
1X
xi
X
X
(40201725
! 3X f
X
X
107
40301726
13X
! SX i
ix
X
40201727
4X
1 2X i
Ix
X • !
40201728
5X
2X1 1
t x
x ; ;
40201731
'
X
X
40201732
5X
X!-
x jx i i
40201733
1
X
X
40201734
X
X
40201735
1 I :
ix
X ' !
40201736
' X i
X
X
40201799
69X
SX 1
i X 1 X ' i
40201301
8X
1 i
X
!
40201803
1 i !
X
X
40201804
I
X
!
40201806
X
X ] 1 i
40201806
9X
2X
X
x j i ;
40201899
1X
4X
X
x !
93
40201901
31X
58X
i
X
x 1
40201903
;
X
x 1
i
40201904
X
x i s ;
40201999
5X
16X
i
X
X | | 1
40202001
7X
4X
i
X
X ' t I
402020Q2
X
XI],.
40202003
X
* t
40202004
i
X
x i i ;
40202005
IX
X
X i 1 . 1
40202031
i
X
X 1 [
40202032
X
X 1 1 i
13-Sep-93
B-50
Page 4$
-------
TEMPORA
„ ALLOCAT
ON DAT)
k SOURC
ES FOR
SCCs I
i
i |
\
1
j | <
FREQUENCY OF OCCl
JRRENCE tN THESE DAT
A SOURCES * j i i
ill !
Is , i
see
DATA
I ill
1 i
PRIORITY
GAP
I
I i
1 ! 1 ;
RANK
see
CODE
CARB
TACB
SOS LMOS
CEM
W-T-E AC ID - M
NAPAR BLS
E/E; CRBr PET
|
1 i
i
40202033
3X
|
! t
ix
X
40202034
|
t i
fx
x 1 : ;
40202099
4X
I 7X
i i
|x i !
40202101
4X
I 7X
i •
Ix j i
40202103
j
i i.
Ix 1 ; i
40202104
i
i !
Ix ]
40202105
I
i i
ix 1 ,
40202106
5X I ! 2>j
1 i
! x ] :
40202107
1X
i
i i
i x i . .
40202108
4X
I 2X
i i
jX S :
40202109
7X
I I I i
! x |
40202131
1X
I xi i i
;x i
40202132
5X ! I 4Xi | ; I i X I
40202133
2X I j
! ' i I ! x t
40202199
sx !' I m
| !
;x '
159
40202201
67X lx ! 15X
: ; ! ! x i
40202202
I- ! X
i ! ] '¦% 1
40202203
12X I i
I ' i ; * !
40202204
1 1
i ' i ;X ' .
40202205
| ] 6X
! 1 ] !x !
40202299
14X ! I 3X
'! ! i !x i
40202301
I I
1 ! ! Ix i • '
40202302
1X I i
1 ! 1 ix i
40202303
I I
: ! 1 | x j x '
40202304
I ! I i ! i ! x ; x
40202305
i \ t t i i
! ! 1 : ! |x t x
40202306 |
2X
1 X i ; ] i x ; X i
40202399
1X
'
; ; ! ; X : X
!
40202401
15X
X 1
1 i | i X IX . ¦
40202402
22X
j
1 ! 1 1 x i x : .
I
40202403
2X
i
i ! 1 " Ix ix
40202404
I
¦¦ r
! |
i X 1 x
I
40202405
I
i
i
1 x ! x
40202406
44X
!
1 x
x ; " :
40202499
36X
!
ix
X
192
40202501
194X
X 1 158X1
1
! x
40202502
17X
X | 5X1
j
ix
;
40202503
3X
>3
I
Ix
40202504
1X
(
|
1 X
40202505
3X
s)d
1
!x
;
40202531
3Xf
1
1 x i '
40202532
2X
6X
•
lx 1 !
40202533
4X
I
1X 1 ! ; ^
40202534
X
i
Ix | ! ! 1
40202535
X
I
X ) 1 1 I
40202536
7X
1
X |l'
40202537
4X
2X
1
X
1 1 :
40202599
18X
35X
I
X
1 1 i
40202S01
16X
1
X
i :
i "• i
40202602
t
\
X
! S ¦ :
40202603
I
X
1 ! !
40202604
j
X
! i 1
40202605
|
X
i j 1
40202606
|
X
1 1
40202S07
1
X
S ! (
40202699
1X
>
1
X
! i ;
13-Sep-93
B-51
Page 5C
-------
TEMPORA
.ALLOCAT
ON DATA SOURCES FOR
SCCs |
I I
1 i ! :
j
! i
1 i ; ' :
FREQUENCY OF OCCURRENCE IN THESE DATA SOURCES • I 1 1 !
I ! I I !
see
DATA
! t
111"'
priority
GAP
! i
l i s
RANK
see
CODE
CARB
TAC0
SOS
LMOS
CEM W-T-E ACID—M NAPAR BLSl E/E: CR8I PET
it I
I
I ; :
40288801
33X
12X1
I ' i I
; • r
40288802
SX
I i ' I
! ! :
40288803
1X
i i I
1 ; ;
40288804
1X
I i ' I
! i i
40288805
1X
j ; |
I ; ¦ :
40290011
z
i i I !¦ ! :
40290012
2
! ! ! I I
40290013
i ax
40290023
! 3X
7x1 j . ! i ! ; .
40299995
19X
i ; 1 ! * ;
4029999S
17X
I 6Xi ' ; ! ! x :
40299997
47X
1 sxl i : ; X :
40299998
! sx
[ 47X1 ! : ; ; * ;
40299999
! 4X
! 13XJ ' ; ' i I
I 40300101
i21X I } ZX , I i ' ;
j 40300102
142X i I' I i - i | i
140300103
i 59X I 1 3X1 i | | i :
f 40300104
i 97X | i 1 : • i !
]40300105
19X I ! 1 i I ill.
T40300106
! 11X
1 2X3 1 ^ t i : ; ¦
140300107
i ssx 1 i 36x1 i | ; ;
I 40300108
hx i i ! 1 : 1 i i
t 40300109
Z
j
i
I s } j j i :
40300110
z
j
i
i . i l I ;
40300111 t
:1X
1 xi j : i : ;
i
40300112
2>3 j
i ;
i
40300113
z
I
1 !
i j :
40300114
z
i
i
1 ; ;
j j !
40300115
z
i
i
[ 1
j j ;
40300116
13X
2X | ;
i i
40300150
¦ 49X
: i
i :
40300151
1 53X
23d | . : [ | i
40300152
1115X
3oxl ! :
! !
40300153
! 23X
1 II !
i i - ; :
40300154
133 X
j j 1 !
I 1 • ;
40300155
z
I
1 if
• 1 1 : :
40300156
|10X
1 3d | . i
It1
40300157
152X
X! |
1 1 ! :
40300158
13X
!
i
| ;
40300159
hx I
I 1
)
i : ;
40300160
13X
i :
1 :
40300161
146X
X
¦
i i
40300198
239X
19X
I
: ; i
40300199
143X
14X
j
: !
40300201
11X
30X
i 1
40300202
373X
25X
1
i 1 [
40300203
SX
7X
I
I ! :
40300204
164X
7X
(
1 ! 1
40300205
z
j
I I
40300207
16X
1
! i
40300208
21X
5
40300209
2X
1
! ¦ 1 ;
40300210
z
1
1 ! !
40300211
z
!
! i ;
40300212
1X
*
40300213
z
1
! i 1
13—Sep-93
B-52
Page 51
-------
TEMPORAL alloc at
ON DAT)
V SOURCES FOR
SCCs
i
|
1
l
j
I
FREQUENCYC
foccI
JRRENCE IN THESE DAT
A SOURCES *
I
I I
II'.:'
sea
DATA
I
i '
I 1
PRIORITY
GAP
I
I
i !
i ! ¦
RANKSCC
CODE
CARB
TAG 8
SOSi LMOS
GEM
W-T-E ACSD-M NAPAN BLSl E/E; CRB: PET.
I I
1 i ¦ ;
40300214
2
|
|
!
40300215
2
I
! |
|. ; ;
40300216
7X
I
I
1 i
40300299
82X
I >1
1
f i
40300302
1X
1 3)4
i
i 1 ;
40301001
14X
1 6X1
i
i 1 i ' - l
40301002
4X
1 1 1 i '
II •
40301003
10X
III i i i 1
|
40301004
j 1X
lli ; i i i
i
40301005 | ! I 1X,
i II ! 1 . 1 1
40301006 I ! I4X
i i 4030100? |
19X
• 5* i I ; :
! 122140301006 !
i SOX
! i 1 = i 1 !
; 140301009
! 29X
i j ; I | j
! 140301010
I SIX ! i l2Xi : I . ! ;
! 140301011
112X ! i . i i I ; ' :
125)40301012
1157X
i 9>3 1 • s j !
140301013
113X
1 j i i i i i : i
|
o
o
8
**
179X
1. I " 1 1 1 :
I 40301015
12X
1 ! * • 1 1 i ;
140301016
12X
! 12X! S ¦ I i I
140301017
4X
! I ! . I j !
140301018
66X
! 9X3 j 1 | I 1
! 140301019
69X
i 32xi i ; 1 i t ;
i
40301020
17X
1 4X
I ! ! i i :
I
40301021
317X
! 28Xl
I : ill
I 201 i40301097
343X
ioxl i ! ! i .
1 !40301098 |
I 99X
i xl • | : I i
I 206!40301099 |
I2439X
7:« I 1 ! ! ¦ ' i
j
40301101 |
113X
i exi 1 • lli
| 199
40301102 |
] 109X
1 1 ! 1 ! 1
i
40301103
87X
III ! ! ! |
40301104
4X .
1 xl ¦ 1 - 1 i |
40301105
8X
1
!
1 I
40301106
3X
III :
40301107
101X
! 23X|
til
40301108
142X
1 2X1
40301109
12SX
1 i
i ! li.::
40301110
240X |
i ;
( 1
40301111
88X I
i .
40301112
18X
40301113
8X
,
40301114
1X
i
! •
40301115
14X
1
i 1
40301116
2X
i
i
; ; ;
40301117
102X
22X
i
! ' :
40301118
4X
!
I- 1 !
40301119
13X
; 1
i 1 t
40301120
5X
19X
1
i ! ' 5
40301130
45X
i
i s ;
40301131
91X
i
i : :
40301132
11X
>
!
i '¦
40301133
ex
j
; ! i
40301134
9X
I
< ! I
40301135
47X
i
i ! !
40301140
125X
!
i t i
13-Sep-93
B-53
Page 5;
-------
TEMPORA
.AILOCAT
ON DAT J
\ SOURCES FOR
sees 1 j
I I I
1 I
1 1 1
FREQUENCY OF OCC
JRRENCE IN THESE DAT
A SOURCES *
1 : :
I
1
1 i : :
SCO
DATA
j
r
!
: j i
PRioRrnf
GAP
|
i 1 i 1
RANt
-------
TEMPORAL allocat
O
z
~
¦-"W-
s, SOURCES FOR
SCCs I
i i ! 1 t
¦
I
I
I I i II!
' I
FREQUENCY OF OCCURRENCE IN THESE DATA SOURCES- III!
I
I i i lit
,
sccj
DATA
I
t
'i 1 III
:
PRIORmt
GAP
[
i 1 ; i ;
RANK
see
CODE
CARB
TACB
SOS LMOS
OEM
W-T-E ACID- M NAPAP BLSi E/E; CRB
PET
|
: i i .
40400160
4X
I
' 1 ! Ix i i
40400161
j
. : 1 ¦ I lx :
40400170
1X
i
! ; ! I ix ^ ;
40400171
i
! i I 1 i 1 * i i
40400199
10X
I XI i s | t ! ! i ' ' •
40400201
4X
! xi I i ' IX
40400202
2X
! 1 i ¦ 1 1 ! x - :
¦ 40400203
29X
i i i ! 1 1 ! x i «
] |40400204
- i 3X
1 xi j l i ix i
I 40400205
! 27X
i
) • 1 i lx i
I 40400206
I 24X
!
I ; ' i 1 *
j • i 40400207
I
1
i , I i ;x
i j 40400208
17X I !
i ! ; ; x
j |40400209
14X 1 !
1 ¦ i ; ! x
140400210
57X ! ! x i . v • . i • lx
140400211 I
! I I
1 ! i ! x '•
40400212 !
! ! :
i : i i lx i ;
40400213 I
i
1 ^ ! ix
40400230
i
I ! i i x ;
40400231
j
!
1 ! ! x
40400240
j 2X
1
1
! : ! ! ix ! !
40400241
I
J
1
I ! !x i
I 149
40400250
1145X
i
l
1 !x j x 1 !
I
40400251
j 43X
i
I
1 ! x •! x"
i
40400254
IIX
I
I
i
j ix ix :
I
40400260
! 9X
i
j
i ! ix
|
40400261
1
1
I ! ! x ' :
i
i
40400270
i
I
I
1 > ix j :
i
40400271
i i
S
! ! ix : i
40400301
1 sex | 1
¦ i ' :
40400302
1213X ! ! I ¦ I ! ! *
40400303
134X i- ! I ¦ I ' ! ' I I ! ¦
40400304
¦37X
! ! ! ' ; ' ! 1 ! ;
j
40400305
17X
[
1 1 j [ 1 ! t . :
40400401
4X
i
1 1 i i* > :
40400402
3X
i
1 : I . I |x I :
40400403
7X
I
! | ! !x
40400404
10X
1
: !x '
40400405
7X
i
! •
i Ix | i
40400406
3X
i
*
j
X | I
40400407
i
I i
i
x !
40400408
IX
i
•
|
x ! 1 ;
40400409
i
|
X i ! !
40400410
23X
i
1
x ! !
40400411
i
j
x j | r
40400412
1
x 1 1 !
40400413
9X
j
X i 1 1
40400414
6X
I
1 1 i
40400497
40X
j
X i | 1
40400498
14X
i '
x r i i
172
40500101
35X
x
I
1
1 i
40500199
5X
32X
1 1
132
40500201
34X
1 21X
I 1
40500202
| 3X
1 l. 1
40500203
z
I
! :
40500211
1X
I
i i >
13—Sep—93
B-55
Page 5*-'
-------
tempohaL allocat
ON DATA SOURCES FOR ISCCs I
i I ' I i i 1
1
[ 1
1 1 ill
" 1
| FREQUENCY OF OCCl
JRRENCE tN THESE DATA SOURCES * 1
I
! ! ' 1 1 ! 1 II!
see
DATA
I
III I I ' ! !
PRIORITY
GAP
I
II! I l ll'i
RANK
see
CODE
CARBt TACBt
SOS LMOSl CEMl W-T-E ACID-M NAPAP BLS! E/E! CRB
PET,
'
i
I I I
1 i i
40500212
)ix
I XI I
1 I
1 176
40500301
85X
X I 43Xi |
1
I ! ! '
40500302
z
i i
i 1
40500303
1X
I
1
i i •
40500304
2
I
1
1 !
40500305
6X
I
(
f
1 I i ; ¦
40500306
2X
III I
- i
40S00307
2
I i j I
1 1 ! i
tea
40500311
34X
x I 31 x! : '
i :
40500312
12X
i 12X1 ! ; : 11
;
40500314
i
! xl 1 : I 1 :
191
405004Q1
146X
1 64X1 I I i ! ;
40500411
! i 7X
1 19X1 ; i ; : :
40500412
I l7X i i 4)4 ! ! ! :
!40500413
!2X f ; l i i ; i
I 40500414
j 4x f i xi i ' 1 ;
as
40500501
i t OX ! X i 8X1 | i i i ;
1
i
40500502
z
1 1 ; ! i : : : : ;
!
40500503
4X i i | j j | j i
!
40500506
IX I ! II'! i I I
i
40500507
z
i i i i i i ;•
I 140500510
z ! i I i ! ! i i i ; :
1. 108140500511
( l'2X I ! 15X3 ! i i i ;
1
40500512
5X j I 9X ! i i |
40500513
isx i I I ; ! i i
40500514
4X I I I I ; • ! I ! :
40500598
sx ! ; 5X5 :
j 117
40500599
10X i 1 9Xt ¦ j ! i I :
I
40500S01
SX
1 3X3 i : i |
j-
40500701
15X
II I • i !
