I Q
w
*1 PROl*-
Air Pollutants Exposure Model Documentation
(APEX, Version 5.2)
Volume I: User's Guide
-------
-------
EPA-452/R-23-009a
June 2023
Air Pollutants Exposure Model Documentation (APEX, Version 5.2)
Volume I: User's Guide
By:
ICF International
Durham, North Carolina
Prepared for:
John Langstaff, Project Officer
Office of Air Quality & Standards
U.S. Environmental Protection Agency
Contract No. EP-W-12-010
Work Order No. WA 4-55
U.S. Environmental Protection Agency
Office of Air Quality Planning and Standards
Health and Environmental Impacts Division
Research Triangle Park, NC
-------
DISCLAIMER
This document has been prepared at least partially by ICF (under EPA Contract No. EP-
W-12-010 [WA 4-55]). It has been subject to the Agency's review and has been approved for
publication as an EPA document. Mention of trade names or commercial products is not
intended to constitute endorsement or recommendation for use.
CONTACT
Questions should be addressed to John E. Langstaff, U.S. Environmental Protection Agency,
C504-06, Research Triangle Park, North Carolina 27711 (email: langstaffiohn@epa.gov).
li
-------
ACKNOWLEDGEMENTS
The primary author of this document is Graham Glen (ICF). It includes contributions from
Melissa Nysewander, Luther Smith, and Casson Stallings (while at Alion Science and
Technology, Inc.); Stephen Graham, Kristin Isaacs, Tom McCurdy (retired), and John Langstaff
(EPA); and by ICF.
111
-------
CONTENTS
CHAPTER 1. INTRODUCTION 1
1.1 Overview of the APEX Model 1
1.2 Nomenclature 3
1.3 Strengths and Limitations of APEX 4
1.3.1 Strengths 4
1.3.2 Limitations 5
1.4 Applicability 5
1.5 Brief History of APEX 6
1.6 Scope and Organization of This Guide 7
CHAPTER 2. SETTING UP AND RUNNING APEX 9
2.1 Downloading and Setting Up APEX 9
2.2 Setting Up an APEX Simulation 11
2.3 Overview of Input and Output Files 12
2.3.1 Input Files 12
2.3.2 Output Files 12
2.4 Overview of Model Settings and Options 14
CHAPTER 3. CHARACTERIZING THE STUDY AREA 21
3.1 APEX Spatial Units 21
3.1.1 Initial Study Area 21
3.1.2 Sectors 21
3.1.3 Air Quality Districts 23
3.1.4 Modeling Commuting 24
3.1.5 Meteorological Zones 25
3.1.6 The Final Study Area 25
CHAPTER 4. APEX INPUT FILES 26
4.1 Input File Formats 26
4.2 Control Options File 29
4.2.1 Input and Output File List Sections of the Control Options File 30
4.2.2 Pollutant Parameters Section of the Control Options File 32
4.2.3 Job Parameter Settings Section of the Control Options File 38
4.3 Population Sector Location File 52
4.4 Air District Location File 53
4.5 Air Quality Data File 54
4.5.1 Air Quality Input Data (Type 1) 55
4.5.2 Air Quality Input Defined as Hourly Distributions (Type 2) 56
4.6 Meteorology Zone Location File 57
4.7 Meteorology Data File 57
4.8 Population Data Files 59
4.9 Commuting Flow File 61
4.10 Commuting Time File 62
4.11 Employment Probability File 63
4.12 Profile Factors File 64
4.13 MET Mapping File 66
4.14 MET Distribution File 71
iv
-------
4.15 Physiological Parameters File 74
4.16 Ventilation File 77
4.17 Profile Functions (Distributions) File 78
4.17.1 Defining a Profile Function 79
4.17.2 Functions for Built-in, User-defined, AQ and Regional APEX Variables 81
4.18 Microenvironment Mapping File 89
4.19 Diary Questionnaire (DiaryQuest) File 92
4.20 Diary Events File 93
4.21 Diary Statistics File 94
4.22 Diary Occupations File 95
4.23 Diary Transitions File 96
4.24 Microenvironment Descriptions File 96
4.24.1 Microenvironment Descriptions Section 97
4.24.2 Parameter Descriptions Section 97
4.25 Prevalence File 105
CHAPTER 5. APEX OUTPUT FILES 106
5.1 Log File 107
5.2 Hourly File 108
5.3 Timestep File 110
5.4 Daily File 112
5.5 Profile Summary (Persons) File 115
5.6 Microenvironmental Results File 118
5.7 Microenvironmental Summary File 121
5.8 Output Tables File 122
5.8.1 Exposure Summary Tables 123
5.8.2 Dose Summary Tables 129
5.8.3 ResponseProb Summary Tables 131
5.9 Sites File 132
5.10 Events File 132
5.11 Multipollutant File 134
5.12 Diary Clustering Files 135
5.13 Sobol Results File 135
REFERENCES 137
v
-------
LIST OF TABLES
Table 2-1. APEX Input Files 13
Table 2-2. APEX Output Files 14
Table 2-3. APEX Settings and Options 15
Table 4-1. APEX Input File Descriptions 27
Table 4-2. Pollutant-specific Job Parameters 32
Table 4-3. Output Parameter Levels in the Output Summary Table 34
Table 4-4. Job Parameters in the APEX Control Options File 42
Table 4-5. CHAD Activity Codes 66
Table 4-6. Available Probability Distributions in APEX 72
Table 4-7. Parameters in the Physiological Input File 74
Table 4-8. Variables that can be Defined in the Profile Functions File 83
Table 4-9. CHAD Location Codes 90
Table 4-10. CHAD Occupation Codes 93
Table 4-11. CHAD Locations used in Constructing the Outdoor Time and Vehicle Time Diary
Statistics Files Supplied with APEX 95
Table 4-12. Microenvironment Parameters for the Factors and Massbal Methods 97
Table 4-13. Keyword Definitions for the Parameter Descriptions Section of the
Microenvironment Descriptions File 100
Table 5-1. APEX Output Files 107
Table 5-2. APEX Variables Written to the Hourly Output File 108
Table 5-3. APEX Variables Written to the Timestep Output File Ill
Table 5-4. APEX Variables Written to the Daily Output File 112
Table 5-5. APEX Variables Written to the Profile Summary File 115
Table 5-6. APEX Variables Written to the Microenvironmental Results File 119
Table 5-7. Format of the APEX Microenvironmental Summary File 121
Table 5-8. Interpretation of the Variables in Exposure Table Type #3 and Other "Person-Days"
Based Tables 126
vi
-------
LIST OF EXHIBITS
Exhibit 4-1. Input Files Section of an Example Control Options File 31
Exhibit 4-2. Output Files Section of an Example Control Options File 31
Exhibit 4-3. Pollutant Parameters Section of an Example Control Options File 38
Exhibit 4-4. Job Parameters Sections of an Example Control Options File 52
Exhibit 4-5. First Part of the Population Sector Location File (2010 Census) 53
Exhibit 4-6. First Part of an Example Air District Location File 53
Exhibit 4-7. First Part of an Example Air Quality Data File (Type 1) 56
Exhibit 4-8. First Part of an Example Air Quality Data File (Distribution Type) 57
Exhibit 4-9. First Part of an Example Meteorology Zone Location File 57
Exhibit 4-10. Part of an Example Meteorology Data File 59
Exhibit 4-11. First Part of a Population Data File (2010 Census, Female Black or African
American, Not Hispanic or Latina) 61
Exhibit 4-12. First Part of the Commuting Flow File (2010 Census) 62
Exhibit 4-13. First Part of the Commuting Time File (2010 Census) 63
Exhibit 4-14. First Part of the Employment Probability File (2010 Census) 64
Exhibit 4-15. First Part of an Example Profile Factors File 65
Exhibit 4-16. First Part of the MET Mapping File 71
Exhibit 4-17. First Part of the Activity-specific MET File 74
Exhibit 4-18. An Example of a Portion of the Physiological Parameters File 77
Exhibit 4-19. The APEX Ventilation Input File for VEMethod=l 78
Exhibit 4-20. The APEX Ventilation Input File for VEMethod=2 78
Exhibit 4-21. Examples of Profile Functions 87
Exhibit 4-22. First Part of an Example Microenvironment Mapping File 92
Exhibit 4-23. First Part of the Diary Questionnaire File 93
Exhibit 4-24. First Part of the Diary Events File 94
Exhibit 4-25. First Part of the Diary Statistics File for Time Spent Outdoors 95
Exhibit 4-26. First Part of a Diary Occupations File 96
Exhibit 4-27. Example of a Microenvironment Descriptions Section of a Microenvironment
Descriptions File 97
Exhibit 4-28. Example of a Parameter Descriptions Section of a Microenvironment Descriptions
File 104
Exhibit 4-29. First Part of an Example Prevalence File 105
Exhibit 5-1. First Part of an Example APEX Hourly Output File 110
Exhibit 5-2. First Part of an Example APEX Timestep Output File 112
Exhibit 5-3. First Part of an Example Daily Output File 115
Exhibit 5-4. First Part of an Example Profile Summary File 118
Exhibit 5-5. First Part of an Example Microenvironmental Results File 121
Exhibit 5-6. First Part of an Example Microenvironmental Summary File 122
Exhibit 5-7. Example of Exposure Table Type #1 in the Output Tables File 125
Exhibit 5-8. Example of Exposure Table Type #3 in the Output Tables File 126
Exhibit 5-9. Example of Exposure Table Type #6 in the Output Tables File 128
Exhibit 5-10. Example of Exposure Table Type #11 in the Output Tables File 129
Exhibit 5-11. Portion of ResponseProb Table 131
Exhibit 5-12. Portion of an Events File 134
vii
-------
LIST OF FIGURES
Figure 3-1. Example of Study Areas, Air Quality Districts, Meteorological Zones, and Sectors22
Figure 4-1. Relationship between Profile Functions and Microenvironmental Descriptions Files
79
viii
-------
CHAPTER 1. INTRODUCTION
1.1 Overview of the APEX Model
The Air Pollutants Exposure model (APEX) is part of EPA's overall Total Risk Integrated
Methodology (TRIM) model framework (EPA, 1999). TRIM is a time series modeling system
with multimedia capabilities for assessing human health and ecological risks from hazardous and
criteria air pollutants; it has been developed to support evaluations with a scientifically sound,
flexible, and user-friendly methodology. The TRIM design includes three modules:
Environmental Fate, Transport, and Ecological Exposure module (TRIM.FaTE);
Human Inhalation-Dietary-Dermal Exposure module (TRIM.Expo); and
Risk Characterization module (TRIM.Risk).
APEX is the inhalation exposure component of TRIM.Expo. The APEX model is a
multipollutant, population-based, stochastic, microenvironmental model that can be used to
estimate human exposure via inhalation for criteria and air toxics pollutants. APEX is designed
to estimate human exposure to criteria and air toxic pollutants at the local, urban, and
consolidated metropolitan level. The current release of the model is Version 5.2 (October,
2019). Human exposure to a contaminant is defined as "contact at a boundary between a human
and the environment at a specific contaminant concentration for a specific interval of time"
(National Research Council, 1991). For air pollutants, the contact boundaries are nasal and oral
openings in the body. Dose is the amount actually received, or absorbed, in the body, leading to
physiological effects. Pollutant exposures are estimated in a microenvironmental model by
treating each individual's activities as a sequence of events, which are periods with known
starting and ending times in particular microenvironments. A microenvironment is a defined
space with relatively homogeneous air pollution concentration for a simulated individual.
"Indoor kitchen," "outdoor parking lot," or "in vehicle" are examples of microenvironments.
The pollutant concentrations in the air in each microenvironment are estimated from ambient air
pollutant concentrations and parameters specific to each microenvironment and each pollutant.
A person's inhalation exposures for a time interval are the pollutant concentrations in the
microenvironment that person for that interval multiplied by the length of the interval.
The APEX model uses the personal profile approach to generate simulated individuals, for whom
exposure time series are calculated. The profile is a description of the characteristics of an
individual that may affect their activities, their locations, or the concentrations in the
microenvironments that they encounter. Each profile is a random sample from the population (or
a specified population group) within the study area. Each profile is generated independently of
the others, and any number may be run, subject only to computer limitations. Typically, the
profile includes demographic variables such as age, gender, and employment, as well as
physiological variables such as height and weight, and finally some situational variables such as
living in a house with a gas stove or air conditioning. The situational variables are used to help
determine the microenvironmental concentrations, and the physiological variables are used in the
determination of ventilation rate and dose. The demographic variables are used in the selection
of activity diaries from EPA's Consolidated Human Activity Database (CHAD, McCurdy et al.,
1
-------
2000; EPA, 2009) to represent the individual. Note: CHAD is a comprehensive database of
human activity studies, which is provided with APEX. However, APEX may utilize other
human activity data at the discretion of the user. Throughout this document "CHAD" will be
used to denote the human activity diaries, although the reader should note that other data could
be used.
APEX calculates the exposure and dose time series for a user-specified number of profiles for
any number of pollutants. For modeling the pollutant CO, APEX contains an algorithm for
estimating the blood dose (percent carboxyhemoglobin, %COHb). APEX also contains an
algorithm for modeling particulate matter (PM) dose. In the case of PM, dose is defined as the
rate of mass deposition in the respiratory system. For modeling any other pollutant, APEX
calculates dose as Exposure* Ventilation (see Volume II for details of the dose algorithms).
Collectively, the APEX profiles are intended to be a representative random sample of the
population in a given study area. To this end, demographic data from the decennial census are
used so that appropriate probabilities for any given geographical area can be derived. In APEX,
the demographic geographical units are called sectors. Using the standard input files provided
with the model, each sector is a census tract. Ambient air quality and meteorology data for the
study area are also required by the model; the area represented by an air quality monitor (or air
quality model grid cell) is called an air district, and the area covered by a meteorological
monitor (or meteorological model grid cell) is referred to as a zone. APEX matches up each
sector in the study area with the closest air district and zone to provide the data necessary to
simulate exposure and dose for an individual.
For each simulated person (profile), the following general steps are performed:
select the profile variables to characterize the person;
construct the event sequence by selecting a sequence of appropriate activity diaries for the
person (using demographic and meteorological variables);
for each pollutant, calculate the concentrations in the microenvironments (using
situational variables);
for each pollutant, calculate the person's exposure and dose for each event; and
summarize the results for that profile.
The APEX model reports the results for each profile on various output files (some of which are
pollutant-specific), described in detail later in this guide. Once all the profiles have been
simulated, the model produces a set of summary tables for each pollutant that indicate the
distributions of exposure and dose across all the profiles.
APEX can be thought of as a simulation of a field study that would involve selecting an actual
sample of specific individuals who live in (or work and live in) a geographic area and then
continuously monitoring their activities and subsequent inhalation exposures to a specific air
pollutant during a specific period of time. The main differences between the model and an actual
field study are that in the model, the:
sample of individuals is a "virtual" sample, created by the model according to various
demographic variables and census data of relative frequencies, in order to obtain a
2
-------
representative sample (to the extent possible) of the actual people in the study area;
activity patterns of the sampled individuals (e.g., the specification of indoor and other
microenvironments, the duration of time spent in each) are assumed by the model to be
similar to individuals with similar demographic characteristics, according to activity data
such as diaries compiled in EPA's Consolidated Human Activities Database (CHAD)
(EPA, 2009; EPA, 2002; McCurdy et al., 2000);
pollutant exposure concentrations and doses are estimated by the model using temporally
and spatially varying ambient outdoor concentrations, coupled with information on the
behavior of the pollutant in various microenvironments; and
various reductions in ambient air quality levels due to potential emission reductions can be
simulated by adjusting air quality concentrations to reflect the scenarios under
consideration.
Thus, the model accounts for the most significant factors contributing to inhalation exposure
the temporal and spatial distribution of people and pollutant concentrations throughout the study
area and among the microenvironmentswhile also allowing the flexibility to adjust some of
these factors for regulatory assessments and other reasons.
1.2 Nomenclature
The terms below are used throughout this guide.
Diary: a set of events or activities (e.g., cooking, sleeping) for an individual in a given
time frame (e.g., a day)
Air quality district: the geographical area represented by a given set of ambient air quality
data (either based on a fixed-site monitor or output from an air quality model)
Event: an activity (e.g., cooking) with a known starting time, duration, microenvironment,
and location (usually home or work)
Microenvironment: a space in which human contact with an environmental pollutant takes
place
Profile: a set of characteristics that describe the person being simulated (e.g., age, gender,
height, weight, employment status, whether an owner of a gas stove or air conditioner)
Sector: the basic geographical unit for the demographic input to and output from APEX
(usually census tracts)
Study area: the geographical area modeled
Study area population: total population of persons who live in the study area
Meteorological zone: the geographical area represented by a given set of meteorological
data (either based on a meteorological station or output from a meteorological model
The labeling conventions below are used in this document.
Input and Output File Names are in italics title case (in some cases, key terms are also
introduced in italics, not capitalized, within a paragraph)
Model Variables are in bold italics
KEYWORDS, which are used in the input files to identify variables and settings, are given
in uppercase bold italics
3
-------
|InPut and output file excerpts are jn a ^Qx surrouncJecJ by a single line
Courier (fixed space) font is used for folder names, paths, and system
commands outside of APEX
1.3 Strengths and Limitations of APEX
All models have strengths and limitations, and for each application it is important to carefully
select the model that has the desired attributes. With this in mind, it is equally important to
understand the strengths and weaknesses of the chosen model. The following sections provide a
summary of the strengths and potential limitations of APEX.
1.3.1 Strengths
APEX simulates the movement of individuals through time and space to estimate their exposure
to individual or multiple pollutants in indoor, outdoor, and in-vehicle microenvironments.
Compared to conducting a field study that would involve identifying, interviewing, and
monitoring specific individuals in a study area, APEX provides a vastly less expensive, more
expedient, and more flexible approach. The model also allows different air quality data,
exposure scenarios, and other inputs and thus is very useful for decision making applications.
An important feature of APEX is its versatility. The model is designed with a great deal of
flexibility so that different levels of detail in the input data can be applied for a variety of
different applications. The input data sets supplied with APEX contain information for several
microenvironments, covering the needs of most applications. The air quality data input to the
model can be in the form of monitoring or modeling data. The data can be customized for
specific locationson roadways; within or geo-political units such as counties; census units such
as tracts; the locations of air dispersion model receptors; or the grid cells of Eulerian model
output. Criteria and hazardous air pollutants can be modeled by APEX.
A key strength of APEX is the way it incorporates stochastic processes representing the natural
variability of personal profile characteristics, activity patterns, and microenvironment
parameters. In this way, APEX is able to represent much of the variability in the exposure
estimates resulting from the variability of the factors effecting human exposure.
Secondly, APEX has the ability to estimate exposures and doses on different timescales (e.g.,
5-minute, hourly, daily, or annual) for all simulated individuals in the sample population from
the study area. This ability allows for powerful statistical analysis of a number of exposure
characteristics (e.g., acute and chronic exposure, correlations with activities and demographics),
many of which are provided automatically by APEX in formatted output tables.
APEX also estimates the exposures of workers in the areas where they work, in addition to the
areas where they live. The pollutant concentrations in these respective locations may be
substantially different from each other.
The use of APEX has been facilitated by the availability of model-ready input files which have
been developed from the databases discussed above: national population demographics and
4
-------
commuting information from either the 2000 or 2010 U.S. Census; CHAD activity data; and
microenvironment definitions.
1.3.2 Limitations
The limitations detailed below have been identified with regard to APEX.
The population activity pattern data supplied with APEX (CHAD activity data) are
compiled from a number of studies in different areas and for different seasons and years.
Therefore, the combined data set may not constitute a representative sample.
Nevertheless, much of CHAD is from random-sample studies of national scope, which
could be extracted by the user if desired to create a representative sample.
The commuting data address only home-to-work travel; travel between sectors for other
purposes is not modeled directly. APEX can model time spent in travel; however, based
on the model settings, using the ambient air quality for one or more study area sectors, or
by using a special roadway algorithm.
APEX creates seasonal or year-long sequences of activities for a simulated individual by
sampling human activity data from more than one subject in CHAD. Thus, uncertainty
exists about season-long exposure event sequences. This approach can tend to
underestimate the variability from person to person because each simulated person
essentially becomes a composite or an "average" of several actual people in the
underlying activity data (which tends to dampen the variability). At the same time, this
approach overestimates the day-to-day variability for any individual if each simulated
person is represented by a sequence of potentially dissimilar activities from different
people rather than more similar activities from one person. These uncertainties have been
reduced with the implementation in APEX of algorithms for combining diaries which
address these limitations to some extent.
The model currently does not capture certain correlations among human activities that can
impact microenvironmental concentrations (e.g., cigarette smoking leading to an
individual opening a window, which in turn, affects the amount of outdoor air penetrating
the residence).
Certain aspects of the personal profiles are held constant, though in reality they change
every year (e.g., age). This is only an issue for simulations spanning several years.
At this point in time, no interactions between pollutants are modeled.
Other data and model limitations exist besides those identified above, including physiological
data and algorithms, meteorological data, and the data and algorithms associated with estimating
concentrations in microenvironments. EPA continues to refine the model and data to reduce
these limitations to the extent possible. The uncertainties which result from these limitations of
APEX have been characterized for an ozone assessment (Langstaff, 2007).
1.4 Applicability
APEX is an advanced air inhalation exposure model which can be used for a range of
applications. APEX can be employed to model episodic "high-end" inhalation exposures that
result from highly localized pollutant concentrations (e.g., residual risk assessments). APEX can
also provide detailed probabilistic estimates of exposure for urban and greater metropolitan areas
5
-------
(e.g., for regulatory analyses supporting national decisions such as NAAQS reviews). APEX is
appropriate for assessing both long-term chronic and short-term acute inhalation exposures of the
general population or of specific segments of the population. The model is designed to look at
the range of inhalation exposures among different groups of people across a population, for a
range of averaging times, in a single simulation. The current version of APEX produces results
for flexible averaging times. By default, APEX produces results for 1 hour, 8 hours, 24 hours,
and annual time periods (or the length of a simulation, if shorter than one year). However,
APEX can optionally model results for timesteps on a much smaller scale (e.g., 5 minutes) by
setting appropriate run parameters and providing air quality data on the appropriate time scale.
Although it is outside the original scope of APEX, the model has been successfully applied to a
set of specific individuals by arranging to have exactly one person per sector and just one
matching set of activity diaries that may be assigned to each person. However, many properties
of these people remain randomly assigned as APEX is basically a stochastic model.
Due to the computational demands (run time and disk space) of running APEX, it is not
appropriate for national-level assessments of population exposure. However, this is not an
inherent limitation in the model code or algorithms.
1.5 Brief History of APEX
APEX was originally derived from the probabilistic National Ambient Air Quality Standards
Exposure Model (pNEM). The NEM series was developed to estimate exposure to the criteria
pollutants (e.g., CO, ozone). In 1979, EPA began to develop NEM by assembling a database of
human activity patterns that could be used to estimate exposures to outdoor pollutants (Roddin et
al., 1979). The data were then combined with measured outdoor concentrations in NEM to
estimate exposures to CO (Biller et al., 1981; Johnson and Paul, 1983). In 1988, OAQPS began
to incorporate probabilistic elements into the NEM methodology using activity pattern data
based on various human activity diary studies in an early version of probabilistic NEM for ozone
(pNEM/03). In 1991, a probabilistic version of NEM was developed for CO (pNEM/CO) that
included a one-compartment mass-balance model to estimate CO concentrations in indoor
microenvironments (Johnson et al., 1992). A newer version of pNEM/03 was developed in the
1990s and applied to nine urban areas for the general population, outdoor children, and outdoor
workers (Johnson et al., 1996a, b, c). During 1999-2001, an updated version of pNEM/CO
(versions 2) was developed that relied on activity diary data from CHAD and enhanced
algorithms for simulating gas stove usage, estimating alveolar ventilation rate (a measure of
human respiration), and modeling home-to-work commuting patterns.
APEX evolved from pNEM to provide greater applicability, flexibility, and accuracy. The
APEX model was substantially different than pNEM, particularly in the use of a personal profile
approach rather than a cohort simulation approach. APEX introduced a number of new features
including automatic site selection from large (e.g., national) databases; a series of new output
tables providing summary statistics; and a thoroughly reorganized method of describing
microenvironments and their parameters. Most of the spatial and temporal constraints were
removed or relaxed in APEX. Several major improvements to APEX have been introduced in
the most recent version, APEX5. Specifically, APEX5 includes the improvements listed below.
6
-------
Multipollutant capability
Algorithms for the assembly of multi-day (longitudinal) activity diaries that model intra-
individual variance, inter-individual variance, and day-to-day autocorrelation in diary
properties.
Methods for adjusting diary-based energy expenditures for fatigue and excess post-
exercise oxygen consumption
New equations for estimation of ventilation
The ability to model commuters leaving the study area
The ability to model air quality and exposure on different time scales
The ability to model person-to-person variability in air quality within an air district
New output files containing diary event-level, timestep level, and hourly-level exposure,
dose, and ventilation data, and hourly-level microenvironmental data
The ability to model the prevalence of disease states such as asthma
New output exposure tables that report exposure statistics for subpopulations such as
children and active people under different ventilation levels.
The ability to model inhaled dose for pollutants
The inclusion of commuting data from the 2000 or 2010 Census
Expanded options for modeling microenvironments
Expanded location options for sampling ambient air quality
The option of running Sobol sensitivity analysis
Model enhancements and other changes are occasionally made to APEX, and thus users are
encouraged to revisit the APEX website https://www.epa.gov/fera/human-exposure-modeling-
air-pollutants-exposure-model for notices of these changes. Comments and suggestions for
improvements to the model or the input data provided with the model should be addressed to:
John Langstaff, U.S. Environmental Protection Agency, Research Triangle Park, North Carolina
27711 (email: langstaffiohn@epa.gov).
1.6 Scope and Organization of This Guide
The documentation of the APEX model is currently divided into two volumes. Volume I: User's
Guide [this document] is designed to be a hands-on guide to the model. It focuses on how to run
the APEX computer model, develop the appropriate input files, and understand the model output
files.
Volume II: Technical Support Document describes the scientific basis of the APEX model and
provides scientific background for the model algorithms. It covers the methods implemented in
APEX for calculating microenvironmental concentrations, modeling ventilation, estimating dose,
and assembling activity diaries.
The remainder of Volume I have been organized into the chapters listed below.
Chapter 2. Setting Up and Running APEX: Provides instructions for setting up APEX and
running single or multiple APEX simulations
Chapter 3. Characterizing the Study Area: Described the procedure for characterizing the
area to be modeled, in terms of APEX input files
7
-------
Chapter 4. Input Files: Provides a description of the format, data, and options for each of
the APEX input files
Chapter 5. Output Files: Provides a description of the format and data associated with
each of the APEX output files
Volume II has the chapters listed below:
Chapter 1, Introduction
Chapter 2, Overview of Model Design and Algorithms
Chapter 3, Using Probability Distributions in APEX
Chapter 4, Characterizing the Study Area
Chapter 5, Generating Simulated Individuals (Profiles)
Chapter 6, Constructing a Sequence of Diary Events
Chapter 7, Estimating Energy Expenditures and Ventilation
Chapter 8, Calculating Pollutant Concentrations in Microenvironments
Chapter 9, Calculating Exposures
Chapter 10, Calculating Dose
Chapter 11, Sobol Sensitivity Analysis
8
-------
CHAPTER 2. SETTING UP AND RUNNING APEX
APEX is written in Fortran using only standard routines and conventions to allow portability to
different operating systems and compilers. APEX has been tested on Windows 10, 7, Vista, XP,
2000, NT, and 98 operating systems, as well as Linux, using Intel Fortran. Other compilers may
produce warnings and/or errors, and may require some small code changes for compatibility.
APEX currently uses ASCII coding internally.
In addition to providing flexibility in modeling options, the APEX code is specifically designed
for fast execution time and reasonable memory requirements.
We recommend running APEX on a computer with at least:
2 GB of RAM;
600 MHz processor; and
1000 MB of available hard drive space.
The input files supplied with APEX will require 500 MB of hard drive space, and the additional
input files created by the user may take up another 1-10 MB of space, or more, depending
mainly on the size of the air quality input files.
APEX run time on a PC with a 2.8 GHz Core i7 8 CPU and 8 GB of RAM, running Windows 7,
is about 1.8 hours for a one-year single-pollutant simulation of 20,000 individuals in a large
metropolitan area. The combined size of the output files from this simulation is small (-50 MB),
unless detailed daily and hourly data are requested, in which case the output files can take up
more than 5,000 MB.
This chapter, which describes the steps involved in setting up and running an APEX simulation,
is organized as shown below.
Section 2.1, Downloading and Setting Up APEX
Section 2.2, Setting up an APEX Simulation
Section 2.3, Overview of APEX Input and Output Files
Section 2.4, Overview of Model Settings and Options
2.1 Downloading and Setting Up APEX
To install APEX, download the APEX installer from http://www.epa.gov/fera. This will install
an APEX folder containing all of the necessary files to run the model, as well as supporting
documentation and supplementary input files. It is suggested that users install to the directory
"C:\APEX" (this is the default directory specified in the installer) as it will make running the
example APEX run easier. The file "unins000.exe" can be used to uninstall APEX from a user's
computer. This will remove the installed files but not files created by the user. See An
Introduction to APEX (EPA, 2017) for more details about installing APEX.
Note that if APEX has previously been installed on a user's computer, users can overwrite the
previous installation of APEX to that same directory or choose to install APEX to a different
9
-------
location on their computer. If users choose to overwrite the previous installation of APEX, the
installer will warn them that "The Folder: [directory] already exists. Would you like to install to
that folder anyway?" Selecting "Yes" will re-download APEX to the location it was previously
downloaded, and all of the files that exist in the APEX directories with names that match the file
names of the incoming installation (i.e., all the files that have their original names) will be
overwritten. If, however, users have generated new files with different names in the existing
folders or additional folders in the "APEX" directory, these files and folders will not be
overwritten with the re-installation of APEX.
To run APEX, you should open a DOS window in the directory your APEX run batch file is, and
then type the batch file name. (If the DOS link puts you in the wrong directory, right click on it,
select properties, and put the directory you want in the "Start in" box.) APEX should run at this
point and its progress will be indicated as it runs. If it fails for some reason, an error message
will appear. You should also look in the log file to check that the run was successful.
A batch file is a text file with the extension .bat. For a single APEX run a batch file (e.g.,
"runAPEX.bat") will contain one line, naming the APEX executable followed by the Control
Options file (COF), for example:
C:\APEX\EXEs\APEX.exe C:\APEX\Input\COF.txt
Then if you type 'runAPEX' in the DOS window, the APEX simulation governed by the COF is
run. You can run multiple APEX simulations using one batch file, for example:
C:\APEX\EXEs\APEX.exe C:\APEX\Input\COFl.txt
C:\APEX\EXEs\APEX.exe C:\APEX\Input\COF2.txt
C:\APEX\EXEs\APEX.exe C:\APEX\Input\COF3.txt
C:\APEX\EXEs\APEX.exe C:\APEX\Input\COF4.txt
C:\APEX\EXEs\APEX.exe C:\APEX\Input\COF5.txt
Then if you type 'runAPEX' in the DOS window, the 5 APEX simulations governed by the
COFs are run, one after the other. After the initialization of the run, APEX will begin
progressing through the simulated profiles. When the model run ends successfully, APEX will
stop with the message "Finished APEX model run."
Note that each of the Control Options files used should contain unique names for the model
output files to avoid overwriting the output from the previous run. As the model run starts and
then progresses, normal status messages will be printed to the screen, in addition to any error or
warning messages that may arise from incomplete or incorrect model settings.
Even if an APEX simulation runs to completion, the user should examine the APEX Log output
file to confirm that the model behaved as expected. The Log file contains information on the
model settings, input parameter values, and input and output file names. The file also contains a
great deal of detailed information about the model run including, but not limited to, summaries
of: 1) the modeled profiles; 2) the final study area (including the final sectors, air quality
districts, and meteorology zones); and 3) the simulated microenvironments. The Log file, which
is discussed in Section 5.1, will also contain a listing of any warning or error messages that
10
-------
resulted from the run. Some of the common error messages are explained in the companion
document An Introduction to APEX (EPA, 2017).
2.2 Setting Up an APEX Simulation
This section describes the steps involved in performing an APEX simulation.
1. Select Model Options
After identifying the scope of the analysis, the user must decide which options to select. To
determine the appropriate options for the application, the user must answer questions such as
those listed below.
How many pollutants do I want to model in a single run?
Do I want to model worker commuting?
How many profiles do I want to model?
How many microenvironments do I want to model?
How should I define my microenvironments?
How should the activity diaries be constructed (i.e., randomly select a new diary every day
for each simulated individual, or construct longitudinal diaries based on diary
properties?)
Which other model settings should I select?
What types of outputs do I want from the model?
What time resolution do I want to use?
These options and others are described in greater detail in Section 2.4.
2. Prepare Input Files
After deciding which model options to use and how to configure them, the next step in
configuring an APEX simulation is to set up the input data files with the necessary data. One of
these files, the Control Options file, is used to specify input and output file names and locations
and the simulation settings. The remaining files contain the input data necessary to run APEX.
The data contained in these remaining files varies depending on the configuration of the scenario
to be modeled and the number of pollutants used. The input files are described in CHAPTER 4.
3. Configure the Simulation Settings
The final step in preparing an APEX simulation is to create the Control Options input file for the
desired simulation settings. This file contains four sections:
input file names and locations;
output file names and locations;
pollutant parameters (including output table specifications); and
job parameters.
A detailed description of the data for each of the sections of the Control Options file is provided
in CHAPTER 4.
11
-------
4. Running APEX
To perform an APEX simulation, the user can run the model as described above in Section 2.1.
2.3 Overview of Input and Output Files
This section provides a brief overview of the input and output files associated with APEX. For
more detailed descriptions of the input and output files, refer to CHAPTER 4 and CHAPTER 5,
respectively. All of the input and output files used by APEX are ASCII text files; they can be
read and/or modified by the user using a text editor or other software. Note, however, that
certain files, such as the commuting flow input file and the hourly and events output files, may
be very large (in some cases several gigabytes), and difficult for some text editors to handle.
2.3.1 Input Files
There are several types of APEX input data files. Most of these files are required; however, the
Diary Statistics, Diary Occupation, Profile Factor and Prevalence files are optional in some
cases. With the exceptions of the Population Data and the Air Quality Data files, only one file
of each type is required for a simulation. The input file paths and names are designated in the
Control Options file using a "keyword" approach. APEX processes the file and searches for
particular keywords followed by an equal sign and one or more values for the keyword. Table
2-1 lists each file type and the keyword that must be used to identify it. CHAPTER 4 provides a
detailed description of the input files and their syntax.
2.3.2 Output Files
APEX utilizes a total of 16 possible output file types. These files contain such information as:
1) a summary of the properties of the simulated persons; 2) hourly or event-level exposures,
doses, and breathing data for the simulated profiles; 3) hourly or daily values of
microenvironmental parameters and pollutant concentrations; 4) dose and exposure summary
tables for the modeled population; and 5) exposure statistics for the modeled microenvironments.
The creation of some of the output files is dependent on settings in the Control Options file,
which also contains the path and file name for each output file.
Table 2-1 lists each of the input data files and their corresponding keywords, while Table 2-2
lists the output data files. If an output file is specified with the same name and location as an
existing file, the old files are overwritten. Therefore, if the user wishes to conduct a series of
model runs, the output files for each run should be named differently or written to a different
directory, or the output from a prior run should be moved elsewhere before the next model run is
submitted.
The APEX output files are covered in detail in CHAPTER 5.
12
-------
Table 2-1. APEX Input Files
Input Files
Pollutant
Specific?3
Control Options File Keyword
Control Options
-
Commuting Time
COMMTIME
Commuting Flow
COMMUT
ME Mapping File for Clustering Diary Assembly
DIARYCLUS
Diary Events (DiaryEvents)
DIARYEVE
Diary Occupation (DiaryOcc)
DIARYOCC
Diary Questionnaire (DiaryQuest)
DIARYQUEST
Diary Statistics (DiaryStat)
DIARYSTA
Diary Raw Transitions
DIARYTRANS
MET Distribution
DISTRIB
Air District Location
DISTRICT
Employment Probability
EMPLOY
Profile Functions
FUNCTIONS
Microenvironment Mapping (MEMap)
MEMAP
Meteorology Data
METEOR
MET Mapping
METMAP
Microenvironment Descriptions
MICROENV
Physiological Parameters
PHYSIOL
Population Data
POP
Prevalence
PREVAL
Profile Factors
PROFILE
Air Quality Data
YES
QUALITY
Population Sector Location
SECTOR
Seed offsets and Sobol grouping
SEED
Ventilation
VENTIL
Meteorology Zone Location
ZONE
a if yes, then a separate file is required for each pollutant modeled.
13
-------
Table 2-2. APEX Output Files
Output Files
Pollutant Specific?3
Control Options File Keyword
Daily
DAILY
Events
EVENT
Hourly
HOUR
Log
LOG
Microenvironmental Results
YES
MICRORES
Microenvironmental Summary
YES
MICROSUM
Profile Summary (Personsj
PERSON
Sites
SITE
Sobol
SOBOL
Output Tables
YES
TABLE
Timestep
TIMESTEP
MultiPollutant
YESb
MULTIPOL
Diary Clustering
CLUSTER
Diary Clustering Weights
CWEIGHT
Diary Clustered Transition
TRANSITIONS
a if yes, then a separate file is written for each pollutant modeled.
b the output is specific to the first two pollutants in a multipollutant run.
2.4 Overview of Model Settings and Options
This section briefly describes the primary settings and options available in APEX. These are
specified by the user in the Control Options file or other input files. There are six general
categories of settings and options in APEX:
general model settings and options;
study area location;
pollutants;
profiles;
microenvironments; and
outputs.
Table 2-3 describes the settings and options in each of these categories, how they are selected or
changed, and the impact of changing a setting or option on the other input files. See CHAPTER
4 for additional details of input files and their content, how to edit or create them, and how they
interact with other files.
"YES" or "NO" settings within the Control Options file are not case-sensitive and may be
abbreviated to a single letter; thus, "Y" or "y" means "YES," and "N" or "n" means "NO."
However, writing these out in full may provide more clarity.
14
-------
Table 2-3. APEX Settings and Options
Setting/Option How Option is Selected Impact
GENERAL MODEL SETTINGS AND OPTIONS
Simulation start/end
dates
Specified in YYYYMMDD format (e.g., 19960704 is July 4, 1996) using
the STARTDATE and END DATEkeywords in the Control Options
file. The user must define the appropriate start and end dates for an
application.
The indicated start and end dates must be included in the date ranges
specified on the Air District Location, Meteorology Zone Location,
Meteorology Data, and^4/r Quality Data files. These files may contain
additional data before and/or after the start and end dates, but must contain
data for the entire period between the specified start and end dates.
Adjust for Daylight
Saving Time (DST)
Specified using the DST AD JUST keyword in the Control Options file. If
this parameter is set to "YES," then the Air Quality Data file will be
adjusted for Daylight Saving Time (DST) in the summer; if it is set to
"NO," no adjustment is made. This keyword should be set to "YES" if the
Air Quality Data file is based on Standard Time yet the activity data are
based on DST. The default is "YES."
Changing this setting means that the Air Quality Data file is based on DST
(which is not usual) or that the activity data are based on Standard Time
(CHAD data are based on current local time, which is typically DST in
summer). Regardless of this setting, the output (hourly exposure and dose)
for all simulated days will contain exactly 24 hours, and all input activity
diaries must contain exactly 24 hours.
Model worker
commuting
Specified using the COMMUTING keyword in the Control Options file.
If this keyword is set to "YES," commuting to work is allowed and the
user must provide a Commuting Flow file in the appropriate format and
employment data must be specified in the Employment Probabilities file; if
it is set to "NO," all workers are assumed to work at home and the user is
not required to provide the Commuting Flow or Commuting Time files.
The Commuting Flow file accompanying APEX contains commuting flows
between all census tracts from the 2010 Census. These commuting data
are sufficient for all applications within the United States in which the
sectors are defined as census tracts. The Commuting Time file specifies
the duration of commutes for each census tract. The default is "NO."
If commuting is modeled, the Diary Questionnaire file must have an
additional column that lists the total commuting time on each diary. Also,
if the user chooses to define sectors as something other census tracts, new
Commuting Flow and Commuting Time files (in addition to a number of
other input files) must be created corresponding to the new sectors.
Air quality rollback
adjustment (for
estimating exposure in
hypothetical scenarios)
Specified using the ROLLBACK keyword in the Control Options file. If
this keyword is set to "YES," the user must specify appropriate values for
the RBTARGET, RBBACKGND, and RBMAX keywords in the Control
Options file; if it is set to "NO," values are not required for these keywords
(and any present will be ignored).
If the ROLLBACK key word is changed to "YES" in the Control Options
file accompanying APEX, the RBTARGET, RBBACKGND, and RBMAX
keywords must be set to appropriate values.
Time resolution
(length of APEX
timestep)
Specified using the TIMESTEPSPERDAY keyword in the Control
Options file, which must be an integer. The timestep can be either smaller
or larger than an hour. However if the timestep is larger than an hour, it
must be an integer multiple of an hour. If it is smaller than an hour, there
must be an integer number of timesteps in an hour. The default APEX
timestep is one hour. If TIMESTEPSPERDAYis not set, then APEX uses
the default, which is 24.
The timestep dictates the required time resolution of the following air
quality input. The time resolution of the Air Quality Data file must match
that indicated by TIMESTEPSPERDAY.
Random number seed
Specified using the RANDOMSEED keyword in the Control Options file.
If RANDOMSEED = 0, the program uses the clock to determine the initial
seed. Otherwise, the seed may range from 1 to 2147483646.
If all input data and settings are unchanged in two model runs, along with
RANDOMSEED, then the output will be identical.
15
-------
Setting/Option
How Option is Selected
Impact
Run Sobol analysis
Specified using the SOBOLRUN keyword in the Control Options file. If
YES, then a Sobol sensitivity analysis is performed (which takes
considerably longer). If "NO" (the default), a regular model run is
performed.
If SOBOLRUN = YES, the SobolVarList keyword is needed. A Sobol run
should have fewer profiles than a regular run, because the program
performs multiple passes through the profiles and creates additional arrays.
STl m AULA LOCATION
Center of study area
Specified as the latitude and longitude of the center of the study area in
decimal degrees using the LATITUDE and LONGITUDE keywords in the
Control Options file. The user must always define the appropriate study
area center for an application. Latitude and longitude are in decimal
degrees, without a compass direction. American locations (west of the
prime meridian) have negative longitude and positive latitude.
If the study area is changed, the user must ensure that the following files
contain appropriate data for the new location: Population Sector Location
file (unless the included file is used), Air District Location file,
Meteorology Zone Location file, Meteorology Data file, and^4/r Quality
Data file.
Radius of study area
Specified as the distance (in km) from the center to the edge of the study
area using the CITYRADIUS keyword in the Control Options file. The
user must always define the appropriate study area radius for an
application.
If the study area is changed, the user must ensure that the following files
contain appropriate data for the new location: Population Sector Location
file (unless the included file is used), Air District Location file,
Meteorology Zone Location file, Meteorology Data file, and^4/r Quality
Data file.
Restrict study area to
selected counties
Specified using the COUNTYLIST keyword in the Control Options file.
The default is "NO." If set to "YES," the "county" is identified by the first
5 characters of the sector ID, which for the default files is the state and
county FIPS code. Use the COUNTY keyword once for each county to be
included. This option operates in conjunction with the study area center
and radius.
None, normally. However, if the user does not use the included Population
Sector Location file, they must ensure that the new Population Sector
Location file provides suitable "county IDs," which consist of the first 5
characters of the sector IDs.
Restrict study area to
selected census tracts
Specified using the TRACTLIST keyword in the Control Options file.
The default is "NO." If set to "YES," the user must list the sector (tract)
ID for the tracts to which the study area will be restricted using the Tract
keyword in the Control Options file. The sector IDs for all census tracts in
the 2010 Census are included in the Population Sector Location file
accompanying APEX.
This is similar to COUNTYLIST, except that "tracts" are identified using
the first 11 characters of the sector name (the standard for U.S. census
tracts). If the user is not using actual census tracts, the TRACTLIST
option will still select all sectors from the Population Sector Location file
that match the first 11 characters (and are within CITYRADIUS of the
study area center).
Locations of sectors
(filename)
Specified as sector IDs and locations (latitude and longitude) in the
Population Sector Location file. The Population Sector Location file
accompanying APEX use the census tracts from the 2010 Census as
sectors. This file also specifies the county associated with each sector (via
the first five characters of the sector ID, which are the county FIPS codes
in the supplied data), which allows the user to restrict the study area to
selected counties. The default files provide sectors with the necessary
population and commuting data for the entire United States.
Sectors in Population Sector Location file must match the sectors in
Population Data files. The default file contains all U.S. census tracts from
the 2010 Census. If commuting is modeled, the sectors on the two
commuting files must match those on the Population Sector Location file
and the Population Data files; if other sector definitions are used, the user
must provide compatible Commuting files.
Locations of air
districts (filename)
Specified in the Air District Location file. The user must always define the
appropriate air districts for an application.
The locations of the air districts must be selected such that they can
provide reasonable estimates of air quality for the sectors and study period
included in the analysis. Data for each AQ monitor for each pollutant in
the simulation must be provided in the Air Quality Data files (one for each
pollutant).
16
-------
Setting/Option
How Option is Selected
Impact
Radius of air district
Using either a single the AirRadius keyword in the Control Options file or
multiple ones via the Air District Location file, the user can specify the
maximum distance (in km) that a sector can be from the nearest air district
to remain in the study. If a sector has no district centers within
AIRRADIUS of its center, that sector is removed from the study area and
is not modeled. Users must always define an appropriate value for this
radius based on their application.
The radius of the air districts must be selected such that they will include
the sectors the user would like to include in the analysis.
Type of air quality data
The keyword MODELAQVAR specifies the type of air quality data to be
used in the simulation. The air quality may be entered as values for each
timestep in the simulation (the default, MODELAQVAR = N) or as
distributions for each hour (MODELAQVAR = Y).
The value of MODELAQVAR dictates the expected format of the^4/r
Quality Data file. See Section 4.5 for details.
Using roadway
concentrations
The user has the option to model roadway concentrations if they set
ROAD WA Y= Y. The ROAD WA Y parameter can be set for each
pollutant, so that in a multipollutant run, it is possible to selectively use
roadway concentrations for each pollutant. If ROADWAY = NO, then all
MEs in the microenvironmental mapping file that are mapped to R will
instead be mapped to O. If the user specifies using roadway locations
based on the last home/work location (ROADLAST = YES), then roadway
locations will be chosen from either the home or work location, whichever
most recently occurred.
Additional concentrations must be specified in the Air Quality Data and
Air District Location files.
Location of
meteorological data
stations (filename)
Specified as zone IDs and locations (latitude and longitude) in the
Meteorology Zone Location file. The user must always define the
locations of meteorological stations for an application.
Data for each meteorological data station specified in the Meteorology
Zone Location file must be provided in the Meteorology Data file.
Radius of
meteorological station
coverage
Using the ZONERADIUS keyword in the Control Options file, the user
can specify the maximum allowed distance (in km) from a sector to the
nearest meteorological station. If all zone centers are further than
ZONERADIUS from the sector center, the sector is removed from the
study area and is not modeled.
The radius of the zones must be selected such that they will include the
sectors the user would like to include in the analysis.
POI.I.l i.W IS
Number of pollutants
The number of different pollutants to be modeled must be specified using
the #Pollutants keyword
Pollutant Names
The user must specify each pollutant with the keyword POLLUTANT.
The pollutant name may contain only alphanumeric characters and the
underscore (" ") character, as it is used to generate filenames.
Must be followed in the Control Options file by the pollutant-specific
parameters and output table levels. Must match the pollutant names in the
specification of the air quality data input files.
Model dose for
pollutant
Specified using the DODOSE keyword in the Control Options file.
Pollutant-specific. If this keyword is set to "YES," APEX will calculate
dose for the pollutant; if it is set to "NO," the dose calculations will be
suppressed.
If DODOSE is set to "YES" and CO is being modeled, the user must
specify the correct values for the ALTITUDE and COHBFACT keywords
in the Control Options file.
PKOI II.KS
Number of profiles
Set to a positive integer using the itPROFILES keyword in the Control
Options file. Users must determine an appropriate value based on the
application.
None.
17
-------
Setting/Option
How Option is Selected
Impact
Modeled populations
(filenames)
Specified in the Control Options file following the specification of the file
names. A population file is required for each combination of gender and
race to be modeled. All gender/race combinations without specified
population files are assumed to have zero populations and not considered
further. Users can select from the sets of available Population Data files
accompanying APEX, or generate their own.
These files must have the same sectors (in the same order) as the
Population Sector Locations file.
Profile function
options (filename)
Specified in the Profile Functions file. The user must develop data
relevant to a particular application prior to performing an APEX
simulation. The Profile Functions file is required.
None.
Employment status
(filename)
Specified in the Employment Probability file for implementation of
commuting and occupation. The file accompanying APEX should cover
all applications where sectors are defined as census tracts. The
Employment Probability file is required.
None.
Occupation status
Occupational status can be modeled by using the Profile Factor file and
specifying that it is an employment-related parameter. The user can match
by occupational status and it can also affect work-related MET values.
Minimum and
maximum ages for
simulated profiles
Specified using iheAGEMIN and AGEMAX keywords in the Control
Options file.
None.
Modeled age groups
Specified in the Population Data files. The files that accompany APEX
define the age groups as single years up to 99, and are sufficient for all
applications where sectors are defined as census tracts.
None.
Size of age window
The AGECUTPCT and AGE2PROBAB keywords in the Control Options
file are used to specify the window around the assigned age of a profile
from which activity data can be selected.
None.
Probabilities for
selecting diaries with
missing characteristics
Using the MISSGENDER, MISSEMPL, MISSAGE, and MISSOCC
keywords in the Control Options file, the user can specify the probability
that activity diary data with missing gender, employment status, age, or
occupation will selected.
None.
Type of diary assembly
Determined by the LONGITDIARYand CLUSTERDIARY keywords. If
LONGITDIARY = YES, longitudinal dairy assembly will be performed
based on the statistic in the Diary Statistics file. If CLUSTERDIARY=
YES, the clustering algorithm will be performed on the entire CHAD
database.
If LONGITDIARY is YES, then the Diary Statistics file must be
designated in the Control Options file, and the DIARYD and
DIARYA UTOC keywords must be set. If CL USTERDIARY = YES then
three output file switches CLUSTEROUT, TRANS OUT, and
CWEIGHTOUT are active, and two clustering parameters
CLUSTERAGES and USEADJACENTmay be set.
Physiological
parameters for the
simulated population
(filename)
Specified in the Physiological Parameters file. The default values in this
file are suitable for most APEX applications. This file is required.
None.
18
-------
Setting/Option
How Option is Selected
Impact
Activity-specific
energy expenditures
for the simulated
population (filenames)
Specified in the MET Mapping and MET Distribution input files. The
default values in these files are suitable for most APEX applications.
These files are required.
None.
Method of selecting
height and weight
Determined by the HTWTMETHOD keyword. If omitted (or set to 2), the
method implemented in 2017 is used, in which an explicit correlation is
maintained (so that for a given age and gender, the taller people are
heavier, based on empirical data from NHANES). HTWTMETHOD = 1
uses the older method, which was the only option previously.
Method of determining
resting metabolic rate
(RMR)
Determined by the RMRMETHOD keyword. If omitted (or set to 2), the
new (2017) method is used. Set RMRMETHOD = 1 to use the older
method.
Method of determining
breathing ventilation
rate VE
Determined by the VEMETHOD keyword. If omitted (or set to 2), the
method implemented in 2017 is used, in which VE is a function of both
V02 and V02max. Set VEMETHOD = 1 to use the older method.
Apply a maximum to
VE?
Determined by the VEMAX keyword. If omitted (or set to YES), then VE
is limited to 150 L/min for events up to 5 minutes, or to 100 L/min for
longer events. This was automatically applied prior to the implementation
of the new VE method. The maximum is not as relevant when using the
new method because it does not generate an excess of high VE values.
Modeling of disease
prevalence (option and
filename)
Determined by the DISEASE keyword. If DISEASE is given a value (a
string of maximum length 12 characters containing the condition name,
spaces allowed) in the input file, APEX will then use data in the
ePrevalence file to assign a YES/NO value to the physiological profile
variable 111, and produce output tables for the subpopulation of modeled
persons with 111 = YES. If disease is not defined, this file is not needed.
If DISEASE is given a value (a string of maximum length 12 characters
containing the condition name, spaces allowed) in the input file, then
APEX requires that a Prevalence file be designated as well.
MICROENVIRONMEMS
Maximum number of
microenvironments
Set to an integer using the #MICRO keyword in the Control Options file;
must not exceed 127.
Number of APEX microenvironments in the Microenvironment Mapping
and Microenvironment Descriptions files must not exceed the specified
value in the Control Options file.
Microenvironment
definitions (filename)
Specified in the Microenvironment Descriptions file. The user must
develop data relevant to a particular application prior to performing an
APEX simulation. This file is required.
Each location referenced in the activity database (e.g., CHAD) must be
mapped to one of the microenvironments specified in the
Microenvironment Descriptions file using the Microenvironment Mapping
file. The user may choose to define custom microenvironmental parameter
definitions that depend on conditional variables. If so, these variables must
be defined on the Profile Functions file.
()l 1 PI I S
Produce hourly outputs
Specified using the HOURLYOL'T keyword in the Control Options file. If
this keyword is set to "YES," the hourly output file named in the Output
File section is created; if it is set to "NO," the file is not created (even if it
is named). The variables to be written are listed using HOURLYLIST.
None.
19
-------
Setting/Option
How Option is Selected
Impact
Produce daily outputs
Specified using the DAILYOUTkeyword in the Control Options file. If
this keyword is set to "YES," the daily output file is created; if it is set to
"NO," the file is not created. The variables to be written are listed using
DA1LYLIST.
None.
Produce
microenvironmental
output
Specified using the MSUMOUT and MRESOUT keywords in the Control
Options file. If these keywords are set to "YES," the Microenvironmental
Summary and/or Microenvironmental Results output files are created; if
they set to "NO," these files are not created.
None.
Produce event output
Specified using the EVENTSOUT keyword in the Control Options file. If
this keyword is set to "YES," the events output file is created; if it is set to
"NO," this file is not created. This refers to activity diary events, which
are of variable duration between one minute and the timestep size.
None.
Produce timestep
output
Specified using the TIMESTEPOUT keyword in the Control Options file.
If this keyword is set to "YES" and the timestep is not one hour, the
timestep output file is created; if it is set to "NO," this file is not created. If
the timestep is one hour, then this file would be identical to the Hourly
output file, so it is not written. The optional variables to be written are
specified using TIMESTEPLIST. The "timestep" is the spacing in time of
the air quality data.
None
Produce multipollutant
table output
Specified using the MULTIPOL keyword on the Control Options file. If
this keyword is set to "YES" and at least two pollutants are modeled, then
the user must also specify level cut-points for each pollutant using the
keywords Multil and Multi2. The cut-points are written on one line in a
comma-separated list. If there are N cut-points, then there will be (N+l)
bins, because the lowest below means "below the first cut-point", and the
last means "equal to or above the last cut-point". Values exactly on a cut-
point go into the higher bin.
20
-------
CHAPTER 3. CHARACTERIZING THE STUDY AREA
An initial study area in an APEX analysis consists of a set of basic geographic units called
sectors^ typically defined as census tracts (see Section 1.2). The user provides the geographic
center (latitude/longitude) and radius of the study area. APEX contains a database with the
central locations of every census tract in the U.S. Each tract is assigned a central point as its
location, as determined by the U.S. Census Bureau. APEX calculates the distances to the center
of the study area of all the sectors included in the sector location database, and then selects the
sectors within the radius of the study area. One can also provide a list of counties or census
tracts as part of the specification of the initial study area. APEX then maps the user-provided air
quality district and meteorological zone data to the selected sectors. The sectors identified as
having acceptable air and meteorological data within the radius of the study area are selected to
comprise a final study area for the APEX simulation analysis. This final study area determines
the population make-up of the simulated persons (profiles) to be modeled.
All geographical data in APEX use latitude and longitude to locate places. The convention is
that both are specified in decimal degrees (do not use degrees, minutes, and seconds), and
longitudes in the western hemisphere are negative, as are latitudes in the southern hemisphere.
Longitude "wraps around" with -180 degrees equal to +180 degrees. The following sections
describe in more detail how a final study area is determined in an APEX simulation analysis.
3.1 APEX Spatial Units
3.1.1 Initial Study Area
The APEX study area has typically been the neighborhood around an emission source, a city, or
a larger metropolitan area. Larger study areas are possible to simulate, depending on computing
capabilities, available data, and the desired precision of the run.
The user defines an initial study area by specifying the latitude and longitude of a central point
(referred to here as the study area central location), together with a radius. The user also has the
option of providing a list of counties or census tracts to be modeled. If present, this list further
restricts the area to be modeled to the counties or tracts to be modeled which are within the
specified study area radius. The final study area is a function of the availability of the user-
supplied demographic data, pollutant concentration data, and the meteorological data within the
initial study area, as determined respectively by population sectors, air quality districts, and
meteorological zones. Figure 3-1 and the subsections below provide additional details about
these geographical units.
3.1.2 Sectors
The demographic data used by the model to create personal profiles is provided at the sector
level. For each sector the user must provide demographic information allowing the
determination of age, gender, race, and work status. This is most commonly done by equating
sectors with census tracts and providing input files with counts at the tract level for each age,
gender, and race combination. The current release of APEX includes input files that already
contain this demographic and location data for all census tracts in the 50 states and Washington,
21
-------
D.C., based on the 2010 Census. Data files based on the 2000 Census are also available. One of
the APEX input files, named Population Sector Location file in this guide, lists the sector ID and
location for all sectors that have associated population data. The supplied Population Sector
Location file has been prepared listing all the census tracts in the 2010 U.S. Census.
Corresponding Population Data files have been supplied as well. This allows the user to model
any desired study area in the country without having to make any changes to these input files.
The 2000 Census files have also been used with APEX. However, one cannot mix 2000 and
2010 files because the lists of tracts are not the same.
hit ial Study Area
(withi n C fty R a dius Study Area Cent er
distance from Study y
Area Center) y
/
Intermediate Study
Area (e.g., Charlotte,
HC C MS A)
i
Zone Center
Zone Radi
I
Zone(Zone
Radius distance
from Zone
Center)
\ \
District Air Monitor
I
\
District Ar Radius
s
\
\
District (Air Radius
. distance from Air Monitor)
X
X
City (Study A-es) Radius
N
X
\
Sector
(census
tract)
Final Study Area
(dark line;the
Sectors comprising
the Ave Districts)
Figure 3-1. Example of Study Areas, Air Quality Districts, Meteorological Zones, and
Sectors
If available, finer scales such as census block groups could be used instead. Also, data could be
aggregated to larger regions such as counties if fewer sectors were desired. Regardless of the
specific meaning for sectors in APEX, the shape of sectors is irrelevant in the sense that the
model only uses the central location each sector, determined by the latitude and longitude for
some representative point. Sector names in APEX are 40 characters or fewer.
In the Control Options file for an APEX run, the user specifies the area to be modeled by
specifying the latitude and longitude of a central location for the study area, along with a radius
(the CITYRADIUS parameter). Optionally, the user may also provide a list of counties or
22
-------
sectors (typically census tracts, but can be census blocks or other geographic units) to be
modeled. If present, this list further restricts the study area.
For each model run, APEX selects the sectors that meet the study area conditions in the
following way. First, the sector location must be within the specified distance (radius) of the
designated center of the study area. Second, if the user provided a specific list of counties
identified by their FIPS codes (using COUNTYLIST and COUNTY in the Control Options file),
then the sector must belong to one of these counties. This is achieved using the first five
characters of the sector ID, which should contain the FIPS code for the county in which the
sector is located. The user may also provide a list of census tracts (sectors) using the
TRACTLIST and TRACT variables, which match the first 11 characters of the sector ID. If
census tract names are not being used as the sector names, the COUNTYLIST (or TRACTLIST)
option will still select those sectors matching the first 5 (or 11) characters with any names on the
designated list.
If no county or tract list is provided, the initial study area is roughly circular, consisting of all
sectors with sector locations within the specified radius. Sectors are never divided; each one is
either entirely inside or entirely outside the study area.
One way to exert greater control over the selection of sectors is to edit the census input files to
eliminate any undesired sectors. However, the Population Sector Location file and the various
Population Data files must all have the same set of sectors in them, so consistent editing is
necessary. The Population Sector Location and Population Data files provided with the current
release of APEX contains data for every census tract in the 50 states and Washington, D.C. from
the 2010 Census.
3.1.3 Air Quality Districts
The spatial units for ambient air quality data are called districts. Ambient air quality data are
provided as time series at specific locations. Previous versions of APEX required hourly time
series, but now the user may specify other timesteps. All the air quality data in a single APEX
run must use the same timestep. The locations could be monitoring sites, political units such as
counties, census units such as tracts, or receptor locations or grid points as used by some air
quality models. As with sectors, each air quality district has a nominal central location indicated
by its latitude and longitude. The air quality district locations are stored in the Air Quality
District Location input file. The user designates the maximum representative radius of the air
quality districts as a modeling parameter cal 1 ed AIRRA I) I US A single AIRRADIUS specified
on the Control Options file applies to all air districts. Otherwise, the user can specify a different
AIRRADIUS for each district on the Air Quality District Location file. The same set of air
quality districts must be used for each pollutant. Users can also indicate air data to be used
specifically for roadway concentrations in the Air Quality District Location file by including
"Road" as part of the district name. These data will be used for microenvironments set to "R
(Road)" or "RW (Road Work)" in the Microenvironment Mapping file instead of using the
ambient AQ data for the person's Home or Work district.
The model checks each air quality district listed in thzAir Quality Data input files (one for each
pollutant) to determine if the district has data covering the entire simulation period, as indicated
23
-------
by the start and end dates. Districts without complete data are dropped. Each air quality district
may have a different period of operation (i.e., different start and/or end dates). When the air data
are read, there can be no gaps (missing data) between the simulation start and end dates, or else
APEX stops, reporting error #4 in the ReadAirQuality module. If the user is supplying
monitoring data as inputs, for example, then any missing values within the simulation period
must be filled in before running APEX.
APEX calculates the distance from each sector location to each air quality district center, and
then assigns the sector to the nearest air quality district, provided it is within the maximum
representative air quality radius. If there is no air quality district within range (that is, all air
quality district centers are further than AIRRADIUS from the sector center), the sector is deleted
from the study area and is not modeled. It is possible, and perhaps even likely, that some air
quality districts in the Air Quality Data input files will have no sectors assigned to them. Such
air quality districts are not used. This feature allows the user to prepare an input file in the
simplest manner, perhaps containing more air quality districts than are necessary. For example,
one might prepare a single Air Quality Data input file for a pollutant for all air quality monitors
in the state of Texas. This same input file could then be run on a study area around Houston,
Dallas, or some other location in Texas, without having to alter the input file.