I
i
40588801
5QX
X 1 35Xl j ! 1 1 ! '
40588802
5X
1 X! ! i 1 !!•¦;.
40588303
3X
i 1 | ' I i 1
40588804
2
III : 1 1 i
40588805
\ sxi | i i i | !
40600101
4X
1 xi | 1 | ( i :
40600126
! 9X] 1 ! I 1 1 ! • •
406001 30
1 9X t i | ! ! " • :
102
40600131
95X
1 9X3
¦ " i
1
! I * i ;
406005 32
18X
1 1
i
1 I x 1
40600133
13X
I xi
1 i 1 1
1 j ; X
40600134
SX
I' !
1
! Ix
40600135
16X
1 10X
1
t
! U
40600136
58X
l
1
U ] !
40600137
3X
I
i
|x ;
40600138
28X ,
I
1
X !
40600139
6X
X ! !
40600140
28X
X
1
* ! !
103
40600141
35X
13X
1
X 1 1
40600142
2X
i
X 1 !
40600143
14X
[
!x 1 1
40600144
3X
1
|x 1 i
40600145
j
1
Ix 1 !
40600146
3X
(
1
I* ! I
40600147
3X
1 1
ix ; i
40600148
I
1 1
j*- 1 i
"13-Sep-93
B-56
Page 5
-------
TEMPORA
. ALLOCAT
ON OAT)
\SOURd
ES FOR
SCCs
" 1
1 1 1 1 !
1
I ! 1 I ' ¦ 1
FREQUENCY C
>F OCCt
JRRENCE EN THESE DATA SOURCES * j 1 >
1 ! 1 i j ;
SCG
~ATA
! 1 i III:
PRIORITi
GAP
I 1 ! it:..
rank!
see
CODE
GARB
TACB|
SOS' LMOS
GEM
W-T-E ACID— M NAPAR BLSl E/E: CRB! PET
i
1 1
1 ! 1
1 I
40600149
4X
1 j
(x ! • |
40600160
1 1
Sx i !
40600161
9X
1 | i lx i 1
40600162
1X
1 1 1 lx J i
406001 S3
I i i i t x i i
40600197
17X
I
f
i
1
i ! t ' «
40600198
7X
i xi
1
i i
40600199
13X
I
i
! [ .
40600231
7X
i
I
!
I i i ;
40600232
I 3X
i
1" i ! ! !
40600233
. hx
i
1 ¦ !
1 i . : i ;
40600234
I 2X
it! 1 -i 1
|40800235
z
I
1 1 i ' ! : : !
I40600236
1X
II! 1 i i :
U0600237 1 z !' i i 1 i : 1 i i
40600238
! '5X
! 1 i i ' ' i i i i i '
40B00239
I sx
1 1 i ! I'll
68
4C6CQ240
I lax
! 1 1 1 II:.
40600241
2 I I j
1 I i
l ' :
40600242
Z
i i I
1 i i
i i
40600243
10X
1
' 1- I 1
• ! i
40600244
4X
1
! i i
| ; ;
40600245
Z
s
1 1 t
;¦ . ,
40600246
10X
1
i i 1
i :
40600248
12X
j
! !
; ; ;
40600249
sx i i
i • i
40600250
2X ! !
1 ! 1
1
40600251
IX ! ! i ! ' 1
I ! :
40600253
2X I ! It ! ¦ !
j
40600254
z
ill! ' !
! : I 1
40600255
z
1
t ! ¦ j ' ! 1 :
40600256
z
1
1 i 1 1 ¦ '
' I '
40600257
IX
1
1 1 j 1 i i
40600259
24X ¦
1 1 1 1 1 !
40600298
22X
1 I - ! 1 ! I
40600299
13X
1 i ill!.
40600301
1X
1 x
1 i i i
40600302
2X
i
i i i I i i
40600305
z
I
i ; : i i . ;
40600306
IX
I
! ! I ! ! ; :
40600307
12X
40600399
3X
; ;
40600401
1X
i
40600402
z
| • ; ;
40600403
z
1 i i
40600499
z
j
1 t I
40688801
3S0X
X
1
; i
40688802
3X
1 1 1 1 !
40688803
1 1 1 . 1 '
40688804
1X
1 1 1 1 . I ' !
40688805
9X
II 1 ' 1 t
40700401
9X
¦ I 1 1 1
40700402
z
1 1 ;
40700497
4X
|
f 1
40700493
5X
i
i 1
40700801
24X
1
( j. :
13-Sep-93
B-57
page 5
-------
TEMPORA
L ALIOCAT
ON datX sourc
ES FOR
SCCs 1 |
i i
ill!
1 '
; i
FREQUENCY OF OCCURRENCE IN THESE DAljA SOURCES * 1 : I !
!
1 1 1
1 : ;
see
DATA
i
1 1 1
[ * ' '
PRIORmi
GAP
i
) 1 1
1 1 ! >•
RANH
see
CODE
CARB
TACB
SOS LMOS
OEM) W-T-E ACID—M NAPAfl BLS
E/E: CRBi PETi
j
i 1 ' 1 1
i ; :
40700802
19X
j
ii ! I |
i ¦'
40700803
7X
t
1 i
i i
40700804
z
t
] 1
I |
i
40700805
1X
1
! i
i |
j
40700806
16X
i
i 1
! !
i
40700807
3X
[
i i
1 !
40700608
3X
!
1 !
I i ¦
40700809
35X
! 3>d
! '. j : :
40700810
17X
i ¦ 3X
j : | j ;
40700811'
19X
i
! ! ! j i ¦ :
I 40700812
5X
l
1 • 1 ! 1
140700813
31X
: ZA 5 • | ! •
[40700814
27X. j • XS i i i !
i40700815
54X ! • 5X > " ! i >
!407CQ816
89X 1 6*3 i ;
I407QQ817
sx i i 1 ! ; . ;
I 40700818
4X ! • ! : ! ; 1
40700897.
101X
: j i = 1 i "
- —
40700898
40X
i •
i ]
40701601
z
1 ¦
! 1 : : :
40701602
z
i "
1 ! i !
40701603
z
1 ;
! !
: i ...
f ; ; i
40701604
z
1
i I
if-:
40701505
16X
i
1 i j i i
40701606
SX
i 1 ! : ' I j ! , i
40701607
z
I I ! i . i :
407Q1608
z
I ' !
! - si
40701609
z
I ; j
40701610
z
I
; ( IS!;
40701611
36X
: ill ;
40701612
12X
I" ; ! i :
40701613
65X
i
1 i i . | 1 : !
40701614
22X
I
I ;
1 1 ¦ : ' :
40701697
48X
i
1
i i r
i ! ; ;
40701698
53X
;
| ;
I ' 1 " : i :
40702001
z
!
| i
1 1 i 1 :
40702002
z
)
f •
111!'
40702003
z
1
1 i
1 i 1 , i
40702004
z
1
i i
! 1
40702097
26X
1
' 1 ;
II.!!
40702098
42X
)
1 j ,
40703201
2X
i
,
i ! ! 1 i
40703202
6X
i
i
i 1 i 1
40703203
13X
I
• i •
i i > !
40703204
20X
i
j
1 i . i
40703205
4X
i
li!;
40703206
7X
i
1 i
¦
40703297
116
1
1
40703298
71X
i
i
4
40703601
22X
i
I | !
40703602
23X
i
40703603
3X
1
i
I 1
40703604
24X
1
i
t
i l !
40703605
15X
1
i
; i ;
40703606
7X
I
I
! i !
40703607
z
1
1
1 1 i
13-Sep-93
B-58
Page 5"
-------
TEMPORAL allocat
ON OAT
h SOURC
ES FOR
sees III I i I !
I
I
! I
) 1 i i
1
FREQUENCY OF OCCURRENCE IN THESE DAT|A SOURCES » !
: |
II t 1 li
i i
scd
DATA
1 1 i 1 fill,
PFutmm
GAP
Ill I I ! I . i ;
ranh
see
CODE
GARB
TACB SOS! LMOSj CEMt W-T-E ACID-M NAPAfl BLSi E/E; CRB PET.
1 i
i
l 1
40703608
z
| |
i
li.;
40703609
11X
i
i
! i
40703610
SX
i
! 1 -i
40703611
z
i
[ I •
40702612
z
1
V
! \ . \
40703613
16X
I
* !
40703614
41X
I
I
i
: :
40703615
|19X
1 1 !
i ; .
40703616
131X
1 ! I
! 1 • !
40703617
lax
! I I !
i i i
(40703618
j 3X
II! !
| S
140703619
!3X i ! i l ' ; •
|40703620
j1 ox i i : i • « ¦
140703621
! !ix i ! ! i ; !
140703622
hex
! I . i ; j :
!40703623
113X
f I i = I :
I 40703624
133X ¦
I I i : \ ^ ¦ i
|40703697
152X
! i i i ! i i
I4Q703698
1138X
I i t i i l :
40704001
i40X
! >2 t ' ! : ;
40704002
113X
i i ! i ¦ I ;
40704003
2
I
»
i
; ; . 1
40704004
133X
I
! ! 1 - i ¦
40704005
Z
I
! I ! i l : !
40704006
2
l
i
i ! i > • ! '
40704007
Z
i
111 1 i ' 1
40704008
3X
ill i 1 ! : ;
,40704009
1X
1' 1 : 1 ! 1
40704010
¦z
I I i ' j ' i '
40704097
16X
1
! i lit
40704098
13X
¦ |
1 1
j ; t
40704401
3X
III 1
1 1 • 1
40704402
4X
j i
1
i
! i ¦ i
40704403
6X
1 1
1
I j .
40704404
8X
ill !
I ! ! i
40704405
8X
i
1 !
1 S * 1 1
40704406
2X
s
j i
i t
40704407
1X
I
1 ! t [ I ' ¦ i
40704408
5X
, i
! i i i i
40704409
1X
1
1 ! 1
40704410
Z
j '
1
1
L i
40704411
IX
1
1
40704412
Z
1'
; i
40704413
1X
1
i i
40704414
2X
1
! I
40704415
z
! 1
40704416
3X
1 i 1
40704417
5X
1 1 ! 1
40704416
13X
1
I 1 1 I
40704419
12X
I
1 1 ' 1
40704420
17X
1
! 1 ' I
40704421
z
i
j I
| i
40704422
2X
j
) 1 : I
40704423
z
! I : 1
40704424
z
! .
II-1
40704497
23X
i
1 1 1 1
13—Sep-93
B-59
Page £
-------
TEMPO RA
.. ALLOCAT
ON DATJ
V SOURC
ES FOR
SCCs
|
i I i
! 1 i
t
i
I I I
i
FREd
3UENCY OFOCC
JRRENCE iN THESE DAltA SOURCES *
i I i
-
! I i
! 1 !
see
DATA
I
|
1
PRIORITY
GAP
I
i
1 ' 1
RANf
see
CODE
GARB
TACBl
sosl
LMOSj OEM
W-T-g ACID—M NAPAR BUS E/Ei CRBi PET:
| I
1
1 ! i
40704498
20X
t
|
i s ;
40704801
2X
I
|
I I !
4Q7048Q2
7X
!
1 1 i !
: i
40704837
23X
I
I
! S 1 1 t
4O704898
20X
I
I !
I | ! 1 !
40705201
3X
i
t i
i i
» ! S
40705202
z
I
I I
til!
40705203
4X
I I
I i i I
40705204
z
I ! 'I i j ! ¦ I
40705205
2X
I
i I ! i I ; ! !
40705208
z
i I i
40705207
z
i
"it i
i : ; :
40705208
j 3X
| | : *
! ; :
40705209
. 3X
I
f
i ¦
1 1 ;
40705210
19X
. i
i
1 1 ' i
40705211 i z
|
i ,
I
! ; i 1
40705212 I 7.
f
[ •
1 i 1 l!
40705213
z
I
I !
! 1 i i i
40705214
z
I
I
t ! ! !
40705215
z
I
! i
i
i
1 1
40705216
6X
I
I !
1
1 1
40705217
7X
I
I I
I
! 1 ' :
40705218
3X
I
I I
I
1 1 !
40705297
66X
I
I i
1
j j ;
40705298
34X
I
I I
!
1 1 1
! • 1
40705601
4X
I
I !
j
j j :
40705602
1X
I
•I !
i
I | ;
40705603
46X
i
i : '
1 1 l i !
!40705604
27X
111 I
1 1 > 1 !
40705605
5X
! I
j
! i ¦ ]
40705606
6X
I
|
(
1 1 t
40705607
IX
I
! i i
40705603
z
I
i
i i '
40705609
15X
I
1
t i :
40705610
13X
i
i
1 I i
40705697
33X
i
1
f 1
40705698
53X
!
i
40706001
1X
i
I
1 I i 1
40706002
z
I
j
! I 1 !
40706003
5X
I
!
!
; ;
40706004
14X
I
j
t .
I
40706005
6X
|
i
I
j
40706006
15X
|
I
I
I
]
40706007
5X
|
t
40706008
SX
I
40706009
4X
J
\
I
40706010
2X
!
40706011
z
I
40706012
2
I •
40706013
21X
i
1
40706014
29X
i
40706015
1X
I
40706016
z
!
40706017
z
I
40706018
2X
1
40706019
5X
I I
13—Sep—93
B-60
Page 5i
-------
TEMPO RA
- ALLOCAT
ON DATASOURC
ES FOR
3CCs 1 1
I
I I ! i i
I
I ! ¦ i I
FREC
WENCY OF OCCURRENCE tN THESE DAT
!A SOURCES * | i | i
-
i
I ! !
see
DATA
!
PRIORITV
GAP
i
. i ;
RANK
see
CODE
GARB
TAC8I
soa
LMOSS
GEM
W-T-e ACID—w
NAPAF
BIS!
E/Ei CRB! PET:
1
| *
I : I
40706020
22X
»
i
i i
I > i
40706021
z
!
i i
|l:j
40706022
2
|
i !
i i ' i :
40706023
12X
1
I i
II ! I
40706024
8X ) i
i i
i i , ;
4C706097
36X j f
i i
I i
I I :
40706096
59X t i
I i
| ( : ;
40706401
27X I I
! I
I I
40706402
2X
I
! i
40706403
8X
•
j I
I !
40706404
Z
i
111 i ! i i
40706497
!9X
i
I : III
i 40706498
! IX
i
• ! i I i ! :
i 40706801
13X -
!
i i. ! ! ;
i40706802
I 19X
• ; i
40706013
hx
I-
i i i i i
40706814
5X
i
! ! I
i i
40706897
40X
I : I ! i
40706898
57X
i
I i
| i I
40707601
z
1
l i
I i : ;
40707602
10X
1
t I
I l ¦ i
40707697
ax
;
! ;
i i : i
40707698
5X
|
l i
i |
i i :
407080Q1
z
1
i I
j I ; ;
40708002
z
!
t i i j i . i
40708097
10X
i i I i ;
40708098
1X
I
i j ! ;
i
40708401
SX
1 ! ! ^ I i i
40708402
4X
! 1- ! !
!
40708403
33X
i 1 ! ¦ !
! I
40708404
125X
! 1 i i i i i ;
40708497
21X
! 2X1 ! ! I i ! ; !
40708498
17X
2Xi
.
| ,, '
I i
40717205
7X
!
i i ' i
40717206
3X
i
II ; :
40717207
4X
12X
I
I I ' . . i
40717208
5X
12X
I
I : i :
40717209 |
8X
!
i ¦ :
40717210
z
I
I
40717211
SX
1
i • ;
40717212
z
!
t
I ; I
40717297
7X
1
I
I i ! ;
40717298
5X
I
I : i ;
40717601
9X
I
I i I !
40717602
ax
I
I I ;
40717603
2X
I
I I
40717604
11X
I
I ! I !
40717605
z
I
! ! i
40717606
z
i
; | |
40717697
34X
!
¦ 1 .
' 1 !
40717S98
32X
I
1 1 1
40718001
3X
i
1 1 :
40718002
z
i f !
40718003
z
I
; i i
40718004
z
I
I 1 1
40718005
z
i
I
i 1 i
13- Sep-93
B-61
Page 6-
-------
TEMPORA
<
O
o
<
ON DAT
\ SOURCES FOR
SCCs I j
1 I" 1 i 1
I I
1 1 1 1 ; i
FREQUENCY C
F OCCURRENCE EN THESE DATiA SOURCES *
1 •
! I if
I
scd
DATA
! I II
i
• PRIORITY
GAP
II I -I
'' !
RANH
see
CODE
CARS
TACbI
SOS
LMOS
CEM W-T-E ACID—M NAPAFI
BLS
e/e: crb pet.
j
! i I
•
4071800S
z
I I t
I
40718007
z
I i I
|
!
40718008
z
i ¦ ! ! !
4071SO09
z
I I I III'!
407 "5 8010
l
)
i I i I" !
40718097
) 122X
I i I ! i ;
40718098
I 12X
! i [ II::
40720801
I Ux
i I | : ;
40720802
z I !
i : j j i i
140720803
! 12X
i ' X)
i ; ' i ! i
• |40720804
I ! 6X
i X i ! I ! ! ;
|40720897
I I 13X
I i I ! i : :
140720898
1 ! 3X
'• I i ;
I40722001
1 1
: ; s ' :
140722002
1 1
Z 1 ]
1 : I ! !
40722003
z
I
i : 1 i 1
40722004
box
i i i i i
40722005
14X
i
I
1 : ; i :
40722006
z
|
|
I ! 1 ! 1 :
40722007
z
j
I
! i ' ' ¦
40722008
! IX
I I
I !
40722009
z
I -
! I
| ! j
i 1 1 : • ;
140722010
1 IX
I
! ii| :
40722097
- 1 15X
I I ! i ! i i . :
40722098
|3X
lit ! I ! ! • !
40722801
: 14X
I I ! s !
!. ; ;
40722802
jl5X
II; ; !
1
40722803
12X
I
1 : !
i
40722804
¦ I4X
j
! i' i
1
40722805
11X
|
i i
40722806
| 2X
|
! ! 1 i i
40722897
]3X
t ! 1 • 1 i ;
40722898
12X
1 i
1 ' :
40723201
z
I
• . . 1 I
1 1 ;
40723202
z
I
i :
1 i !
40723297
z
I i
1 i ! 1
40723298
z
1 !
1 | ; ! i
40781201
z
1 i
1 1 r !
40731202
z
i |
i !
i I
40781601
1X
1 !
1 ¦ i
40781602
25X
1 i
1 :
1 J :
40781603
1X
i
! ;
40781604
25X
[
; 1
4078160S
27X
i-
! j
40781606
5X
i
! i
40781607
2X
1
! 1 i
40781699
121X
1
1 1 1
40782001
66X
I
1 ! ;
40782002
7X
1
i S ;
40782003
1X
|
1 1 ¦
40782004
z
|
1 1 :
40782005
6X
1
! 1 1
40782006
17X
1
i i
40782007
10X
I
| i •
40782008
z
1
! i i
40782009
3X
i
1 i 1
13-Sep~93
B-62
Page 5
-------
TEMPORAL ALLOCAT
ON OATi
\SOURCES FOR
SCCs
t 1
! 1
j |
II : :
FREC
iUENCY OF OCCURRENCE EN THESE DAT
A SOURCES* III1
I i
j
|
i i
see
DATA
1
j
I i 1 I
PRIORITY
GAP
II! !
1 1 • !
RANK
see
CODE
GARB
TACB
SOS! IMOS CEMi W-T-E ACID-W
NAPAP BLS| E/Ei CR8I PET;
I ! ! ! ¦ .
II ! ' ¦ l
40782010
2
|
I I ! :
1 1
40782011
Z
I
i i j :
1 1 :
40782099
54X
! I . ; I
1 ! ¦
40782401
z
i
' ;
i" i
40782499
z
I
! I
i -• !
; I : :
40783201
1X
i
i
i j
I ]
! i
40783202
5X
I
| t
1 ( .
40783203
SX
I
! !
i 1 ;
!40783299
2X
|
| t
i i i : j
I 40784801
10X
I
I i
;
|40784899
17X
I-
!
|40786001
6X
|
1 i
j 40786002
11X
i i.
I i 40786003
! 7X
; i t
I 40786004
]2X
i. 1 < !
|40786099
119X
1 j : j • !
140786401
z
!
i 1 ' i i i
j
I 40786499
3X
i 1 i i i
s
j 40787201
1X
i
; : t.
:
40787299
6X
I
i ! | •
t
40799997
382X
l
" 1 1 1 ! : !¦ '
40799998
1596X
I x
40899995
) 188X
i
i ! i i i i . !
40899937
96X
1
• ¦ i } :
40899999
I 259X
1
1 j I'll .
L
49000101
z
I
t
i ; ! •1 •
49000102
z
j
1
! i ; :
I
49000103
z
\
{
! i I ! •
49000104
z i r , ¦
; < : ! ;
49000105
z
I
! i 111 ;
49000199
7X
ill; 1 i ! :
49000201
3X
1 38X1 | j | I !
49000202
5X
1 4Xi | ! 1 II::;
49CCC203
2X
1
1 1 1 III:!
49000204
1 4X
! ! 1 1 i ¦ i
49000205
2X
1 33X
1 1 1 | | i . ! I
49000206
6X
49000299
135X ,
1
j ;
1 i
49000301
!
I '
! lx
49000302
1
j \
i l'x :
49000303
1
! i
\ ' X ;
49000304
1
j x ;
49000399
26X
!
X . !
49000401
z
!
! 1 i
49000402
z
1
1 :
49000403
z
|
! ! i
49000404
2
1
1 1 !
49000405
z
1
i S 1
49000439
10X
1 '
I 1 i
49000501
2
1
i 1 1
49000502
Z
i
i
49000503
z
1
1
s i I
49000S04
z
1
1
: i j
49000599
1X
1
1
i i i i
49090011
2
1
1
i ¦ j i
49090012
Z
1
1
i i i i
13-Sep—S3
B-63
Page 6
-------
TEMPORA
L ALLOCAT
ON DATA SOURCES FOR
sees I I
1
1
t !
I
! |
I.I
1 1 :
I
FREQUENCY OF OCCt
JRRENCE IN THESE DAllA SOURCES *
: ;
I
. j
! 1 !
, I '
see
DATA
I
I
^ 1
i
PR'.ORfTi
GAP
|
i 1
i :
RANK
see
CODE
CARB
TAG a
SOS LMOS
CEM
W-T-E ACID —M NAPAP
BIS! E/Ei CRBi PET
1 . : ,
49090013
16X
: 1 i i ;
49090021
z
1
1 1 : i
49090022
z
,
i i
1 1
49090023
27X
|
I I
1 ( !
49099998
60X
| 15X1
| . ;
i 1 : !
49099999
167X
X j 12X1
i :
I r
56
50100101
15X
1 8X
3x |30x i
| | i ;
144
50100102
4X
| XI
3x i 26x
! ! !
50100103
|
3x I ! 1
79
50100201
13X
!
x ! ! ! ; ;
50100202
I 4X
x !' : ! i ;
50100401
> 1X
j
3x ; I ! 1 ;
50100505 j
hex
1 11Xl 3X i ' 1
50100506 j
132X
i »3x ! ; :
50100507 | ' | [
1 13x I i !
50100508
|
! |ax i ! i !
50100510
11X
1 13x 1 : 1 ! 1
50100511
I
! 3X X ! • !
'50100512
IIX
1 . ¦ 13x ! ¦ 1 ! ! ' '
50100515
i
] 13x | !
1 1 ¦ i
:5010051o
1
! j 3x !
j i ;
50100517
i 13X !
i l
50100601
2X
II:
i I
[50100602
2
ill 1 ! i 1
|50100S03
z
1 i 1 ! 1 i i
150100604
z
ill ; I i i
48l50100701
17X
• 241 x! I i | | i ¦ : .
I
50100702
1X
I i" I ! :
50100703
11X
I i l i :
50100704
IX
I i ; I : ;
50190005
2x I : I'll':
50190006
2x | :
j ! ; '
50190010
2x
j
I j
50200101
81X
sx
9x !
1 r
50200102
13X
5Xi 3X
15x j
¦II 1 !
50200103
9X
loxi 2X
I
i ! :
1 1 ;
50200104
z
1
i
1 !
1 i !
50200105
2X
xi
s I I i ; i
50200201
12X
1 i ¦ i
1 ! , ;
50200202
2X
' 1
i
| j : s :
50200301
1X
9»
I
1 1 i i i
50200302
1X
SX
!
I I ! i
50200505
57X
8X
X
I
I
1 1 i ! ;
50200506
4X
i
1
1 ' t 4
50200601
z
i
i ( ;
50200602
1X
154X
i
: i •
50200901
z
i
j !
50290005
z
i
i i i
50290006
z
i
1 ;
50290010
z
i
50300101
S5X
17X
16x I
50300102
63X
1 9X
54x !
\ 1
50300103
6X
1 SX
1
i
: i :
50300104
3X
|
1
1 1
150
50300105
2X
1
1
! !
50300106
25X
1 X
1
1 1 i
13-Sep-93
B-64
Pagt &
-------
TEMPORAL allocat
ON DAT* SOURCES FOR
SCCs
i i
1
! 1
:
FREQUENCY OF OCCURRENCE ttJ THESE DATA SOURCES *
1
|
j . ; ,
SCQ
DATA
i
'
PRIORiTV
GAP
I
|
HANK
see
CODE
C ARB
TACB!
sosl LMOS
CEMl W-T-E ACID—M NAPAR
BIS
E/E! CRB PET
i
|
i
1 (
50300107
i
i
2X5
1 r
50300108
| 2X
1 i | j 1
it ill
50300109
z
i
1
1 !
50300201
I 2CX
i
1 j
50300202
I 2X
i
I 1
[50300203
z
I
I j
1 1
!50300204
z
I
i I
| |
i 50300205
2
I
! i ill
150300501 |
lex
1 1 1 1 1
(50300506 |
! 5X
3X
1 ! ' i i
|50300601 |z
I
It ' i . i i
I50300602 )
15X
¦ | . ; i |i . ¦
i i50300603 |
I1X
I ' i : , | i ;
150300701 I | !36X
i XI
! i 50300S01 i | i 64X I
| i50300810
! 12X
50300820
lex
j
i
50300830
iix
I
i ;
j
50300901
z
!
I
; !
1 i
50390005
z
i
|
1 i 1 1 i
503S0006
I 4X
I
! i 1 i 1 . ;
50390010
z
|
I
! ! I i I
I
|
: 1 II
|'
I
1 1 1 ' I 1
i
I
! : ' i
I ! * X is equ
lavent tc
one occurrence in the applicable data source !
1 I z denotes
a data gap for the applicable SCO ¦ j [ ' :
13-Sep-93
Page 6-
-------
APPENDIX C
BLS DATA
C-l
-------
1
i
i
I i
5353
si < et el
2 3 s S
a a
lit
iZZ
1 i
»
s
15
2
5;
3 S
S d
5
d
d d
d
ts£
d d
*»
*.
55
s
a
S3
S
a
d S
«s
<0 4
«
5
« 5
3S
2
2
SH
2
2
H 5
d rf
d
d
3 3
el
- is I
3 2*2
Si
i
S3S
Hi
ass
5 3 3
13 3
9 9?
pt
t f a
i S S
9
a S ®
f S fi
lis
SIS
3 3 2
i 1 a
ill
1 i %
sie
«4
8 £ t
i S2
122
SSS
£ S £
8 S 5
is 2 Si
III
S i ;
sip
is:
ill
is;
St 3 2
£ £ 2
133
s 3 a
1
i a i
333
333
a
3 3
3 3 3
3
a a a
3 S 3
*33
s
i 5
* 3 *
• •3
=
i - —
• H
i 2 s
•
313
||2
333
333
*5£
3 S 3
S3?.
333
a as
is.
a s;
s s s
s g a
tst
SSi
ft 3 n
is£
3 3 8
a 2 ?.
S s 5
3 3 3
511
8 8 8
8 8?.
4s*
8 f 3
3 3 3
ass
333
3 S
3
3 3 3
3 3 3
I i
3
3 3 3
Si 3
33 $
S9i
as;
ss
1
3 3 g
ass
a
a s s
££3
S3 S
£ 3 8
£* £
3
SS3
3*3
i i
si t*
833
?tal
*41
2*1 %*n
i 33
S i s
5* S
3 S S
?.l ?
Ill
8 3
t
5
a
s
a s
i
R
3
s
£ 3
3
?
i
S
3 3
? 9 S
333
13 3
3 3
3
ill
3 3
3
3 1
3
331
333
3 3 1
3 2
sss
f « ^
^ s
S
S R8
s ?
5 ?.
£
3?,g
ssg
|| i
9S?
* 2
S - •
-d ¦»
3SS
*15
¦
i sf
2
i
S?!S
3 3 3
3 ? |
3 3
333
s
a
a
a *
3SS
i
s
I
S S
S ? s
s
"
•
?2s
313
§ « S
*22
3 3 1
3 * 3
* 4 ^
3 3 3
3*3
113
las
s is
3 3 3
a a 3
sua
3 3 3
sss
313
33 1
3
e a 3
S«3
S 5? 3
i a 3
E
" H SI
*¦
g s a
ial
? ii *
21:
3 £ 1
*p:;
ll8
333
3 3 3
3 3 3
3 9 3
339
3 8 S
13 3
3 3 i
3
511
t a i
SHS
333
s a *
3 a s
3 SI S
3 5 3
S3?
? t m
tti
Ss-
i c i
2 fi :
2Kb
sc;
p a
s s i
sss
S R 2
3 3 3
SSi;
? t s
9 3 9
1 lis
2 3 SS
31 S
323
3 2 S
I |1
ils
3 3
3 3
3
9
g
3
. 3 3
3
§1
a
f 3 S
3 8 S
S 1
S 2
3
2
s
3 S
9
32
o
S 2 S
3S3
£ i
; £
5
2
Si
s:
£ 2
3£
i
a;
lis
13 3
a a a
3 12
SB?
123
5 2 3
323
|H
all
2S;
Si;
• iS
£ i 5
i i >
3 31
3SS
3 3 3
332
333
ISi
ill
g a a
i 2 i
SIS
1 2 £
3 3 3
isi
38 3
iSi
3 3 3
3 2 3
~
3 3 1
3 11
3 3
i
ISi
15 3
1 £
s
^ ~ 5
?! 2 ¦
S S
3 1
3
SSi
a 3
i
If 1
3 i
a
i 3 a
IS
a
lis
r- «*
$
si £
«
?rJ5.
\ 3 3
111
3 3
3
3 3
3
3 3 §
333
3 33
3 3 3
1 i
a
5 3
S
s 3 a
33 a
a g a
3 3 3
t *J
S
~ ¦
a
S ^ I
SiJS
S5w'»
113
'is
8 3 1 3 13
3 3S 5 3 5
£ i * t 2 5
sss
U3
SSS
111
55?
>>>
% J SJ
a - a
I '
i
§'
*
'¦*
P
X
a
3
- i'
1 2
1
0 |
& 1
c
5 '¦*
5
ass i£* ssi ass ass at> in sis
Ipl III SIS IIS 311 III III III
lis
MS
* s
C-2
-------
1 i
5S52
d d d c
SSSS s£5£
d d d d dd4
is;
HI
is?
lis
119
a(
s~y
in
Hil
3.3 B
S3*
ss -
III
ill
Hi
111
111
lis
in
ill
Ssa
StSS
III!
ill *M
3sS sal
S 5 3 =*s
sis
lii
I||
195
211
i i'i
lii
til -==
i?g ill
in
5 » s
in
ill
in
111
ill
Si*, st
lis
ill
S 5 i
sag
8 So
all
111
S 3 *
5
tt
it
|l
i|
i s
ii
ii
«
1 5 a
sit
ill
ill
11 A
5 J J
|il§
liS'I
ill
3 si a
;as
ill
I S 9
5 = 5
S-*
*3 3
| i||
ilii
nitt
Si #
333
f,i,x
ill
S 11
}.$ =
Ill
a a e
a s a ¦
- 5 * *
9SS
see
a a a
|s|
Q ( a
135
Hi
® 5 3
III
*4 S3 ri
fill
111
5-3
is:
Us
SIS
111
ISI
III
• 2 * ?.
pi
b a s
ill
lis
5Sa
3
3
i
3 a
Ii
S 5
SiS;
HI. ISI:
lii;
HI.