By default, APEX will assign each person within a sector the corresponding appropriate ambient
values from the sector's matching air district. Thus, for each timestep in the simulation, all
persons within the same district will have the same outdoor air quality value. However, APEX
can optionally model person-to-person variation in air quality within an air district. In this case,
an optional form of the Air Quality Data file is provided, which lists air quality distributions for
each hour for each air district. Each person in the district is assigned a randomly-sampled value
from the appropriate hourly distribution. Note that this option can only be used when the APEX
timestep is equal to 1 hour (the APEX default).
3.1.4 Modeling Commuting
APEX models commuting by assigning a work sector to each employed individual based on
commuting data for that individual's home sector. Two commuting data files are required.
These are files consisting of: 1) commuting flow data (the Commuting Flow file), and 2)
commuting time data (the Commuting Time file). Nationwide files are supplied with APEX.
The nationwide input files use census tracts as the sectors. The Population Data files and the
two commuting files must refer to the same census (for example, the 2010 Census), as the list of
tracts must match in all three files. APEX extracts the flows for the selected home sectors from
the Commuting Flow file and derives profile level commuting times from the Commuting Time
file.
In APEX5, a new option has been added which allows the user to use the national Commuting
Flow file with finer sectors, such as census blocks or block groups. The reason is that this is the
only input file that provides a sector-to-sector mapping. An exhaustive listing of sector pairs
may become unmanageable.
24
-------
To use the national Commuting Flow file with finer sectors, it is necessary that each sector name
be the same length, and that the first 11 characters of the name indicate the proper census tract.
If APEX detects that the commuting flow place names are shorter than the sector names, then it
automatically performs this mapping. The same destination probabilities are applied to each
sector within a given commuting tract. Once a destination tract is chosen, each sector within that
tract (and also within the study area) is equally likely to be selected as the work sector.
This capability has not been extended to other input files that list sectors. All the other such files
are linear in the number of sectors, and even with very fine sectors, these files are still of
reasonable length.
3.1.5 Meteorological Zones
Another spatial unit in APEX is the meteorological zone, which is the equivalent to air quality
districts but for meteorological data. Most of the rules that apply to air quality districts also
apply to meteorological zones. Each meteorological zone is associated with a central location
(specified by latitude and longitude), a maximum representative radius given by ZONERADIUS,
a START DATE, and an END DATE. The start and end dates may differ for each
meteorological zone and must encompass the entire simulation period, otherwise the
meteorological zone is deleted. If there are missing data between the start and end dates of the
simulation, APEX stops, reporting error #4 in the ReadMeteorologyData module.
APEX calculates the distance from each sector location to each meteorological zone center and
assigns each sector to the nearest meteorological zone if within range {ZONERADIUS),
otherwise the sector is deleted.
3.1.6 The Final Study Area
The final study area consists of all the sectors within CITYRADIUS of the study area central
location, restricted to the listed counties or tracts (if provided), that have both an air quality
district and a meteorological zone within specified ranges. If both tracts and counties are listed,
then the resulting study area is the union of the two lists. Sectors for which a valid air quality
district (that is, within a distance AIRRADIUS) or a valid meteorological zone (within
ZONERADIUS) cannot be found are discarded from the final study area. The study area
population is the total population in the input Population Data files that reside in these sectors.
25
-------
CHAPTER 4. APEX INPUT FILES
This chapter provides the details necessary for creating and modifying the APEX input files.
The first section describes the general format and properties that pertain to all of the APEX input
files, while the remaining sections cover each input file in detail.
4.1 Input F ile F ormats
The APEX input files and their descriptions are given in Table 4-1. Some input files are not
required if certain features of the model are turned off. For example, the Diary Statistics file is
not needed if the D&A longitudinal diary assembly is not being used, the Prevalence file is not
needed if disease is not defined, and the Commuting Flow and Commuting Time files are not
needed if commuting is not considered. All input files are ASCII text files that can be edited
using a text editor. Each input line of these files is categorized into one of four types, as noted
below.
1. Keyword (or variable or parameter) line: Keywords are used in the Control Options file to
indicate to APEX where the input files are located and what values should be assigned to
certain variables. A keyword line always contains an "=" sign. The part of the line to the
left of "=" is called the "keyword" and the part to the right is called the "value." The
keyword must start with a letter and must match the spelling sought by the program code,
after which the keyword may contain additional letters, blanks, or commas. Keywords
are not case sensitive. APEX uses the keyword to identify and set the input values. The
values may be character, logical, or numeric values, or file names.
2. Numeric line: Any line beginning with a digit (0-9) is recognized as a numerical data line
by APEX. Non-digits may appear later in a numeric line.
3. Character line: A line that begins with a character but does not contain an "=" sign is
recognized as a character data line.
4. Comment line: Any blank lines and any lines beginning with "!" generally are regarded as
comment lines by APEX and are used by the user to help document the file. However,
comment lines should not be inserted in the middle of a block of data. That is, if the
computer code is expecting to read a long series of numbers without a break, then
comments may break the flow.
The keywords and input values are not case sensitive, except as noted. Additionally, each line
on an input file is processed independently by APEX. Continuation of data values across
multiple lines is not permitted unless specifically noted for a particular file. APEX uses "list" (or
"free") format for all input values. This means that the values or data do not have to be fixed in
specific positions on an input line. Multiple items on an input line can generally be separated by
either a blank or comma, although certain items (as noted below) insist on one or the other.
Words on numeric and character input lines should not contain internal blanks, as these will be
interpreted as delimiters between input fields. This does not apply to keyword lines, as those
lines have only two fields (separated by the "=" sign), so either or both sides may contain
internal blanks. Keywords can have additional characters after the required characters given in
the table (e.g., "commut" can be "commuting").
26
-------
Table 4-1. APEX Input File Descriptions
Input File
Keyword
Description
Control Options
(Required) Specifies the overall settings (or parameters) for an APEX
simulation, including input and output file names, job settings and
switches, and output table levels.
Commuting Time
COMMTIME
(Optional) Provides a distribution of recorded commuting times for
workers in each home census tract. This file is only required when work
commuting is modeled (COMMUTING = Y).
Commuting Flow
COMMUT
(Optional) Provides probabilities of a worker commuting to various
destination census tracts from any given home tract. This file is only
required when commuting is modeled (COMMUTING = Y).
Diary Cluster
Mapping
DIARYCLUS
(Optional) Provides the mapping from activity location codes in the Diary
Events file (e.g., from CHAD) to user-defined microenvironments for the
purposes of calculating transition probabilities used in the clustering
method of diary assembly (needed only when CLUSTERDIARY=Y).
Diary Events
DIARYEVE
(Required) Provides the 24 hour event descriptions (i.e., start time,
duration, activity, and location) for all the diary days in the original
activity database. This file contains the same list of diary IDs as the
Diary Questionnaire file, in the same order, but it has multiple records
(one per event) for each ID.
Diary
Occupations
DIARYOCC
(Optional) Contains an occupation for all CHAD activity diaries. If
provided, these occupations will overwrite the default CHAD occupations
listed on each diary.
Diary
Questionnaire
DIARYQUEST
(Required) Provides personal and other information (e.g., day type,
gender, age, race, occupation) relating to each 24 hour activity record
from the original CHAD activity database. If commuting is used, this file
must also contain commuting times.
Diary Statistics
DIARYSTA
(Optional) Contains the value of the key statistic for all CHAD activity
diaries, used in the D&A method of longitudinal diary assembly.
Statistics files are included with APEX for outdoor time and time spent in
vehicles. Users could construct other statistics files from CHAD.
(Required if LONGITDIARY = Y).
Diary Transitions
DIARYTRANS
(Optional) Contains a list of diary IDs that belong to the distinct persons.
Multiple diaries from the same person are used to calculate empirical
transition probabilities when the CLUSTERDIARY = YES option is
active.
MET Distribution
DISTRIB
(Required) Provides distribution types and parameters for calculating the
metabolic equivalent of task (MET) value for each distribution number in
the MET Mapping file. A MET value is a dimensionless ratio of the
activity-dependent energy expenditure rate to the basal or resting energy
expenditure rate.
Air District
Location
DISTRICT
(Required) Provides the site IDs and locations (degrees latitude and
longitude) of air quality monitoring or modeling locations. The file is
used along with the user-defined AIRRADIUS to define the geographical
area covered by the air quality data. Sectors within the study area that are
close enough to an air district that has data between the simulation start
and end dates are retained for modeling.
Employment
Probability
EMPLOY
(Required) Contains employment probabilities by age group, gender, and
study sector. The default file is based on the tracts from the 2010 U.S.
Census. For other definitions of sectors, the user would have to supply a
new employment file.
Profile Functions
FUNCTIONS
(Required) Contains user-defined functions for several model variables,
which in turn can be used by the model for a variety of purposes,
including calculating microenvironmental concentrations.
27
-------
Input File
Keyword
Description
Microenvironment
Mapping
MEMAP
(Required) Provides the mapping from activity location codes in the
Diary Events file (e.g., from CHAD) to user-defined microenvironments
in the Microenvironment Descriptions file.
Meteorology Data
METEOR
(Required) May contain temperature, wind, humidity, and precipitation
data for the meteorological stations and dates indicated in the
Meteorology Zone Location file. These data can be used to determine
window positions, group activity pattern pools, and microenvironmental
parameters in APEX.
MET Mapping
METMAP
(Required) Maps each activity codes present in the Diary Events file to an
MET distribution. (A MET is a dimensionless ratio of the activity-
dependent energy expenditure rate to the basal or resting metabolic rate).
The MET distributions are defined by number in the MET Distribution
File.
Microenvironment
Descriptions
MICROENV
(Required) Contains the definitions of the microenvironments and the
microenvironment parameters used to determine the exposure
concentrations in microenvironments. The data on this file will change
greatly from one pollutant to another, so no default version of this file is
provided.
Physiological
Parameters
PHYSIOL
(Required) Contains tables of age- and gender-specific physiological
parameters.
Population Data
POP
(Required) Contains information on the population (by age group) in each
study sector. Each race/gender combination being modeled has its own
file.
Prevalence
PREVAL
(Optional) Contains prevalence rates (probabilities) for disease (or any
other condition) for different age/gender cohorts. This file is optional
only if the Control Options file variable DISEASE is not defined.
Profile Factors
PROFILE
(Optional) Contains details of a user-specified profile factor that can vary
by age group, gender, and study sector, and can be used to specify
microenvironmental parameters.
Air Quality Data
QUALITY
(Required) Provides the air quality data for each air monitoring/modeling
location listed in the Air District Location file. The time resolution on
this file depends on the Control Options file setting
TIMESTEPSPERDA Y. There is one Air Quality Data file per pollutant.
Optionally, the file may include distributions for hourly air quality values,
see Section 4.5 for details.
Population Sector
Location
SECTOR
(Required) Provides the IDs and locations (in degrees latitude and
longitude) of sectors (e.g., census tracts). The file is used along with the
user-defined CITYRADIUS and other data to select the sectors within the
modeled area.
Seed offsets and
Sobol grouping
SEED
(Required) Contains variable-specific offsets for random number seeds.
This file also contains the Sobol grouping numbers for each random
variable, but these are used only if SOBOLRUN = Yes. If not a Sobol
run, these numbers must still be present for file formatting reasons, but
their values do not matter, and may all be zero (for example).
Ventilation
VENTIL
(Required) This file contains regression parameters used to estimate total
ventilation (VE) from MET. The format of the ventilation file is different
for VEMETHOD = 2 versus VEMETHOD = 1.
Meteorology Zone
Location
ZONE
(Required) Provides the site IDs and locations (degrees latitude and
longitude) of the meteorological stations. The file is used along with the
user-defined ZONERADIUS to determine the area covered by the
meteorological data. Start and end dates indicate the dates during which
the data for a particular location are valid.
28
-------
The following sections discuss the details of APEX input files and provide several examples.
Note that these example files in this and the next chapter are provided for illustration purposes
only. These are provided for the purpose of highlighting various aspects and options of APEX.
Most of these examples are only portions of the necessary input files. Thus, these example files
will not work as an actual set of input files. Users are encouraged to view the example input files
(which can be downloaded separately) for a complete set of input files.
4.2 Control Options File
The Control Options file is APEX's master file. The Control Options file names all the other
input and output files, sets model parameters, and controls formats of output files. APEX only
processes keyword lines in this file. Any other types of input lines are ignored. However, the
very first line of this file (even if it is a comment beginning with !) is saved to be used as part of
the header that is written to each output file for audit trail purposes. Therefore, it is helpful for
this line to include information that describes or identifies the simulation.
When creating the Control Options file, the rules listed below should be used.
The very first line of the file should identify the specific simulation (up to 224 characters
in length) (it does not need to start with !)
Keywords (or parameter or variable names) are placed to the left of the equal sign in a
keyword line, and are not case sensitive
Parameter values are to the right of the equal sign
Yes/no parameters are not case sensitive and may be abbreviated to "Y" or "N"
Lines may appear in any order after the first line, with the exceptions listed below
o Lines using the COUNTY keyword, which must immediately follow the line with
the COUNTYLIST keyword
o Lines using the Tract keyword, which must immediately follow the line with the
TRACTLIST keyword
o Lines using the Block keyword, which must immediately follow the line with the
BLOCKLIST keyword
o Lines using pollutant-specific parameters or table levels, which must immediately
follow the line with their corresponding POLLUTANT keyword
Lines for specific keywords may be omitted if defaults are allowed and are acceptable
Only one equal sign is allowed per keyword line
Anything after an exclamation mark in a line is treated as a comment and ignored
Any unexpected line without an equal sign treated as a comment and is ignored
The information on the Control Options file is not case-sensitive, except that quoted
strings are left as they are. On Windows systems, file names are not case sensitive.
It is useful to keep a copy of the Control Options file associated with each simulation to provide a
record of the input and output files and model settings associated with the simulation and to
make it easier to run the model again, either with or without modifications.
29
-------
We describe the Control Options file in terms of four sections of the file: input files, output files,
pollutant parameters (including output table levels), and job parameters. Organizing the Control
Options file in this manner is not required. The details of each section are discussed below.
4.2.1 Input and Output File List Sections of the Control Options File
In the Input Files section of the Control Options file (Exhibit 4-1), the user needs to specify the
names and path names of all of the input files. The details on the content and format of these
input files are provided in the subsequent sections of this chapter.
The keywords for these files were given in Table 2-1. The keyword may be extended, as long as
the listed keyword is contained within the text. For example, "Employ" could become
"Employment," but "MEMap" cannot be "MicroenvironmentMap." The keyword FILE must
appear (with a blank space before it) right after each of the file keywords and before the "=".
Full paths including the drive letter are required. If any of these files are not found at the
specified locations, then APEX will print an error listing the file that is missing.
The Air Quality Data files are the only input files that are pollutant-specific with one file for
each pollutant modeled. Each AIR QUALITY FILE keyword must be followed by a comma and
the name of its corresponding pollutant (i.e., the pollutant names must match the names given by
the POLLUTANT keyword in the Control Options file; see Section 4.2.2). Exhibit 4-1 provides
an example of designating Air Quality Data files for a two-pollutant scenario (CO and ozone).
The example in Exhibit 4-1 has 10 population data files. The number of population files could
change, depending on how the user classifies the population. For example, the user could
provide two population files, one for all females and one for all males.
For the population input files, the keywords POP and FILE must appear at the beginning of the
keyword part of the keyword input line in the Control Options file, followed by a comma and
GENDER and another comma and RACE. GENDER currently must be either male or female
and it can be shortened to M or F. If the population files provided with APEX are to be used, the
RACE must be White, Black, Asian, NatAm, or Other. If the user provides different population
files, RACE can be customized, however, the first 5 characters of each race must be unique. It is
necessary for RACE to match the designation in the header of the population files, or an error
will result. Further information on population files is given in Section 4.8. It is not necessary to
specify all genders and race combinations for APEX to run. However, the model assumes that
any missing gender/race combinations have zero population. A warning message is returned if
one gender for a race is present but the other is missing.
In the Output File section of the Control Options file, illustrated in Exhibit 4-2, the user needs to
specify the keywords (see Table 2-2), names, and paths for the output files. If the user turns off
the hourly file creation, event file creation, or microenvironmental summary file creation, the
corresponding output files will not be generated, and file names do not need to be specified. The
Microenvironmental Summary, Microenvironmental Results, and Tables files are pollutant-
specific; one of each of these files will be created for each pollutant. However, only one
filename for each type has to be defined in the Control Options file - output filenames for each
pollutant are constructed by appending the pollutant name (as defined using the Control Options
30
-------
file POLLUTANT keyword) to the end of the filename base. Further details on output files are
given in CHAPTER 5.
! INPUT FILES
Zones file
=
C:\APEX\Input\METsites.txt
Air Quality file, Ozone
=
C:\APEX\Input\AirQuality ozone.txt
Air Quality file, CO
=
C:\APEX\Input\AirQuality co.txt
Districts file
=
C:\APEX\Input\AQdistricts.txt
Meteorology file
=
C:\APEX\Input\METdata h.txt
Functions file
=
C:\APEX\Input\ProfileFunctions.txt
Microenvironment file
=
C:\ APEX\Input\MicroDescriptions.txt
MEMap file
=
C:\APEX\Input\ME Mapping.txt
DiaryEvent file
=
C:\APEX\Input\CHADEvents.txt
DiaryQuest file
=
C:\APEX\Input\CHADQuest.txt
METMap file
=
C:\ AP EX\Input\CHADMap.txt
MET Distribution file
=
C:\APEX\Input\MetsDists.txt
DiaryStat file
=
C:\APEX\Input\CHADSTATSoutdoor.txt
Physiology file
=
C:\APEX\Input\Physiology.txt
Ventilation file
=
C:\APEX\Input\Ventilation.txt
Diarycluster file
=
C:\APEX\Input\Cluster ME Mapping.txt
Prevalence file
=
C:\APEX\Input\Asthma.txt
Seed offset file
i
C:\APEX\Input\Groupings.txt
! POPULATION INPUT FILES
Pop file, Female, Asian
=
C:\APEX\Input\pop fa.txt
Pop file, Female, Black
=
C:\APEX\Input\pop fb.txt
Pop file, Female, Natam
=
C:\APEX\Input\pop fn.txt
Pop file, Female, Other
=
C:\APEX\Input\pop fo.txt
Pop file, Female, White
=
C:\APEX\Input\pop fw.txt
Pop file, Male, Asian
=
C:\APEX\Input\pop ma.txt
Pop file, Male, Black
=
C:\APEX\Input\pop mb.txt
Pop file, Male, NatAm
=
C:\APEX\Input\pop mn.txt
Pop file, Male, Other
=
C:\APEX\Input\pop mo.txt
Pop file, Male, White
=
C:\APEX\Input\pop mw.txt
Sectors file
=
C:\APEX\Input\pop geo.txt
Employment file
=
C:\APEX\Input\Employment.txt
Commuting file
=
C: \APEX\Input\Commuting2000.txt
CommTime file
=
C:\APEX\Input\CommutingTimes2000.txt
Exhibit 4-1. Input Files Section of an Example Control Options File
! OUTPUT FILES
log file
=
C:\APEX\Output\log.txt
hourly file
=
C:\APEX\Output\hours.txt
daily file
=
C:\APEX\Output\days.txt
events file
=
C:\APEX\Output\events.txt
persons file
=
C:\APEX\Output\psum.txt
microsum file
=
C:\APEX\Output\msum.txt
microres file
=
C:\APEX\Output\mres.txt
tables file
=
C:\APEX\Output\tables.txt
site file
=
C:\ AP EX\Output\sites.txt
Sobol file
=
C:\APEX\Output\Sobol.txt
Exhibit 4-2. Output Files Section of an Example Control Options File
31
-------
If the user is intending to run APEX multiple times, then the output files from each run should be
given different names; otherwise, the later runs will overwrite the earlier ones. It may be
convenient to create a folder for each run. However, the inputs often refer to the same file in
multiple runs. This avoids having a separate copy of a large file (like CHAD, or the commuting
databases) for each run.
4.2.2 Pollutant Parameters Section of the Control Options File
Several Control Options file keywords described in Table 4-2 are pollutant-specific.
#POLLUTANTS determines the number of different pollutants being modeled, and must precede
all of the pollutant-specific keywords. The POLL UTANT keyword is used to: 1) assign a name
to each pollutant being modeled, and 2) designate the start of a keyword list for another
pollutant. The pollutant name may be up to 40 alphanumeric characters long and may contain
underscores When modeling particulate matter (PM), each discrete size of PM must be
modeled as a separate pollutant. All PM pollutants must start with the characters "PM." The
SIZE and DENSITY parameters must be defined for these pollutants.
Table 4-2. Pollutant-specific Job Parameters
Keyword
Type
(length)
Description
#.POLLUTANTS
Integer
(Optional) The number of pollutants in the simulation. Any number of
pollutants may be modeled - the maximum is limited only by the available
system memory.
Default = 1
POLLUTANT
Char(40)
(Required) Pollutant name. There must be one POLLUTANT statement for
each pollutant modeled, which must be immediately followed by the other
pollutant-specific job parameters and table levels. If the pollutant is a PM size
category, then the pollutant name must start with the characters "PM."
INPUTUNITS
Char(40)
(Optional) Pollutant concentration units used for the input data for the pollutant
(ppm, ppb, or (ig/m3). The last of these is indicated by "UGM3". Default is
ppm (parts per million).
OUTPUTUNITS
Char(40)
(Optional) Pollutant concentration units used for the output data for the
pollutant (ppm, ppb, or (ig/m3). The last of these is indicated by "UGM3".
Default is ppm (parts per million).
#SOURCES
Integer
(Optional) Largest number of sources in any one microenvironment for the
pollutant. Any number of sources may be modeled - the maximum is limited
only by the available system memory. Default = 0.
PPMFACTOR
Real
(Optional) Units conversion factor, the number of |ig/m3 that equal 1 ppm. For
example, when modeling CO, set PPMFactor = 1145 (because 1 ppm = 1145
ug/m3 for CO at typical indoor temperature and pressure). PPMFactor would be
the same for a given chemical regardless of the choice of INPUTUNITS or
OUTPUTUNITS, although in principle it would be reduced when modeling a
high-altitude study area. See discussion below.
DODOSE
Char(l)
(Optional) Y = perform dose calculations, N = don't perform dose calculations.
Like all Y/N flags, the words YES and NO may be spelled out. DODOSE = N
may save some job execution time if doses are not of interest. Default = No.
32
-------
Keyword
Type
(length)
Description
SIZE
Real
(Only used for modeling particulates) Aerodynamic diameter (particle size) in
micrometers for a particle pollutant. This parameter is not required for gaseous
pollutants.
DENSITY
Real
(Only used for modeling particulates) Density (in g/cm3) of a particle pollutant.
This parameter is not required for gaseous pollutants.
ALERTTHRESH
Real
(Optional, but default is zero) Timestep exposure threshold for alerting the user
that a simulated person has high exposure. The number of exceedances and the
time over the threshold are reported, for each person that goes over the limit. If
this is not set then every person will exceed the threshold.
ROADWAY
Char(l)
(Optional) If "Y", then APEX will use roadway concentrations while in
roadway microenvironments. Default = No.
ROADLAST
Char(l)
(Optional) If "Y", then APEX will use roadway concentrations based on the last
event occurring in either the Home or Work location. Default = No.
HOURL YFEVE1
Char(l)
(Optional) If "Y", then the constant intra-individual variability term used by the
%AFEV1 model will be sampled hourly, otherwise it will be sampled daily.
Default = No. See Section 7.4 in Volume II for details.
HOURL YFEVE2
Char(l)
(Optional) If "Y", then the ozone-dependent intra-individual variability term
used by the %AFEV1 model will be sampled hourly, otherwise it will be
sampled daily. Default = No.
In addition to the parameters listed above, the output table level specifications discussed below
are pollutant specific.
The role of PPMFACTOR was restricted starting with APEX version 4.69, so earlier Control
Options files might need to be altered to run correctly with the newer code. It now always
represents the conversion factor for gaseous pollutants between molar fraction and mass density
at a typical indoor temperature and pressure. It is the number of micrograms per cubic meter
equivalent to 1 part per million volume, and it is independent of the choice of INPUTUNITS or
OUTPUTUNITS. It is required in two cases: 1) if one (but not the other) of INPUTUNITS and
OutputUnits are "UGM3", or 2) if some microenvironments contain emission source (ESUM)
terms and INPUTUNITS are not "UGM3". In other cases it is not required and not used. For
particulate pollutants, both INPUTUNITS and OUTPUTUNITS must be "UGM3" as molar
volume is not well defined, and PPMF ACTOR is not relevant.
In the Pollutant Parameters section, the user specifies the levels of each of the parameters used in
the creation of the output summary tables for each pollutant. These specification parameters
include: PERCENTILES, DAVGEXP, DM1HEXP, DM8HEXP, DMTSEXP; SAVGEXP,
TIMEEXP, TSEXP, DAVGDOSE, DM1HDOSE, DM8HDOSE, DMEHDOSE, DMTSDOSE,
H EHDOSE, SAVGDOSE, TIMEDOSE, TSDOSE, TSMULTI, and RESPONSEPROR. The
table specifications for each pollutant must come after the corresponding POLLUTANT
keyword (and prior to the next occurrence of "Pollutant" in a multi-pollutant run). Each
parameter is identified by a single keyword, and the values are a list of numbers ordered from
smallest to largest and separated by commas. All the values are read as real numbers, although
33
-------
the decimal points are optional if the values happen to be integers. Items in each list must be
separated by commas. Except for the PERCENTILES, all these parameters are used to bin
exposures or doses into categories in order to create output tables. Note that there is always one
more bin than there are number of values in the list, since the first bin is less than the first value
in the list and the last bin is greater than or equal to the last number in the list. The specific
meanings of the parameters are explained in Table 4-3. These parameters, with the exception of
PERCENTILES, are optional; if they are omitted, then the corresponding table is simply not
written in the output file. See CHAPTER 5 for more information on the APEX output tables.
Table 4-3. Output Parameter
^evels in the Output Summary Table
Table
Parameter
Keyword
Data
Type
Description
Percentiles
PERCENTILES
Real
(Required) This does not refer to a separate table, but
specifies the population percentiles to be analyzed in the
other exposure and dose tables. Each requested percentile
occupies one row in each table. Values can include up to
one digit beyond the decimal point (e.g. the 99.5 or 99.9
percentile).
Daily Average
Exposure Cut
points
DAVGEXP
Real
(Optional) This parameter specifies cut-points for daily
average exposure for binning all the person-days in the
simulation period.
Daily Max 1-Hour
Exposure Cut
points
DM1HEXP
Real
(Optional) This parameter specifies the daily maximum 1-
hour exposure cut-points for binning all the person-days in
the simulation period. (Note: 1-hour tables are not
generated with the APEX timestep is greater than one
hour.)
Daily Max 8-Hour
Exposure Cut
points
DM8HEXP
Real
(Optional) This parameter specifies daily maximum 8-hour
average exposure cut-points for binning all the person days
in the simulation period. It is similar to DM1HEXP except
for the longer averaging time. (Note: 8-hour tables are not
generated with the APEX timestep is greater than one
hour.)
Daily Max
Timestep
Exposure Cut
points
DMTSEXP
Real
(Optional) This parameter specifies daily maximum
timestep exposure cut points for binning all the person days
in the simulation period. It is equivalent to DM1HEXP,
but for timesteps. (Note: If using the default timestep of
one hour, then only the hour tables are generated - the
timestep tables are not.)
Simulation
Average Exposure
Cut points
SAVGEXP
Real
(Optional) This parameter specifies cut-points for average
exposure over the simulation period. The cut points are
used to bin all simulated persons created in a run.
Exposure Cut
points
TIMEEXP
Real
(Optional) This parameter specifies the exposure cut points
for summing time spent at various exposure levels. The
time is expressed in minutes and is summed across all
profiles. TIMEEXP is used in two tables. (Exposure
Tables Type 1 and 2; see discussion of Tables file in
CHAPTER 5)
Timestep
Exposure Cut
points
TSEXP
Real
(Optional) This parameter specifies timestep exposure cut-
points for counting multiple exceedances of timestep levels
over the simulation (Exposure Table Type #9; see
discussion of Tables file in CHAPTER 5)
34
-------
Table
Parameter
Keyword
Data
Type
Description
Daily Average
Dose Cut points
DAVGDOSE
Real
(Optional) This parameter specifies cut points for the daily
average dose. The cut-points are used to bin all the
person/days in the simulation period.
Daily Max 1-Hour
Dose Cut points
DM1HDOSE
Real
(Optional) This parameter specifies cut points for daily
maximum 1-hour dose. The cut points are used to bin all
the person-days in the simulation period.
Daily Max 8-Hour
Dose Cut points
DM8HDOSE
Real
(Optional) This parameter specifies cut points for daily
maximum 8-hour dose. The cut points are used to bin all
the person-days in the simulation period.
Daily Max End-
of-hour Dose Cut
points
DMEHDOSE
Real
(Optional) This parameter specifies cut points for daily
maximum end-of-hour dose. The cut-points are used to bin
all the person/days in the simulation period. Note that
DMEHDOSE uses the instantaneous level at the end of
each hour, whereas DM1HDOSE uses the time-averaged
level over each hour. For CO, these two statistics track
each other fairly closely. For other pollutants, the end-of-
hour dose is just the dose on the last event of the hour.
Daily Max
Timestep
Exposure Cut
points
DMTSEXP
Real
(Optional) This parameter specifies daily maximum
timestep dose cut points for binning all the person days in
the simulation period. It is similar to DM1HDOSE except
that that the time period considered is a timestep rather than
an hour. (Note: If using the default timestep of one hour,
then only the hour tables are generated - the timestep tables
are not.)
Hourly End-of-
hour Dose
HEHDOSE
Real
(Optional) Similar to DMEHDOSE, except that instead of
using just the highest single end-of-hour dose on each day,
it collects results for all 24 end-of-hour doses on each day.
As with the other keywords, the values specified here refer
to the cut points used for tabulating the dose results.
Simulation
Average Dose Cut
points
SAVGDOSE
Real
(Optional) This parameter specifies cut points in dose for
the Average Dose over the entire simulation. The cut-
points are used to bin all the persons (or profiles) created in
the APEX run.
Dose Cut points
TIMEDOSE
Real
(Optional) This parameter specifies cut-points in dose for
summing time spent at various dose levels. Apart from the
statistic, the tables resemble the TIMEEXP tables.
Timestep Dose
Cut points
TSDOSE
Real
(Optional) This parameter specifies timestep dose cut-
points for counting multiple exceedances of timestep levels
over the simulation (Exposure Table Type #9; see
discussion of Tables file in CHAPTER 5)
Time Step
Multiple
Exceedance Cut-
points
TSMULTI
Real
(Optional) This parameter lists the number of exceedances
to use as cut-points in Exposure Table Type #9 and Dose
Table Type #5 (multiple exposure or dose exceedances of
timestep values of the simulation; see discussion of Tables
file in CHAPTER 5). For example, if the user may want to
track the number of persons that have 1, 10, 50, and 100
exceedances of the levels indicated by TSEXP and
TSDOSE over the course of the simulation.
35
-------
Table
Data
Parameter
Keyword
Type
Description
Risk Levels
RESPONSEPROB
Real
(Optional) These are lists of risk probabilities, with one
probability for each exposure bin. Multiple lists may be
provided. The first 20 characters after the "=" is a label,
followed by comma-separated risk values starting 22 places
after "=". Each risk probability is multiplied by the
number of persons in the corresponding exposure bin (see
RESPONSEVAR). A useful practice is to provide a final
"ResponseProb=" line labeled "Bin counts" with a
probability of 1.0 for each bin, to provide a summary line
with the number of persons in each exposure bin.
Risk Exposure
Metric
RESPONSEVAR
Char(4)
(Optional) One of "DM1H", "DM8H", or "DMTS".
APEX will use the exposure levels of the corresponding
exposure table to calculate the number of persons expected
to have a positive response (based on the
RESPONSEPROB probabilities), for each exposure bin.