S 5
SS
uaa*:
> ? > >
1311 Hll
?«S
a-jaia
5?
so z
I if I
m B *
jS S £
i 5 at
< £ -5- t-
s 1 a
s s i
§ 2
| |
2 1 2
1 <
2 c
j= &
n *
1 1
C -«
C-3
-------
1 1
1
1
i
l
1
I
I
I 1
ll ill! I!!!
IIII
III ill! 1111 IIII IIII IIII IIII IIII
| 1III fill
III!
in isii mi mi mi nil mi mi
| IIII III!
III! III! III! 111! III! fill nil IIII III!
| 1111 IIII
III!
in mi mi mi mi mi nil mi
1 1 III
Ill
155
!!
111
IIS III
III ill
HI II!
¦i HI III
III
II
HI
11!
ill
HI
HI
Hi SSI
i III I!?
HI
ii
II!
Ill
ill
!S1
111
?sj
HI IS!
s IS! HI
III
I!
HI
HI
HI
HI
11!
HI HI
I III III
SII
IS
IS!
111
ill
IIS
Hi
IH 111
! Ill III
III
11
Ill
III
HI
III
111
111 !!S
s 111 11!
11!
ll
III
HI
ill
III.
III
ill IIS
i III III
IS!
II
III
HI
HI
Ill
III
III 111
i III III
!I!
I!
HI
HI
HI
111
HI
HI HI
i HI II!
IIS
II
IIS
ill
HI
HI
HI
SSI ¦ 155
i IIS SII
III
IS
III
III
III
II!
III
IS! Ill
i 111 III
11!
II
I!!
111
111
!H
Hi
HI IH
1 lis us
Ill
Si
til
111
Hi
!!!
HI
HI HI
lis II;
III 11?
Sis
111
S S
n
ill
CIS
III
til
Hi
£2 s
III
U;
III
ii: fcSs
III 111
iihlH!
MINI
I 11 M i
I I i 5 I i
Hi in
ill it I
S! I I
z
5 !
r ?;
i I
f t f I I
= i i i i
,111111111
i! II I if is I I! I
I i
I i M I i I! I i 11 I
5i ? I 1 I 5 I 3 i i I 5
is ii is is s | ri i
mini
C-4
-------
APPENDIX D
GARB DATA
D-l
-------
z-a
l 5
E
u
? ? I : i
r ' ¦ 1
n
1 ? r < ^ ? p f ¦
) ; S f C ! I f I
;u;!u S[ S
3 C s ©
> > > >
5 5 5 =
P O P O P Bp
I it s! £!
3eo*«7
rr
i
¦ ?
u
r r
< I
ft
| r
i
*s"
:il]
4 = s J
t£8*
ru-
£ I \ £
-tit
( . {¦ r
p'*
' s
i
i
?
f -:
e 5 ;
>-F t
{ I f » g F
if
' } 1
i . Ef .
5 L i ; i
iXfrf *
n
? 5.
i?
|
e
?
a
fi
P?
£ <7
?r
; ru ? r
rlitH
s |i sM
I ^Isr|
i 3 11 i ¦
KlHf
Vf!
| - g P P r g
iiiiii
e o g b e
till'
P P
!!
I
£ £¦
i
: I
p p p
| | £
p p p
|2|
p p p
III
pee
eee
eee
o e e
*
s
Z"
a
t
1 1
5 5 5
5 5 5
I
I
p p p
|S|
p p p
2||
p p p
2 2|
eee
eee
o o e
fi
e
>»
2
p p p
III,
p p p
2 2|
p p p
2 2 2
e e e
P C B
eee
e b e
e
?
3
p p p
IIS
p p p
£ | |,
5 5 5
p p P
£ 2 2
eee
e e o
o e e
2
~
p p P
III
p p p
III
p p p
|2|
o e e
b o e
see
...
e
p p P
||S
p p p
III
p p p
2 2 £
eee
eee
eee
s
» • »
p p p
III
p p p
||£
5 5 5
p p p
£||
b e o
e p e
eee
COO
« S S
» » »
5 5 5
o e 2
P
sow
e o £
p
e e m
s
p p p
2 £ 8
5 55
p p p
III
p p p
III
A 2
p
BOM
P
e o x
e e £
a
p p p
III
» » »
p p p
2 2 2
p p p
£2|
p
o e n
P
POM
e e £
P
e e m
s
p p p
III
p p p
III
p p p
|2|
e o 5
e e £
P
e e •-
P
O P N
s
p p p
|S|
p p p
III
p p p
2||
e b £
P
e £
P
COM
=
p p p
III
p p p
2 £ 2
p p p
£||
p p p
|2|
p p P
2||
p p p
2||
B P P
e p e
e e o
p p e
5 5 5
5 5 5
5 5 5
e e o
P P o
o e e
eee
=
p p p
III
P p p
III
p p p
2 2 2
5 5 5
eee
B B B
eee
eee
=
p p p
III
p p p
III
p p p
III
eee
e o e
& e o
-
p p p
III
p p p
2||
p p p
||2
eee
Bee
eee
eee
p p p
2||
p p p
III
p p p
|E|
pep
o p e
"•
eee
=
p p p
III
p p p
2||
p p p
2 E 2
5 5 5
p p p
III
p P p
S|2
S « E
p p p
2||
eee
eee
e o o
eee
s
p p p
2 2 2
p p p
III
p p p
| 2 £
e p e
eee
o e e
eee
E
5 5 5
p p P
2 £ £
p p p
2 2 2
5 5 5
Hi
P' P o
eee
...
pee
5 5 5
P P p
||£
5 5 5
p p p
||2
p p p
2 2£
eee
...
o e e
p p p
III
p p p
£||
p p p
||2
e P P
eee
eee
eee
=
p p p
£ 2 2
p p p
E||
p p p
|2|
eee
eee
e e P
e p p
li
5 5 5
i i £
i i i
11 i
e e -
© o —
e e -
e e -
lr
Hi
ill
in
p p i
£ 2 !
6*
*
B
s
s E
: B p
: I E
ffi:
uU
si It
f =E 5
N r s
rlif
rf
ni
FM
E
*
§3
£P
-------
APPENDIX E
SOS DATA
E-l
-------
(Page 2A)
(Complete this page for each Process Code}
jcess Code
DATE
ACTUAL OPERATING SCHEDULE
INPUT/PRODUCTION
RATE
GAL.
voc
EMISSIONS
LBS.
HR/DAY | START TIME
END TIME
"July 15
24
00:01
24:00
90.59
178.33
July 16
24
00:01
24:00
76.89
129. 18
July 17
24
00:01
24:00
77.76
137.04
July 18
8
.. 07:00
15:00
7 . 12
28.46
t
July 19
IS
09:00 '
24 :00
28 .00
61.28' (
July 20
24
I
00:01 ( 24:00
95. !8
20 1. 1 1 |
July 21
24
00:01
24 :Q0
102.66
205.98
July 22
24
00:01
24:00
106.94
180.78 !
July 23
24
00:01
24 :00
105.40
181.25
July 24
24
00:01
24:00
91.03
183.77
July 25
17
07:00
23:00
. 7.62
32.05 !
July 26
-0-
-
-0-
-0- j
July 27
24
00:0 1
24 :00
106.82
172.58
July 28
24
00:01
24 :00
126.47
' 305.97
July 29
24
00:01
24:00'
96.06
210.20
July 30
24
00:01
24 :00
102.41
194.31 J
July 31
16
07:00
23:00
29.4 1
83.74
August 1
9
14:00 23:00
1
5.37
19. 15
August 2
-o-
-
-
-0-
-o-
August '3
24
00:0!
24 :QQ
98.16 -
1
196.. 99 |
August 4
24
00:01
24 :00
107.68
189.88
August 5
24 •
00:01 ,
24 :00
83.68
12 1.86
August 6
24
00:01
24 :00
112.8!
240. 13
August 7
24
¦ 00:01
24 :00
58.84
76.50
August 8
16
07:00
23:00
15.79
48.62
August 9
7
08:00
15:00
9.32
26.46
August 10
24
00:01
. 24:00
87 .78
144 .73
E-2
-------
Lv i L. 1 I I^WWWSJW 1111
(Page 2A continued)
(Complete this page for each Process Code)
cess Code LIQ
. DATE
' ACTUAL OPERATING SCHEDULE
INPUT/PRODUCTION
voc
HR/DAY
START TIME
END TIME
RATE
EMISSIONS
•ugust 11
24
00:01
¦ 24:QQ
90.3
113.82
^ugust 12
24
00:01
24:00
112.9
244. 10
vjgust 13
24
00:0!
24:00
99.96
200.54
august 14
24
00:01
24 :G0
83.92
172.21 ¦
August 15
¦ ! 6
07:00
23:00
"52.46
162.20
Xugust16
-0-
-
-0-
-0-
August 17
16 '
07:00
23:00
83.33
138.62
August 18
16
07:00
23:00
62.24
132.07
August 19
16
07:00
23:00
70.64
•
95.87 J
August 20 -
16
07 :00
23:00
ait. 86
15 1.07 1
August 21
16
07:00
23:00
70.68
118.2 1 i
August 22
7
16:00
23:00
6.74
9. 13 1
August 23
16
07:00
23 :Q0
7.95
30.25
August 24
| 16
07:00
23 :0Q
100.03
209.6
August 25
16
07:00
23:00
125.49
3 14 .8
August 26
16
07:00
23:00
82.77
143.56
August 27
16
o
o
r-~
o
1
1 23:00
89.46
174.33
August 28
16
07:00
| 23:00
84.29
153. 16
August 29
16
07 :00
23:00
59. 16
157.45
August 30
i6
07:00
23:00
37.41
1 18.'50
August 31
16
07:00
23:00
I 10.85
230.25
E-3
-------
SCS - GAILY PROCESS 1NFCFMA7K3N
facility name
L9cfch*«d
COUNTY PLANT
1180 0027
CITY
Manefta
1S92
DATS
Jwly 15
July 19
July 17
July 19
July 19
Juty 20
Ju*21
July 22
Jwfy23
July 24
Ju* 25
July 26
July 27
July 29
Juty2«
July 30
July 31
August 1
August 2
August 3
August 4
August 5
Aug us! 6
Auguai 7
August 8
August 9
August 10
August 11
August 12
August 13
August 14
August 15
August 18
August 17
August 16
August 19
August 20
August 21
August 22
August 23
August 24
August 25
August 26
August 2?
August 2S
August 29
August 30
August 31
weekday
Operas Hours
start stop
0
0
0
SATURDAY
Op«f«cing Houfs
SWT STOP
PROCESS
UO
SUNDAY
OperaTing Heurs
STAflT STOP
24
24
24
24
24
24
24
24
24
24
24
24
23
24
24
24
24
24
24
24
24
24
24
23
23
23
23
23
23
23
23
23
23
22
0*1* Production R*t«
WEEKDAY SATURDAY SUNDAY
90.59
76, as
77.78
15
95.16
102,69
106.34
105,4
91.03
106.62
126.47
96.C6
102.41
23.41
96.16
107.66
33.66
112.41
56.34
57.78
90.3
112.9
99 36
53.92
S3 33
62.24
70.64
54.66
75.66
100 03
125.49
62-77
69.46
64.29
7.12
26
3.67
6.74
AVERAGE 2.47C565 21.S47Q5 9,265714 21,85714 7.75 21,25 31C6.23 154.76 82.85
ASSIGNED 2 -24 9 22 8 21 0.929531 0 $46255 0.024712
profiles:
weekday
Taction
0
3
C 0*5454
G.04S-4S4
0.045454
0.045454
0.C45-154
O.C4$454
0,045^54
0.045454
0.045454
11 0.045464
12 0.045454
13 0.0*5454
14 0.045454
15 0,045454
16 C.G4S4S4
17 0.045454
0.045454
0.045454
Q.C4*ft54
0.045454
Q.C45454
0.045454
Saturday
fraction
0
0
0
0
0
0
0
0
0
0.076323
9.076323
0.076S23
0.076923
0.076923
0.07BS23
0.076S23
Q.076S23
0.076923
0.076923
0.076323
0.078323
0.076323
0
0
Sunday
fraction
0
0
Q
0
0
0
0
0
0,076323
0076323
0.075323
0.C7S923
0.076323
0.C75923
0.078323
0;076323
0076923
0,076323
0 075923
0.575S23
0.07SS23
0
0
0
I
dai^r
w««May: 0.31429
Sax: 0.00356
Sunday* 0.0013
0.99983
E-4
-------
APPENDIX F
LMOS DATABASE FORMAT
F-l
-------
ructure for database: C:\LMOS\DAY_PT.DBF
.tiber of data records: 64543-
-e of last update : 07/01/93
aid
• Field Name
Type -
Width
Dec
Index
1
STID '
Character
2
N
2
CYID
Character
3
N
3
FCID .
Character
15
• N
4
STKXD
Character
. 12
N
5
DVID
Character
12
N
6
PRID
Character
12
N
7
POLID '
Character
5
N
8
DATE
Character
8
N
9
TMZN
Character
3
N
10
DYSP1
Numeric
7
4
N
11
DYSP2
Numeric
7
4
N
12
DYSP3
Numeric
7
4
N
13
DYSP4
Numeric
7
4
N
14
nvoDt:
W fcaF •*»
Numeric
7
4 •
N
15
DYSP6
¦ Numeric
7
4
N
16
DYSP7
Numeric
7
4
N
17
DYSP8
Numeric
7
4
N
18
DYSP9
Numeric
7
4
N
19
DYSP10
Numeric
7
4
N
20
DYSP11
Numeric
7
4
N
21
DYSP12
Numeric
7
4
N
22
DYSP13
Numeric
7
4
N
23
DYSP14
Numeric
7
4
N
24
DYSP15
Numeric
7
4
N
25
DYSP16'
Numeric
7
4
N
26
DYSP17
Numeric
7
4
N
27
DYSP18
Numeric
7
4
N
28
DYSP19
Numeric
7
4
N
29
DYSP20
Numeric
7
A
N
30
DYSP21
Numeric
7
4
N
31
DYSP22
Numeric
7
4
N
32
DYSP23
Numeric
7
4
N
3 3
DYSP24
Numeric'
4
N
Total **
241
F-2
-------
nature for database: C:\LMOS\PROCESS.DBP
•Jaer of data records: 6616
e of last update ; 07/01/93
Id Field Name
Type •
Width
, Dec
Index
1 STID
Character
2
N
2 CYID
Character
. 3
N
3 FCID
Character
15
N
4' STK1D
Character
12
' N
5 ' DVID
Character
12
N
6- PRID
Character
12
N
7 . SCC
Character
8
N
Total **
65
-------
11 s.s|»i|IississSs;Ssi
>o«ooaaat«0«sedaoo'oe
liillllillplllllslll
ijiiiSiiiiiiil
iliilliiiiililliiiiil
oOcoc
^VWUBBMBSBBOOOClOVOO
I illllllSlllljflllill
I j'llllllllill IIIIIIII
;»ese««>.eo(ioi>oaooe»'
'•iii'A'.
ieeeeda.oa-egtDdeec"
1ifllllilllllllllill!
xsDOSBH^tfl.soiloBeoe*
. oeoooB.oo-oc^ctDoeo'*
i illlllillllllliillll
. o a o* o o <* - e o - e o * e 0 9 e o •*
i i||lilll|plllllllll
iiillllillilliiiliiii
. ® Cto t>' aa-o O ¦.«.'« a o O o ©"»-
iiiiiiiiisiii
| » iide»eiiMgtpaeogeio
iilllllljpjipilslll
• e e o 9 o » . a n* -< e o o a o o o o o
Iilllllillllllliillll
;«aeeesBefieada
i iljlllliitllllllllll
.ssdBsd.oiitoo'dedggod
illllllllillllliilll
;ooooeooooeoe9e«Qoeo
i!II!IIIeIIII|I!!s!§|
i
: c»0
i iMMiMIMfUiiiiu
¦ e»»»09«ago«t»ggsetie
=itlsSlz:iiHI|ii;ils5
» © o o D oi o o «> o o o c' O O O O D o" o"
iillllllllfllllllllll
¦ ft«ddtoiiaso»«»»d«9e»
*!llllilil|lflil|l|||
Miiliiliillliliillii
l|
3.;slSIIJIIIIIIIilffll '
£|ipSIS!S|i!£|ii8pSpS| y
32:
sssisss:sis;::::::;
li
n\
iiiiii
!!!!!!
S2S5S2
^llllllllltllllllill
:SSSSS2SSSS222SSS22S
2!
li
iillillll i
5 sL
I c:ssss?::::$£issli:i:
5 s:s2ss2s:ss:sssi5s:s;
EEEEt-EitEEEEEE-EEE
!Is;SIlI:;S!:!l
iii
I °!
F-4
-------
APPENDIX G
CEM DATA
G-l
-------
Typical Kentucky CEM Data Output
{KYCEM92)
KENTUCKY DIVISION OF AIR POLLUTION CONTROL
CONTINTJDUS EMISSION MONITORING SYSTEM
RAWEATA REPORT
PAGE: 003
DATS! 07/19/93
C72-2 46 0-00 0 5
CO. NAHZj TV*-ENVIRONMENTAL AFFAIRS
PLANT-NAME-. SHAWNEE PLANT
COMP STAT: VIOLATION# PROCEDURAL
PERMIT NUMBER] O92049
STACK-JEM 03
UNIT-ID J I
PARAMETER« 42401 502 TOlTSi SOOHDS PER MILLION 3TU
ANALYSIS METHODs COAL ANALYSIS
COLLECTION METHODi EXTRACTIVE
DATA FOR: AC3CST
LAST CATS DATED;
1912
1230?
01
02
03
94
OS
06
06
IARTIH0 SODil...
10 11 12 13 14
IS 1$
17
21
22
WJH DAILY
23 OSS MEAN
01
82
03
04
05
06
0?
99
09
10
11
12
13
14
15
IS
1?
13
1*
20
21
22
23
24
25
26
27
28
29
30
31
.50 0.56 0.
.5a 0.50 0,
.5fl 0.53 0,
,57 0.56 0,
.57 0.52 0,
,41 0.51
.54 0.54
.55 0.43
.55 0.59
.52 0.53
.SO 0.65
.53 0.48
.51 0.57
.71 0.64
.53 0.56
.55 0.56
.53 0.53
.54 0.55
.58 5.55
.$7 9.68
.52 0,55
.55 0.53
.52 0.57
.54 0.50
.54 0.55
.53 0.52
.51 0.53
.SS 0.54
.54 0.54
.55 0.55
.57 0.56
54 0.53
48 0.48
54 0.54
56 0.53
50 0.53
63 0.54
56 0.55
43 0.44
66 0.70
54 0.54
5? 0.50
47 0.59
56 0.57
53 0.55
52 0.56
47 0.59
49 0.56
47 0.56
55 0.55
60 0-SS
54 0.53
55 0.57
53 0.48
4 9 0.55
55 0.53
55 0.55
54 0.54
57 0.53
55 0.55
54 0.55
50 0.56
0.5Q
0.47
0.54
9.54
9.S2
0.55
0.56
0 .43
0.53
0.55
0.64
0.65 G
0.46 0
0.54
0.52
0.53
0.56
0.57
9.55
0.57
0.55
0.53
0.44
0.53
9.56
0.54
0.54
0.57
0.S4
0.56
9.54
.53 0.
,47 0.
,57 0.
.49 0.
.55 0
,57 0,
.50 0.
,44 0.
,52 0
,54 0
66
.52
. 5 6
60
34
54
54
56
53
56
55
45
53
56 a
54 0
53 0
54 0
53 0
58 0
53 G
55 0
46
47
52
51
51
75
40
42
49
54
65
55
46
66
49
47
50
35
48 0
47 0
46
54
59
36
49
33
54 0
55 0,
47 0,
46 0.
47 0
53 0.
55 0,
45 0,
47 0,
34 a.
60 a.
40 0.
55 O,
51 0.
64 0,
62 0
47 S,
56 0,
51 0,
56 0.
54 0.
66
76
54 0
46 0
93 0
52 0,
53 0,
56 0.
55 0.
42
5S
76
63
55
25
54
54
53
61 0,
55 0,
53 0,
91 0,
52 0.
53 0.
55
88
61
64
62
51
45
89
11
54
40
92
53 0.
53 0,
55 0,
56 0,
57 0,
3S 0.
55 0.
28 0,
54 9.
94 0.
56 0,
54 0,
56 0.
28 3,
56 0.
84 9.
58
53
56
60
87
49
55 9
65 0
62 0
54 0
94 0
80 0
55 0
57 9
64 0
51
54
57
53
55
39
56
47
52
72
75
52
51
46
53
52
48
55
63
55
51
S5
52
71
31
54
75
59
55 0.
55 0,
46 9
52 0.47
55 0.53
58 0.54
54 9.57
06 9.46
35 0.59
51 0.57
55 0.62
59 0,53
61 0.60
75 0.68
54 9.55
54 0.55
59 9.74
65 0.49
52 9.52
63 0.52
49 0.53
60 9.58
56 9.S3
54 0.S7
54 0.54
56 0.53
61 0.55
30 9.4?
54 9.59
56 0.59
64 0.50
55 O.Sl
54 0.57
39 9.47
49 0.48 0
53 0.54 0
57 0.40 0
59 5.€9 9
58 0.51
56 0.53
58 0.52
55 0.52
56 0.35
61 0.54
50 9.49
53 0.55
47 0.50
32 0.45
56 0.53
54 9.54
60 0.51
53 0.59
4 0.53 0
54 0.64 9
63 5.53 9
69 3.57 9
58 0.45 0
60 0.44 0
58 G-51 0
55 0.53 0
63 0.56 0
48 0.54
51 0.53
60 9.65
46 9.43
47 0.25
53 0,55
50 0.56
54 9.47
34 9.42
54 9.41
46 9.22
55 0.56
SB 0.50
61 0.59
55 9.S3
66 0.44
48 0.50
44 0.35
52 0-54
79 9.51
55 9.43
58 0.44
53 0.54
54 9.58
65 0.81
66 9.64
38 0.28
61 0.54
56 0.56
57 0.52
53 9.54
0 . 54
0.53
5.40
0.52
0.56
0.57
0.55
0.44
0.52 0.
0.64 0.
0.33 0,
49 a
56 0.
56 D,
54 5,
9.63 D
0.56 D
0.34 0
0.57 9
0.46 5
0.42 0
0.37 0
0.55 9
0.58 0
0.91 1
0.71 9
0.34 0
0.53 0
0.52 0
0.56 0
0,57 0
53
54
49
79
49
52
52
68
60
73
42
57
52
55
56
51
38
56
58
58 0
.54 9,53
.56 9.55
.53 0.48
.35 9.63
.52 9.53
.S3 0.S4
.45 9.33
.45 9.57
.55 9.57
.85 9.82
.64 9.48
.56 0.50
.56 O.S0
.57 9.56 0
.53 9.55 9
.59 0.58
.34 0.62
.56 0.53
.61 0.71
.38 0.39
.64 0.64
.64 Q.SS
.53 9.54
.53 0.45
.54 9.60
.47 9,51
.64 0.56
.56 0.53
.56 8.54
.54 0.53
.55 9.53
.52 24
.52 24
.59 24
.54 24
.40 24
.52 24
.56 24
.64 24
.57 24
.77 24
.49 24
.55 24
.72 24
.58 24
.54 24
.55 24
. 62 24
.56 24
.73 24
.58 24
.49 24
.56 24
.54 24
.59 24
-B1 24
.63 24
.56 24
.55 24
.52 24
.55 24
,55 24
0.52
0.52
0.54
9.53
9.52
9.53
0.52
0.50
0-53
9. 65
9.57
9.53
0.55
9-56
0.54
9.55
0.57
0.51
0.58
9.56
0.57
0. S3
0.53
9.58
0.62
0.58
0.56
9.61
9-54
0.54
0.57
N 31 31 31 31 31 31 31 31 31 31 31 31 31 31 31 31 31 21 31 31 31 31 31 31
MH 9.56 0.55 9.54 0.55 0.54 0.54 0.59 0.56 0.56 0.59 0,64 0.59 9.57 0.56 9.55 0.54 0.52 5.54 0.59 0.53 0.57 9.54 0.55 0.S7
MAX 0.80 9.68 0.68 0.79 9.65 0.-66 0 .75 9.75 0.79 1.22 1.25 0.94 0.91 1.06 0.74 0.S9 9.67 9.70-0.81 9.91 1.04 9.35 0.82 9.77
MAXIMUM OBSERVATION
MONTHLY MEAN
NUMBER D? OBSERVATIONS..
1.25
0.5S
744
G-2
-------
Typical Pennsylvania CEM Data
Output
01/09/93-
-0027-
¦0022-
'0031'
-0026"
•0023"
•0G26'
'0029-
-0041-
•0041"
•0046'
'0048'
-0046-
¦0044-
¦0046*
• 0047•
•0048'
-0051-
¦0051'
¦0044'
¦0049'
"0046"
¦0042'
¦0046'
¦0045
01/10/93'
•0045-
-0025'
-0029-
•0022'
•0022*
•0030'
-0047-
-0054'
•0056"
•0056"
•0057'
¦0055'
•0053-
¦0056-
•0055'
¦0057"
¦0058"
•0057'
¦0061'
¦0059'
•0059'
•0057'
•0051'
•0048
01/11/93-
-0036'
-0030*
'0029-
•0022-
•0029'
-0041'
¦0043-
-0052-
•0053'
•0053'
•0055-
-0052-
•0053'
¦0053'
¦0054'
¦0049'
¦0048-
¦ 0055 •
•0059"
•0056-
•0054'
•0056'
•0053 •
"0049
01/12/93 •
•0047-
-0046-
¦0050'
¦0051"
•0051-
¦0049-
'0052-
¦0057-
•0054-
- 0057 *
-0057-
-0055'
-0049-
¦0051'
* 0051 ¦
¦0050'
•0053'
• 0057 •
•0056'
¦0055*
•0060'
¦0056'
•0053*
•0054
01/13/93-
•0026'
¦0021"
¦0030"
-0024*
-0024-
•0033'
•0045'
¦0050«
'0047 •
•0049*
•0054-
'0056*
-0053-
'0056'
"0057"
-0056-
-0056*
•0054'
•0050*
•0050'
•0064'
•0064'
•0062'
•0054
01/14/93-
¦0056-
¦0051-
-0055'
¦0054'
¦ 0055¦
- 0058•
¦0071'
¦0065-
•0058"
•0042'
¦0046'
•0045*
-0057-
¦0061'
¦0065'
•0059"
-0070-
•0069'
•0067'
¦0065"
¦0066"
•0065'
•0063*
'0042
01/15/93-
'0044*
•0045-
•0048-
¦0045-
'0051¦
¦0067-
•0062-
-0068-
•0066'
•0065"
¦ 0068•
-0069-
'0064'
¦0067-
•0063'
-0060-
•0058"
¦0059'
• 0059•
¦ 0058¦
¦0050-
•0060'
•0063 •
•0055
01/16/93-
-0033'
-0025'
'0034'
¦0030'
-0033'
-0051-
• 0057•
•0057 •
¦0060'
•0063"
¦0063 •
-0062-
-0061-
¦0061'
'0056'
• 0052*
• 0052 •
'0056*
•0056'
•0056'
¦0059'
'0057'
•0054"
•0053
01/17/93'
•0049-
¦0029-
"0033"
•0026*
¦0031-
¦0031-
•0031'
¦0046'
•0044-
• 0047•
•0050'
-0050'
•0044-
-0044-
•0045'
¦0047'
•0044'
¦0047'
•0045-
¦0046'
•0047"
¦0048'
•0047*
•0044
01/18/93-
•0034-
•0031'
¦0039-
-0036"
* 0040 "
'0044'
*0047*
¦0055'
•0056'
•0060*
•0063-
-0062*
•0060"
¦0059-
•0059'
• 0068•
•0064'
•0068'
•0059'
¦0053'
•0060"
'0052'
•0050-
•0049
01/19/93'
-0051-
-0054-
-0053-
•0056-
¦0064'
-0065-
-0050"
¦0067-
•0062-
•0061"
¦ 0059 •
•0063-
-0060-
¦ 1121¦
¦ 1121•
¦0061-
¦0047'
•0057-
'0061•
'0057'
¦0059'
¦0063'
•0061"
•0055
01/20/93-
•0050'
•0045-
-0052-
"0053-
•0052'
•0056-
•0048-
-0054-
•0051"
•0051"
¦0053*
'0053'
•0043"
"0051'
•0059'
•0050"
•0049'
¦0052-
¦0064'
•0055"
¦0053*
¦0051'
-0044"
¦0043
01/21/93-
-0042-
¦0032'
•0040"
"0038-
¦0040-
-0049'
•0051"
¦0047-
•0034'
•0035'
•0032"
-0035*
-0035-
¦0035'
¦0036'
-0035-
•0036'
• 0042•
•0049"
"0058-
•0053"
'0064'
"0057-
¦0051
01/22/93'
¦0056-
¦0048-
•0056"
¦0040'
¦0054-
-0055-
'0058*
¦0059'
•0054'
•0059-
•0063"
•0060'
-0054-
-0058"
¦0061'
'0063•
•0066-
¦0065'
•0065'
¦0065'
•0060"
¦0063'
-0055-
¦0053
01/23/93'
•0050'
'0047'
¦0053"
-0055-
-0055*
•0051-
-0057-
-0062-
•0062'
•0061-
•0063'
•0063'
-0052-
•0043-
¦0046*
¦0047"
'0036:
•0046'
•0046'
¦0050"
¦0057'
¦0042'
-0047-
'0055
01/24/93'
-0055-
¦0049'
- 0045•
•0043"
-0042-
•0041*
'0040'
¦0046-
-0054-
•0049'
¦0050'
•0053'
¦0052-
•0055"
¦0054'
¦0052"
'0056*
•0058"
¦0059'
¦0060'
¦0061-
¦0061'
•0053'
¦0035
01/25/93-
-0036'
•0036'
¦0043'
-0042-
¦0043"
•0043'
'0050-
¦0060'
•0066'
•0069'
•0056*
•0058'
¦0061'
¦0053"
¦0050"
¦0050"
•0057'
¦0062'
•0064'
'0063'
'0066'
•0063-
-0060-
¦0052
01/26/93-
•0048-
¦0052-
-0060'
•0064-
•0066'
•0062'
'0058'
•0056'
•0064'
• 1115'
•0061'
•0059'
-0056-
•0056"
•0055"
¦0054"
¦0054-
•0054"
•0058'
'0057'
¦0053'
•0055'
"0054-
¦0048
01/27/93'
•0050-
¦0049'
- 0054•
-0053'
•0053-
-0051-
¦0057'
¦0061'
•0057-
• 0059•
•0061"
¦0062'
•0056-
•0057'
¦0060"
•0058"
¦0059-
•0060'
•0061."