Multi-pollutant
summary
MULTIPOL
Real
(Optional) This parameter specifies that a file be prepared
and output containing information on the simultaneous
exposure levels for two pollutants. Can only be used in a
run with 2 or more pollutants. If used, MULTI1 and
MULTI2 are also required.
Multipollutant
#1 levels
MULTI1
Real
(Optional) A comma-separated list of exposure cut-points
for the first named pollutant in the run, similar in format to
the TIMEEXP list. The values may differ from those in
TIMEEXP
Multipollutant
#2 levels
MULTI2
Real
(Optional) A comma-separated list of exposure cut-points
for the second pollutant in the run, similar in format to the
TIMEEXP list. The values may differ from those in
TIMEEXP
Hours in Running
Averages
RUNHOURS
Integer
(Optional) The number of hours used in computing the
running averages for EVR, Exposure, Dose, and
HomeAmb. Default value is 8. Can use only if
TIMESTEPSPERDAY>= 24. Note that the keywords for
affected variables such as DM8HEXP and DM8HDOSE
still contain an '8' even if RUNHOURS is set to a different
value.
The final entry in Table 4-3 is RunHours, which indicates the number of hourly values to use in
Running averages for exposure and dose. The default is 8 hours. The keywords on the Control
Options file (such as DM8HExp and DM8HDose) still contain the number "8" even if RunHours
is set to a different value. RunHours must be a whole number, and running averages cannot be
used in APEX runs with timesteps greater than one hour. See section 9.4 of the APEX Technical
Support Document for more information.
The following example using a Control Options file excerpt (Exhibit 4-3) provides an illustration
of a pollutant parameters section for the simulation of two pollutants: ozone and CO. Note that
while the input units for ozone are ppb, the output is in ppm, and therefore the table cut-points
are also in ppm. PPMFACTOR is not relevant for ozone because there are no
microenvironment-specific sources and can be omitted. The user is alerted by a note on the log
file for each simulated person who encounters ozone levels over 0.16 ppm. The
RESPONSEPROB levels are reported for ozone exposure. This example has seven risk
36
-------
probabilities because the DM8HEXP table for ozone has six cut-points (and therefore seven
bins). There are two sets of RESPONSEPROB probabilities, and a "Bin counts" line.
37
-------
! POLLUTANT PARAMETERS
#Pollutants
= 2
Pollutant
= Ozone
DoDose
= NO
RoadWay
= YES
RoadLast
= NO
InputUnits
= PPb
OutputUnits
= ppm
PPMFactor
= 1.
#Sources
= 0
AlertThresh
= 0. 16
Percentiles
= 10, 25, 50,
75, 90, 95, 99
TimeExp
= 0.03, 0.04,
0.05, 0.06, 0.07, 0
. 08
DMIHExp
= 0.03, 0.04,
0.05, 0.06, 0.07, 0
. 08
DM8HExp
= 0.03, 0.04,
0.05, 0.06, 0.07, 0
. 08
DAvgExp
= 0.03, 0.04,
0.05, 0.06, 0.07, 0
. 08
SAvgExp
= 0.03, 0.04,
0.05, 0.06, 0.07, 0
. 08
ResponseVar
= DM8 H
ResponseProb
= LO10
: 0,0.08,0.
15,0.25,0.40,0.55,0.75
ResponseProb
= L015
: 0,0.02,0.
04,0.07,0.10,0.15,0.25
ResponseProb
= Bin Counts
: 1,1,1,1,1
,1,1
Pollutant
= CO
DoDose
= YES
RoadWay
= NO
RoadLast
= NO
InputUnits
= PPb
OutputUnits
= ppm
PPMFactor
= 1145.0
#Sources
= 1
Percentiles
= 10, 25, 50,
75, 90, 95, 99
TimeExp
= 2, 4, 6, 8,
10, 15, 20, 25, 30,
35, 40, 45, 50 , 60
DMIHExp
= 5, 10, 20,
30, 40, 50, 75
DM8HExp
= 3, 6, 7, 8,
9, 10, 11, 12, 13,
14, 15, 18, 20, 25
DAvgExp
= 2, 4, 5, 6,
7, 8, 9, 10, 11, 12
, 14, 16, 18, 20
SAvgExp
= 0.5, 1, 1.25, 1.5, 1.75, 2, 2.5
, 3, 4, 5, 6, 8, 10
DMIHDose
= 0.5, 1.0, 1
.25, 1.5, 1.75, 2.0,
2.25, 2.5, 2.75, 3.0
DM8HDose
= 0.5, 1.0, 1
.25, 1.5, 1.75, 2.0,
2.25, 2.5, 2.75, 3.0
DMEHDose
= 0.5, 1.0, 1
.25, 1.5, 1.75, 2.0,
2.25, 2.5, 2.75, 3.0
H EHDose
= 0.5, 1.0, 1
.25, 1.5, 1.75, 2.0,
2.25, 2.5, 2.75, 3.0
DAvgDose
= 0.5, 0.75,
1.0, 1.25, 1.5, 1.75
, 2.0, 2.25, 2.5, 2.75
SAvgDose
= 0.4, 0.5, 0
.6, 0.7, 0.8, 0.9, 1
.0, 1.2, 1.4, 1.6, 1.8
AlertThresh
= 100.
Exhibit 4-3. Pollutant Parameters Section of an Example Control Options File
4.2.3 Job Parameter Settings Section of the Control Options File
In the Job Parameter Settings section of the Control Options file, the user can specify a number
of different job parameters for APEX runs. Table 4-4 provides a description of the keyword,
data type, and uses of these job parameters. As with Input and Output Files, the keyword is the
part of the Parameters input line that is necessary to allow APEX to identify the parameter. Data
type must be integer, real, or character. Each character variable has a specified length; input
values longer than allowed will be truncated to this length, and values shorter than allowed are
38
-------
simply padded with blanks. In all cases in this section, with the exception of COUNTY or
TRACT, if the same keyword appears more than once, then the last occurrence overwrites the
others. Exhibit 4-4 shows an example of this section of the Control Options file.
When APEX runs, the values of all the job settings (including the default settings for parameters
not explicitly set), will be printed to the Log file.
#PROFILES: This is the number of simulated persons to randomly generate and assess. Each
one is independent of the others, and is followed for the duration of the simulation, from the
STARTDATE to the ENDDATE (see below). The APEX execution time is essentially linear
in the number of profiles, until a limit is reached when the data arrays must be stored in virtual
memory, which slows it down substantially. If more profiles than this are desired, it is best to
run APEX multiple times with the same inputs except for the RANDOMSEED. Then the output
can be combined into what is essentially a single large run.
RANDOMSEED. APEX uses many random number seeds internally, but all are derived from
this one. The Fortran random number generator is known to produce correlated streams of
numbers in certain cases for simple offsets, so a second level of randomization was added to
APEX to eliminate this. Effectively, this extra randomization amounts to a total scrambling of
the list of 2,147,483,646 possible seeds, using a different random number generator. For this
reason, simply incrementing RANDOMSEED by one, between two runs, will change the seeds
in an unpredictable way, resulting in virtually no chance for correlations between runs.
Two APEX runs with the same inputs and same RANDOMSEED will produce identical output.
If only RANDOMSEED is changed between runs, the two runs may have their output combined
into a larger sample. Using the same seed with slightly different inputs in multiple runs allows a
sensitivity analysis of the importance of the differing inputs. If different seeds were used in this
case, then stochastic differences between the runs may obscure the effects of the changed inputs.
START DATE and END DATE: Dates in APEX are specified using an 8-digit string, with the
year occupying the first 4 digits, the next 2 digits for the month, and the last 2 for the day.
Months and days less than 10 require a leading zero as a place holder. This format is shown
symbolically as: YYYYMMDD. While month/day/year or day/month/year formats are more
common, the APEX order is used so that numerical sorting results in chronological sorting. The
end date may not be less than the start date. If the two are the same, the simulation consists of a
single day. Often, APEX is run for a calendar year (e.g., from 20140101 to 20141231). APEX
may be run over longer periods, but the profiles will remain at their initial ages throughout the
simulation. APEX knows which years contain February 29. Dated inputs such as air quality
data must include the entire simulation period, although these files can contain data that extend
beyond the start and end dates.
TIMESTEPSPERDAY. APEX uses three levels of time resolution within a day: the hour, the
timestep, and the diary event. The air quality data is input as time series data with a fixed
timestep, controlled by the parameter TIMESTEPSPERDAY. The default is 24, representing
hourly data. The meteorological data must be hourly, regardless of TIMESTEPSPERDAY. If
the timestep is less than one hour, then there must be a whole number timesteps in one hour to
permit the determination of hourly output data. If the timestep is longer than one hour, then
39
-------
there must be a whole number of hours in one timestep, and a whole number of timesteps in a
day. Thus, if TIMESTEPSPERDAYis less than 24, it must one of 12, 8, 6, 4, 3, 2, or 1.
The activity diary events may range from 1 to 60 minutes in length, and always break at the top
of each hour. With hourly air quality data, which also change at the top of each hour, each diary
event is associated with one (constant) air concentration, which simplifies the exposure
calculations. In this case, multiple diary events within the same hour use the same ambient air
concentration, although the microenvironmental concentration may change. When using the
default hourly timestep, there is no point in requesting APEX to output both hourly files and
timestep files, since the information on them would be the same.
If the timestep is less than one hour, any diary events that cross a timestep boundary are split. If
the timestep is greater than one hour, then by insisting it must be a whole number of hours
ensures that no diary event will cross a timestep boundary, and the calculation of (say) average
temperature over the timestep is simple, given hourly temperature data. There must be a whole
number of timesteps in one day to permit aggregation to daily totals.
DSTADJUST. Every day in an APEX simulation is 24 hours long, as is every activity diary. If a
particular location uses daylight saving time (DST), one spring day is 23 hours long and one fall
day is 25 hours long. Air quality data are reported as a time series, typically in Standard Time
throughout the year. Suppose a particular pollutant spikes at 8 a.m. every day, due to traffic.
The air quality data will show regular spikes every 24 hours, except for a 23 hour spacing on the
first day of DST, and a 25 hour spacing on the last day, because the traffic follows human
behavior (local time) patterns. When matched with APEX diaries, the spikes will appear to
occur at 7 a.m. in the summer. This may affect the exposure calculations, since there may be
fewer commuters available to be exposed at 7 a.m. If DSTADJUST = Yes, then APEX
duplicates one hour of air quality data at the start of DST in the spring, and deletes one hour at
the end of DST in the fall, effectively shifting the summertime air quality data so the spikes
always occur at 8 a.m. APEX time; that is, 24 hours apart every day of the year. The start and
end dates for DST are coded into APEX and change each year. U.S. states may opt out of DST,
but if they use it, they must follow the national dates. At present, only Arizona and Hawaii do
not use DST.
COUNTYLISTTRACTLIST, and CITYRADIUS: The user may specify the study area in three
ways: using counties, tracts, and a target circle. The basic geographical unit in APEX is the
"sector", which in a standard run is the same as a census tract. The APEX sector names (or ID)
may be up to 40 characters in length, but census tract names are just 11 characters long. A
"county" is defined by the first five characters of the sector ID. For example, "06037" is Los
Angeles County when using census tract names. If the user supplies a list of their own sector
names, APEX will still match the first five characters if COUNTYLIST = Yes in the Control
Options file. If COUNTYLIST = Yes, then each county in the study area is listed on a separate
line as "County = xxxxx". For example, County = 06037 selects Los Angeles County. All
counties must be listed on consecutive lines; once a different keyword is encountered, the county
list is assumed to have ended.
If the county resolution is too coarse, the option TRACTLIST = Yes may be used. This operates
just like COUNTYLIST, except that the first 11 characters of the sector name are matched. If
40
-------
both COUNTYLIST and TRACTLIST are used, then sectors that match either one are included
in the study are (the union of the lists is used).
The study area is also defined by a specific latitude and longitude and must lie within a given
radius (in km) of this point (CITYRADIUS). This defines a circular area. All sectors in the
study area must lie within this circle (that is, the intersection of this circle and the county/tract
list is used). If the user wants to effectively use just the county/tract list, then specify a very
large radius. If a small radius is used, it may select only some of the sectors in a listed county. If
no sectors remain, an error occurs.
KEEPLEAVERS: The default target population in APEX is all persons who live inside the
study area. If commuting is modeled, some people who live inside the study area will work
outside of it. The air quality at these work places is unknown to APEX, and is assumed to be
related to the average concentration over all air districts in the study area at the same point in
time. Calling this average Cavg, the ambient concentration C for work outside the study area is:
C = LEA VERMULT*Cavg + LEA VERADD For KEEPLEA VERS = No, those persons are
still modeled, but are excluded from all output tables. This ensures that the random numbers
assigned to each individual are not affected by the KEEPLEAVERS status, if comparisons are
being made to earlier runs using the same random number seed but altered job settings. The
point is that sometimes the corresponding person in the two runs may stay inside the study area
in one run but leave in the other, depending on which job settings were changed. For
KEEPLEAVERS = Yes, the target population is all persons who live in the study area. For
KEEPLEAVERS = No, the target population for the output tables is all persons who both live
and work inside the study area.
NEARBYRADIUS. APEX assigns a home sector (and also a work sector if employed) for each
simulated person. Certain diary events like shopping or restaurants may be assigned to "other"
sectors. Every sector has a list of nearby sectors, where "nearby" means within a distance of
NEARBYRADIUS, as measured (in km) between the nominal centers of each sector. The
default is 20 km. If there are multiple nearby sectors to the current location, one is selected at
random for diary events in "other" locations. The air quality at the "other" sector might (or
might not) be different from the "base" sector, depending on the mapping of sectors to air quality
districts.
COMCUT1, COMCUT2, COMPROBAB1, COMPROBAB2 Each simulated individual who
works is assigned a commuting time. APEX uses this parameter to match the commute time on
each diary day to the target time of the person. For example, if a person's commute time is 30
minutes and COMCUT1 = 10 minutes, then diaries with a commute time between 20 and 40
minutes inclusive will be given full (100%) weight. Diaries outside of this window, but inside
COMCUT2 of the target time, receive a weight of COMPROBAB1. All other diaries in the
appropriate pool receive a weight of COMPROBAB2. These weights are used when randomly
selecting diaries. Do not set all four variables to zero, or APEX may stop prematurely due to the
lack of any possible matches.
41
-------
Table
-4. Job Parameters in the APEX Control Options File
Keyword
Type
(length)
Description
SI Ml I.A I ION PARAMITKRS
#.PROFILES
Integer
(Required) Number of profiles to simulate.
RANDOMSEED
Integer
(Optional) RANDOMSEED > 0 is specified by the user,
RANDOMSEED = 0 gets seed from clock. RANDOMSEED may be
any natural number up to 2147483646. Default = 0.
END DATE
Integer
(Required) Simulation end date in YYYYMMDD format.
FIRSTPROFILE
Integer
(Optional) First profile number to simulate. For example, this can be
used for skipping to a particular person's profile when performing
repeated runs using a controlled RANDOMSEED. Default = 1.
START_DATE
Integer
(Required) Simulation start date in YYYYMMDD format (e.g.,
19960704 for July 4, 1996).
TIMESTEPSPERDA Y
Integer
(Optional) Number of timesteps in a day. This setting dictates the
required time resolution of the air quality input data, as well at the
resolution of calculated exposures and doses. Default = 24.
OCCFACTOR
Char(l)
(Optional) This parameter tells APEX that the Profile Factor file
contains a parameter that is related to occupation. If this parameter is set
to "Y", then the profile factor will only be applied to employed
individuals, and an extra factor group will be made for all unemployed
profiles. Default = No.
S I I l)Y ARI A P \RAMi: i l!RS
LOCATION
Char(40)
(Optional) Study area location (for output labeling only; not used
internally). Default = empty character string.
LATITUDE
Real
(Required) Latitude in decimal degrees for the center of the study area.
Note that latitude south of the equator is negative.
LONGITUDE
Real
(Required) Longitude in decimal degrees for the center of study area.
Note that longitude between 0 and 180 degrees west of the prime
meridian is negative (e.g., in the United States).
ALTITUDE
Real
(Optional) Altitude of study area in feet. The altitude is assumed
constant for the study area. It is used in the Coburn-Forster-Kane (CFK)
equation for determining blood COHb concentration. Only necessary
when simulating CO dose. Default = 0.
DSTADJUST
Char(l)
(Optional) Y = adjust air quality data for Daylight Saving Time (DST),
N = do not use DST. Default = Yes.
CI TYRADIUS
Real
(Optional) Radius of study area in km. The population sectors (e.g.,
census tracts) with centers (or representative locations) within this radius
will be automatically selected for modeling. If the default is used, then
COUNTYLIST should be specified. Default = 99999.
AIRRADIUS
Real
(Optional) Maximum representative radius (km) of air quality data.
Each sector uses the nearest air quality site, as long as it is within a
distance AIRRADIUS. This term can be specified via the Air District
Location File; however, if AIRRADIUS is in the Control Options file, it
will overwrite those found in the Air District Location File. Default =
99999.
MODELAQVAR
Char(l)
(Optional) Dictates the expected format of the Air Quality Data file.
The default is ModelAQVar = N, when APEX expects AQ values for
each timestep. If MODELAQVAR = Y, then APEX expects AQ
distributions for each hour of the simulation. Default = No.
ZONERADIUS
Real
(Optional) Maximum representative radius (km) of temperature data
collected at a weather station. Default = 99999.
42
-------
Keyword
Type
(length)
Description
COUNTY
Char(5)
(Optional) FIPS code for listed county (or other relevant portion of the
sector ID if the supplied sector files are not used). COUNTY is used
only if COUNTYLIST = Y. Repeat this line for each additional county
code.
COUNTYLIST
Char(l)
(Optional) COUNTYLIST = Y means that the study area is composed of
sectors in the listed counties (previous variable) and within
CITYRADIUS; otherwise the study area is restricted to sectors within the
specified CITYRADIUS only or defined with the TRACTLIST. (May be
used in conjunction with TRACTLIST, final study area is union of tracts
and counties listed). Default = No.
TRACT
String
(Optional) Sector ID for a listed sector (usually census tract). Tract is
used only if TRACTLIST = Y. Repeat this line for each additional
sector to be used.
TRACTLIST
Char(l)
(Optional) TRACTLIST = Y means that the study area is composed of
the sectors (usually census tracts) listed using the TRACT keyword
which are within CITYRADIUS; N = the study area is restricted to
sectors within the specified CITYRADIUS only or defined with the
COUNTYLIST. (May be used in conjunction with COUNTYLIST, final
study area is union of tracts and counties listed). Default = No.
SCENARIO
Char(40)
(Optional) Scenario description (for output labeling only; not used
internally). Default = blank character string.
NEARBYRADIUS
Real
(Optional) Radius (km) for use when randomly selecting nearby tracts to
the home or work location (For use with the NW, NH, L locations).
Default = 20.
RESAMPLEN
Char(l)
(Optional) This allows the user to select either a new random nearby
tract for each day, or use a single tract for the whole simulation. Default
= No.
MICROENVI RON MEN TA L PARA METE RS
micROS
Integer
(Required) Number of microenvironments defined in the
Microenvironment Mapping file and on the Microenvironment
Descriptions file.
COMM1 11NG PAKAM1.1 I KS
COMMUTING
Char(l)
(Optional) COMMUTING = Y allows a simulated person to commute to
a work sector. COMMUTING = N means any work occurs in the home
sector. If "Y", a work sector is randomly selected for each simulated
profile based on the probabilities in the commuting database. Default =
No.
KEEPLEAVERS
Char(l)
(Optional) KEEPLEA VERS = Y means that people who live in the
study area but work outside it area still modeled. Default = No.
LEAVERADD
Real
(Optional) Additive concentration term applied when working outside
study area (only used if KEEPLEA VERS = yes). Default = 0.
LEAVERMULT
Real
(Optional) Multiplicative factor for city-wide average concentration,
applied when working outside study area (only used if KEEPLEA VERS
= yes). Default = 0.
COMCUT1
Real
(Optional) The width (in minutes) of the window of commuting times
within which all times will be weighted by 100%. Default = 0.
COMCUT2
Real
(Optional) The width (in minutes) of the second commuting time
window. This parameter works similar to the previous commuting time
window, COMCUT1. For example, if a person's commute time target is
60 minutes, COMCUT1 = 10 min, COMCUT2 = 20 min, and a diary has
a commute time of 77 minutes (within ±20 min but not within 10 min),
then that diary will be given a weight of COMPROBAB1. Default = 0.
43
-------
Keyword
Type
(length)
Description
COMPROBAB1
Real
(Optional) The weight given to diaries with commuting times in the
window between COMCUT1 and COMCUT2 on either side of the target
commuting time assigned to the current profile. Default = 0.
COMPROBAB2
Real
(Optional) The weight given to all diaries that fall outside the
COMCUT1 and COMCUT2 windows. For example, with a target of 60
minutes and COMCUT2 = 20 min, this probability is assigned to all
diaries with <40 minutes or >80 minutes of commuting time. Default =
0.
DIARY Sl.l.i:( 1 ION parami: I i:rs
AGE2PROB
Real
(Optional) Diary probability factor for "shoulder" ages. This parameter
allows an optional shoulder window of ages outside the primary age
window. The shoulders have the same width in years as the main age
window, so in the example under AGECUTPCT the shoulders are ages
2029 and 51-60. TheAGE2PROBAB parameter operates like
MISSAGE, by suppressing the selection probability in the shoulders. If
AGE2PROBAB = 0 then shoulder ages are never selected. Default = 0.
AGECUTPCT
Real
(Optional) Width of main age window (%). Each simulated profile
(person) is assigned a specific year of age, but the activity diaries
assigned to this person do not need to match this age exactly. A window
is created around this target age, of size equal to AGECUTPCT percent
of the target age. If the profile age is 40 and AGECUTPCT = 25, then the
age window is ten years wide (25% of 40) and diaries for persons from
30 to 50 years of age inclusive are permitted to be selected. The age
window is always at least 1 year wide, even when using the default.
Default = 0.
AGEMAX
Integer
Maximum age for simulated profiles (persons). Each profile is assigned
a specific age, used for selecting diaries and for physiological variables.
This age does not change over the simulation, even if it is one year or
longer.
AGEMIN
Integer
Minimum age for simulated profiles (persons).
MISSAGE
Real
(Optional) Diary probability factor for missing age. Some of the
supplied CHAD diaries are for persons of unknown age. This factor
operates just like MISSGENDER and MISSEMPL to lower the
selection probability for such diaries. Default = 0.
MISSEMPL
Real
(Optional) Diary probability factor for missing employment. Some of
the supplied CHAD diaries are for persons of unknown employment
status. Like MISSGENDER, this factor lowers the selection probability
for such diaries. If MISSEMPL = 0, then such diaries will never be
selected. Default = 0.
MISSGENDER
Real
(Optional) Diary probability factor for missing gender. Some of the
supplied CHAD diaries are for persons of unknown gender. All APEX
profiles are assigned gender, however, and the CHAD diaries are selected
from those of the same gender or from the unknowns. MISSGENDER is
used as a multiplicative factor to reduce the probability of selecting
diaries of unknown gender. If MISSGENDER = 0, then diaries with
missing gender will never be selected. If MISSGENDER = 1, then such
diaries are equally likely to be selected as diaries of the correct gender.
MISSGENDER can also be set to values between zero and one.
Allowing small but nonzero values for MISSGENDER and the other
MISS parameters may prevent empty diary pools. Default = 0.
44
-------
Keyword
Type
(length)
Description
MISSOCC
Real
(Optional) Diary probability factor for missing occupation. This
parameter will only be used if diaries are to be weighted via occupation,
which requires the inclusion of a Profile Factors file, and OCCFACTOR
= Y and USEOCCGROUPS = Y. Most diaries in CHAD have missing
occupation; MISSOCC provides a weight for these diaries. If MISSOCC
= 0, then no diaries other than those matching the occupation of the
individual will be selected. Age, gender, employment and occupation
each produce a probability, which are then multiplied together to
determine the overall diary selection probability for that profile. Default
= 0.
USEOCCGROUPS
Char(l)
(Optional) This parameter tells APEX to match diaries based on
occupation group. To use this option, occupations must be specified in
the Profile Factors file. Default = No.
DOSI. PARAMETERS
COHBFACTOR
Real
(Only needed when modeling CO) Convergence parameter for COHb
algorithm. This is a safety factor that limits the permitted error in
determining the solution to the CFK equation. Larger factors mean
greater accuracy but slower evaluation. Numerical tests indicate that
factors in the range of 2-3 are optimal for most purposes. Only
necessary when simulating CO dose.
LOCATION PARAMETERS
UOTHER
Integer
(Optional) Number of other districts to use in calculating the air quality
for diary events with location = 0 ("Other") when
SAMPLEOTHERLOCS is used. (The probability of the person's home
district being one of these districts is given by HOMEPROBAB.)
Default = 1.
CUSTOMWORK
Comma-
delimited
list
(Optional) List of CHAD activity codes that will be assigned to location
= W (Work). Default is all CHAD activity codes < 11000.
HOMEPROBAB
Real
(Optional) Probability (0-1) of a person's home district being one of the
districts used to calculate the air quality for diary events with location =
O ("Other") when SAMPLEOTHERLOCS is used. Default = 0.
SAMPLEOTHERLOCS
Char(l)
(Optional) If SAMPLEOTHERLOCS = Y, a random list of air districts
will be selected for each person for calculating the air quality for diary
events with location = O ("Other"). The number of districts selected for
each person is given by #OTHER, and the probability of the person's
home district being the list is given by HOMEPROBAB. Default = No
ROLLBACK PARAMETERS
RBBACK
Real
(Optional) Rollback background concentration. Use same units as
INPUTUNITS. The background is the part of the air quality that is not
affected by Rollback controls. Default = 0.
RBMAX
Real
(Optional) Rollback maximum concentration. Use same units as
INPUTUNITS. Default = 0.
RBTARGET
Real
(Optional) Rollback target concentration. Use same units as
INPUTUNITS. The target must be less than the maximum for any air
quality improvement to occur. If RBTARGET = RBMAX, the air quality
is unchanged by rollback. If RBTARGET = 0, the air quality always
stays at RBBACK. Default = 0.
ROLLBACK
Char(l)
(Optional) ROLLBACK =Y uses air quality rollback adjustments. The
default is ROLLBACK = N. ROLLBACK adjusts the ambient air quality
data before the exposure calculations occur. The purpose is to determine
exposure in hypothetical scenarios where the ambient concentrations
have been reduced by various controls. Default = No.
45
-------
Keyword
Type
(length)
Description
DIAGNOSTIC PARAMETERS
DEBUGLEVEL
Integer
(Optional) A value > 0 results in more information being written to the
log file than for a value of zero. For production runs with a large number
of profiles, use DEBUGLEVEL = 0, or else the log file will become too
Inrec. DEBUGLEVEL may also be 1 or 2. Default = 0.
i.oc; i ii.i swik mi s
LOGDISTRICT
Char(l)
(Optional) Y = the name and location of each of the air districts will be
written to the Log file. Both a preliminary list (all the air districts in the
Air District Location file that are within the study area and have data for
the entire simulation period) and a final list (those required to simulate
the final list of study sectors) are printed. Default = Yes.
LOGPOPULA TION
Char(l)
(Optional) Y = the following population information will be written to
the Log file for each study area sector: The total population of the sector
(TotalPop); the base population for the study (StudyPop), which will be
smaller than TotalPop if only certain age ranges are being considered; the
total population of workers in the sector (Workers); the sector population
of workers who work inside the study area (Worklnside); and the final
population (FinalPop) for the simulation, which may be smaller than
StudyPop if the workers who leave the sector are excluded (if
KEEPLEA VERS = NO) Default = Yes.
LOGPROFILES
Char(l)
(Optional) Y = the following population information will be written to
the Log file for each study area sector: The total population of the sector
(TotalPop); the base population for the study (StudyPop), which will be
smaller than TotalPop if only certain age ranges are modeled; the total
number of workers in the modeled age range who live in the sector
(Workers); the population of these workers who work inside the study
area (Worklnside); and the final study population of the sector
(FinalPop), which may be smaller than StudyPop if the commuters who
leave the study area are not modeled (if KEEPLEA VERS = NO).
Default = Yes.
LOGSECTORS
Char(l)
(Optional) Y = the name and location of each study sector will be
written to the Log file. Both a preliminary list (all the sectors
geographically within the study area) and a final list (those sectors within
the study area having available air quality and temperature data) are
printed. Default = No.
LOGTABLES
Char(l)
(Optional) Y = all the tables that are written to the Tables file are also
written to the Log file. Default = No.
LOGZONES
Char(l)
(Optional) Y = the name and location of each of the meteorological
zones will be written to the Log file. Both a preliminary list (all the air
districts in the Meteorology Zone Location file that are within the study
area and have data for the entire simulation period) and a final list (those
required to simulate the final list of study sectors) are printed. Default =
Yes.
OUTPUT FILE SWITCHES AND KEYWORDS
CLUSTEROUT
Char(l)
(Optional) Y \\ rile llie cluster assignments lor all diaries mi ilie
database to an output file. Default = No.
46
-------
Keyword
Type
(length)
Description
CUSTOMSAMPLE
Comma-
separated
list of
integers
(Optional) The profiles designated by CUSTOMSAMPLE are written in
addition to the profiles specified by the EVENTSAMPLE variable. If
both EVENTSSAMPLE and CUSTOMSAMPLE are set, then all the
EVENTSSAMPLE events are written as before and any additional
CUSTOMSAMPLE events are written in the appropriate place in the
numerical profile order. Writing of CUSTOMSAMPLE events is
dictated by the value of the EVENTSOUT variable, so no events will be
written if EVENTSOUT=N, even if a CUSTOMSAMPLE is specified.
If neither CUSTOMSAMPLE nor EVENTSAMPLE is set, then events
are written as dictated by the default EVENTSAMPLE value (if
EVENTSOUT = Y). If the user wishes to write only the
CUSTOMSAMPLE events, then EVENTSAMPLE should be set to 0.
Default = empty list.
CWEIGHTOUT
Char(l)
(Optional) Y= Write the cluster weights for all diary pool-cluster
combinations to an output file. Default = No.
DAILYLIST
comma
or space-
separated
strings
(Optional) List of keywords indicating which variables are to be written
to the Daily output file. See Section 5.4 for details. Only relevant if
DAILYOUT = Yes.
DAILYOUT
Char(l)
(Optional) Y= the Daily output file containing values of daily
parameters (exposures, doses, etc.) is created. Otherwise it is not written.
Default = No.
EVENTSAMPLE
Integer
(Optional) Dictates which profiles have their event data written to the
Events file. If EVENTSAMPLE = K, then K diaries are written out,
spaced evenly through the run. For example, UPROFILES = 10,000 and
EVENTSAMPLE = 10 prints profile #1000, 2000, etc. This only has
effect if EVENTSOUT = Yes. Default = 10.
EVENTSLIST
comma
or space-
separated
strings
(Optional) List of keywords indicating which variables are to be written
to the Events output file. See Section 5.10 for details. Only relevant if
EVENTSOUT= Yes.
EVENTSOUT
Char(l)
(Optional) Y = the output file containing the event-level model outputs
for each simulated individuals is written. Otherwise, the file is not
written. Default = No.
HOURLYLIST
comma
or space-
separated
strings
(Optional) List of keywords indicating which variables are to be written
to the Hourly output file. See Section 5.2 for details. Only relevant if
HOURLYOUT = Yes.
HOURLYOUT
Char(l)
(Optional) Y= the Hourly output file containing values of hourly
parameters (exposures, doses, etc.) is created. Otherwise it is not written.
Default = No.
MRESHOME
Char(l)
(Optional) If = Y, then only values associated with "home" locations
will be written to the Microenvironmental Results file. Otherwise, values
will be written for each of the locations specified in the
MicroenvironmentMapping file. Default = No.
MRESLIST
comma
or space-
separated
strings
(Optional) List of keywords indicating which variables are to be written
to the Microenvironmental Results output file. See Section 5.6 for
details. Only relevant if MRESOUT= Yes.