'0059'
•0059"
•0059'
"0054"
¦0048
01/28/93'
-0048-
•0040'
•0046-
¦0045"
•0046-
-0052'
•0060'
¦0061-
- 0057 •
•0063*
• 0062•
•0059"
¦0054'
•0059"
¦0060'
•0061•
¦0062*
•0061'
¦0061*
¦0063*
¦0069'
•0060-
•0061"
¦0050
01/29/93-
• 004 8 -
-0051-
-0055'
•0054-
-0054-
•0056-
-0055'
¦0067-
•0061'
*0062-
•0068'
•0064•
¦0058'
'0062'
¦0062'
¦0062-
¦0065'
•0066'
¦0062'
•0053'
•0054'
•0064-
"0062"
¦0055
01/30/93'
•0056-
-0053-
- 0055•
-0057-
•0057"
-0052-
-0050-
-0050'
•0054"
•0052*
•0054'
•0053•
•0050-
-0054-
•0054"
"0051"
•0053'
• 0052 •
¦0056'
•0054'
•0054"
-0055-
•0047'
¦0051
01/31/93"
-0049-
•0043 '
-0050-
-0059-
¦0054'
-0052'
-0053-
-0050-
-0049'
'0044'
•0040-
•0046*
-0047-
•0050'
¦0056'
-0051-
¦005l-
¦0052*
¦0056'
' 0057•
¦ 0060'
¦0061*
¦0053'
¦0039
02/01/93-
¦0034-
•0032-
•0038'
•0037-
¦0038-
-0051-
-0050'
-0059"
¦0063-
¦0062'
• 0050 •
•0047-
-0047-
•0051'
¦0054*
•0054¦
* 0045•
¦0046'
"0061"
•0051"
-0057'
¦0061¦
'0054•
¦0055
02/02/93'
•0059-
¦0052-
•0054"
- 0053•
•0051-
-0058-
-0067-
¦0063 '
•0052-
"0053-
-0055-
-0052'
•0052-
-0064-
*0059'
¦0060"
•0051•
•0048'
¦0050*
¦ 0050'
•0050'
¦0052'
-0055-
¦0049
02/03/93'
-0051 '
•0050-
- 0053 •
•0053'
-0057-
-0053-
•0051¦
¦0050'
-0050'
¦0051-
¦0051-
•0050'
- 0049•
•0047-
•0046'
•0042"
¦0042'
•0041'
•0045'
•0050'
¦0050*
•0049-
-0044-
¦0044
02/04/93-
-0044-
¦0037'
'0039*
-0034"
¦0038*
¦0042-
•0049'
-0060-
•0050*
-0059-
•0062*
•0062'
¦0055-
-0050'
•0076'
-0070-
•0055'
•0053'
•0055"
•0054'
¦0052"
•004 4*
•0051'
•0047
02/05/93'
-0041"
-0037'
¦0043"
-0042'
•0045"
¦0048-
-0053-
'0057'
-0054-
¦0055-
- 0053 •
•0052*
• 004 8•
•0050'
¦0050'
"0047-
¦0043-
•0046*
•0054'
•0054"
¦0054'
-0052-
¦0051'
¦0030
02/06/93 -
•0037'
•0031-
-0037-
¦0036-
•0040«
¦0042-
-0048-
•0046"
-0052-
-0057-
•0057'
-0054'
• 0045•
•0043-
•0040'
'0038'
•0038'
•0051"
•0054'
¦0052'
•0054•
-0056*
¦0055*
¦0055
02/07/93-
¦0049"
-0042'
¦0048-
-0049-
•0050'
-0052'
' 0051'
-0050*
•0049'
-0052'
•0050*
• 0050"
•0047'
-0045"
•0035'
¦0035-
¦0037'
" 0 040"
•0043*
¦0043'
•0045'
•0042'
¦0039'
¦0032
02/00/93-
¦0029-
-0029*
'0036'
•0036'
•0045'
-0043-
•0045*
¦0055'
•0053'
¦0057*
•0056'
•0054-
•0056*
•0060'
¦0060'
"0053-
•0051-
¦0050'
¦0056*
¦0054'
¦0053'
•0056'
•0046'
¦0051
02/09/93-
¦0052'
•0049-
-0053-
"0055*
-0057-
•0058*
'0053-
¦0057-
¦0061*
•0062'
•0050-
*0051'
'0052'
•0051"
•0049'
-0051-
*0059*
¦0060'
¦0056'
'0060'
'0059'
•0056*
•0050*
¦0043
02/10/93-
'0039*
-0035-
•0037*
•0037*
-0047"
•0055-
-0062'
-0054-
¦0052'
-0053-
•0054*
"0054-
-0060-
'0058*
•0056'
-0052-
¦0051-
¦0055*
¦0063"
'0061'
"0056•
'0059'
¦0058'
¦0054
02/11/93'
'0038'
-0035-
-0037-
¦0031-
¦0033*
-0040-
-0049-
•0056'
•0059"
-0060-
•0060*
"0055"
'0057'
¦ 0060•
•0060"
¦0060'
¦ 0058 ¦
¦0061'
•0060'
'0056*
•0057*
¦0058'
¦ 0058¦
¦0056
02/12/93-
-0045-
•0047'
•0047-
-0045-
¦0045*
-0052'
•0052'
• 005 B •
• 0056¦
-0055'
•0057'
-0057-
¦0054'
•0057'
¦0054¦
¦0046'
¦004 8'.
•0057'
¦ 0050•
•0058'
"0056"
•0054'
¦0053'
¦0040
02/13/93-
-0030-
-0033'
-0038-
'0035'
¦0041-
-0050-
¦ 0053 1
•0044 -
¦0045-
-0052-
-0055'
"0052-
¦0048'
•0040'
"0049"
•0046'
¦0048-
•0052*
¦0052*
•0040"
•0049'
•0047'
¦0049*
•0033
02/14/93 -
-0026'
•0027-
-0032-
-0028-
¦0031'
•0031-
1002 8•
•0027'
•0032*
¦0041-
¦0038-
-0041-
¦0032'
•0031*
"0032-
•0029'
•0034'
¦0046*
•0047'
•0044'
¦0042'
•0042-
¦ 0039•
¦0029
02/15/93-
-0026-
-0023 '
•0030-
•0026*
¦0030"
-0034-
10037 -
•0034-
•0035'
-0040'
•0040'
•0040'
-0039-
•0042'
•0042'
•0042'
¦0041"
•0045*
•0053"
•0054"
•0055"
-0052'
¦0044'
•0032
02/16/93-
¦0025-
-0025-
•0032"
•0032'
¦0034*
-0035-
•0036'
•0037-
-0037-
-0043-
'0044 '
•0049'
-0065-
•0062'
¦0058*
•0052'
•0052"
¦0055'
¦0052'
"0053-
-0048-
•0047*
¦ 004 9¦
•0050
02/17/93'
•0048-
-0045*
-0046-
¦0041 "
'0030-
-0053-
-0055-
-0056-
•0054-
-0055*
•0056-
¦0053'
•0048-
-0053-
•0052*
•0052"
• 0052•
-0053-
•0040'
¦0045'
•0044'
¦0046-
¦0047-
•0047
02/18/93'
-0042-
•0046-
-0042'
-0048-
-0049-
¦0052'
•0052*
-0059-
'0057-
- 0053•
-0057-
-0061-
¦0046*
-1114-
¦0042-
¦ 004 B•
•0050'
¦0047'
•0052-
¦0060-
¦0053'
•0056*
¦0058*
•0052
02/19/93-
-0054-
•0051-
-0053-
•0055-
¦0055-
-0054-
-0054'
•0056'
-0052-
¦0053'
•0051'
•0053-
¦0051"
•0049'
¦0057'
• 0059¦
•0059'
-0061-
- 0061•
"0057"
¦ 0059•
-0056-
-0057-
•0053
02/20/93'
•0056'
-0052*
-0053-
-0053'
¦0052'
¦0053-
¦0052-
•0054'
•0052-
-0053-
-0056"
'0055'
•0053'
¦0051'
•0054*
•0051¦
•0053'
-0053-
¦0053*
"0054"
'0055*
¦0052-
•0054*
•0051
02/21/93-
-0049'
-0039'
'0041"
¦0036-
¦0036"
-0038-
•0046"
¦0049'
-0045'
¦0046'
•0046'
•0047*
•0045'
¦0043'
•0052'
¦0055'
•0056'
-0054'
•0051•
"0051'
•0049'
¦0046'
•0041'
•0044
02/22/93 ¦
•0042-
•0040-
•0044'
•0042'
•0042'
•0041'
•0060'
•0060'
•0065-
-0067-
•0066*
-0063'
¦0061"
-0058-
•0052'
¦0058'
¦0051-
•0052"
•0049'
•0049'
•0047-
•0043'
¦0044"
-0046
02/23/93-
•0044-
•0041*
•0046"
-0046-
-0052"
-0057'
•0053-
•0047-
•II21-
-1121'
• 1121•
¦ 1121 •
'0047'
•0048"
¦0050'
¦ 0054 •
¦0057'
"0057"
•0057'
• 0052•
•0054"
• 0057 •
•0057-
•0055
02/24/93-
•0061"
* 1114'
•0061"
¦0067'
•0063'
¦0061-
-0059-
¦0060'
•0072-
-0071-
¦0068'
'0066'
•0060*
¦0061"
•0062'
•0055"
¦0058'
"0060"
¦0059"
•0061'
•0062-
¦0059'
•0058'
•0059
02/25/93•
-0062'
•0061 '
¦0054'
-0055*
•0060¦
•0054'
•0053*
-0054¦
'0067'
•0069'
•0062-
¦0051'
•0051'
•0058'
¦0053-
•0055'
¦0055"
"0053"
•0053"
•0051*
¦0051-
•0051'
•0056'
•0052
02/26/93'
-0040-
-0043-
-0045-
•0042'
-0041"
-0057-
¦0060*
•0059'
•0060-
•0057'
' 0059 •
•0061-
¦ 0059•
•0059'
¦0055'
"0046"
¦0048"
"0046"
*0047*
'0046'
¦ 0046•
'0046•
¦0050'
•0051
02/27/93-
•0047'
•0040'
•0049-
•0057"
'0053'
'0050'
¦0053*
•0053•
•0051'
•0049*
¦0048'
•0044-
¦0039*
• 0040•
•0034'
¦0035'
¦ 0041 "
•0038'
•0048*
•0050'
"0051-
¦0049-
¦0039'
•0040
02/28/93'
-0043-
-0037*
•0044'
¦0044'
¦0047-
¦0050'
¦0050'
•0049"
-0050-
•0050'
'0042'
•0044-
• 0045•
-0048"
'0036'
¦0030"
¦0033'
¦0034*
¦0040'
¦0041"
"0041*
-0042-
¦0041'
•0042
03/01/93-
-0051'
•0048-
•0056-
-0054-
¦0052'
•0052-
* 0059•
•0055"
•0058'
•0057-
¦0054'
'0049'
¦0047'
•0053'
•0053'
¦0047'
¦0052"
•0051'
"0051"
•0054"
•0040'
•0052-
-0052-
•0049
03/02/93-
•0048-
•0044*
¦0051"
•0047-
-0045-
¦0045-
•0049"
•0056-
• 1121•
•1121 •
' 1121¦
• 0053 •
•0052'
"0063"
•0060-
-0060-
•0057'
¦0059'
•0054"
•0052'
¦0053"
-0050'
•0053*
¦0052
03/03/93-
•0050-
-0043'
•0050-
¦0049'
¦0046'
•0049"
- 0058'
•0057-
•0060-
•0064'
•0065"
•0064"
¦ 0062•
"0057"
•0050"
•0056'
¦0053"
'0052*
•0052'
'0046'
•0047'
'0049'
•0057'
•00 50
03/04/93-
-0056-
-0039'
•0044"
•0042-
-0044'
¦0045*
•0063'
* 0060•
¦0038'
•0035'
•0043'
•0050*
-0058'
•0059'
¦0057*
•0053'
¦ 0055•
•0050'
•0059*
"005B•
•0059'
• 0059•
¦0053'
•0051
03/05/93 '
• 004 8'
-0044-
•0054-
•0050'
¦0053*
• 0053'
•0052'
•0053'
¦0061-
¦0057'
• 0057•
•0059-
-0056-
•0059'
•0059"
¦0060-
•0059'
-0059-
• 0060•
•0061-
¦0059'
• 0055•
•0058'
-0055
03/06/93 -
•0055-
-0044-
*0048*
'0041'
'0049-
-0051"
-0052-
•0047•
•0050-
¦0054-
-004 6*
'005 0'
-0053¦
•0051'
•0055"
'0052'
•0050*
-0056-
¦0061'
¦0068'
•0065'
¦0061'
•0060'
-0055
03/07/93-
*0050*
-0046"
•0055-
¦0051-
¦0053'
- 0054 •
-0049-
•0044•
-0046'
•0040"
•0046*
•0042'
¦0039'
¦0040'
¦0042*
-0046'
¦ 0043 •
-0047-
• 0050•
"0056*
'0061*
•0058-
•0053'
-0042
03/08/93-
¦0044-
*003 8'
-0046'
•0042"
-0043-
'0046'
-0054 •
•0056•
-0056-
•1121"
'0056*
¦1121 •
•0050'
¦0053'
¦0051'
¦0051-
•0052*
-0051'
¦0051-
"0057-
•0055'
- 0052•
¦ 005B•
-0059
03/09/93'
•0056-
-0052-
¦0054*
¦0046-
¦ 0044 •
•0048-
•0040-
• 0053 •
•0059-
•0058-
•0055"
¦0052"
¦0050-
•0054'
¦0054'
•0053*
-0057-
•0055'
¦0055*
"0057"
•0050'
¦0052'
¦0053'
-0050
03/10/93-
¦0049-
-0050'
•0054'
"0049-
-0057-
-0058-
•0054-
¦0057"
-0056-
-0056-
-0055-
•0054'
'0051*
¦0053'
¦0056'
•0056-
-0052"
¦0049"
¦0055'
•0056*
•0055"
• 0054 •
-0044-
-0046
03/11/93 '
-0052*
-0045-
-0056"
-0054'
-0053•
-0052-
¦0050-
-0061-
•0067-
•0065'
¦ 0065 *
•0063'
'0055'
'0053'
•0060"
¦0045'
¦0048-
¦0056'
¦0056'
•0059"
•0059'
•0060*
-0059'
•0043
03/12/93-
¦0046'
-0053-
-0055-
-0051'
¦0049"
-0054'
"0054 *
•0050'
• 0050•
• 1114 ¦
¦0062-
•0063'
' 0 061•
¦0062-
•0055-
•0051-
• 0057 •
•0062-
"0057*
•0059'
• 0060•
•0059-
¦0057-
-0036
03/13/93-
•0022*
¦II13-
'II13 •
•II13-
-1113-
-1113 "
* II13 •
• II13 ¦
•II13-
¦ 1113 •
¦ 1113 ¦
• 1113'
'1113 '
•II13-
" 1113 •
" 1113 "
-1113 -
' II13 '
¦ 1113 ¦
• 1113 •
• 1113 •
-1113
03/14/93-
•TI13-
-1113 -
• 1113 •
-1113-
'1113 •
• III 3 •
¦ 1113 •
-1113 •
• 1113 -
¦ 1113 ¦
•1113"
• 1113 •
• 1113 •
" 1113 •
•1113'
• 1113 -
¦ 111 3 -
• 1113 '
'1113'
• 1113 ¦
"1113 *
•1113
03/15/93-
¦ 1113 ¦
• 1113'
•1113-
-1113 -
- II13 ¦
¦II13-
¦1113-
• 1113 -
¦ 1113 -
'1113 *
• 1113 ¦
-1113 •
'1113 '
• 1113 •
• 1113 ¦
•1113'
•1113'
' 1113 -
• 1113 1
'1113 •
¦ 1113 "
-1113
03/16/93'
• II13 •
• 1113 ¦
* 1113 '
'1113'
-1113-
¦ 1113 ¦
-1113 -
• 1113 •
-1113 -
' 1113 "
* 111 3 •
-1113 -
-1113 '
• 1113 •
¦ 1113 -
¦ 1113 •
• 1113 ¦
'1113'
'1113 •
• 1113 •
'1113 '
-1113
03/17/93-
-1113-
• III3 ¦
-1113-
"1113-
' II13 •
'1113 '
' 1113*
• II13 •
• 1113 •
¦iin-
-1IJ3'
' 1113 '
• 1113 '
• 1113-
-1113-
' 111 3 '
¦ 1113*
-1113-
-1113-
• 1113-
' 1113 "
' 1113
-------
Typical Ohio CEM Data Output
Montgomery County North Incinerator Source NQQ3 - Hourly S02 Emissions
| | Hour
Hfly Avg S02!
grapples/I
I' S02
Mo
Da
Yr| Ending
lbmr|
hour!
Ton/grap
Ton/tirj
Ibfl"
7
S
92)22:00
8.681
11
1.14
12.51
0.69
7
s
92(23:00
5.931
101
1.14
11.4|
0.52
7
St 92)00:00
9.39)
iat
1.14
14.8)
0.63
7
5! 32101:00
8.201.
10|
1.14
11,41
0.72
7
6] 92102:00
a.sij
101
1.14
11.41
0.75
7
S
92)03:00
7.68)
10|
1,14
11.4]
0.67
7
S| 92|04:00
4.24!
14
1.14
18.0!
0.27
7
S| 92)05:00
6.88!