MRESMICROS
comma-
separated
list of
integers
(Optional) A comma-separated list of integers that indicate the
microenvironments for which data will be written to the
Microenvironmental Results file. Only relevant if HOURLYOUT = Yes.
47
-------
Keyword
Type
(length)
Description
MRESOUT
Char(l)
(Optional) Y = the Microenvironmental Results file will be created.
Otherwise, the file is not written. Default = No.
MSUMOUT
Char(l)
(Optional) Y = the Microenvironmental Summary file will be created.
Otherwise, the file is not written. Default = No.
PSUMLIST
comma
or space-
separated
strings
(Optional) List of keywords indicating which variables are to be written
to the Profile Summary output file. See Section 5.5 for details. By
default, only 12 variables are printed, but the user may add up to 59
others using this option.
SOBOLRUN
Char(l)
(Optional) Y = perform Sobol sensitivity analysis, using the input
variable groupings on the Seed file. See CHAPTER 11 in Volume II for
details. Default = No.
SOBOLVAR
comma
or space-
separated
strings
(Optional) List of APEX output variables subject to Sobol analysis. See
CHAPTER 11 in Volume II for details. Default = none.
TABLESLIST
comma
or space-
separated
strings
(Optional) List of keywords indicating which variables are to be written
to the Tables output file. See Section 5.8 for details. Default = none.
TIMESTEPOUT
Char(l)
(Optional) Y= the Timestep file will be created. Otherwise, the file is
not written. If the default timestep (1 hour) is used, then this file will not
be written because it will be identical to the Hourly file. Default = No.
TIMESTEPLIST
comma
or space-
separated
strings
(Optional) List of keywords indicating which variables are to be written
to the Timestep output file. See Section 5.3 for details. Only relevant if
TIMESTEPOUT = Yes.
TRANSOUT
Char(l)
(Optional) Y= Write the pool-cluster transitions for all diaries in the
database to an output file. Default = No.
TABLE PARAMETERS
ACTIVE
Real
(Optional) Threshold median daily PAI (MET) value for defining active
persons. Simulated individuals having median PAI equal to or greater
than this value over the simulation period will be included in the "active
persons" subgroup in the output exposure tables. Default = 0 (this will
catch all person-days).
CHILDMAX
Integer
(Optional) Maximum age for inclusion in the "child" and "active child"
population subgroups in the output exposure tables. Default = 0.
CHILDMIN
Integer
(Optional) Minimum age for inclusion in the "child" and "active child"
population subgroups in the output exposure tables. Default = 0.
HEA VYEVR1
Real
(Optional) This parameter sets the threshold for equivalent ventilation
rate defining one-hour heavy exertion. It is used in generating the APEX
output tables for one-hour exposures under heavy exertion. Default = 0.
HEAVYEVR8
Real
(Optional) This parameter sets the threshold for equivalent ventilation
rate defining eight-hour heavy exertion. It is used in generating the
APEX output tables for eight-hour exposures under heavy exertion.
Default = 0.
HEAVYEVRTS
Real
(Optional) This parameter sets the threshold for equivalent ventilation
rate defining timestep-level heavy exertion. It is used in generating the
APEX output tables for timestep exposures under heavy exertion. Thus,
this value should be dependent on the length of timestep used. Default =
0.
48
-------
Keyword
Type
(length)
Description
MODEVR1
Real
(Optional) This parameter sets the threshold for equivalent ventilation
rate defining one-hour moderate exertion. It is used in generating the
APEX output tables for one-hour exposures under moderate exertion.
Default = 0.
MODEVR8
Real
(Optional) This parameter sets the threshold for equivalent ventilation
rate defining eight-hour moderate exertion. It is used in generating the
APEX output tables for eight-hour exposures under moderate exertion.
Default = 0.
MODEVRTS
Real
(Optional) This parameter sets the threshold for equivalent ventilation
rate defining timestep-level moderate exertion. It is used in generating
the APEX output tables for timestep exposures under moderate exertion.
Thus, this value should be dependent on the length of timestep used.
Default = 0.
i.o\(;iTi dinai. diary selection parameters
DIARYAUTOC
Real
(Optional) Lag-1 autocorrelation statistic for the D&A longitudinal diar\
assembly algorithm. Provides a target for the autocorrelation in the key
diary statistic. Default = 0.
DIARYD
Real
(Optional) Provides a target D statistic for the D&A longitudinal diary
assembly algorithm. The D statistic reflects the relative importance of
within person variance and between person variance in the key diary
statistic. Default = 0.
LONGITDIARY
Char(l)
(Optional) Y = APEX will use the D&A longitudinal diary assembly
algorithm to construct the activity diaries for the simulated persons,
based on the statistics in the DiaryStat file. In this case, DIARYAUTOC,
DIARYD, and the name of the diary statistics file must all be designated
in the Control Options file. Default = No.
CLUSTERDIARY
Char(l)
(Optional) Y = APEX will use the clustering algorithm for longitudinal
diary assembly based on the transitional probabilities calculated by
analyzing the CHAD input files. Default = No
CLUSTERAGES
comma-
separated
list of
integers
(Optional) Age cut-points for binning the empirical transition
probabilities. A list of N cut-points gives N+l bins. The cut-point age
goes into the higher bin. Default = 0 (all ages in one bin).
USEADJACENT
Integer
(Optional) Minimum number of diaries in a pool age bin combination
for restricting the empirical transition probabilities to consecutive
calendar dates (no gaps allowed). Groups with fewer examples also use
non-adjacent cases.
PHYSIOLOGY PARAMETERS
DISEASE
Char(12)
(Optional) Provides the name of a condition or disease. If set, then
APEX expects the Prevalence file to be defined as well, and a
subpopulation of persons with the condition will be modeled, resulting in
exposure summary tables corresponding to the subpopulation. The tables
will be labeled using this variable; spaces are allowed. Default = none.
HTWTMETHOD
Integer
(Optional) Selects the method of setting height and weight for each
profile. HTWTMETHOD = 1 is the method used by APEX prior to
2017. HTWTMETHOD = 2 has distributions with explicit correlation
between height and weight, with parameters matched to NHANES data.
Default = 2.
MODEVRMETHOD
Integer
(Optional) Selects the method of determining the cutoff for moderation
exertion (8-hour average only) for output tables. MODEVRMETHOD =
1 applies the same cutoff (given by ModEVR8) to everyone, while
MODEVRMETHOD = 2 assigns a value for each profile using a
distribution from the physiology input file.
49
-------
Keyword
Type
(length)
Description
VEMAX
Char(l)
(Optional) Y = apply an upper limit to VE, which is 100 L/min for
events over 5 minutes, and 150 L/min for shorter events. N = calculate
VE without explicit truncation. Default = Y.
VEMETHOD
Integer
(Optional) Selects the method of determining VE on each diary event.
VEMETHOD = 1 is the method used in APEX prior to 2017.
VEMETHOD = 2 is a new method that uses both V02 and V02max.
Default = 2.
! PARAMETER SETTINGS
! SIMULATION
PARAMETERS
#Profiles
=
40000
RandomSeed
=
0
Start date
=
20040401
End date
|
=
20040930
! STUDY AREA
PARAMETERS
Location
=
Description of Location of the Study Area
Latitude
=
33.7629
Longitude
=
-84.4004
Altitude
=
150.
DSTadj ust
=
YES
CityRadius
=
100.
AirRadius
=
25.
ZoneRadius
=
100.
CountyList
=
YES
TractList
=
NO
NearbyRadius=
5.
ResampleN
=
NO
County
=
01017
County
=
13013
County
i
13015
! MICROENVIRONMENT PARAMETERS
#Micros
i
=
12
! COMMUTING PARAMETERS
Commuting
=
YES
KeepLeavers
=
YES
LeaverMult
=
0.0
LeaverAdd
=
0.0
ComCutl
=
O
O
ComCut2
=
20.0
ComProbabl
=
0.20
ComProbab2
1
=
0. 05
! DIARY SELECTION PARAMETERS
AgeMin
=
0
AgeMax
=
99
ChildMin
=
5
ChildMax
=
18
MissGender
=
o
o
MissEmpl
=
o
o
MissAge
=
o
o
50
-------
MissOcc
lO
O
II
AgeCutPct
= 20.0
Age2Probab
LO
o
o
II
UseOccGroups
= YES
OccFactor
1
= YES
! DOSE PARAMETERS
COHbFact
i
= 2.5
! LOCATION PARAMETERS
CustomWork
=
SampleOtherLocs = YES
#OtherDistricts = 2
HomeProbab
1
= 0
! ROLLBACK PARAMETERS
Rollback
= NO
RBtarget
= 5.0
RBbackgnd
= 0.0
RBmax
|
II
M
O
O
! DIAGNOSTICS
PARAMETERS
DebugLevel
i
= 0
! LOG FILE SWITCHES
LogDistrict
= NO
LogPopulate
= NO
LogProfiles
= NO
LogSectors
= NO
LogTables
= NO
LogZones
i
= NO
! OUTPUT FILE
SWITCHES AND KEYWORDS
EventsOut
= YES
EventSample
= 2
Customsample
= 3092
MResOut
= NO
MSumOut
= NO
HourlyOut
= NO
DailyOut
= YES
PSumList
= AVGEXP,MAXEXP,AVGEXP,MAXEXP
HourlyList
= CONC1, AMB, EXP, EVR, VE, VA,
EE, MET, EF, DFEV1
DailyList
= MAX1DOSE MAX8DOSE MAX1FDOSE
AVGDOSE
MResList
= VOL, AER, RR, PRX, PEN, CSUM,
AMB
MResHome
= YES
MResMicros
1
= 1,2,8,12
! TABLES PARAMETERS
HeavyEVR1
= 30
HeavyEVR8
= 99
ModEVRl
= 16
ModEVR8
= 13
ActivePAI
|
= 1.76
! LONGITUDINAL
DIARY PARAMETERS
LongitDiary
= YES
DiaryAutoC
= 0.2
51
-------
DiaryD = 0.5
!
! CLUSTERING DIARY PARAMETERS
ClustDiaryA = NO
ReRunClus = NO
ClustDiaryB = NO
Exhibit 4-4. Job Parameters Sections of an Example Control Options File
4.3 Population Sector Location File
The Population Sector Location file provides the latitude and longitude of a representative
location such as the geographic center of all the sectors (e.g., census tracts) to be included in the
population data files. Each line includes a Sector Name, Latitude, and Longitude. The sector
name (also called "ID") may be any string, numeric or character, and is stored as a character
string (up to length 40). The string may contain any characters except "!" or embedded spaces.
When census tracts are used as the sectors, the names are composed entirely of numerical digits,
and the term "ID" is common in such cases. The sector names must match those in the
Commuting Flow and Commuting Time files (if worker commuting is being modeled), although
APEX now optionally accepts a coarser level of spatial resolution for the former. The names are
case-sensitive, so the values in the two files must match exactly. Tract-level 2010 Census
demographic files covering the U.S. are provided with APEX.
The population sector location file is used along with the user-specified CLTYRADLUS to
automatically select population sectors within the study area (after also addressing an optional
county test and ensuring suitable air district and meteorology zone data). APEX calculates the
distance between the location of a sector and the center of the study area and then compares it
with the CLTYRADLUS. Sectors with a distance from the study area center greater than the city
radius will not be included in the exposure assessment.
The tract-level population sector location file supplied with APEX contains the 11-character
names and latitudes and longitudes of the corresponding year U.S. census tracts. APEX expects
that the left-most five characters of a sector name will be the state and county FIPS code, or the
county-level code used in the COI/ATFlist (if the study area is limited in that way). APEX
reads counties and tracts in character format, so for example Los Angeles County is 06037, with
the leading zero retained.
If not modeling actual United States counties and/or tracts, the user may create custom names for
their sectors that follow the same rules. The "county" refers to the first 5 characters of the sector
name, and the "tract" refers to the first 11 characters (i.e., the county name plus six more). The
COUNTYLLST and TRACTLLST options can be used for subsetting customized sector files to
smaller study areas.
The latitude and longitude should be in decimal degrees. At least three digits should be provided
after the decimal point to prevent significant rounding error. Note that the longitude west of the
prime meridian (e.g., United States locations) should be negative. Exhibit 4-5 provides an
example of the first few records of this input file. These tracts are all in county "01001".
52
-------
e! TractID
Latitude
Longitude
01001020100
32.47711 -
86.4903033
01001020200
32.47576 -
86.4724678
01001020300
32.47402 -
86.4597033
01001020400
32.47108 -
86.4446805
01001020500
32.45892 -
86.4218165
01001020600
32.44735 -
86.4768023
01001020700
32.43052 -
86.4369107
01001020801
32.41172 -
86.531683
01001020802
32.54713 -
86.531596
Exhibit 4-5. First Part of the Population Sector Location File (2010 Census)
4.4 Air District Location File
The Air District Location file provides the Site Name, Latitude, Longitude, air data Start Date,
air data EndDate, and optionally, Air Radius, for all air quality (modeling or monitoring) sites
included in the Air Quality Data file (Section 4.5). As for sector names, the site name (or ID)
may be any string, numeric or character, and is stored as a character string (up to length 40), but
must not contain an ! character or embedded spaces. Latitude and longitude are in decimal
degrees. The start and end dates are in YYYYMMDD format (for example, 19951231 is
December 31, 1995). If the user wishes to define unique air district radii for each district, instead
of supplying a single one using the ALRRADLUS parameter in the Control Options File, then the
user can supply a sixth column in this file, which is the AirRadius in km. If no AirRadius values
are set either in Control Options File or on the Air District Location, each air monitor will have
an unlimited effective radius, and each sector will use the nearest monitor.
It is a good practice to insert comments on the first few lines of each input file to indicate the
source or type of data used. See Exhibit 4-6 for an example of the first few records of an Air
District Location file. The variables are site name, latitude, longitude, start date, end date, and
effective radius. The designated start and end dates for the simulation must be entirely covered
by the date range indicated on this file, or else the monitor will be discarded.
! Hourly ozone air quality districts for an example metropolitan area
! This file contains the locations of 105 air quality districts
! Created on November 4, 2005
Roadwayl 33.500000 -85.300000 20040301 20041031 25
Roadway2 34.500000 -85.300000 20040301 20041031 25
0000100010 34.371470 -85.461103 20040301 20041031 30
0000100009 34.194947 -85.461103 20040301 20041031 30
0000100008 34.018423 -85.461103 20040301 20041031 35
0000100007 33.841899 -85.461103 20040301 20041031 35
0000100006 33.665375 -85.461103 20040301 20041031 30
0000100005 33.488851 -85.461103 20040301 20041031 30
0000100004 33.312327 -85.461103 20040301 20041031 25
0000100003 33.135804 -85.461103 20040301 20041031 25
0000200011 34.547994 -85.239577 20040301 20041031 10
Exhibit 4-6. First Part of an Example Air District Location File
53
-------
APEX uses the Air District Location file to determine the "air district" or geographical area
represented by the ambient air quality data for a specified location. All pollutants must use the
same air districts and thus there is only one file of this type. APEX first compares the start and
end dates for each air quality site with the start and end dates for the APEX exposure simulation.
Only the sites with air quality data covering the entire simulation period are accepted. If a site is
encountered with incomplete data (gaps) between the listed start and end dates, APEX prints a
warning to the log file and stops execution. For example, the warning message is "No
appropriate districts found" if the start date of the simulation is before the first date on the
concentration input file. Air quality data in the file for dates before or after the simulation period
are simply ignored. If the user wishes to specify a roadway air district, then "road" must appear
within the district name. Consequently, "road" cannot appear in the name of a regular AQ
district.
APEX then calculates the distance of an air district location from the study area center and
compares it with the sum of CityRadius and AirRadius. This allows air quality data to be used
from sites a little outside the study area, in case they happen to be the nearest to some population
sectors. Only the sites with a distance less than this sum are retained for further calculations.
APEX then calculates the distances of each site from the sector locations. Sectors within a
distance AirRadius of an air site are included in the final study area, and use the nearest site for
their ambient air data. Each sector is assigned to only one air district. Sectors within the study
area that lack a matching air district are not included in the simulation. If no sectors remain, the
model stops during initialization and does not simulate anyone.
Not all air districts on the air quality input file need sectors assigned to them. Such air districts
are simply not included in the modeling. This feature allows the user to prepare an input file in
the simplest manner, perhaps containing more air districts than are necessary. For example, a
single input file could be prepared for all air districts in a given state. This same input file could
then be run on several study areas in the state without having to alter the air quality input file.
Internally, APEX refers to air quality districts by a sequential index (district #1, #2, etc.) that is
assigned when the district-sector mapping is established. The Log file for the model run reports
the names and locations for each air quality district number (these are also in the Sites output
file). Note that district #1 for a particular study area might not always mean the same location on
the ground for all model runs. For example, if a series of runs for different years in Denver were
performed, different monitors might be online during different years, in which case district #1
might change meaning from year to year. This can be avoided by preparing an Air Quality Data
input file (see next section) that has complete data for all air quality districts for all years being
modeled, in which case the mappings should remain the same from year to year.
4.5 Air Quality Data File
The Air Quality Data file provides air concentration data for air sites listed in the Air District
Location file for a given pollutant; there is one file of this type for each pollutant in the
simulation. Only keyword or numeric input lines are processed; other types of input lines are
ignored in this file with the exception of the first line, which (even if it is a comment) is always
54
-------
echoed to the header in each output file. Therefore, the first line should contain information
describing the content of the file.
There are two different types of air quality (AQ) data files that may be used in APEX. The first
type of file simply contains values of the AQ data for each air district for each timestep (e.g.,
hour) in the simulation. The second type of AQ data file contains distributions that allow for
person-to-person variability in the AQ data for each hour of the simulation. This type of file may
only be used when the APEX timestep is 1 hour. APEX expects type 1 by default; if a type 2 file
is to be used, the user must set the Control Options file flag MODELAQDATA = YES.
The formats of the two types of files are described in detail below.
4.5.1 Air Quality Input Data (Type 1)
Type 1 is the APEX default. Within this file, the data for each site begins with a header section
containing the site Name (see Exhibit 4-7). Recall that these site names must match those in the
Air District Location file exactly; the names are case sensitive and must not contain an ! or
embedded spaces. The sites can be in any order in this file. APEX locates the air data set by
matching a site name in the Air District Location file with the site name in this file. There can be
no missing data within the simulation period. If monitoring data contains gaps or missing
values, then these should be filled in by the user prior to running APEX.
Each of the subsequent numeric records includes a list of Timestep Average Air Concentrations
followed by a Date. The date should be in YYYYMMDD format (e.g., 20010507 is May 7,
2001). Air quality data should be in the units specified in the Control Options file for the
pollutant. The data values can be either comma or space delimited. Note that the length of each
data line in an air quality file should not exceed 5000 characters. For example, if the APEX
timestep is one hour, then each numeric record will list 24 hourly average concentration values
followed by a date. Thus, if the APEX timestep is 5 minutes, then each line will have 288 5-
minute averages followed by the date. An example of the beginning portion of this type of file is
depicted in Exhibit 4-7. The ellipsis on each line indicates data that were removed in this exhibit
so the final date entry on each line could be seen.
55
-------
Ozone air quality data for an example metropolitan area
For 105 air quality districts, for the period 03/01/04 to 10/31/04
Created on November 4, 2005
Name = Site0000100003
0.01553 0.01825 0.02621 0.02989 0.02975 0.02650 0.02310
0.03822 0.03738 0.03749 0.03754 0.03687 0.03550 0.03240
0.00577 0.00570 0.00528 0.00477 0.00394 0.00453 0.00430
0.01456 0.01828 0.01916 0.01810 0.01547 0.00925 0.00591
0.03354 0.03244 0.02412 0.01705 0.01293 0.01076 0.01066
0.03891 20040301
0.00948 20040302
0.01169 20040303
0.03326 20040304
0.02849 20040305
Exhibit 4-7. First Part of an Example Air Quality Data File (Type 1)
4.5.2 Air Quality Input Defined as Hourly Distributions (Type 2)
AQ input data defined as hourly distributions can be used to model person-to-person variability
within an hour within an AQ district. Consequently, data of this type can only be used if the
APEX timestep is equal to 1 hour (that is, TIMESTEPSPERDAY = 24, the APEX default).
Within this input file, the data for each site begins with a header section containing the site ID or
Name (see Exhibit 4-7). Recall that these site IDs must match those in the Air District Location
file exactly; the IDs are case sensitive and must not contain an ! or embedded spaces. The sites
can be in any order in this file. APEX locates the air data set by matching a site name in the Air
District Location file with the site name in this file. There can be no missing data within the
simulation period.
In this type of AQ file, each numerical record begins with a date and an hour number, followed
by any APEX distribution definitions. See Table 2-1 and Volume //for a discussion of available
probability distributions in APEX. If this type of input is to be used, the Control Options file
flag MODELAQVAR must be set to Y, otherwise an APEX error will result. An example of the
first part of an AQ data file (distribution type) is shown below in Exhibit 4-8. In this example,
the AQ value for each hour is defined by a normal distribution. The ambient AQ value for the
hour for will be sampled from this distribution for each person in the air quality district.
56
-------
! Hourly ozone air quality distributions for an example metropolitan area
! This file contains data for 127 air quality districts, for the period 01/01/04 to 12/30/04
! Created on February 26, 2008.
! Format is Date Hour DistributionDef
! Where DistributionDef is any standard APEX distribution definition
Name =0000200006
! Date Hr Distribution
20040101
1
Normal
0
01066
.00005 .
. 0
0 . 10
Y
20040101
2
Normal
0
01121
.00005 .
. 0
0 . 10
Y
20040101
3
Normal
0
01184
.00005 .
. 0
0 . 10
Y
20040101
4
Normal
0
01067
.00005 .
. 0
0 . 10
Y
20040101
5
Normal
0
01231
.00005 .
. 0
0 . 10
Y
20040101
6
Normal
0
01515
.00005 .
. 0
0 . 10
Y
20040101
7
Normal
0
01537
.00005 .
. 0
0 . 10
Y
Exhibit 4-8. First Part of an Example Air Quality Data File (Distribution Type)
4.6 Meteorology Zone Location File
The format and use of the Meteorology Zone Location file is analogous to the Air District
Location file. Each record represents one site, and contains five values: Site ID, Latitude,
Longitude, StartDate, and End Date. In the same way, the Site LD may be any string up to 40
characters long; it cannot contain an ! or embedded spaces. The IDs must match those in the
Meteorology Data file exactly; the IDs are case sensitive. The site selection process is also
analogous to that described above for the Air District Location file. The file is used to map the
set of meteorology data collected at a weather station to sectors within its zone radius for
exposure calculations. An example file is provided in Exhibit 4-9. Similar to air districts, zones
within the sum of CITYRADIUS and ZONERADIUS are used. Study area sectors for which no
meteorology data are available are not included in the simulation.
APEX makes an internal list of meteorological zones that have sectors assigned to them and
assigns them sequential numbers for convenience. This mapping is reported in the Log file and
Sites file, which are output from each model run.
! Example
APEX Meteorological
Station
Locations (Zones) File
! Createc
11/4/05
03812
35.4333
-82.5333
20040101
20041231
03813
32.7000
-83.6500
20040101
20041231
03816
37.0667
-88.7667
20040101
20041231
03820
33.3667
-81.9667
20040101
20041231
03856
34.6500
-86.7667
20040101
20041231
03870
34.9000
-82.2167
20040101
20041231
03937
30.1167
-93.2167
20040101
20041231
Exhibit 4-9. First Part of an Example Meteorology Zone Location File
4.7 Meteorology Data File
The Meteorology Data file provides hourly temperature and meteorological data for the sites
listed in the Meteorology Zone Location file. Only numeric input lines or lines containing the
keyword "name" followed by an equal sign are processed. All other types of input lines are
ignored.
57
-------
The meteorology sites may be in any order in this file. The section of data for each site must
begin with the "name" keyword input line. An example is shown in Exhibit 4-10 (some times
have been removed for display, as indicated by ellipses). The site names (site IDs) must match
those in the Meteorology Zone Location file exactly; the IDs are case sensitive and must not
contain an ! or embedded spaces.
APEX matches a site name in the Meteorology Zone Location file with the data set site name to
locate its data in this file. If desired, the user can add more comment lines in the header section
of a data set.
Temperatures can be used to assign activity diaries to days (via the profile function Diary Pools,
see Section 4.17), and any meteorological variable present in the file may be used as conditional
variables for microenvironment parameters (see Section 4.17.2 and Volume II).
The "site name" input line is followed by the meteorological data. Each data line may contain the
data listed below.
Date (YYYYMMDD)
Hour (1-24)
Temperature (degrees Fahrenheit)
Relative Humidity (percent)
Precipitation (character code, 1 or 2 characters)
Wind speed (km per hour)
Wind Direction (degrees clockwise from north)
The data are not required to be in fixed columns, but must be separated by whitespace only.
The numerical data may be integer or real (decimal)they are truncated to integers when the file
is read.
The precipitation code may be any string of 1 or 2 characters. The codes used for precipitation
must match those used in the Profile Functions file (see Section 4.17).
Only date, hour, and temperature are required. To use an optional variable such as wind speed,
then precipitation and humidity must also be present. This does not imply that the user must
make use of the precipitation data in the model run (e.g., to set microparameter distributions, see
Section 4.17). Therefore, a dummy code could be entered for precipitation in this case.
Each data set should cover the entire exposure simulation period, but it may extend further.
Thus, the user may prepare a file with a full year or many years of data for each site and then use
the same meteorology file for a series of different simulation periods. There can be no missing
data within the simulation period.
58
-------
! Hourly Met
eorologi
cal Data
! Date
Hr
Temp
Humidity
Prec
Windspeed
Direction
name=03812
20040101
1
64 . 0
30. 0
RA
12 . 0
180. 0
20040101
2
64 . 1
30. 1
CL
12 . 3
182 . 5
20040101
3
64 . 7
31. 8
RA
15.5
215.3
20040101
4
65.3
33.3
CL
18 . 7
246.7
20040101
5
65. 9
34 . 8
CL
21.6
276.3
20040101
6
65.7
34.2
CL
20.4
263. 8
name=03 813
20040101
1
66. 1
30.4
CL
12 . 7
187 . 4
20040101
2
67 . 6
34 . 1
CL
20.1
261. 1
20040101
3
67 . 7
34.2
CL
20.4
264.2
20040101
4
67 . 0
32 . 4
CL
16.9
228 . 6
20040101
5
66. 4
31. 0
CL
14 . 0
199. 8
20040101
6
66. 9
32 . 3
CL
16.7
227 . 0
Exhibit 4-10. Part of an Example Meteorology Data File
4.8 Population Data Files
Each Population Data file contains sector-level data for a single gender/race combination. Ten
gender/race specific population data files for all year 2010 census tracts have been prepared and
provided with the APEX release. However, user-defined population data files may be
constructed, if the format given below is followed.
The population files contain the population counts for each sector contained in the Sector
Location file. In general, each population file is for a single race/gender combination, although
composite files containing more than one gender or race can be used. The population counts are
given by age group. The age groups are designated in the first part of the file (i.e., the descriptor
records).
Four descriptor records must appear in each population file. These records must appear
immediately after any header comment records (which start with "!") and before the population
data records (i.e., the actual population counts). The data on these four records are read starting
to the right of the '=' sign, if present. Text descriptors to the left of the '=' signs are optional.
The contents of these four records must be as shown below.
Descriptor record 1: Gender ("Female" "Male" or "All"), Race (5 characters), #Age
(number of age groups)
Descriptor record 2: Race description (may contain blanks, up to 200 characters)
Descriptor record 3: Age Group Minimum
Descriptor record 4: Age Group Maximum
The fields in descriptor records 1,3, and 4 are space-delimited. Gender must be "Female"
"Male" or "All." The 5-character label for race also appears as a column header on the Profile
Summary output file. If the population files provided with APEX are to be used, the Race must
be White, Black, Asian, NatAm, or Other, which may be shortened to W, B, A, N, or O. If the
user provides the population files, Race could be different. For example, if one file each is given
59
-------
for all males and all females, Race could be specified as All. However, it is necessary for Race
to match the designation in the Control Options file, or a fatal error will result.
The race description is not used, but is echoed in the log file for the benefit of the user. Only the
shorter 5-character race label that is given on the first line is written to the other output files, to
save space.
The next two records specify the minimum and maximum ages for the age groups. The ages
must be delimited by a single space. Note that all the population data files must contain the same
number of population groups, and furthermore, all the group age limits (minima and maxima)
must match as well, or APEX will exit with a fatal error. The Employment Probability file,
Prevalence file, and Profile Factors file can have different age groupings. The Population Data
files provided with APEX contain single-year age groups.
The actual population data follows the descriptors records. Each population record has the
Sector ID, (which must match the IDs in the Sector Locations file exactly, and thus can be any
alphanumeric string of 40 or fewer characters without embedded spaces or an !) followed by a
Count for each population age group (youngest first). The counts are the number of people in a
given age group living in the sector; they must be delimited by a single space. Each Population
Data file used in a model run must have a record for each sector listed in the population Sector
Location file or a fatal error will result. The sectors do not necessarily have to be in the same
order in every population file in order for APEX to run, however, a warning message will result
if APEX finds that the order of the sectors in any population file differs from the order of the
sector list. A single error message will be written for each population file having out-of-order
sectors, no matter how many differences are found. APEX will exit with a fatal error message if
a sector in the final list of study area sectors cannot be found in a population file.
Exhibit 4-11 provides an example of a portion of a Population Data file. The ellipsis on each
line indicates data that were removed for brevity in this exhibit.
60
-------
! Population file by
census
tract,
extracted
from 2010
census
! File prepared by Redhorse
Corporation and
ICF
International
r
January 2014.
! The 99 to 99 age group
includes people
older
than 99
Gender,Race,#Age =
Female,
Black,
100
Race description = Black
or
African
American
Alone, Not Hispanic or Latina
Age group minimum = 0
1
2
3
4 5
6 7
8
9
10 11
12 . . .
9f
99
Age group maximum = 0
|
1
2
3
4 5
6 7
8
9
10 11
12 . . .
9f
99
01001020100 1
1
0
2
2 0
1 2
1
1
5 1
2 ...
0
0
01001020200 4
11
5
6
4 10
5 10
13
12
7 11
2 ...
0
0
01001020300 3
7
2
5
4 3
2 3
6
12
3 8
4 ...
0
0
01001020400 2
2
0
0
2 2
1 2
3
2
1 1
1 ...
0
0
01001020500 9
9
14
7
15 13
11 11
14
17
16 9
15 . . .
0
1
Exhibit 4-11. First Part of a Population Data File (2010 Census, Female Black or African
American, Not Hispanic or Latina)
4.9 Commuting Flow File
The Commuting Flow file provides cumulative fractions of the population in a home sector that
commute to different work sectors. An example portion of this file is provided in Exhibit 4-12.
After the header lines, this file is composed of sections, each starting with a home place and an
arbitrary number of work places. The names of these places must be matched to those contained
in the Sector Location file, as discussed below. The first record of each section lists the Home
Sector ID followed by two values of -1. These have no meaning; it is simply used by APEX to
recognize the beginning of a new data section (i.e., a new home). After the home record, each of
the work places for that home is listed. Each work record contains the Work Place ID, a
Cumulative Fraction of the home sector population commuting to this work place, and the
Distance (km) between the home and work places. The cumulative fraction for the last work
place in each group should always be equal to 1. APEX uses this file to determine which work
sector a simulated individual may commute to by using the cumulative fractions as commuting
probabilities.
The commuting place names must all be of the same length. These either match the population
sector names, or can be shorter. If shorter, APEX assumes that each one applies to all population
sectors that start with the commuting place name. In this case, the destination place is
determined using the probabilities on the commuting flow file, then a single population sector is
selected at random from all that map to the same commuting place.
If the population sectors are U.S. census tracts, or subdivisions of tracts, then the national
database of Commuting Flow data may be used. Otherwise, the user must prepare a replacement
file using the same format, unless commuting is not modeled.
The default Commuting Flow database contains all the U.S. census tracts and their associated
work tracts. For the 2000 Census files, the mean number of associated work tracts per home
tract was 79, with a minimum of 1 and a maximum of 413. The 2010 database uses the same
format, but the list of home and work tracts is compatible with the 2010 population files (and is
different from those used in 2000).
61
-------
APEX U.S. Tract-Level Commuting File for use with 2010 Census
Prepared by Alion Science and Technology, March 2014
ID CumulFraction Distance(km)
01001020100 -1.00000
01001020200 0.19075
01001020400 0.27168
01101000200 0.34104
01101003302 0.40462
01001020500 0.46243
01001020100 0.50867
01001020700 0.55491
-1.0
28 . 3
4 . 4
21.0
6.9
0.0
7 . 6
1.8
Exhibit 4-12. First Part of the Commuting Flow File (2010 Census)
4.10 Commuting Time File
The Commuting Time file provides the distribution of the duration of one-way commuting times
for all workers in all tracts. An example portion of this file is shown in Exhibit 4-13. The first
column lists all census tracts {Home Sector ID), and the following columns show the number of
people in each of fifteen bins, which for the 2000 Census were:
1. Total: Workers 16 years and over
2. Total: Did not work at home
3. Total: Did not work at home: Less than 5 minutes
4. Total: Did not work at home: 5 to 9 minutes
5. Total: Did not work at home: 10 to 14 minutes
6. Total: Did not work at home: 15 to 19 minutes
7. Total: Did not work at home: 20 to 24 minutes
8. Total: Did not work at home: 25 to 29 minutes
9. Total: Did not work at home: 30 to 34 minutes
10. Total: Did not work at home: 35 to 39 minutes
11. Total: Did not work at home: 40 to 44 minutes
12. Total: Did not work at home: 45 to 59 minutes
13. Total: Did not work at home: 60 to 89 minutes
14. Total: Did not work at home: 90-120 minutes
15. Total: Worked at home (0 commuting time)
APEX uses these bins to create a cumulative probability distribution of one-way commuting
time, which it uses in conjunction with commuting distance information, to assign a profile-level
commuting time variable to each employed person in the population. These data are from the
2000 or 2010 Census, and as such, include all census tracts included for that year. Take care to
note that the census data for the bins shown here are one-way commuting times, while the times
in the Diary Questionnaire file should indicate the total daily commuting time. APEX accounts
for the difference internally, as the time from the census commuting bin is doubled before
matching to activity diary times.
62
-------
! APEX U.S. Tract-Level Commuting File from 2010 American Community Survey
! Prepared by WGG at Alion Science and Technology, March 24 2014
! Variables: Tract, allworkers, allnonhome, timebins 1-9, work-at home
! New headers in March 2014: first is #bins (do not include work-at-home)
! Second header has bin boundaries (lowest time in each bin)
! The final bin may be open-ended (if so, APEX assumes 30 min width)
number of bins = 9
boundaries =
0
5
15
20
30
45
60
75
90
01001020100
905
890
40
200
80
135
335
60
10
25
0
15
01001020200
810
810
10
110
115
210
265
90
0
0
10
0
01001020300
1540
1520
45
270
260
505
355
55
15
0
15
25
01001020400
2190
2155
55
700
215
585
495
45
30
4
15
40
Exhibit 4-13. First Part of the Commuting Time File (2010 Census)
The 2010 Commuting Time file uses different time bins. There are only 12 numbers per tract, as
follows:
1. Total: Workers 16 years and over
2. Total: Did not work at home
3. Total: Did not work at home: Less than 5 minutes
4. Total: Did not work at home: 5 to 14 minutes
5. Total: Did not work at home: 15 to 19 minutes
6. Total: Did not work at home: 20 to 29 minutes
7. Total: Did not work at home: 30 to 44 minutes
8. Total: Did not work at home: 45 to 59 minutes
9. Total: Did not work at home: 60 to 74 minutes
10. Total: Did not work at home: 75 to 89 minutes
11. Total: Did not work at home: 90-120 minutes
12. Total: Worked at home (0 commuting time)
4.11 Employment Probability File
A nationwide Employment Probability file was prepared for ages 16 and above, covering all of
the tracts from the 2010 Census. Each record (tract) contains 26 probabilities13 age groups
each for males and females. The age groups in the provided file are comprised of the following:
16-19, 20-21, 22-24, 25-29, 30-34, 35-44, 45-54, 55-59, 60-61, 62-64, 65-69, 70-74, and 75
and older.
The employment probability age groups do not have to match the population file age groups,
providing increased flexibility in the demographic inputs to APEX. As a result, users may create
their own employment files, as long as the file format is followed. The ages in the Employment
Probability file may extend beyond those in the population files, but APEX will never generate a
profile outside of the ages in the Population Data files.
An example portion of the Employment Probability file is provided in Exhibit 4-14. The file
contains optional header lines, followed by three required lines. The first required line reports
the Gender for each column of data, the second line reports the age group minimum (MinAge),
63
-------
and the third line reports the age group maximum (MaxAge) Below that, each line starts with
the Sector ID, followed by a vector of decimal probabilities (one per column). The first item on
each line below the header lines is the Sector ID, followed by the eight employment probabilities
for that sector. Each probability in the national file is calculated by dividing the number of
employed persons by the total sector population for the specified age range and gender.
Whenever the total sector population for a particular age range and gender is zero, then
obviously, the employed persons must also be zero. These data are reported as zero probabilities
in the file. It should not matter what values are assigned, since no simulated persons of that type
should ever be generated by the model. Users should be aware that a custom Employment
Probability file must be created if custom Population Data files are used. The sectors in the
Employment Probability file must include all those in the study area. If the file includes extra
sectors not on the population files, the extras must be outside the study area. The ellipsis on each
line indicates data that were removed for brevity in this exhibit.
Note that any ages not covered by one of the employment age groups will automatically have an
employment probability of zero. In the example below this would apply to persons younger than
age 16.
! Employment probability fractions by gender and age group from 2010 census.
! File prepared by Redhorse Corporation and ICF International, January 2014.
Gender= M M M M
MinAge= 16 20 22 25
MaxAge= 19 21 24 29
01001020100 0.00000 0.56098 1.00000 0.73913
01001020200 0.00000 1.00000 0.85714 0.78947
01001020300 0.37903 0.35000 1.00000 1.00000
01001020400 0.26866 0.76923 0.68293 0.82178
01001020500 0.51103 1.00000 0.87903 0.84733
01001020600 0.06667 1.00000 0.68750 1.00000
01001020700 0.00000 1.00000 0.00000 1.00000
01001020801 0.13830 1.00000 0.00000 0.56250
F F
70 75
74 200
0.00000 0.00000
0.00000 0.00000
0.11538 0.00000
0.10577 0.02459
0.00000 0.00000
0.00000 0.00000
0.30769 0.06977
0.00000 0.00000
Exhibit 4-14. First Part of the Employment Probability File (2010 Census)
4.12 Profile Factors File
The Profile Factors file allows the user to specify particular profile factors by age, gender, and
sector. These profile factors can then be used to select microenvironmental scaling factors for
each group. The groups can be applied to all individuals, or to employed individuals only.
The Profile Factors file is indicated by the keyword PROFILE in the Control Options File. If a
Profile Factors file is supplied, APEX will automatically use it. Groups are assigned in a very
similar manner as in the Employment Probability file. Apart from comment lines, the Profile
Factors file starts with the following lines. The first line uses the syntax "Level = Number " to
indicate how many factor groups are included in the file. In the example below (Exhibit 4-15),
five groups are specified. The next three lines define the age-gender categories. The second line
indicates the gender, and the next two lines indicate the minimum and maximum ages for each
category. All modeled ages must be defined. The data section for each group begins with the
keyword (Name = Name). Each sector ID starts a new line, followed by a list of probabilities
64
-------
one for each age/gender category. Finally, an "END" must be inserted after every
sector/probability block to inform APEX that this block is complete.
If the profile factors are tied to employment, then groups will only be applied to employed
individuals. Unemployed individuals will automatically be assigned to Group 1, while Group 2
will be defined by the first block listed in the Profile Factors file, Group 3, the next block, and so
on. To specify employment groups, a new keyword, OCCFACTOR, was added to the Control
Options file. Set OCCFACTOR = YES to use this option. When matching diaries based on
occupation, it is important to be careful to match the name of the group to the name of the diary
occupations.
APEX uses these probabilities to select a group for each individual. These probabilities do not
have to equal oneAPEX will automatically scale these values before using them to randomly
assign a group to each individual. Groups are stored as a profile variable that may be printed in
the Profile Summary file by using the keyword FGROUP. The FGROUP number corresponds
to the order of the groups listed in the Profile Factors File. The name of each profile group can
also be printed using the GROUPNAME keyword.
Finally, these groups can be used as a conditional in the Microenvironmental Descriptions file.
To specify profile factor groups, the user should use the keyword FACTORGROUP as the
Conditional name. Microenvironmental parameters must be defined for all groups, even for
unemployed individuals, when using employment-related groups.
! Profile Factors
fractions
oy gender and
age group
! Prepared by Alion Science
and Technology
, Inc. for EPA in November 2011
Levels = 5
Gender= M
M
M
F
F
F
MinAge= 0
46
78
0
46
78
MaxAge= 4 5
77
99
45
77
99
N ame = 0-0.5 km
36025991300 0.71
0.74
0. 11
0. 00
0.38
0.56
36025991400 0.62
0.70
0. 12
0. 82
0.34
0.42
36105950200 0.76
0.57
0. 04
0.81
0.56
0. 95
36105950300 0.49
0. 14
0. 11
0. 11
0. 96
0.39
36105950400 0.66
0.86
0.37
0.27
0. 95
0. 96
END
N ame = 0.5-2 km
36025991300 0.45
0.22
0.26
0.55
0.75
0.53
36025991400 0.71
0. 93
0.50
0. 09
0. 95
0.53
Exhibit 4-15. First Part of an Example Profile Factors File
The Profile Factors file can also be used to match diaries by occupation. If occupations are
defined in the file {OCCFACTOR = Y), and diary matching is selected (USEOCCGROUPS =
Y), then APEX will weight diaries by occupation. Each diary is already assigned an occupation
in the standard diary input file, but the user can choose to define a different list of occupations in
the Profile Factors file. If the pre-assigned diary occupation is not on the Profile Factors list,
the occupation for that diary is changed to be missing ("X").
65
-------
APEX gives full selection probability weight (100%) to diaries that have the same occupation as
the individual's occupation. APEX will not select diaries with a different occupation and will
give a user-specified weight (MISSOCC) to diaries that have missing occupation. MISSOCC
has a default value of zero, so if this is not changed, then exact occupation matching will be
forced. Setting MISSOCC to a non-zero value may avoid situations of diary pools with no
possible matches for some occupation, which would cause APEX to stop without completing the
run. Occupation has no influence on the selection of diaries for individuals who are not
considered to be workers.
Different occupations have different MET distributions for work activities. Under normal diary
matching (i.e., not using OCCFACTOR and USEOCCGROUPS), these occupations are the pre-
assigned ones on each diary. Hence, the occupation (and the MET values while at work) for a
simulated individual may change from day to day, whenever the selected diary changes. When
matching diaries by occupation, each profile is assigned one occupation, which determines the
work-time MET values on all simulation days.
The MET Mapping file maps each CHAD (or other database) activity code to an internal APEX
distribution number in order to calculate the energy expended by a simulated person for each
diary event. Energy expenditures are used for estimating activity level and ventilation for each
simulated person. These quantities are used for used in creating tables of exposures at different
exertion levels and for estimating pollutant dose. A MET value is a dimensionless ratio of the
activity-dependent energy expenditure rate to the basal or resting energy expenditure (metabolic)
rate, and the CHAD activity code is an identifier associated with each diary event that indicates
the type of activity being performed. The current CHAD activity codes are listed in Table 4-5.
4.13 MET Mapping File
Table 4-5. CHAD Activity Codes
Code Activity Description
Code Activity Description
10000 Work and other income producing
activities, general
10100 Work, General
10110 Work, general, for organizational activities
10111 Work for professional/union organizations
10112 Work for special interest identity
organizations
10113 Work for political party and civic
participation
10114 Work for volunteer/helping organizations
10115 Work of/for religious groups
10116 Work for fraternal organizations
10117 Work for child/youth/family organizations
17121 Passive, sitting
17140 Create art, music, participate in hobbies
17141 Participate in hobbies
17142 Create domestic crafts
17143 Create art
17144 Perform music / drama / dance
17150 Play, unspecified, general
17151 Play, unspecified, low level
17152 Play, unspecified, moderate level
17160 Use of computers
17170 Participate in recess and physical education
17180 Other sports and active leisure, general
17200 Passive leisure, general
17201 Indoor passive leisure
17210 Watch
10118 Work for other organizations
10120 Work, income-related only
10130 Work, secondary (income-related)
10200 Unemployment
10300 Breaks
66
-------
Code
11000
11001
11100
11101
11110
11120
11121
11122
11130
11131
11200
11210
11220
11230
11231
11240
11300
11301
11310
11320
11330
11331
11332
11333
11340
11341
11342
11343
11344
11345
11350
11360
11370
11400
11401
11410
11411
11412
114201
11421
11422
11500
11600
Activity Description
Code Activity Description
Household activities, general
Other household
Prepare food, general
Washing
Prepare and clean-up food
Cooking
Baking
Fry, grill, saute
Simple food preparation
Cutting/chopping
Indoor chores, general
Move things
Put things away
Straighten up
Make bed
Clean-up food
Outdoor chores, general
Shoveling
Clean outdoors
Chop wood
Garden
Harvest
Watering
Weeding
Lawn/grass
Lawn watering
Lawn weeding
Mowing
Pruning
Use rake/leaf blower
Load/unload
Mechanical chores
Move objects
Care of clothes, general
Fold/sort
Wash clothes
Hand wash/dry
Laundry
Maintain clothes
Mend/sew
Press/iron/steam
Build a fire
Repair, general
17211 Watch adult at work
17212 Watch someone provide childcare
17213 Watch personal care
17214 Watch education
17215 Watch organizational activities
17216 Watch recreation
17220 Listen to radio/listen to recorded music/
watch T.V.
17221 Listen to radio
17222 Listen to recorded music
17223 Watch TV
17230 Read, general
17231 Read books
17232 Read magazines / not ascertained
17233 Read newspaper
17240 Converse / write
17241 Converse
17242 Write for leisure / pleasure / paperwork
17250 Think and relax
17260 Other passive leisure
17300 Other leisure
17400 Walk, bike, or jog (not in transit), general
17410 Bike, general
17411 Bike
17412 Cycles, other
17413 Tricycle
17420 Run or jog, general
17421 Run around, casual
17422 Running, vigorous/sustained
17430 Walk, general
17431 Crawl
17432 Use of walker
17433 Walk dog
17434 Walk for chores
17435 Walk inside
17500 Participate in sports, general
17501 Archery
17502 Equestrian sports
17503 Frisbee
17504 Gymnastics
17505 Skateboarding
17506 Skating
17507 Track
17510 Combat sports
67
-------
Code
11610
11620
11630
11640
11641
11642
11650
11700
11710
11720
11800
11900
11901
11902
11903
11904
11910
11911
11912
11913
11914
11915
11920
11921
11922
12000
12100
12200
12300
12400
12500
12600
12700
12800
13000
13100
13200
13201
13202
13203
13210
13211
13212
Activity Description
Code Activity Description
Repair of boat
17511
Boxing
Paint home / room
17512
Fencing
Repair / maintain car
17513
Martial arts
Home repairs, general
17514
Wrestling
Home improvement/ construction,
17520
Racquet sports
moderate level
Home maintenance, low level
17521
Badminton
Other repairs
17522
Racquetball
Care of plants, general
17523
Squash
Care of plants, low level
17524
Tennis
Care of plants, moderate level
17530
Team sports
Care for pets/animals
17531
Baseball
Clean house, general
17532
Basketball
Collect/empty trash
17533
Cheerleading
Mop
17534
Dodgeball
Sweep
17535
Football
Vacuum
17536
Hockey
Clean house, heavy
17537
Kickball
Clean bathroom
17538
Lacrosse
Clean carpet
17539
Rugby
Clean floors
17541
Soccer
Clean kitchen
17542
Softball
Wash windows
17543
Volleyball
Clean house, light
17550
Water sports
Dust
17551
Surfing
Use aerosol cleaner/freshener
17552
Swimming
Child care, general
17560
Winter sports
Care of baby
17600
Play games, general
Care of child
17610
Active games
Help / teach
17611
Fighting
Talk /read
17612
Running games
Play indoors
17613
Trampoline
Play outdoors
17620
Board games/card games
Medical care-child
17621
Board games
Other child care
17622
Bingo
Obtain goods and services, general
17623
Card games
Dry clean
17630
Dress-up/make believe
Shop/run errands, general
17640
Low energy games
Errands for children or pets
17641
Arts and crafts
Shopping, general
17642
Play with books
Shop at mall or superstore
17643
Writing/drawing
Shop for food, general
17650
Outdoor play
Grocery shopping
17651
Playground/swings
Shop for meals/snacks
17660
Play with animals
68
-------
Code
13220
13230
13300
13400
13500
13600
13700
13800
14000
14001
14100
14110
14111
14112
14120
14121
14122
14200
14201
14210
14300
14400
14500
14600
14700
15000
15100
15110
15120
15130
15140
15200
15300
15400
15500
16000
16001
16002
16100
16200
16210
16300
Activity Description
Code Activity Description
Shop for clothes or household goods
Run errands
Obtain personal care service
Obtain medical service
Obtain government / financial services
Obtain car services
Other repairs
Other services
Personal needs and care, general
Wake up
Shower, bathe, personal hygiene
Shower, bathe, general
Bathe
Shower
Personal hygiene, general
Use restroom
Wash hands/teeth/face
Medical care, general
Use nebulizer/oxygen machine
Feel sick
Help and care
Eat
Sleep or nap
Dress, groom
Other personal needs
General education and professional
training
Attend full-time school
Attend day-care
Attend K-12
Attend college or trade school
Attend adult education and special training
Attend other classes
Do homework
Use library
Other education
Social activities, general
Gamble
Go to park or festival
Attend sports events
Participate in social, political, or religious
activities
Practice religion
Watch movie
17670
Puzzles
17671
Jigsaw puzzle
17672
Word puzzle
17680
Toys
17681
Toy balls
17690
Video games
17691
Active video games
17692
Computer games
17700
Active leisure, general
17701
Camping
17702
Caving/rock climbing
17703
Climb trees/structures
17710
Dance
17720
Hiking
17730
Horseback riding
17740
Water recreation
17741
Boating
17742
Recreational swim
17743
Scuba diving
17800
Exercise, general
17810
Cardiovascular exercise
17811
Aerobics
17812
Bike for exercise
17813
Run or jog for exercise
17814
Swim for exercise
17815
Walk for exercise
17820
Strength/stretching
17821
Lift weights
17822
Physical therapy
17823
Stretching
18000
Travel, general
18010
Travel by bus, general
18020
Travel by foot, general
18030
Travel by motor vehicle, general
18031
Drive a motor vehicle, general
18032
Ride in a motor vehicle, general
18040
Wait, general
18100
Travel during work, general
18110
Travel during work by bus
18120
Travel during work by foot
18130
Travel during work by motor vehicle
18131
Travel during work, drive a motor vehicle
69
-------
Code
Activity Description
Code
Activity Description
16400
Attend theater
18132
Travel during work, ride in a motor vehicle
16500
Visit museums
18140
Travel during work, wait
16600
Visit
18200
Travel to/from work, general
16700
Attend a party, general
18210
Travel to/from work by bus
16701
Attend a party, dance
18220
Travel to/from work by foot
16702
Attend a party, eat/drink
18230
Travel to/from work by motor vehicle
16703
Attend a party, sit/stand
18231
Travel to/from work, drive a motor vehicle
16704
Attend a party, talk
18232
Travel to/from work, ride in a motor vehicle
16705
Attend a party, walk
18240
Travel to/from work, wait
16800
Go to bar / lounge
18300
Travel for education, general
16900
Other entertainment / social events
18310
Travel for education by bus
17000
Leisure, general
18320
Travel for education by foot
17010
Indoor leisure
18330
Travel for education by motor vehicle
17111
Hunting, fishing, hiking
18331
Travel for education, drive a motor vehicle
17112
Golf
18332
Travel for education, ride in a motor vehicle
17113
Bowling / pool / ping pong / pinball
18340
Travel for education, wait
17114
Yoga
U
Uncertain
17120
Participate in outdoor leisure
X
Missing
Each of the CHAD codes is mapped to an internal APEX distribution number; activities that
have identical energy expenditure associated with them map to the same distribution. The
distributions themselves are defined by number in the MET Distribution File (Section 4.14).
Each line of the MET Mapping file contains the variables described below.
Activity code: This maps the CHAD activity to the internal APEX distribution number.
Age category: Some MET distributions differ for persons of different ages. This variable
maps the age groups to the correct distribution number. The age category given in this
file is a label representing the age group. APEX will assign distributions as shown
below.
o Age is "0": APEX will use for persons of all ages
o Age is "20": APEX will use for persons age 0 to 25
o Age is "30": APEX will use for persons age 26 to 39
o Age is "40": APEX will use for persons age 40 and older
Occ: The MET distributions for the "Work" CHAD activity differ based on the
occupation of the profile. This variable maps the different occupations to the correct
distribution number. If the user has defined their own occupational groups for each
profile, then profile occupations will be used instead of diary occupations.
APEXDist #: This is an internal index used by APEX to access the distribution. These
values range from 1 to a maximum of 256.
Notes: Description of the activity being modeled by the MET distribution. This is for the
convenience of the user and is not used internally by APEX.
An example portion of the MET Mapping file is given in Exhibit 4-16.
70
-------
! CHAD MET Distribution Mapping file for input to APEX
! Created 9-28-2015 by ICF International, with updates to many activity codes and !
distributions
Activity Age Occ. APEX Dist # Notes
10000
0
ADMIN
7
Work
and
other
income
producing
activities,
general
10000
0
ADMSUP
7
Work
and
other
income
producing
activities,
general
10000
0
FARM
23
Work
and
other
income
producing
activities,
general
10000
0
HSHLD
13
Work
and
other
income
producing
activities,
general
10000
0
LABOR
85
Work
and
other
income
producing
activities,
general
10000
0
MACH
176
Work
and
other
income
producing
activities,
general
10000
0
PREC
82
Work
and
other
income
producing
activities,
general
10000
0
PROF
67
Work
and
other
income
producing
activities,
general
10000
0
PROTECT
67
Work
and
other
income
producing
activities,
general
10000
0
SALE
67
Work
and
other
income
producing
activities,
general
10000
0
SERV
75
Work
and
other
income
producing
activities,
general
10000
0
TECH
81
Work
and
other
income
producing
activities,
general
10000
0
TRAMS
11
Work
and
other
income
producing
activities,
general
10000
0
X
66
Work
and
other
income
producing
activities,
general
10100
0
X
66
Work,
general
10110
0
X
66
Work,
general,
for organizational activities
10111
0
X
66
Work
for
profes
sional/union organizations
10112
0
X
66
Work
for
special interest identity organizations
10113
0
X
66
Work
for
political party and civic participation
10114
0
X
66
Work
for
volunteer/ helping organizations
10115
0
X
66
Work
of/for religious
groups
10116
0
X
66
Work
for
fraternal organizations
10117
0
X
66
Work
for
child
/ youth / family
organizations
10118
0
X
66
Work
for
other
organizations
10120
0
X
66
Work,
income-related
only
10130
0
X
66
Work,
secondary
¦ (income-related)
10200
0
X
66
Unemployment
10300
0
Any
93
Breaks
11000
0
Any
14
Household activities,
general
11001
0
Any
1
Other household
11100
0
Any
125
Prepare food, general
11101
0
Any
52
Washing
11110
0
Any
4
Prepare and clean-up
food
11120
0
Any
125
Cooking
Exhibit 4-16. First Part of the MET Mapping File
The user should not change the MET Mapping file unless the user has developed their own
activity codes, or if they have defined their own occupation groups. User-defined occupation
groupings must match the occupations listed for activity code 10000.
4.14 MET Distribution File
The METDistribution file provides the actual distributions for calculating the MET value for
each diary event (activity). The distributions are defined by APEX distribution numbers as given
in the MET Mapping file. A MET value is a dimensionless ratio of the activity-dependent
energy expenditure rate to the basal or resting energy expenditure (metabolic) rate, and the
CHAD activity code is an identifier associated with each diary event that indicates the type of
activity being performed. In general, the user should not change the distributions in this file, as
these data were developed from extensive experimental data on human energy expenditures.
71
-------
The distribution definitions use the standard APEX distribution format (i.e., a distribution shape,
followed by 4 distribution parameters, upper and lower truncation bounds, and a resampling
flag). The 4 parameters used are dependent on the shape of the distribution. Each data line in
this file provides the information shown below in list format.
Dist#: This is an internal index used by APEX to access the distribution. These values
range from 1 to a maximum of 256. This matches the distribution numbers used in the
MET Mapping file.
Shape. This variable gives the type of the MET distribution.
Pari . Parameter 1 of the MET distribution.
Par2. Parameter 2 of the MET distribution.
Par3 . Parameter 3 of the MET distribution.
Par4. Parameter 4 of the MET distribution.
LTrunc. Lower truncation point of the MET distribution.
UTrunc. Upper truncation point of the MET distribution.
ResampOut: Distribution resampling flag.
General Use. Text description of the general use of the particular distribution in APEX.
(Optionalnot used by the model code).
Volume II provides complete details for defining probability distributions in APEX; a summary
of the available distributions is presented in Table 4-6.
ResampOut is a Yes/No variable. If it is set to "Y," then generated random values outside the
truncation limits are effectively thrown out and replaced by new samples, repeated until falling
within the bounds. If ResampOut = N, such values are shifted to the appropriate truncation
point. This may result in a "spike" of probability at LTrunc and/or UTrunc. Both methods
ensure that all returned values equal or exceed LTrunc and are less than or equal to UTrunc.
Both methods, however, may alter the statistical properties of the returned values. For example,
truncation will always reduce the variance and may in some cases alter the mean as well. The
default is ResampOut = Yes.
Table 4-6. Available
robability Distributions in APEX
Distribution
Shape
Pari
Part
Par3
Par4
LTrunc
(optional)
UTrunc
(optional)
ResampOut
(optional)
Beta
Beta
Minimum
Maximum
Shapel (si)
>0
Shape2 (s2)
>0
Lower
truncation
limit
Upper
truncation
limit
Resample outside
truncation? (Y/N)
Burr
Burr
Scale(b)
>0
Shapel(sl)
>0
Shape2 (s2)
>0
Shift(a)
Lower
truncation
limit
Upper
truncation
limit
Resample outside
truncation? (Y/N)
Cauchy
Cauchy
Median
Scale (b)
>0
1 .ower
truncation
limit
Upper
truncation
limit
Resample outside
truncation? (Y/N)
Discrete
Discrete
This type of distribution has no parameters, ralher llie keyword is simply followed by a list of up to 100
specific values. One of these values is selected al random, with equal probability for each. Duplicate values
are acceptable.
Exponential
Exponential
Decay
constant,
k > 0
Shift (a)
Lower
truncation
limit
Upper
truncation
limit
Resample outside
truncation? (Y/N)
Extreme
Value
Evahie
Scale (b)
>0
Shift (a)
Lower
truncation
limit
Upper
truncation
limit
Resample outside
truncation? (Y/N)
72
-------
Distribution
Shape
Pari
Par2
Par3
Par-t
LTrunc
(optional)
UTrunc
(optional)
ResampOut
(optional)
Gamma
Gamma
Shape (s)
>0
Scale (b)
>0
Shi It (a)
Lower
truncation
limit
Upper
truncation
limit
Resample outside
truncation? (Y/N)
Logistic
Lgt
Mean
Scale (b)
>0
Lower
truncation
limit
Upper
truncation
limit
Resample outside
truncation? (Y/N)
Lognormal
Lognormal
Geometri
c mean
(gm) of
unshifted
distributio
n
Geometric
standard
deviation
(gsd)> 1
Shi It (a)
Lower
truncation
limit
Upper
truncation
limit
Resample outside
truncation? (Y/N)
Loguniform
LUniform
Minimum
>0
Maximum
>0
Lower
truncation
limit
Upper
truncation
limit
Resample outside
truncation? (Y/N)
Normal
Normal
Mean
Standard
deviation
Lower
truncation
limit
Upper
truncation
limit
Resample outside
truncation? (Y'N)
OffOn
OffOn
Probabilit
y of being
0 (0-1)
Pareto
Pareto
Shape (s)
>0
Scale (b)
: 0
Shi It (a)
1 ,o\\er
truncation
limit
I pper
truncation
limit
Resample outside
truncation? (Y/N)
Point
Point
Point
Value
Triangle
Triangle
Minimum
Maximum
Peak
Lower
truncation
limit
Upper
truncation
limit
Resample outside
truncation? (Y/N)
Uniform
Uniform
Minimum
Maximum
Lower
truncation
limit
Upper
truncation
limit
Resample outside
truncation? (Y/N)
Weibull
Weibull
Shape (s)
>0
Scale (b)
>0
Shift
Lower
truncation
limit
Upper
truncation
limit
Resample outside
truncation? (Y/N)
Periods must be used as placeholders in the file if a parameter is not needed for a particular
distribution (except for the Discrete distribution).
See Volume II for further information about the use of MET probability distributions in APEX.
A portion of this file is shown in Exhibit 4-17. For brevity in this exhibit, the General Use
descriptions of the corresponding activities have been truncated with ellipses.
73
-------
APEX MET Distribution File
Created 9-28-2015 by ICF International, with updates to many activity codes and
distributions
For use with CHAD mapping file
!Dist#
Shape
Pari
Par2
Par3
Par4 LTrunc
UTrune
ResampOut
General Use
1
Exponential
0.3
3
1
9
Y
Other household
2
Exponential
0.4
2.6
1
6
Y
Clean outdoors
3
Exponential
0.7
3.5
1
6
Y
Paint home/room
4
Exponential
1.1
1.9
1
4
Y
Prepare and clean-up.
5
LogNormal
0.9
1.1
0
0.8
1.1
Y
Sleep or nap
6
LogNormal
1.2
1.4
0
0.9
2.3
Y
Listen to radio/...
7
LogNormal
1.7
1.2
0
1.4
2.7
Y
Work and other...
8
LogNormal
1.8
1.4
0
1.4
4
Y
General education...
9
LogNormal
2 . 6
1.7
0
1.1
6
Y
Play games, general
10
LogNormal
3
1.2
0
2.5
5
Y
Help and care
11
LogNormal
3
1.5
0
1.3
8.4
Y
Work and other...
12
LogNormal
3
1.5
0
1.5
8
Y
Go to bar/lounge
13
LogNormal
3.5
1.2
0
2.5
6
Y
Work and other...
14
LogNormal
3.6
1.5
0
1.8
7.3
Y
Household activities.
15
LogNormal
3.9
1.4
0
2
9
Y
Participate in...
16
LogNormal
4
1.6
0
1.9
8.4
Y
Walk, general/Crawl...
Exhibit 4-17. First Part of the Activity-specific MET File
4.15 Physiological Parameters File
Thq Physiological Parameters file provides age- and gender-specific distributions for a number
of physiological parameters (see Exhibit 4-18). The parameters are listed in Table 4-7. See
Volume //for details of these parameters and the equations in which they are used in APEX.