17
1.14
19,4|
0.35
7
61 92106:00
4.87)
12!
1.14) 13.7!
0.36
7
S| 92)07:00
4.771
10|
1.141 11.41
0.42
7
6
92(08:00
5.£0|
12|
1.141 13.7|
0.43
7
6[ S2:09:00
4.04|
8!
1.14| 9.1 [
0.44
7
6| 92(10:00
NO 0ATA
I
10|
1.14| 11.4(NO DATA
7
6
92111:00
5.231
10|
1.14
11.4!
0.46
7
6
92)1200
7.471
10)
1.14| 11.4|
0.65
7
61 92(13:00
1Q.30|
12|
1.14| 13.71
0.75
7
S
92)14:00
9.481
71
1.14| 8.0j
1.19
7
S| 92115:00 ,
1.63;
9
1.14
10.3!
0.16
7
s
92|16:00
3.37|
12
1.141 13.7!
0.25
7
S| 92| 17:00
7.08;
9|
1.141 10.3!
0.6S
7
6| 92118:00
11.77!
13|
1.14} 14.01
0.79
7
6
92)19:00
14.87j
81
1.14| 9.K
1.S3
7
6| 92(20:00
8.311
9|
1.14J . 10.3 i
0.81
7
6
92121:00
11.401
101
1.14[ 11.41
l.CO
7
6
92I22.CC
12.601
9!
1,141 10.3!
1.23
7
6
92(23:00
14.44)
11
1.14[ 12,51
¦ 1.15
7
7| 92100:00
13.08i
121
1.14| 13.71
0.96
7
7| 92)01:00
7.76|
. 91
1.14) 10.3!
0.76
7
7
92(02330
12.08)
10|
1.14| 11.41
1.06
7
71 92)03:00
8.80)
111
1.14| 12.5!
0.70
7
71 92104:00
4.34)
9|
1.14| 10.31
0.42
7
7| 92105:00
S.79I
11
1.141 12.51
0.46
7
7
92.06:00
8.751
. 12!
1.14) 13.71 •
0.64
7
7| 92)07:00
9.031
10|
1.14| 11.4)
0.79
7
7
92)08:00
10.251
61
1,14| 6.8|
1.50
7| <7
92I09-.00
5.651
131
1.141 14.8!
0.37
71 7
92)10:00
NO CATA
1
9j
1.14) 10,3| NO DATA
7| 7
92)11:00
9.341
1SI
1.141 18.2!
0.51
7| 7
92)12:00
14.141
7!
1.14) a.oi
1.77
7
7
¦ 92)13:00
3.541
10|
1,14) 11,41
0.32
7 7
92)14:00
12.1S t
9|
1.14) 10.3)
i.ia
7
7
92115:00
4.14|
4
1.14
4.61
0.91
7
7
92(16:00
4.21|
9
1.14
10.3j
0.41
7
7
92)17:00
7.65)
7
1.14
8.01
0.96
7
7
92)18:00
8.721
10
1.14
11.41
0.77
7| 7
92(19:00
10 4Si
111
1.14
12.5|
0.84
7) 7
92)20:00
12.29!
121
1.14! 13.7;
0.90
7
7
92(21:00
13.941
13|
1.14
14.85
0.94
7) 7
92122:00
17.95)
151
1.141 17.1)
1.05
7) 7
92123:00
10.781
91
1.14j 10.3J
1.05
7
31 92(00:00
8.68]
7
1.14| 8.01
1.09
7
8] 92)01 SO
18.57j
- IS
1.14|. l7-1)
1.09
7
8
92102:00
1S.13|
11
1.14
12.51
1.21
7
8 ) 92)03:00
14.38)
12
1.14| 13.7J
1.06
7
8
92)04:00
7.91 (
11
1.141 12.5)
0.63
7
8
92)05:00
15.00!
14)
• 1.14| 16.0|
1.05
7
81 92)06 £0
19.331
8)
1.14
9.1|
2.13
7
8] S2)07:CO
8.83|
121
1.14
13.7!
0.65
71 8! 92108:00
15.38!
111
1.14; 12,5.
1.31
G-4 -
-------
( KYCEM02 )
KEIITUCKY Division OF aip. pollution control
CONTINUOUS EMISSION MONITORING SYSTEM
RAWDATA REPORT
PAGE: 053
DATE: 07/19/93
10-1-3880-0002
CO. NAME: LOUISVILLE GAS i ELECTRIC
PLANT-NAME: TRIMBLE CO GEN STATION
COMP STAT: UNKNOWN COMPLIANCE STAT
PERMIT NUMBER: 092030
STACK-ID: 01
UNIT-ID: 1
PARAMETER: 42603 NOX UNITS:
ANALYSIS METHOD: COAL ANALYSIS
COLLECTION METHOD: EXTRACTIVE
DATA FOR: OCTOBER
LAST DATE UPDATED:
1992
93043
POUNDS PER MILLION BTU
DAY
00
01
01
0.55
0.57
02
0.48
0.49
03
0. 52
0.51
04
0. 62
0. 62
OS
0. 60
0.61
06
0. 53
0.54
07
0.50
0.51
08
0.57
0.58
09
0.58
0.59
10
0.55
0.55
11
0 . 59
0.61
12
0.51
0.52
13
0 . 54
0.54
14
0.52
0.52
15
0.65
0.66
16
0. 64
0. 62
17
0. 60
0.61
18
0.66
0.66
19
0.63
0.63
20
0.00
0.00
21
0.58
0 . 57
22
0 . 59
0.59
23
0 . 54
0 . 54
24
0.56
0.57
25
0 . r. 11
O.'.lt
26
<1 . GZ
1). <. 1
27
0.56
0.55
28
0.56
0.58
29
0.56
0.56
30
0.S5
0.57
31
0.58
0.58
02
03
04
OS
06
07
08
. . S TAR T 111 G
09 10 11 12
H O U P.. NUM DAILY
13 14 15 16 17 18 19 20 21 22 23 OBS MEAN
.59 0.58
.49 0.49
.49 0.48
.£3 0.63
.62 0.63
.55 0.56
.53 0.54
.59 0.60
.60 0.60
.55 0.55
.63 0.64
.52 0.51
.55 0.57
.53 0.54
.65 0.64
.60 0.59
.63 0.62
.66 0.66
.¦62 0.60
.39,0.00
.58 0.58
.61 0.62
.57 0.58
.56 0.57
.'.n 0.59
.6-1 0.6-1
.53 0.53
.59 0.61
.56 0.57
.58 0.59
.58 0.58
0.56
0.49
0.48
0.64
0.64
0.56
0.55
0.61
0.60
0.56
0.64
0.50
0.59
0.56
0.64
0.59
0.62
0.65
0.60
0.46
0.57
0.6-1
0.58
0.57
5(1
C I
54
61
0.58
0.60
0.58
0.54
0.48
0.48
0.64
0.65
0.56
0.55
0.61
0. 60
0.56
0. 63
0.50
0. 61
0.57
0.64
0.59
0.62
0.65
0.59
0.45
0. 58
0.65
0.55
0.58
0 .'.) o
0 .
0.
0.
0.
0.
0.
0.54
0.49
0.48
0.65
0.65
0.57
0.55
0.61
0.60
0.57
0.S5
0.51
0.48
0.64
0.66
64
51
61
59
64
61
62
64
58
55
58
54
59
61
58
64
53
60
61
66
63
62
62
58
55
•,'j
54
61
58
61
59
0.59
0.65
0.53
0 . 5H
O.'.O
U.MJ
0.54
0.61
0.57
0.61
0. 59
0. 60
0 . 63
0.52
0.58
0 . 57
0.61
0.57
0.62
0.56
0.61
0.59
0.57
0.53
0.48
0.64
0.65
0.59
0.52
0.59
0. 61
0.59
0.65
0.55
0.60
0.62
0.64
0.65
0.62
0.62
0.59
0 . 54
0.61
0.61
0.61
0.62
0.56
0.61
0.59
0.58
0.55
0.48
0.63
0.62
0.58
0.51
0.58
0.63
0.60
0.65
0.56
0.60
0.62
0.62
0.63
0.62
0.61
0.62
0.53
0. 60
0.58
0.50
0.59
0.63
0.47
0.62
0.62
0.56
0.62
0.60
0.57 0.55 0.53 0.52
0.50 0.48 0.48 0.47
0.51 0.50 0.50 0.49
0.48
0. 64
0.59
0.56
0. 50
0.59
0.64
0.60
0.66
0.58
0. 59
0. 60
0.59
0.62
0.61
0.63
0.53
0.58
0.56
0.50
0 . 59
0 . 66
0. 00
0.60
0.59
0.57
0.64
0. 60
0.4 9
0.64
0.56
0.56
0.50
0.61
0.64
0.60
0.67 0,
0.60 0.
62
64 0.63
56 0.56
57 0.56
54 0.58
62 0.61
64 0.64
60 0.61
0.60
0.56
0.63
0.60
0.62
0.54
0.58
0.54
0.52 0
0.59 0
0.68 0
0.40 0
0.58
0.59
0.57
0.65
0.59
65
62
61
60
55
63
62
0
0
0
0
0,
0
0
0
63 0.62
60 0.62
65 0.66
59 0.60
54 0.56
59 0.58
66 0.70
lib
.62
.62
.64
. 65
. 60
.55
.62
.64
53 0
57 0,
59 0,
58 0.
66 0.
60 0.
58
59
59
65
62
0.63
0.57
0.55
0.61
0. 60
0.64
0.63
0.58
0.62
0.58
0.67
0.60
0.54
0.61
0.67
0.62
0.61
0.66
0.60
0.58
0.58
0.69
0.57
0.59
0.59
0.59
0.63
0.63
0.63
0.57
0.53
0.61
0.57
62
63
58
61
0.56
0.68
0. 61
0.54
0. 61
0. 67
0. 00
0.54
0.59
0.56
0.61
0. 5fl
0.63
,58
.59
,59
57
0.61
0.63
0.63-
0.56
0.52
0. 59
0.57
0.61
0.62
0.58
0.60
0.55
0.68
0.60
0.54
0.61
0.69
0. 00
0. S3
0. 59
0.55
0. 64
0. 60
0. 63
0.59
0.60
0.59
0.56
0.60
0. 63
0.64
0.55
0.S2
0.57
0.56
0.60
0.60
0.57
0.59
0.55
0.68
0.60
0.56
0.62
0.67
0. 00
0.52
0.59
0.54
0 . 62
0.59
0. 63
0 . 59
0.59
0.59
0. 56
0.61
0.63
0. 47
0.4 9
0.60
0.64
0.54
0.52
0.57
0.55
0.59
0.59
0. 55
0. 57
0 . 54
0.67
0.58
0.59
63
66
00
5 1
58
5 J
59
0.59
0 . (, .1
0.5 9
0.59
0.59
0. 56
0.60
0.63
0.47
0.49
0.60
0.65
0.53
0.52
0.57
0.53
0 . 59
0.59
0.55
0. 56
0.53
0.67
0.59
0. 61
0.64
0.64
0.00
0.55
0.57
0. 53
0.56
0. 60
0. 65
0. 57
0.58
0.59
0.56
0.61
0.63
0.47
0.50
0.61
0.65
0.53
0.52
0.57
0.52
0.57
0.59
0.54
0.57
0.52
0.66
0.60
0.63
0.65
0.64
0.00
0.57
0 . 58
0 . 52
0.55
0.60
0 . 64
0 . 56
0.58
0.59
0.55
0.61
0.64
0. 47
0.52
0.63
0.65
0.53
0.51
0.56
0.51
0.55
0.S9
0.53
0.57
0.52
0.65
0. 62
0.62
0. 65
0.64
0.00
0.47
0.52
0.63
63
54
50
56
0.59
0.60
0.52
0.57
0. 60
0. 63
0. 56
0.S8
0.59
0.55
0.60
0. 64
0.53
0.54
0.58
0.52
0.57
0.53
0.63
0.64
0.61
0.66
0.62
0.00
0.60
0.61
0.52
0.56
0.59
0.62
0.57
0.57
0.58
0.55
0.59
0.64
0.47
24
0.52
0.52
20
0.50
0.62
19
0.53
0.62
24
0.64
0.53
24
0.59
0.50
24
0.54
0.55
24
0.55
0.54
24
0.58
0.54
24
0.60
0.58
24
0.59
0.51
24
0.60
0.5S
24
0.56
0.52
21
0.56
0.64
23
0.62
0.64
24
0.62
0.59
24
0.59
0.66
24
0.63
0.62
24
0.64
0.00
24
0.38
0.59
24
0.48
0.59
24
0.60
0.53
24
0.58
0.56
2 4
0.56
0.59
24
0.58
0.61
2 4
0.62
0.57
2 4
0.55
0.56
24
0.57
0.58
24
0.59
0.55
24
0.57
0.58
24
0.61
0.64
24
0.61
11 31 31 31 31 31 31 31 31 31 31 29 28 29 29 30 30 30 30 31 31 31 31 31 31
MM 0.55 0.56 0.57 0.56 0.58 0.58 0.58 0.59 0.59 0.59 0.57 0.58 0.60 0.61 0.60 0.57 0.57 0.56 0.56 0.56 0.56 0.56 0.56 0.55
MAX 0.66 0.66 0.66 0.66 0.65 0.65 0.65 0.66 0.65 0.65 0.66 0.68 0.66 0.70 0.69 0.68 0.69 0.68 0.67 0.67 0.66 0.65 0.66 0.66
MAXIMUM OBSERVATION 0.70
MONTHLY MEAN 0 . 57
NUMBER OF OBSERVATIONS.. 7 31
0.64
-------
Montgomery County North Incinerator Source N0Q3 - Hourly S02 Emissions
1 1
Hour
Hrly Avg S03
grapples/
S02
Moi 0a| Yr
Ending
Ib/hr
hour
Ton/grip
Tort/hr
Ib/T
7
5| 92
22:00
8.68
11
1.141 12.5
0.69
7
5 92
23:00
5.93
10
1.141 11.4
0.52
7
6| 92
00:00 ,
939
13
1.14j 14.8
0.63
7
S| 92
01:00
8.20
10
1.14
11.4
0.72
7
61 92
02:00
8,51
10
1.14
11.4
0.75
7
6| 92
03:00
7.68
10
1.14
11.4
0.67
7
S| 92
04:00
4,24
14
1.14
16.0
0.27
7
Sj 92
05:00
6.88
17
1.14
19.4
0.35
7
6| 92
06:00
4,87
12
1.14
13.7
0.36
7
6j 92(07:00
4,77
10
1,14
11.4
0.42
7
61 92! 08:00
5.90
12
1.14
13.7
0.43
7
6 92(09:00
4.04
8
1.14
9.1
0.44
7
S| 92110:00
NO OATA
10
1.14
11.4(NO DATA
7
6
92
11:00
5.231 10
1.14
11.4) 0.46
7
6
92
12:00
7.471 10
1.14
11.4| 0.6S
7
S
92
13:00
10.30| 12
1.14
13.71 0.75
7
6
92
14:00
9.48) 7
1.14
8.01 1,19
7
S
92
15:00
1.63] 9
1.14
10.3| 0.16
7
6
92
16:00
3.37| 12; 1.14
13.7| 0.25
7
6
92
17:00
7.08! 9' 1.141 10.3| 0,69
7
6
92
18:00
11.77! 131 1,14i 14.8; 0.73
7
S
92
19:00
14.87 j 81 1.14
9.1( 1.63
7
6
92
20:00
8.31| 9i 1.14
10.3j 0.81
7
S
92
21:00
11.401 10| 1.14
11.4} 1.00
7
6
92
22:00
' 12.601 9| 1.14
10.3! 1J23
7
6
92
23:00
14.44| 111 1.14
12.51 1.15
7
7
92
00:00
13.0SS 121 1.14
13.71 0.96
7
7
92
01:00
7.761 9) 1.14
10.3| 0.76
7
7
92
02:00
12.08) lOj 1.14
11.-4} 1.06
7
7
92
03:00
8.801 1,1| 1.14
12.5
0.70
7
7
92
04:00
4.341 9| 1.14
10,3
0.42
7
7
32
05:00
5.79) 11) 1.14
12.5
0,46
7
7
92
06:00
8.75| 121 1-H
13.7
0.64
7
7
92
07:00
9.03| 10| 1.14
11.4
0.79
7
7
92
08:00
10.25) 6| 1.14| 6.8
1.50
7
¦ 7
32
09:00
5.561 131 1.14] 14.8
0.37
7
7
92
10:00
NQ OATA | 9j 1.14| 10.3
NO DATA
7
7
92
11:00
9.341 16! J.14| 18.2
0.51
7
7
92
12:00
14.141 7| 1.14
8..01 1.77
7
7
92
13:00
3.641 10| 1.14
11.41 0.32
7
7
92
14:00
12.151 91 1.14
10,3j 1.18
7
7
92
15:00
4.14| 4j 1,14
4.6( 0.91
7
7
92
16:00
4.211 si 1.14
10.31 0.41
7
7| 92
17:00
7.65 j 7| 1.14
8.0) 0.96
7
71 92
18:00
8.72! 101 1.14
11.4| 0.77
7
7| 92
19:00
10.491 111 1.14
I2.5j 0.84
7
.-7i 92
20:00
12.291 12| 1.141 13.7| 0.90
7
7
92
21:00
13.941 13! 1.14} 14.8) 0.94
7
7
92|2200
17.95! 15
1.14| 17.1 ( 1.05
7
7
92
23:00
10.78! 9
1.14
10.31 1.0S
7
S
92
00 00
8.68
7
1.14
8.0| 1.09
7
8
92
01:00
18.57
• 1S
1.14
17,11 1.09
7
8
92
02:00
15.13
11
1.14
12,5) 1.21
7
8
92
03:00
14.33
12
1.141 13.71 1.05
7
8
92
04:00
7.91
11
1,14
12.5i 0.63
7
8
92
05:00
16.80
14
1.14
16.01 1.05
7
a
92
06:00
19.38
8| 1.14
9.1| 2.13
7
3
92
07:00
8.83
12| 1.14
13.71 0.65
7
a; 92
08:00
16.381 111 1.14! 12.5. 1.31
G-6
-------
PtNNBOl FAOU1Y; BOUSBflU-E
CEM10 - QO2<*0 10
S02IMPPM
DATE
f Z
WXllSl
0000
oooo
oooo
0>W>2
0000
oooo
oooo
07AO*2
0000
oooo
oooo
OT/CM/B2
0000
oooo
oooo
07jQQft2
0000
oooo
oooo
0/«6.V2
0000
OOOI
oooo
GMW
0000
oooo
MOO
0?/Qe«5!