The variables in Table 4-7 that have RMR = 1 in their descriptions are required when setting
RMRME THOD = 1 on the Control Options file. The ones with RMR = 2 are required when
using RMRMETHOD = 2, which is the default method. Similarly, the variables with HtWt = 1
are required when HTWTMETHOD = 1 is specified on the Control Options file, and those with
HtWt = 2 are required when using the default method (which is HTWTMETHOD = 2).
Table 4-7. Parameters in the Physiological Input File
Variable
Description
Units
NV02Max
Normalized maximum oxygen uptake
ml-02/(min-kg)
(Note: while the APEX inputs for
NV02Max are in ml-02/(min-kg),
APEX outputs V02Max in the
Profile Summary file in L-Ch/min)
Bm
Body mass (HtWt=l)
kg
LogBM
Natural logarithm of body mass (kg) (HtWt=2)
Height
Height (HtWt=2) (Note: units different from other
HtWt method)
cm
HtWtCorr
Pearson correlation between height and logBM
(HtWt=2)
-
RMRInt
Intercept of resting metabolic rate regression (both
RMR= 1 and RMR2, but with different values)
MJ/day
(Note: while the APEX inputs for
RMR are in MJ/day, APEX outputs
RMR in the Profile Summary file in
kcal/min).
RMRSlp
Slope of resting metabolic rate regression (RMR=1)
MJ/(day-kg)
74
-------
Variable
Description
Units
RMRErr
Standard deviation for resting metabolic rate
regression (both RMR=1 and RMR=2)
MJ/day
RMR BM
RMR regression coefficient for BM (RMR=2)
-
RMRLBM
RMR regression coefficient for Log(l+BM)
(RMR=2)
-
RMR Age
RMR regression coefficient for Age (RMR=2)
-
RMRLAge
RMR regression coefficient for Log(l+Age)
(RMR=2)
-
RMR Ht
RMR regression coefficient for Height (RMR=2)
-
RMRLHt
RMR regression coefficient for Log(l+Height)
(RMR=2)
-
Hmg
Blood hemoglobin density
g/dl
BSAExpl
Exponent 1 for calculating body surface area
-
BSAExp2
Exponent 2 for calculating body surface area
-
MaxOxD
Maximum oxygen deficit
ml/kg
BldFacl
Blood volume factor 1
ml/lb
BldFac2
Blood volume factor 2
ml/inches3
Heighlnt
Intercept of height regression (HtWt=l)
inches
HeightSlp
Slope of height regression (HtWt=l)
children under 18: inches/(year of
age)
adults: inches/lb (lbs body weight)
HeightErr
Standard deviation of height regression (HtWt=l)
inches
ECF
Energy conversion factor
L-02/kcal
RecTime
Time required to recover maximum oxygen deficit
hours
Endgnl
Endogenous CO production rate 1
ml/min
Endgn2
Endogenous CO production rate 2 (used for women
in 2nd half of menstrual cycle)
ml/min
B1-B9
Model parameters for calculation of %AFEV1
(note:B7 not currently used)
-
FEVU
Uncertainty term used in calculation of %AFEV1
-
FEVE1
1st error term used in calculation of %AFEV1
-
FEVE2
2nd error term used in calculation of %AFEV1
-
FEVBMI
Cut point for body mass index in regression
-
FEVSlp
Slope of age fit in %AFEV1 calculation
-
FEVInt
y-intercept of age fit in %AFEV1 calculation
-
ModEVR
Distribution of moderate 8-hour average exertion
cutoff
L/(min-m2)
Distributions for the above parameters are assigned to persons of every age and gender
combination in the Physiology file. The distributions are defined in the APEX distribution
format (i.e., a distribution shape, followed by 4 distribution parameters, upper and lower
truncation bounds, and a resampling flagsee Volume II). Thus, each data line contains the
information listed below.
Variable keyword
AgeMin (minimum age) for the current parameter distribution definition
AgeMax (maximum age) for the current parameter distribution definition
Gender for the current parameter distribution
Shape (type of distribution)
Pari (parameter 1) of the distribution, which depends on shape
75
-------
Par2 (parameter 2) of the distribution, which depends on shape
Par3 (parameter 3) of the distribution, which depends on shape
Pari (parameter 4) of the distribution, which depends on shape
LTrunc (lower truncation point of the distribution)
UTrunc (upper truncation point of the distribution)
ResampOut (distribution resampling flag)
Thus, each line of the Physiology file can define the distribution for a range of ages, but only a
single gender. The physiological parameters must be defined for both genders for all ages 0-100
years, with the exception of ENGN2, which need only be defined for females. An APEX fatal
error will result if not all data are provided. In general, the distributions in this file should not be
changed from their default values, as they were derived from available physiological data.
See Table 4-6 for the available distribution shapes and required parameters. Periods (".") must
be used as placeholders if a parameter is not needed for a particular distribution.
76
-------
APEX Physiology
Data
Input File
Reference: Isaacs,
K.
and L. Smith, Alion
Science
and Technology.
"New Values for
Physiological
Parameters
for the Exposure
Model
Input File Physiology.txt,"
December 20,
2005.
Memorandum
to T. McCurdy, US
EPA
Updated
09/29/2010
(Data reorganized, some
added;
existing
data unchanged
from
05/04/06
version)
Updated
02/04/2011
with dFEVl model parameters from McDonnell, Stewart & Smith,
2010
"Predictions of
ozone
-induced lung function..."
Updated
4/11/11
to
correct the
Slope for RMR regression for males
ages 0-2
to 0
.249
from
0.244, per Table 1
in
Layton (1993) .
Updated
8/22/12
with
the B9 threshold parameter for dFEVl,
MSS model
parameters
from MSS 2012
Table 2
Threshold
model
Updated
12/31/12 with
age term
extensions
for the
MSS 2012
model
Updated
4/8/14
to
add
the FEVCUT parameter
for APEX v4 63 (note that
the definition
of FEVE2
changes
with v463,
per the McDonnell et a]
. 2013
model)
Height and weight
data updated
on 11/15/2016 to include correlation,
using
data
from Jonathan
Cohen at
ICF
Variable
AgeMin
AgeMax Gen
Shape
Pari
Par2
Par3 Par4
LTrunc UTrunc
ResampOut
NV02Max
NV02MAX
0
0
M
Normal
48.3
1.7
44.3
52
2
N
NV02MAX
1
1
M
Normal
48.6
2 . 0
43 . 8
53
3
N
NV02MAX
2
2
M
Normal
48.9
2 . 4
43 . 4
54
4
N
NV02MAX
3
3
M
Normal
49.2
2 . 7
43 . 0
55
4
N
NV02MAX
4
4
M
Normal
49.5
3 . 0
42 . 5
56
5
N
NV02MAX
5
5
M
Normal
49.8
3 . 3
42 . 1
57
6
N
NV02MAX
6
6
M
Normal
50 . 1
3 . 7
41.6
58
6
N
NV02MAX
7
7
M
Normal
50 . 4
4 . 0
41.2
59
7
N
NV02MAX
8
8
M
Normal
50 . 8
4 . 3
40.8
60
8
N
NV02MAX
9
9
M
Normal
51. 1
4 . 6
40.3
61
8
N
Body Mass
BM
0
0
M
Lognormal
7 . 8
1.301
0
3 . 6
11
8
N
BM
1
1
M
Lognormal
11. 4
1. 143
0
8 . 2
16
1
N
BM
2
2
M
Lognormal
13 . 9
1. 146
0
9.8
20
9
N
BM
3
3
M
Lognormal
16.0
1. 154
0
11.7
23
7
N
BM
4
4
M
Lognormal
18 . 5
1. 165
0
11. 1
28
1
N
BM
5
5
M
Lognormal
21. 6
1.234
0
13 . 7
42
4
N
BM
6
6
M
Lognormal
23 . 1
1.213
0
16.1
41
1
N
BM
7
7
M
Lognormal
27 . 1
1.216
0
19.3
46
8
N
BM
8
8
M
Lognormal
31.7
1.302
0
19.1
6 6
2
N
BM
9
9
M
Lognormal
34 . 7
1.265
0
24 . 0
69
9
N
BM
10
10
M
Lognormal
38 . 3
1.280
0
24 . 3
72
9
N
BM
11
11
M
Lognormal
44.1
1.308
0
26.2
83
8
N
BM
12
12
M
Lognormal
48.0
1.315
0
27 . 7
94
8
N
Intercept for RMR
regression for RMRMethod=l
RMRINT
0
2
M
Point
-0.127
RMRINT
3
9
M
Point
2 . 110
RMRINT
10
17
M
Point
2 .754
RMRINT
18
29
M
Point
2.896
Exhibit 4-18. An Example of a Portion of the Physiological Parameters File
4.16 Ventilation File
The Ventilation file contains a set of regression parameters used by APEX to estimate ventilation
quantities from the event MET. This is a small file containing the parameters for each of four
age groups (Exhibit 4-19). This file should not be edited except by advanced users who
understand the APEX ventilation algorithm.
For more information on the ventilation algorithm and the derivation of the values in this file, see
Volume II and Graham and McCurdy (2005).
77
-------
! APEX Ventilation Data
File
! MinAge MaxAge bO
sebO
bl
sebl
b2
seb2
b3
seb3
eb
ew
R2
0 19 4.4329
0.0579
1.0864
0.0097
-0.2829
0.0124
0.0513
0.0045
0 . 0955
0.1117
0 . 925
20 33 3.5718
0.0792
1.1702
0 . 0067
0.1138
0.0243
0 .045
0.0031
0.1217
0.1296
0.8927
34 60 3.1876
0.1271
1.1224
0 .012
0.1762
0.0335
0.0415
0.0095
0.1260
0.1152
0.8922
61 100 2.4487
0.3646
1.0437
0.0195
0.2681
0.0834
-0.0298
0.0100
0.1064
0.0676
0.8932
Exhibit 4-19. The APEX Ventilation Input File for VEMethod=l
The newer VEMETHOD = 2 uses an altered format, shown in Exhibit 4-20. In this case, the
same regression applies to everyone.
! Ventilation file - ventilation regression parameters for method 2
! Reference: "Analysis of the VE/V02 relationship in the Adams data set"
! Graham Glen, ICF, September 2016
! MinAge MaxAge Intercept L0G(V02) (V02/V02max)**4 eb ew
0 100 3.300 0.8128 0.5126 0.09866 0.07852
Exhibit 4-20. The APEX Ventilation Input File for VEMethod=2
4.17 Profile Functions (Distributions) File
The Profile Functions input file defines functions for variables associated with each simulated
profile. There are four types of functions that can be defined, as indicated below.
Functions for built-in APEX variables: These are variables that are predefined in APEX,
and whose values under different circumstances can be customized by the functions
defined in this file. Most of these variables are also "conditional variables" since
microenvironmental parameters can depend on their values.
Functions for creating user-defined APEX conditional variables: These are generic
variables that the user may define and then use in calculating microenvironmental
parameters. The names of these variables have no set intrinsic meaning in APEX; they
can be used to represent whatever the user wishes. Up to eight of these variables may be
defined in APEX.
Functions for creating user-defined APEX conditional variables that vary by region: These
are generic variables that the user can define and then use for calculating
microenvironmental parameters. These conditional variables can vary by region (either
county or sector) and thus may be evaluated differently for individuals who reside in
these different regions. Up to five of these functions may be defined.
Functions for creating user-defined APEX conditional variables that vary with ambient air
quality: These are also generic variables that the user can define and then use for
calculating microenvironmental parameters. These variables vary by ambient air quality,
and thus are recalculated during each timestep. Up to five of these functions may be
defined.
The relationships among the different functions that can be defined in the Profile Functions file
and the microenvironmental descriptions are shown in Figure 4-1. The built-in and user-defined
functions are used to define a set of conditional variables, Vc, which are functions of input
78
-------
APEX variables (Vi). These conditional variables are used in determining microenvironmental
parameters. Microenvironment parameters are quantities that appear in the equations for the
microenvironmental concentrations. The relationship between the conditional variables and the
microenvironment parameters are described in the Microenvironmental Descriptions file (see
Section 4.24).
Figure 4-1. Relationship between Profile Functions and Microenvironmental Descriptions
Files
4.17.1 Defining a Profile Function
The general procedure for defining a profile function is as shown below.
1. A function definition begins with its name on the first input line.
2. The user may add as many comment lines as necessary to describe the profile function or
units of the involved parameters.
3. If the function is of type regional (RegionalConditionall through RegionalConditional5),
then a statement is required to define how the regions are defined, either by county or
sector, and how many different regions are being modeled.
79
-------
4. The number of subsequent input lines varies with the number of input variables required
to define the function. At least one (and usually two) input lines are needed for each input
variable of the function. In addition, at least two lines are also needed for the function
result. For each input variable (table dimension), the first line starts with the keyword
Input, followed by the indexing number of the variable in the function, the Type of Input
Variable, and the Number of Values (Nvals) allowed for the input variable. At the end
of this input line, the user may add comments in double quotes to explain input variables.
The lines directly following define the input variable dataspecifically, they define how
the input variable is grouped into integer categories for indexing the table of results. The
Type Of Input Variable must be one of the following:
Probability,
Realrange,
Intrange,
Intvalue,
Intindex
Conditional, or
Regionindex
Probability means the result is randomly determined, using fixed probabilities for each
outcome. The input variable data for Probability is a list of the Nvals fixed
probabilities. The sum of the probabilities must equal 1. RealRange represents a set of
discrete categories, each consisting of a range of real numbers. In this case, the
categories are defined by (Nvals-1) cut-points. If the input variable falls exactly on a
cut point, it falls into the higher bin. IntRange is similar, except that each category
consists of a range of integers. IntValue denotes that each possible value that the input
variable may take on is listed on the data line. Intindex signifies that the input variable
is an integer and is to be used to index the table of results directly (e.g., a value of 3
means that it uses the third cell of a table dimension). Thus, this type of input variable
does not require a second line. Conditional refers to conditional probabilities that
depend on the values of other input variables. Only one Conditional input variable is
allowed in a function, and it comes last in the function specification. A table of
probabilities follows. The number of entries in the probability table must be equal to
the product of the number of category combinations for the other inputs and the number
of possible function results. Regionindex is an option for the RegionalConditional
only. This option uses two columns of input dataa region and an indexand maps
regions to the index to be used directly in the Microenvironmental Results file. The
index must be between 1 and the total number of regions used.
The examples provided in the sections that follow illustrate the appropriate use of these
input variable types. Exhibit 4-21 contains examples of Probability, IntRange,
IntValue, and Conditional. An example of Intindex occurs in the DiaryPools
definition just below Exhibit 4-21. RealRange works exactly like IntRange, except the
cut-points are real numbers (that is, containing decimal points) instead of integers.
5. After all the input variables, except for Regionindex, are specified, the next line must
contain the keyword Result, followed by a type (Integer, Real, or Histogram) and the
80
-------
number of possible results (Nresults). Regionlndex variables must be an integer.
6. The table results are then listed in order in subsequent lines. If the result type is
designated as Integer, the results must be a list of integers of length Nresults. If the type
is Real, then the list of results must contain Nresults real numbers. If the result type is
Histogram, the results are a series of (Nresults+1) cut points that define Nresults bins. If
there are multiple inputs, the indexing follows the Fortran convention of increasing the
leftmost (first defined) variable over its range, then the second variable, and so on.
7. The profile function ends with a new line that has a # sign.
The types of profile functions are discussed in detail below along with examples. Note that
when preparing or editing a profile functions file, be careful not to use Tab to separate the items
on a line. APEX explicitly searches for blanks (spaces) as delimiters, and does not recognize
Tabs as such.
4.17.2 Functions for Built-in, User-defined, AQ and Regional APEX Variables
The built-in APEX variables for which functions can be assigned are given in Table 4-8. All of
these variables are conditional variables which can be used to define microenvironment
parameters, with the exception of the variable DiaryPools. Note that a few other APEX
variables, such as gender, can be used as conditional variables (see Section 4.24.2). DiaryPools
is the only function that APEX requires to be defined, as it is used in the selection of appropriate
CHAD diaries for different days in the simulation. The input variables required for each of these
functions are hard coded; the required inputs for each variable are listed in the table. Some
conditional variables in this file must be defined if others that require them are defined. For
example, AC Home must be present if WindowRes is defined.
Several generic user-defined conditional variables are listed in the table as well. These functions
take a single input variable, which must be defined by fixed probabilities for each of the function
results (categories). The four types are distinguished by the sampling properties of their input
variable: Daily Conditional requires a variable that is sampled daily; ProfileConditional requires
a variables assigned once per profile, such as age or gender; RegionalConditional requires a
region-dependent variable, where regions are defined as groups of home sectors; and finally,
AQConditional variables depend on the ambient air quality (and so can change their status over
time, although not necessarily daily).
Each of the functions in the table returns an integer category for each combination of input
parameters. For the conditional variables, these category numbers can be used in defining the
microenvironment parameters in thq Microenvironment Descriptions file (see Section 4.24.2).
Note that one conditional based on the Profile Factors file, FactorGroup, can be used in the
Microenvironment Descriptions file, but is defined in its own file, rather than in the Profile
Functions file.
Several examples are shown in Exhibit 4-21. The first is the definition for a function for
AvgTempCat. It returns an integer category number for the average temperature, which will be
used in the definition of one or more microenvironment parameters. The first and only input
variable defines the integer ranges (via IntRange) for the three categories of average
temperature. In this case, the ranges are < 50 degrees, 50-77 degrees, and > 78 degrees. The
81
-------
function reads the daily average temperature and determines which category it falls in. The
resulting categories are 1, 2, and 3. If the average temperature were 69 degrees, then the
AvgTempCat function would return a value of 2.
The second example is a definition for WindowRes. The first input variable is ACHome, and
the categories for it are defined by its two possible integer values (via IntValue), as 1 (has AC)
or 2 (does not have AC). The second input variable is the maximum daily temperature; the
categories for it are defined via IntRange in a manner similar to that demonstrated in the first
example. The third input variable for this function as indicated in Table 4-8 is the average daily
temperature. In this case, AvgTemp is not used to determine WindowRes, and leaving it out of
the function definition means that by default means that all possible values for AvgTemp are
assigned to the first bin. The fourth and final input variable is a Conditional probability. Each
row lists the probabilities for each possible output, for one combination of inputs. The first
probability in each row is the chance of YES (which equals 1 in APEX), while the second
probability is the chance of NO (which equals 2). Some functions might have more than two
possible outcomes, so there could be more than two probabilities on each line. Thus, the first
row of the table contains probabilities for WindowRes = 1 (YES/OPEN) and WindowRes = 2
(NO/CLOSED) for AC Home = 1 (YES), MaxTemp<56, and any AvgTemp value. The second
line has the same probabilities for AC Home = 2 (NO), MaxTemp<56, and any AvgTemp value.
The last line has the probabilities for AC Home = 2 (NO), MaxTemp>78, and any AvgTemp
value. In this example, when it is hot outside, the window status is just 10% likely to be open
when AC Home = YES, but 90% likely when AC Home = NO.
82
-------
Table 4-8. Variab
es that can be Defined in the Profile Functions File
Conditional Variable
Purpose
Input Variables
Number of Categories
Function
Reevaluated
TempCat
Binning hourly
temperatures into
categories
INPUT 1: Temperature on hour of simulation
any number
hourly
HumidCat
Binning hourly
humidities into
categories
INPUT 1: Humidity on hour of simulation
any number
hourly
WindCat
Binning hourly wind
speeds into categories
INPUT1: Wind speed on hour of simulation
any number
hourly
DirCat
Binning hourly wind
directions into
categories
INPUT1: Wind direction on hour of simulation
any number
hourly
PrecipCat
Assigning precipitation
codes to categories
INPUT 1: Precipitation code on hour of simulation
any number (equal to or
less than the number of
precipitation codes in
the Meteorology Data
file)
hourly
MaxTempCat
Binning daily maximum
temperatures into
categories
INPUT 1: Temperature on hour of simulation
any number
daily
AvgTempCat
Binning daily average
temperatures into
categories
INPUT1: 24-hour average temperature on day of
simulation (AvgTemp)
any number
daily
DiaryPools
(required)
Assigning diary pools
INPUT 1: Maximum temperature on simulated day
(Max Temp)
INPUT2: Average temperature on simulated day
{AvgTemp)
INPUT3: Day of the week
any number
daily
PoolTrans
Grouping of diary pools
for empirical cluster
transition probabilities
INPUT1: Pool number on "to" day
INPUT2: Pool number on "from" day
(#pools)A2
once per run
HasGasStove
Probability of having a
gas stove
INPUT1: Probabilities for the 2 results
2 (Y/N)
once per profile
83
-------
Conditional Variable
Purpose
Input Variables
Number of Categories
Function
Reevaluated
HasGasPilot
Probability of having a
pilot light, conditional
on Hasgasstove
INPUT 1: Has Gas Stove (Y/N)? (HasGasStove)
INPUT2: Conditional probabilities for the result
categories for both HasGasStove = Y and
HasGasStove = N
2 (Y/N)
once per profile
AC_Home
Probability of having
different types of home
air conditioning or
ventilation
INPUT1: Fixed probabilities for the types of air
conditioning/ventilation (the number of types is user-
defined)
any number
once per profile
ACjCar
Probability of having
A/C in car
INPUT1: Probabilities for the 2 results
2 (Y/N)
once per profile
WindowRes
Probability of residence
windows being open or
closed, conditional on
AC_Home,
MaxTempCat, and
AvgTempCat
INPUT 1: Type of home A/C (AC Home)
INPUT2: Max. temperature on day of simulation
(Max Temp)
INPUT3: Average temperature on day of simulation
(Avgtemp)
INPUT4: Conditional probabilities for the result
categories for every combination of inputl-input3
categories
2 (Y/N)
daily
WindowCar
Probability of car
windows being open or
closed, conditional on
ACCar. MaxTempCat,
and AvgTempCat
INPUT 1: Has car A/C (AC_Car)
INPUT2: Max. temperature on day of simulation
(Max Temp)
INPUT3: Average temperature on day of simulation
(AvgTemp)
INPUT4: Conditional probabilities for the result
categories for every combination of inputl-input3
categories
2 (Y/N)
daily
SpeedCat
Probability of average
speed categories for
vehicles
INPUT1: Fixed probabilities for the result categories
any number
daily
DailyConditionall
Generic daily
conditional variable #1
INPUT1: Fixed probabilities for the result categories
any number
daily
DaifyConditional2
Generic daily
conditional variable #2
INPUT1: Fixed probabilities for the result categories
any number
daily
DaifyConditional3
Generic daily
conditional variable #3
INPUT1: Fixed probabilities for the result categories
any number
daily
84
-------
Conditional Variable
Purpose
Input Variables
Number of Categories
Function
Reevaluated
ProfileConditionall
Generic profile
conditional variable #1
INPUT1: Fixed probabilities for the result categories
any number
once per profile
ProfileConditional2
Generic profile
conditional variable #2
INPUT1: Fixed probabilities for the result categories
any number
once per profile
ProfileConditiona3
Generic profile
conditional variable #3
INPUT1: Fixed probabilities for the result categories
any number
once per profile
ProfileConditional4
Generic profile
conditional variable #4
INPUT1: Fixed probabilities for the result categories
any number
once per profile
ProfileConditional5
Generic profile
conditional variable #5
INPUT1: Fixed probabilities for the result categories
any number
once per profile
RegionalConditionall
Generic regional
conditional variable #1
INPUT1: Fixed probabilities or region index for the
result categories, defined for each region (sector or
county) modeled
any number
once per profile,
based on
profile's home
sector
RegionalConditional2
Generic regional
conditional variable #2
INPUT1: Fixed probabilities or region index for the
result categories, defined for each region (sector or
county) modeled
any number
once per profile,
based on
profile's home
sector
RegionalConditional3
Generic regional
conditional variable #3
INPUT1: Fixed probabilities or region index for the
result categories, defined for each region (sector or
county) modeled
any number
once per profile,
based on
profile's home
sector
RegionalConditional4
Generic regional
conditional variable #4
INPUT1: Fixed probabilities or region index for the
result categories, defined for each region (sector or
county) modeled
any number
once per profile,
based on
profile's home
sector
RegionalConditional5
Generic regional
conditional variable #5
INPUT1: Fixed probabilities or region index for the
result categories, defined for each region (sector or
county) modeled
any number
once per profile,
based on
profile's home
sector
AQConditionall
Generic air quality
conditional variable #1
INPUT1: Fixed probabilities for the result categories,
defined for each air quality bin
any number
once per time
step based on
current location
85
-------
Conditional Variable
Purpose
Input Variables
Number of Categories
Function
Reevaluated
AQConditional2
Generic air quality
conditional variable #2
INPUT1: Fixed probabilities for the result categories,
defined for each air quality bin
any number
once per time
step based on
current location
AQConditional3
Generic air quality
conditional variable #3
INPUT1: Fixed probabilities for the result categories,
defined for each air quality bin
any number
once per time
step based on
current location
AQConditional4
Generic air quality
conditional variable #4
INPUT1: Fixed probabilities for the result categories,
defined for each air quality bin
any number
once per time
step based on
current location
AQConditional5
Generic air quality
conditional variable #5
INPUT1: Fixed probabilities for the result categories,
defined for each air quality bin
any number
once per time
step based on
current location
86
-------
AvgTempCat
! Temperature ranges (categories) in Fahrenheit
INPUT1 INTRANGE 3 "AvgTemp"
50 78
RESULT INTEGER 3 "TempCatA"
12 3
#
WindowRes
! Home windows open(l) or closed
INPUT1 INTVALUE 2 "AC_Home"
1 2
INPUT2 INTRANGE 3 "MaxTemp"
56 80
INPUT4 CONDITIONAL 12
0.2 0.8
0.2 0.8
0.5 0.5
0.7 0.3
0.1 0.9
0.9 0.1
RESULT INTEGER 2
1 2
#
DailyConditional3
! DailyConditional3 - Penetration values for vehicle micro
INPUT1 PROBABILITY 4
0.2 0.5 0.2 0.1
RESULT INTEGER 4
12 3 4
#
RegionalConditionall
! Has attached garage
BY Sector 14
INPUT1 PROBABILITY 2
01017953800
0
05
0. 95
01017953900
0
05
0. 95
01017954000
0
05
0. 95
01017954200
0
05
0. 95
01017954300
0
05
0. 95
01017954400
0
05
0. 95
01017954500
0
05
0. 95
13013180101
0
8
0.2
13013180102
0
8
0.2
13013180201
0
8
0.2
13013180202
0
8
0.2
13013180300
0
8
0.2
13013180400
0
8
0.2
13013180500
0
8
0.2
RESULT INTEGER 2
1 2
J
Exhibit 4-21. Examples of Profile Functions
Note: There may be spaces between the word, "INPUT," and the following number.
The third example is a definition for a user-defined conditional variable DailyConditional3. In
this case, the user wants four categories of a variable (penetration) for a given microenvironment
87
-------
and assigns each category a probability of being selected on a given day. All user-defined
conditional variables are designated in an analogous manner. User-defined conditional variables
require a Probability input. Note that the probabilities for the four categories in the example sum
to one. The resulting category number is saved to the profile on each day and can be used to
determine the microenvironment parameters (Section 4.24.2).
The final example is a function definition for a regional conditional variable,
RegionalConditionall. In this example, it is being used to describe the differences in housing
conditions (i.e., presence of an attached garage) in different sectors of the study area. The BY
statement indicates how the prevalence vary regionally, either by county or by sector (in this
case, sector). This line must additionally contain the number of regions (i.e., counties or sectors)
that will be used (in this case, 14). After this BY line, the probability input for each sector or
county is listed. APEX matches these regions to the appropriate study area sector (or sectors, in
the case of a county), and uses them when assigning the value of RegionalConditionall through
RegionalConditional5 to each profile. The "By county" option matches the first 5 characters of
the sector name. An APEX warning will result if a listed region does not match up with any
study area sector, and APEX will give an error message and stop if there exists a study area
sector for which there is no corresponding region. The result of this function is that profiles in
each sector will be assigned an attached garage (the RESULT, 1 = yes, 2 = no) based on their
sector's listed probabilities.
The DiaryPools mapping is required. A typical example is the following:
DiaryPools
! Group activity diaries into pools
TABLE
INPUT1 INTRANGE 3 "MaxTemp"
4 5 6
12 3
12 3
12 3
12 3
12 3
4 5 6
#
In the above example, there are three bins for MaxTemp, one for A vgTemp, and seven for
DayOfWeek. The 21 results could all be placed on one line, but this layout is clearer. The first
row is Sunday, for low, medium, and high MaxTemp. If there were instead (say) two AvgTemp
bins, then there would be six numbers for Sunday, the first three for the low AvgTemp bin and
the last three for the second bin. Although there are 21 "results", there are only six different
diary pools. Later on, these pools will be subset by gender and age before choosing diaries for
particular individuals.
The minimum number of categories for all the variables defined in the Profile Functions file is
one, in which case all profiles will have the same value for the variable. In the case of all
functions EXCEPT DairyPools, having one category is the default case and can be implemented
55 84
INPUT2 INTRANGE 1
INPUT3 INTINDEX 7
RESULT INTEGER 21
'AvgTemp"
'DayOfWeek"
'Pool number
88
-------
by simply omitting the function definition from the Profile Functions file. DiaryPools, however,
is required to be defined in the file. Therefore, if one wishes to define only a single diary pool,
this must be done explicitly by setting all the RESULT values for the function equal to one. For
example:
DiaryPools
! Group all activity diaries into one pool
TABLE
INPUT1 INTRANGE 1 "MaxTemp"
INPUT2 INTRANGE 1 "AvgTemp"
INPUT3 INTINDEX 7 "DayOfWeek"
RESULT INTEGER 7 "Pool number"
1111111
There is no explicit upper limit on the number of categories, and in practice, it is only limited by
what is convenient. See Section 6.1 in Volume //for more information on diary pools.
4.18 Microenvironment Mapping File
Thq Microenvironment Mapping file provides the mapping of the Location Codes (e.g., for
CHAD) to Microenvironments defined in APEX. The current CHAD location codes are given
in Table 4-9, and an example portion of a Microenvironment Mapping file is provided in Exhibit
4-22. This file only allows comment lines and keyword input lines, except for the first two
header lines. Each keyword input line begins with a location code followed by a short
description; an "=," an integer that designates a microenvironment defined in the
Microenvironment Descriptions file; and a character variable that assigns the location code as
belonging to a "Home," "Work," "Other," "Road," "Road Work," "Near Work," "Near Home,"
"Last," or "Unknown" location (the codes are H/W/O/R/RW/NW/NH/L/U, respectively).
CHAPTER 8 in Volume II explains how APEX calculates air concentrations. The location
determines the choice of ambient air data, and the microenvironment determines the rules for
modifying that ambient data (see Section 4.24).
"Road" specifies that air concentrations are drawn from the roadway district assigned to the
home location, while "Road Work" draws air concentrations from roadways near the work
location. If multiple pollutants are used, not all of them are required to use roadway
concentrations. This option can be set for each pollutant. However, because APEX requires all
pollutants to use the same district information, AQ data are required in all districts for all
pollutants. If the user chooses to not use these data, all "Road" designations will then be
replaced by "Other."
The "Near Home" and "Near Work" locations are randomly sampled within a certain radius of
the home and work tract. These can be used to simulate movements outside the standard home
and work locations. These tracts can either be sampled at each time step, or once for each person
by using the RESAMPLEN keyword in the Control Options file. By default, APEX does not
resample. The NEARBYRADIUS keyword controls how large of a radius is used when
sampling nearby tracts. By default, this value is set to 20 km. If the user selects the "Last"
location (L), then APEX will use either the NH or NW location, based on the location last
visited. Since the Last location depends on the events of the individual, this location is treated
differently than the others, and is not included in the MicroenvironmentalResults file. In the
89
-------
events file, locations mapped to the Last location are designated by "99." If the simulation
includes people who leave the study area, all near work locations will use the ambient
concentrations of the work location.