OQOO
flrtTQ
OOOO
niixm
0000
oooo
QflOfl
01 i10,92
ooco
OOOO
OOCB
07/Mlfi^e2
oooo
moo
ooco
vrmm
oooo
oooo
oooo
07f«6«2
MDOU
oooo
oooo
owiMwr
won
oooo
oooo
orwwu
0009
oooo
OQ00
G7SWS
OOOO
oooo
aaao
mnzm
oooo
oooo
QCOO
07/nm
oooo
oooo
oooo
97/24,93
oooo
con
oooo
(ttrftm
oooo
oooo
oooo
onwsn
oooo
uooo
ouno
Qiumi
OUOO
WOO
ouuo
OlfMM
<•
oooo
enrjvivi
t**JO
cmoo
(M)S
OlfM/92
oooo
oooo
OOOO
aifiim
OOOO
«K»
oooo
«8i0IfflZ
OOCO
oooo
oooo
ocuazm
oooo
oooo
oooo
oomm
oooo
oooo
OOOQ
mifUfaz
ooao
oooo
OOOO
oe mm
oooo
oooo
OOOQ
c*5.«r»
oooo
oooo
OOOO
oe.«?jV2
oooo
oooo
oooo
eew/R
oooo
oooo
oooo
OM»/92
oooo
oooo
oooo
oe/ifii«2
oooo
oooo
oooo
W3/II/W
0000
oooo
oooo
04/12/92
oooo
oooo
oooo
oooo
OOCti
OOOQ
OOfM/92
qooq
oooo
oooo
ob/is/mi
oooo
oooo
oooo
IW12.V2
0QP0
oooo
OOOQ
oo/ijfla
oooo
OOOQ
OOOQ
os/i
OOOO
oooo
oooo
oooo
oooo
OQOO
oooo
oooo
oooo
oooo
ooco
oooo
oooo
oooo
oooo
oooo
oooa
oooo
woo
flnfirt
oooo
0009
oooo
oooo
oooo
oooo
ocm
oooo
oooo
oooo
OOOQ
!»¦>
/yw?
oooo
oooo
00OQ
oooo
oooo
ana
oooo
oooo
/ft/*)
oooo
oooo
ODOO
oooo
oooo
OOOO
oooo
13000
oooo
oooo
OOOQ
oooo
oooo
oooo
oooo
OOOQ
oooo
oooo
oooo
oooo
oooo
oooo
oooo
oooo
OOOQ
oooo
oooo
oooo
oooo
oooo
oooo
oooo .
oaoo
oooo
oooo
oooo
oooo
OOOQ
ocoo
oooo
oooo
oooo
oooo
oooo
oooo
oooo
oooo
OOOO
oooo
oooo
oooo
ocoo
oooo
oooo
oooo
oooo
oooo
OOCKI
oooo
oooo
oooo
oooo
oooo
oooo
oooo
oooo
oooo
OOOQ
oooo
oooo
oooo
oooo
oooo
GQtiO
oooo
oooo
oooo
oooo
OOOQ
oooo
OOOQ
oooo
oooo
oooo
oaoo
oooo
oooo
oooo
oooo
oooo
oaoo
oooo
oooo
oooo
QDOO
oooo
oooo
oooo
oooo
oooo
oooo
oooo
OOOO
00t»
oooo
oooo
oooo
oooo
oaoo
oooo
oooo
oooo
oooo
oooo
tioao
oooo
oooa
oooo
oooo
oooo
oaoo
oooo
oooo
oooo
oooo
0009
oooo
QOOQ
oooo
oooo
oooo
oooo
oooo
OOOO
oooo
oooo
oooo
oooo '
oooo
oooo
oooo
oooo
OOOO
oooo
oooo
oooo
oooo
oooo
oooo
oooo
ooao
oooo
WOO
ooco
I***)
0>xjO
oouo
oooa
oeuo
woo
oooo
oooo
O0"0
DOOO
0>KK>
o>m
oooa
OUOO
OUOO
oouo
oooo
OOUO
OUO0
Ouu
Wrti
oouo
ax io
OuuO
oooo
oooo
oooo
uuuo
a*SO
f*KK>
OUOO
OUOO
eaao
oooo
oooo
oouo
oooo
OUOO
oooo
oaoo
QOuO
oooo
0000
oooo
oooo
oaoo
oooo
oooo
oooo
oooo
woo
oooa
ooco
oooo
oooo
oooo
oooo
OOOQ
Oooo
OOOO
oooo
OOOO
oooo
O000
OAjO
oooo
oooo
oooo
oooo
oooo
oooo
oooo
oooo
oooo
oooo
oooo
oooo
oooo
oooo
oooo
oooo
oooo
oooo
oooo
oooo
oooo
oooo
OOOQ
ooco
oooo
oooo
oooo
oooo
ooao
oooo
oooo
oooo
oooo
oooo
oooo
oooo
oooo
oooo
oooo
oooo
oooa
oooa
oooo
oooo
OOOO
€000
oooo
oooo
oooo
OOCO
oouo
oooo
oooo
oooo
oooo
oooo
oooo
oooo
cooa
oooo
oooo
oooo
QOOQ
oooo
oooo
oogo
oooo
OOOO
oouo
oooo
oooo
0000
oooo
oooo
oouo
OUOO
ocoo
oooo
OOOQ
oooo
OOOO
OOOQ
oooo
oooo
oooo
OOOQ
oooo
ooco
oooo
OOOO
oooo
oooo
oooo
oooa
oooo
oooo
OOOO
oooo
OOOO
OOOQ
oooo
oaoo
oooo
oooo
oooo
oooo
oooo
oooo
oooo
oooo
fltltVt
oooa
OOOO
oooo
oooo
oooo
oooo
OUOO
oooo
oooo
oooo
oooo
oooo
oooo
oooo
OOOO
oooo
oooo
ft¥»
oaxt
oooo
oooo
„ oooo
OuOO
oooo
oooo
oooo
oooo
oooo
oooo
oooo
oooo ,
oouo
oooo
oooo
ooao
OOUO
oooo
onao
OOOQ
oooo
oouo
UUM
auoO
oouo
oooo
oouo
12 O 14 IS !fl 17 IS 19 20 21 22 23
oooo
oooo
oooo
OOOO
oooo
. oooo
oooo
oooo
aooo
oooa
oooo
c>coo
oooo
oooo
oooo
QOOQ
oooo
oooo
oooo
oooo
ooco
oooo
oooo
oooo
oooo
0000
QOOQ
oooo
oooo
oooo
oooo
oooo
oooo
oooo
oooo
ttXtt
oooa
oooo
oooo
Oaoo
oooo
oooo
oooo
oooo
OOOQ
oooo
oooo
oooo
oooo
OOOQ
OOOO
oooo
OOCO
oooo
oooo
oooo
oooo
oooo
oooo
oooo
120
U20
IE20
1120
B20
H20
oooo
oooo
oooo
OOQO
Qryrt
OOOO
K2©
(120
820
1120
1L20
l(20
oooo
aooo
moo
(XVK)
oooo
oooo
oooo
oaoo
ooco
oooo
oaoo
ooao
oooo
oooo
oooo
oooo
cooo
oooo
oooo
oaoo
oooa
oooo
oooo
(yvyi
oooo
oooo
myi
woo
oooo
oooo
oooo
woe
COOO
oooo
ftyn
oooo
OOQO
qnprt
oooo
ocoo
oooo
OGOO
ooco
oooo
oooo
oooa
1 oooo
oooo
oooo
oooo
cooo
oooo
cooo
OOOD
oooo
oooo
OOOQ
OOQO
oooo
ooao
oooo
OOOQ
oooa
onto
oooo
oooo
OOOO
oooo
oooo
oooa
aooo
- oooo
oooo
COO0
oooo
oooo
oooo
oooo
0000
oooo
ooco
oooo
oooo
oooo
oooo
OOQO
oooo
oooo
ooao
oooo
oooo
onao
ocssa
0000
OUOO
DODO
oooo
oooo
oooo
oooo
oooo
oooo
r.jOG
oooo
oooo
oooo
OOQO
oooo
oooo
0000
ocoo
OOOQ
oooo
cooo
QOOQ
oooo
oooo
cooo
nooo
oooo
oooo
oaoo
oooo
OOOO
OOQO
oooo
oooo
oooo
oooo
oooo
OOOQ
oooo
oooo
aooo
oooo
oooo
oooo
oooo
oooo
ocoo
OOOQ
{WKVi
ooao
oooo
oooo
oooo
oooo
cooo
oooo
cooo
oooo
oooo
OOOQ
oaoo
oooo
oooo
0000
oooo
oooo
oooo
oooo
oooa
oooo
QOOQ
oooo
OOCO
oooo
oooo
oooo
OOOO
OOOO
oooo
oooo
OOOQ
oooo
oooo
oaoo
OOOO
oooo
oooo
oooa
oooo
oooo
ooao
oooo
OOOO
OOOD
ocoo
oooo
£000
oooo
OQUO
oooo
oooo
OOCO
oaoo
ooao
oooo
GOOD
oooo
QOGG
oooo
oooo
oooo
oooo
oooo
OOOO
oooo
ooao
oooo
oooo
OOOQ
0000
0009
oooo
oouo
0009
oooo
OUOO
oooo
oooo
aauo
OOOO
oaoo
aooo
oooo
oooo
1W00
oogo
oooo
OOQO
oooo
oooo
OUOO
OUU.*
OOUO
01)00
(WOO
oooo
(»JU0
ouuo
OOOQ
oooo
ocoo
unrm
OUUO
oooo
oooo
OOOO
OOOO
OUOO
OuuO
OUOO
oooa
OuOO
OOOQ
ouuo
OOOO
OUOO
cara
ocoo
oooo
ouuo
UaA
OOuG
00130
oooo
oooo
oooo
oooo
oooo
oaoo
oaoo
oooo
oooo
UOliO
OOOO
oooo
oooo
oooa
oooo
oooo
oooo
can
ooao
oooo
oooo
oouo
OOOO
OOQO
oooo
oooo
oooo
oaoo
oooo
oooo
OUOO
oooo
oooo
oooo
oooo
OOOO
OOQO
oooo
oooo
oooo
oooo
oooo
oooo
oooo
aoe»
0000
oooo
OOOO
oooo
oooo
ocoo
oooo
OOOQ
cooo
OCOO
oooo
oooo
OOOQ
oooo
oooo
oooo
oooo
oooo
CJ000
OOOO
ooco
OOOO
oooo
OOOO
ocoo
oooo
oooo
oooo
(HUl
ncoa
oooo
oooo
OOOO
ooao
oooo
oooo
oooo
oooo
ooco
oooo
oooo
OOOQ
oooo
oooo
OOOQ
oooo
aam
oooo
oooo
oaoo
oooo
OOGO
oooo
oooo
oooo
oooo
QOOQ
oooo
aooo
OCttJ
oooo
oooo
oooo
OOQO
OOUO
oooo
ooco
oooo
OOOO
OOOQ
QO00
ooao
oooo
oaoo
oooo
oooo
OUOO
OOUO
oooo
oooa
OOOD
oooo
OUOO
oooo
0009
oooa
cooo
mypp
OOOQ
OOOQ
oooo
oouo
OOOO
OOOO
OuuO
oouo
OOOQ
oooo
aooo
ooco
oooo
OOOO
oooo
COOO
OOOO
oooo
OUOO
oooo
OOOO
oooo
OOOD
oooo
OOOO
OOOQ
ooao
OOOO
oooo
oooo
oooa
oooo
OOOO
oooa
oooo
OOQO
oooo
oooo
ooao
oooo
oooo
OOOO
ano
oooo
OOOO
oooo
oooo
oooo
oooo
OOOQ
OOQO
oooo
OOQO
OOOQ
oooo
oooo
oooo
oaoo
oooo
oooo
oooo
0000
ODOO
oooo
oooo
oooa
oaoo
oooo
oooo
oooo
woo
OOQO
oooo
OOOQ
OOCO
oooo
oooo
oooo
oaoo
oooo
OOOQ
oooo
oooo
oooo
oouo
ooco
OCtiO
ooco
ooao
oooa
oaoo
OOQO
ocoo
ocoo
oooo
oooo
oooo
oooo
oooo
oooo
oooo
oouo
(*XO
oooo
Quoo
oooo
oooo
oooo
ocm
oooo
OOOO
oooo
oooo
OCKJO
(JimY)
oooo
oooo
oooo
oooo
oooo
-------
APPENDIX H
WASTE-TO-ENERGY DATA
H-l
-------
lllli III III
Mil ill III
Ill
;;Pi I
II
III I
I iii
Ill
III
1 III
111
III III III
III
III I
I III
III
III III III
III
III I
I III
III
"III ill ill
II!
III I
1 III
III
JII III III
Ill
jit I
I III
III
3 5 3 3 5 5 =55
111 111 III
III
jli I
I III
III
jli III III
III .
JII I
I III
III
3ss
HI iii ill
;;;
ill
Hi 1
I III
HI
JII III III
III
JII I
I III
III
jli III III
III
JII I
1 III
11!
JII III III
11!
. JII !
I III
HI
=Il! III III
Ill
JII I
1 III
III
=I1! Ill III
III
jli I
I III
Hi
JII III ill
s;:
ill
CCS -
III I
l , III
Hi
jll III ill
III
)n i
I III
III
Hi III III
III
)ll I
I III
III
"ill III III
iir
JII I
I III
HI
)ll III III
in
)ll I
I III
III
JII III III
II!
JII I
I III
III
)ll III III
Ill
)» I
I 111
III
JII III III
III
III l
I ill
111
fill III III
ill
fill I
I JII
HI
fill III III
III
[ill I
1 III
HI
fill III III
HI
* c s
nii i
I ill
ill
jliilllliiiilill!
pi!? I
II ill! Iii!
Ill I i i., ill 1 i ¦
' I •
------- |