The supplied file contains microenvironment assignments for the CHAD location codes. The
user must assign each location code to microenvironments defined in thq Microenvironment
Descriptions file by specifying the microenvironment number in the APEX Micro column. The
file must contain assignments for all CHAD location codes, or APEX will exit with a fatal error.
A zero in the APEX Micro column will result in no exposure in that particular CHAD
microenvironment location. A value of -1 means that APEX will use whichever
microenvironment was previously in use in the composite diary time series for an individual
(typically used for CHAD locations "U" and "X").
The DiaryCluster file has the same layout as the Microenvironment Mapping file, except it maps
to the axes used by the clustering method for longitudinal diary assembly. This file is required
only if the clustering algorithm is being used (i.e., if ClusterDiary = Y). This file is used for
creating the transitional probabilities of individuals moving throughout different
microenvironments. The microenvironments defined in the DiaryCluster file need not be the
same as in the Microenvironment Mapping file. The method currently uses up to five axes for
clustering; that is, the diary times spent in those five "cluster axes" locate each diary as a point in
a five dimensional space, and then clusters of diaries are defined, one per axis. For purposes of
defining air concentrations, an APEX run that uses this diary clustering file may use a
Microenvironment Mapping file with more (or fewer) than five microenvironments.
Table 4-9. CHAD Location Codes
Loc.
Code Description
Loc.
Code Description
X No data
U Uncertain of correct code
30000 Residence- general
30010 Your residence
30020 Other residence
30100 Residence-indoor
30120 Your residence-indoor
30121 ... kitchen
30122 ... living room or family room
30123 ... dining room
30124 ...bathroom
30125 ...bedroom
30126 ... study or office
30127 ...basement
30128 ... utility or laundry room
30129 ... other indoor
30130 Other residence-indoor
30131 ...kitchen
30132 ... living room or family room
30133 ... dining room
30134 ...bathroom
30135 ...bedroom
31210 Walk
31230 In stroller or carried by adult
31300 Waiting for travel
31310 ... bus or train stop
31320 ...indoors
31900 Travel-other
31910 ... other vehicle
32000 Non-residence indoor- general
32100 Office building/ bank/ post office
32200 Industrial/ factory/ warehouse
32300 Grocery store/ convenience store
32400 Shopping mall/ non-grocery store
32500 Bar/ night club/ bowling alley
32510 Bar or night club
32520 Bowling alley
32600 Repair shop
32610 Auto repair shop/ gas station
32620 Other repair shop
32700 Indoor gym /health club
32800 Childcare facility
32810 ... house
32820 ... commercial
90
-------
Loc.
Code
30136
30137
30138
30139
30200
30210
30211
30219
30220
30221
30229
30300
30310
30320
30330
30331
30332
30340
30341
30342
30400
31000
31100
31110
31120
31121
31122
31130
31140
31150
31160
31170
31171
31172
31200
Description
Loc.
Code
Description
... study or office
32900
Large public building
... basement
32910
Auditorium/ arena/ concert hall
... utility or laundry room
32920
Library/ courtroom/ museum/ theater
... other indoor
33100
Laundromat
Residence- outdoor
33200
Hospital/ medical care facility
Your residence- outdoor
33300
Barber/ hair dresser/ beauty parlor
... pool or spa
33400
Indoors- moving among locations
... other outdoor
33500
School
Other residence- outdoor
33600
Restaurant
... pool or spa
33700
Church
... other outdoor
33800
Hotel/ motel
Residential garage or carport
33900
Dry cleaners
... indoor
34100
Indoor parking garage
... outdoor
34200
Laboratory
Your garage or carport
34300
Indoor- none of the above
... indoor
35000
Non-residence outdoor- general
... outdoor
35100
Sidewalk- street
Other residential garage or carport
35110
Within 10 yards of street
... indoor
35200
Outdoor public parking lot /garage
... outdoor
35210
... public garage
Residence- none of the above
35220
... parking lot
Travel- general
35300
Service station/ gas station
Motorized travel
35400
Construction site
Car
35500
Amusement park
Truck
35600
Playground
Truck (pickup truck or van)
35610
... school grounds
Truck (not pickup truck or van)
35620
... public or park
Motorcycle or moped
35700
Stadium or amphitheater
Bus
35800
Park/ golf course
Train or subway
35810
Park
Airplane
35820
Golf course
Boat
35900
Pool/ river/ lake
Boat- motorized
36100
Outdoor restaurant/ picnic
Boat- other
36200
Farm
Non-motorized travel
36300
Outdoor- none of the above
91
-------
! APEX Microenvironment Mapping File
! Mapping of CHAD activity locations to APEX
!CHAD Loc. Description
micr©environments
Micro #
Location
Notes
U
Uncertain of correct code
=
-1
U
Unknown
X
No data
=
-1
U
Unknown
30000
Residence, general
=
1
H
Home
30010
Your residence
=
1
H
Home
30020
Other residence
=
1
H
Home
30100
Residence, indoor
=
1
H
Home
30120
Your residence, indoor
=
1
H
Home
30121
..., kitchen
=
1
H
Home
30122
..., living room or family room
=
1
H
Home
Exhibit 4-22. First Part of an Example Microenvironment Mapping File
4.19 Diary Questionnaire (DiaryQuest) File
The Diary Questionnaire file provides the personal information component of each 24-hour
activity diary (Exhibit 4-23). Each record contains values for the variables listed below.
CHADID (9 characters)
Weekday (day of week: MON, TUE, ..., SUN, Missing (X))
Sex (Male (M), Female (F), Missing (X))
Race (White (W), Black (B), Asian (A), Hispanic (H), Other (O), not available (X))
Employed (Yes (Y), No (N), Missing (X))
Age (integer years)
MaxTemp (maximum hourly temperature for this diary day; degrees F)
AvgTemp (daily mean temperature for this diary day; degrees F)
Occupation (code; see Table 4-10)
QCMiss (missing time: the total number of minutes associated with events in the Diary
Events file for which the activity and/or location codes are missing for this diary day)
RecCount (record count: the number of records in the Diary Events file corresponding to
this diary day)
Commute (commuting time; only required if commuting is modeled: the total time in
minutes spent commuting on this diary day)
The user should not change this input file unless the CHAD database has changed or other
activity data are to be used instead. In the latter case, the input file format restrictions must be
met, the CHAD coding conventions must be used, and the other CHAD files must be modified to
be consistent with this file. Note that this file has one record per CHADID, whereas the CHAD
Diary Events file has RecCount of records per CHADID. The Commute column is only
required when commuting is modeled. If a diary is missing the average or maximum
temperature, weekday, or is missing time greater or equal to three hours, then the diary will be
dropped.
92
-------
! CHAD Questionnaire
File
! based on the
110116
CHAD
master files
! Created
July 26, 2017 by JEL
! chadid, weekday,sex
, race
employed
age
maxtemp
avgtemp
occupation
qcmiss
reccount
commute
BAL97 001A,TUE,
F,
W,
N,
77,
43,
34,
X,
45,
29,
0
BAL97 00 IB,WED,
F,
W,
N,
77,
51,
41,
X,
135,
28,
0
BAL97 001C,THU,
F,
W,
N,
77,
57,
47,
X,
15,
30,
0
BAL97 001D,FRI,
F,
W,
N,
77,
45,
36,
X,
0,
28,
0
BAL97 0 0IE,TUE,
F,
W,
N,
77,
47,
39,
X,
0,
27,
0
BAL97 00 IF,WED,
F,
W,
N,
77,
36,
29,
X,
0,
28,
0
BAL97 001G,THU,
F,
W,
N,
77,
38,
29,
X,
0,
26,
0
BAL97 0 01H,FRI,
F,
W,
N,
77,
43,
36,
X,
0,
28,
0
BAL97 0 011,TUE,
F,
W,
N,
77,
41,
39,
X,
15,
28,
0
BAL97 001J,WED,
F,
W,
N,
77,
54,
44,
X,
15,
28,
0
BAL97 001K,THU,
F,
W,
N,
77,
48,
40,
X,
0,
30,
0
BAL97 001L,FRI,
F,
W,
N,
77,
42,
36,
X,
30,
30,
0
BAL97 0 0 6A,WED,
M,
W,
N,
80,
51,
41,
X,
0,
31,
0
BAL97 0 0 6B,THU,
M,
W,
N,
80,
57,
47,
X,
60,
36,
0
BAL97006C,FRI,
M,
W,
N,
80,
45,
36,
X,
75,
31,
0
Exhibit 4-23. First Part of the Diary Questionnaire File
Note: In Exhibit 4-23, spaces (not tabs) may be used as delimiters.
Table 4-10. CHAD Occupation Codes
Code
Description
ADMIN
Executive, administrative, and managerial
PROF
Professional
TECH
Technicians
SALE
Sales
ADMSUP
Administrative support
HSHLD
Private household
PROTECT
Protective services
SERV
Service
FARM
Farming, forestry, and fishing
PREC
Precision production, craft, and repair
MACH
Machine operators, assemblers, and inspectors
TRANS
Transportation and material moving
LABOR
Handling, equipment cleaners, helpers, and laborers
X
Missing
4.20 Diary Events File
The Diary Events file provides descriptions of events in each day for all the diary days in the
CHAD database. Events may last from one minute to one hour in duration. Each record
includes the variables listed below.
CHADID
StartTime (the time the event began; HHMM, with 0000 representing midnight)
Duration (the duration of the event, in minutes)
Act (activity code; see Table 4-5)
Loc (location code; see Table 4-9)
The diary events file should be generated from the CHAD database at the same time as the Diary
Questionnaire file to ensure that the CHAD IDs are in the same order. Each diary day begins
and ends at midnight, and there should be exactly 24 hours of data per diary. See Exhibit 4-24
93
-------
for an example of a portion of this file. See the previous section on the Diary Questionnaire file
if user-supplied data are to be provided.
! CHAD Events File
! based on the
110116 CHAD master files
! Created
July
26, 2017
by JEL
! chadid, starttime,
duration,
act, loc
BAL97 0 01A, 0000,
60,
14500,30125
BAL97 0 01A,0100,
60,
14500,30125
BAL97 0 01A, 0200,
60,
14500,30125
BAL97 0 01A, 0300,
60,
14500,30125
BAL97 0 01A, 0400,
60,
14500,30125
BAL97 0 01A, 0500,
60,
14500,30125
BAL97 0 01A, 0600,
60,
14500,30125
BAL97 0 01A, 0700,
30,
14500,30125
BAL97 0 01A, 0730,
30,
14400,30121
BAL97 0 01A, 0800,
60,
16000,30122
BAL97 0 01A, 0900,
60,
14500,30125
BAL97 0 01A, 1000,
30,
14500,30125
BAL97 0 01A, 1030,
30,
X, X
BAL97 0 01A,1100,
45,
14500,30125
BAL97 0 01A,1145,
15,
X, X
BAL97 0 01A, 1200,
60,
14500,30125
BAL97 0 01A, 1300,
60,
14500,30125
Exhibit 4-24. First Part of the Diary Events File
Note: In Exhibit 4-24, spaces (not tabs) may be used as delimiters.
4.21 Diary Statistics File
The Diary Statistics file contains a diary statistic for each diary in the CHAD database. This file
is used in constructing multi-day (longitudinal) diaries in APEX from the CHAD one-day diaries
using the D&A method. Refer to Volume II for information on how to construct this file.
APEX has three options for assembling simulation-length diaries. The first method is to
randomly pick a new day-long diary from CHAD for each day in the simulation. For the second
option, APEX has a longitudinal diary assembly algorithm for selecting diaries based on some
key statistic of each CHAD diary. This algorithm requires the selection of a diary based on some
key diary statistic relevant to the pollutant being studied. For example, the statistic may be time
spent outdoors or time spent in a vehicle. The third method creates longitudinal diaries via
transitional probabilities calculated for clusters derived from the CHAD inputs files. Details of
both longitudinal diary algorithms are provided in Volume II.
The Diary Statistics file must contain the CHAD ID for each diary and the value of this statistic
(ID and statistic separated by a comma or a space, one diary per row). The order of the CHAD
IDs in this file must be the same as on the Diary Questionnaire file, or an error will result.
Two Diary Statistics files have been generated from CHAD and are included starting in the
APEX Version 4 release. These files are for time spent outdoors and time spent in vehicles. The
files were constructed by summing up the time spent in locations considered "outdoors" or "in
vehicle" in each CHAD diary. Table 4-11 gives the CHAD location codes that were used to
94
-------
generate these files. Users may construct other files from the CHAD database. An example
portion of a diary statistic file is shown in Exhibit 4-25.
The use of the second longitudinal algorithm is invoked by setting the Control Options file
keyword LONGITDIARY = YES. If LONGITDIARY = NO, the Diary Statistics file is not
needed, and need not be specified in the Control Options file.
Table 4-11. CHAD Locations used in Constructing the Outdoor Time and Vehicle Time
CHAD Location IDs Considered "Outdoors"
CHAD Location IDs Considered
"In Vehicle"
30332, 30342, 30320, 30200, 31310, 35000-36300
31000-31172
CHAD Longitudinal Activity Statistics File for Time Outdoors
Prepared by EPA for EPA
July 26, 2017 by JEL
ent outdoors (minutes)
! Created
! CHAD ID,
time
BAL97 0 01A,
0
BAL97001B,
165
BAL97001C,
0
BAL97001D,
0
BAL97001E,
0
BAL97001F,
0
BAL97001G,
0
BAL97 001H,
0
BAL 97 0 011,
0
BAL97001J,
0
BAL 9 7 0 01K,
0
BAL 9 7 0 01L,
0
BAL97006A,
75
BAL97006B,
270
BAL97006C,
90
BAL97006D,
60
BAL97006E,
30
BAL97006G,
135
BAL97006H,
135
BAL 97 0 0 61,
60
BAL 9 7 0 0 6 J,
90
Exhibit 4-25. First Part of the Diary Statistics File for Time Spent Outdoors
4.22 Diary Occupations File
The Diary Occupations file contains a diary occupation for each diary in the CHAD database.
This file can be used at any time, but is most useful when defining an occupation-related Profile
Factor. Using this file will overwrite all occupations found in the Diary Questionnaire file. If
occupation is a factor, and if an occupation is found in a diary that is not used in the Profile
Factors file, then that diary occupation will be set to missing (X).
The Diary Occupations file must contain the CHAD ID for each diary and the name of the
occupation (ID and occupation separated by a comma or a space, one diary per row). The order
95
-------
of the CHAD IDs in this file must be the same as on the Diary Questionnaire file. An example
portion of a Diary Occupations file is shown in Exhibit 4-26.
! CHAD Optional Occupation file
! Created 11/10/2011
BAL97 0 01A,SALE
BAL97001B,ADMIN
BAL97001C,TRANS
BAL97001D,FARM
BAL97 0 0IE, TECH
BAL97001F, PROTECT
BAL97001G, PROF
BAL97001H, OTHER
BAL97001I,LABOR
BAL97001J,OTHER
BAL97001K,LABOR
BAL97001L,ADMIN
BAL97006A,TRANS
Exhibit 4-26. First Part of a Diary Occupations File
4.23 Diary Transitions File
This file is used only with the CLUSTERDIARY = YES option. It has the same records as the
Diary Questionnaire file, one for each CHAD ID. The fields are: CHADID, Study, Base, Year,
Month, Day, Wkday, DayNum, and Sasdate. The most important data relate each CHADID to a
Base and a SASDate. The Base is unique to each distinct person, so multiple diaries from one
person share the same Base. The SASDate increases by one each time a day passes, which
makes it easy to decide which diaries from the same person are on consecutive days.
4.24 Microenvironment Descriptions File
Thq Microenvironment Descriptions input file serves two purposes. First, it lists the names and
numbers assigned to each microenvironment, and it defines the methods by which pollutant
concentrations are calculated. Second, this file tells APEX how to define the parameters that are
required to calculate these concentrations. These are called "microenvironment parameters,"
abbreviated as "MP." With the exceptions of air exchange rate and microenvironment volume,
the MPs are specific to each combination of pollutant and microenvironment. The layout of the
Microenvironment Descriptions file typically makes a clear distinction between the two sections
using headers lines which start with !, which are therefore not processed by APEX. An example
of the first section is shown in Exhibit 4-27, while an example of a "Parameter Description"
section is shown in Exhibit 4-28. The examples shown in these figures will be discussed in
detail below.
Important: make sure that this input file (like the others in APEX) does not contain tabs. Be
sure to use spaces instead.
APEX is a microenvironmental human exposure model, and the definitions and properties of the
microenvironments are crucial to an APEX run. The user may spend more time preparing this
input file than on any other file.
96
-------
4.24.1 Microenvironment Descriptions Section
In the "Microenvironment Descriptions" section of the Microenvironment Descriptions file, the
user specifies a Micro (microenvironment number), a Name, and a calculation Method for each
microenvironment, as shown in Exhibit 4-27. The Micro cannot exceed the number of
microenvironments specified in the Control Options file, nor can it exceed 127. It also has to
correspond with each of the microenvironment numbers in thq Microenvironment Mapping file.
A microenvironment Name may be a word up to 40 characters (without spaces). The calculation
Method could be either MassBal or Factors. In the MassBal method, the concentration in a
microenvironment is calculated using a mass balance approach, while in the Factors method the
microenvironment concentration is assumed to be a linear function of ambient concentration.
See Volume II for further description of the MassBal and Factors methods.
Micro
Name
Method
1
Indoor-Residence
MASSBAL
2
Indoor-Other
MASSBAL
3
Outdoor
FACTORS
4
Near-road
FACTORS
5
Vehicle
FACTORS
Exhibit 4-27. Example of a Microenvironment Descriptions Section of a Microenvironment
Descriptions File
4.24.2 Parameter Descriptions Section
The "Parameter Descriptions" section of thq Microenvironment Descriptions file consists of the
specification of probability distributions for the MP that are required for calculating pollutant
concentrations in the microenvironments. See Volume II for further information on the MassBal
and Factors concentration calculation methods. Three types of MP can be defined for the
Factors method and eight for the MassBal method. In each method, some of the MP may be left
at default values, and do not be explicitly defined if the defaults are acceptable to the user. The
parameters and their default values (if present) are given in Table 4-12. Air exchange rate and
volume are not pollutant-specific, so they are only defined once. Otherwise, there must be one
definition for each MP for each pollutant for each microenvironment, with the exception of the
two pollutant source types, concentration (CSource) and emissions (ESource), which permit
multiple sources in the same microenvironment.
Table 4-12.
Microenvironment Parameters for the Factors and Massbal Methods
Calculation
Method
Parameter Type
Parameter
Code
Parameter
Units
Parameter Default Value
Factors
Proximity
PR
None
1
Penetration
PE
None
1
Csource
CS
ppm, ppb, or
l-ig/m3 (same as
INPUTUNITS)
0
MassBal
Proximity
PR
None
1
Penetration
PE
None
1
Decay Rate
DE
1/hr
0
97
-------
Calculation
Method
Parameter Type
Parameter
Code
Parameter
Units
Parameter Default Value
Air Exchange
Rate
AE
1/hr
none
Volume
VO
m3
None (only needed if ESource is
present)
MeanR
MR
1/hr
AE+DE
Csource
CS
ppm, ppb, or
l-ig/m3 (same as
INPUTUNITS)
0
ESource
ES
M-g/hr
0
As mentioned above, not all of the parameters must be explicitly defined for each
microenvironment. If the default values in Table 4-12 are acceptable for a microenvironment,
then a given parameter definition may be omitted from the input file. For Factors, default values
exist for all the parameters. If no parameters are defined for microenvironments using the
Factors method, then the microenvironment concentration is always equal to the current ambient
concentration. For a MassBal microenvironment, the air exchange rate parameter must always
be defined as it has no default value. The volume parameter does not have a default either, but it
is only used if ESource terms exist for that microenvironment and may be omitted otherwise.
All other parameters are optional.
The proximity and penetration factors are used to model the ambient pollutant concentrations
immediately outside and inside a microenvironment. The air exchange rate and volume variables
define the air flow rate in and out of the microenvironment as well as the microenvironment air
volume. The decay rate defines the rate of removal of pollutant from the microenvironment via
various means. The parameter MeanR is a factor that describes the removal of pollutant by both
air flow and decay. The CSource and Esource terms are concentration and emission pollutant
sources, respectively. See Volume II for a detailed description of these parameters and the
microenvironmental concentration equations.
As part of the estimation of microenvironment concentrations, each MP for each pollutant is
given a value for each hour of the simulation, for each profile generated. This value may or may
not be different from the values at other hours, depending on how it is defined. Some MPs, such
as house volume, should logically remain constant throughout the simulation, while others may
change seasonally, daily, or hourly. Values may recur in patterns, such as the same set of 24
hourly values for some parameter may recur each Saturday during the winter season. These
patterns are determined from the four mapping options and the three resampling options
specified in each microenvironmental parameter definition.
The definitions for the microenvironment parameters may appear in any order in the
Microenvironment Descriptions file. Therefore, the user (for example) may choose to group
definitions by microenvironment or by pollutant. Each definition should be separated from the
next either by blank lines or by comment lines (starting with an exclamation point) to aid in
98
-------
clarity. A parameter description consists of keywords and distribution definitions, described in
the following sections.
Keywords
The first part of a microenvironment parameter description is a list of settings, each described by
a keyword. The first keywordfor a given MP should be MP#. The various keywords serve a
number of purposes, including specifying:
Which microenvironment is being considered;
Which pollutant is being considered (not needed for air exchange rate or volume);
Which parameter is being defined for that microenvironment (the parameter code);
The source number for the current parameter (if it is ESource or CSource);
How that parameter varies over hours in the day, days of the week, or months of the year;
Whether the parameter depends on any conditional variables;
Whether the parameter is correlated with any other parameter (by being sampled using the
same random numbers);
Which group the MP belongs to (for Sobol analysis); and
Whether or not a new value of parameter is generated for each hour, for each day, and for
the workplace
The keywords and their descriptions are provided in Table 4-13.
The conditional variable keywords must be either one of the conditional variables listed in Table
4-8; i.e., TempCat, HumidCat, WindCat, DirCat, PrecipCat, MuxTempCat, AvgTempCat,
HasGasStove, HasGasPilot, AC Home, ACCar, Window Res, WindowCar, SpeedCat,
DailyConditionall-DailyConditional3, ProfileConditionall-ProfileConditional5,
RegionalConditionall-RegionalConditional5, A QConditionall-A QConditional5, Gender,
Employed, FactorGroup, Or PopCat. All variables, with the exception of the last four, must be
defined in the Profile Functions file in order to be used as a conditional variable in a
microenvironmental parameter description. PopCat is the "population category," or gender/race
combination (e.g., "white males" is a population category). Therefore, Gender and PopCat
should not both be used as conditional variables for the same MP.
In APEX, the user has the option of correlating the random samples for microenvironmental
parameters. Such correlation would make sense, for example, when the value of the parameter is
assumed to be mainly a function of the properties of a simulated individual's home and the
pollutants have similar properties (for example, if the pollutants are all particles). In addition, in
some cases it may be that the same parameters may be correlated in different
microenvironments. APEX uses a simple method of correlating MPsby sampling them using
the same random numbers. This results in values being selected for correlated parameters at the
same percentile from the appropriate distributions. The percentiles will correspond each hour as
long as the 2 (or more) parameters use the same conditional variables, time and area mappings,
and resampling rates and thus have the same number of required distributions and samples.
Otherwise, the samples get out of phase and any correlation is lost. APEX checks that the
conditionals, mappings, and resampling are the same when correlating parameters, and writes a
99
-------
warning if they are not. APEX will still run, but the user should be aware that the correlation is
lost.
Correlation is handled by an optional keyword in the MP definition, CorrNum. Each subset of
MPs that the user desires to be correlated (sampled at the same percentile each hour) are assigned
a unique integer 1-N, where N is the total number of correlated subsets.
All of the keywords for the MP come at the beginning of the microenvironmental parameter
definition. Except for the Conditional and Resamp variables, APEX examines only the first 5
characters of each keyword to decide what it is; therefore, the user may extend them any way
they like. After the definition of all the keywords, the next line should be the header line for the
data section; that is, the section that contains the actual distribution definitions for the MP. The
header line must begin with the word Block, as APEX recognizes this word as indicating the end
of the keyword section (see Exhibit 4-28 for an example of an appropriate header).
Table 4-13. Keyword Definitions for the Parameter Descriptions Section of the
Microenvironment Descriptions File
Keyword
Abbrev.
Description
MM
MP#
This number is used for tracking this particular MP. Every MP must be
assigned a distinct number, but they do not have to be consecutive.
Microenvironment
Number
Micro
These numbers must match the microenvironment numbers in the
"Microenvironment Descriptions" section.
Pollutant
Pollu
Integer corresponding to the pollutant being considered. (Number
corresponds to the order of the pollutant definition in the Control Options
file). The default is 1, so this is not needed in a single-pollutant APEX run.
Not used for AER and Volume definitions (ignored if defined).
Parameter Code
Param
A parameter code such as PR (proximity) or PE (penetration) provided in
Table 4-12, used to specify the parameter type.
Correlation
Number
Corrn
Integer number corresponding to correlation subset. Each subset of MPs that
the user desires be correlated (sampled at the same percentile each hour) are
assigned a unique integer 1-N, where N is the total number of correlated
subsets. If using this option with Sobol analysis, be certain that the correlated
MPs are assigned to the same Sobol group.
Source Number
Sourc
Numbers multiple sources in the same microenvironment. Not needed if there
is only one source present (or none at all).
Hours - Block
Hours
This variable is used to map hours of a day to different time blocks. A "time
block" is a group of hours for which the same MP distribution(s) will be used.
The input line always contains a list of 24 integers, representing 24 hours a
day. The first hour is midnight to 1 a.m. and the 24th is 11 p.m. to midnight.
The position of an integer in the input line represents the hour in a day. The
integer represents the number of a time block that an hour belongs to. The
hours in a time block do not need to be consecutive, nor does a time block
have to have the same number of hours. If this line is missing, the default
value is that all 24 hours are in a single time block-block #1.
Weekday - Daytype
Weekd
This variable is used to map days in a week to different day types. A "day
type" is a set of days for which the same MP distribution(s) will be used.
Seven integers must be given in this input line. The position of an integer in
the input line represents a day, beginning on Sunday and ending on Saturday.
The integer represents the day type a day belongs to. If this variable is not
defined, all days of a week will belong to day type #1.
100
-------
Keyword
Abbrev.
Description
Month - Season
Month
This variable is used to map months of a year to different seasons. A
"season" is a set of months for which the same MP distribution(s) will be
used. Twelve integers must be given in this input line. The position of an
integer represents a month of a year, beginning in January and ending in
December. The integer represents the season that a month belongs to. If this
line is missing, all 12 months belong to season #1.
District - Area
Distr
This variable is used to map air districts to larger areas. The number of
integers in this line must match the number of air districts in the study area.
This variable is a holdover from APEX2 and should not be used unless really
necessary. The user could delete this line or place the same number of 1 in
this line as the number of air districts.
Condition # 1
Condi #1
Choice for the first conditional variable. A conditional variable is a variable
whose value affects the choice of MP distribution(s). If not used, this line
may either be omitted or the value set to zero.
Condition # 2
Condi #2
Choice for the second conditional variable.
Condition # 3
Condi #3
Choice for the third conditional variable.
ResampHours
Resamph
Either "YES" or "NO". If "YES", a random value is selected from
distribution for a parameter in each hour within a time block. If "NO", a
random value is selected for a parameter for a time block and used for every
hour within the time block. The default value is "NO".
ResampDays
Resampd
Either "YES" or "NO". If "YES", a random value is selected from a
distribution for a parameter for each day within a day type. If "NO", a
random value is selected for a day type and used for every day within the
same day type. The default is "NO".
ResampWork
Resampw
Either "YES" or "NO". If "YES", a separate set of random values is selected
from a distribution for the workplace. If "NO", the same set of random values
are used (for the same day and hour) both for home and at work. The default
is "NO".
ResampTime Step
Resampt
Either "YES" or "NO". If "YES", a new random value is generated each
time step, if longer than one hour. For timesteps equal to or less than one
hour, Resampt is equivalent to Resamph. The default is "NO".
Sobol Group
Sobol
Group number of this MP for Sobol analysis. MPs in the same group are still
sampled independently; this does NOT correlate the inputs. Grouping the
inputs reduces the number of passes through the code needed to obtain
sensitivity indices. Two indices (main and total) are generated for each
group. No default exists. This only needs to be specified if the Sobol method
is being used.
The first four keywords (and lines) for each MP on the Misdescriptions input file should be:
1) MP# =
2) Micro =
3) Pollu =
4) Param =
For a run with just one pollutant, the third line may be dropped. Otherwise, it is crucial that
these lines appear before the "Param" line because the data for that parameter are stored in arrays
based on the most recently read values for MP#, micro, and pollutant. Fortran reads the input
file one line at a time and remembers the most recent entries.
If sources are used, ensure that the #sources parameter is set in the pollutant-specific section of
the Control Options file. This value should be the maximum source number used with anymicro
101
-------
for this pollutant. The same source number (e.g. 1) can be re-used for different micros and/or
different pollutants. Reporting the #sources on the Control Options file allows Fortran to allocate
memory for the source data.
Distribution Definitions
The last part of an MP definition lists the probability distributions for the MP at different times
or under different circumstances during the simulation. Sets of distribution data may exist for all
possible combinations of the user-specified cases of the seven indexing variables defined below.
Block, time block (as described by Hours - Block)
DayType. day type (as described by Weekday - DayType)
Season , season of the year (as described by Month - Season)
Area, air quality area (as described by District - Area)
CI: conditional variable # 1
C2: conditional variable # 2
C3: conditional variable # 3
The above labels are listed in the header line for the data section (which starts with 'Block').
Each subsequent line lists seven indices (which reference combinations of the above), followed
by a distribution. Each possible combination of indices requires one line. For most MPs, most
or even all of the indices have just a single value (which is 1).
The number of cases for the indexing variables Block, DayType, and Season are specified by
mappings in the keyword section, and is 1 by default. For example, the number of time blocks
would be the highest integer indicated in the Time - Block mapping. For the conditional
variables, MaxTempCat. A vgTempCat, HasGasStove, HasGasPilot, AC Home, AC Car.
WindowRes, WindowCar, SpeedCat, DailyConditionall-DailyConditional3,
ProfileConditionall-ProfileConditional5, RegionalConditionall-RegionalConditional5, or
AQConditionall-AQConditional5, the number of cases is determined by the number of Results
indicated on the Profile Functions file (Section 4.17). If fewer than 3 conditional variables are
used for an MP, the ones not used are given an index of 1. For Gender, there are always 2 cases.
For PopCat, the number of cases is indicated by the number of population groups (population
files) defined on the Control Options file (Section 4.2), and the groups are indexed in the order
they appear in the file (for example, if the population file for white females happened to be
defined first in the Control Options file, then that group would correspond to the case PopCat =
1). For FactorGroup, the number is determined by the number of groups in the corresponding
Profile Factors file. If the groups are tied to employment, then an additional group is defined to
be the unemployed. For example, if your Profile Factors group defines probabilities for 8
occupations, 9 cases will need to be defined here; the additional one will be applied to
unemployed people.
The user specifies the MP distribution using the standard APEX distribution format (a
distribution shape, followed by 4 distribution parameters, upper and lower truncation bounds,
and a resampling flag). The 4 parameters used are dependent on the shape of the distribution.
See Volume //for a complete discussion of the use of probability distributions in APEX. Thus
the following data must be present in each specification.
102
-------
Shape. This variable gives the type of the distribution
Pari: Parameter 1 of the MP distribution. Depends on type.
Par2. Parameter 2 of the MP distribution. Depends on type.
Par3 . Parameter 3 of the MP distribution. Depends on type.
Par4. Parameter 4 of the MP distribution. Depends on type.
LTrunc: Lower truncation point of the distribution
UTrunc: Upper truncation point of the distribution
ResampOut: Distribution resampling flag
See Table 4-6 for the available distribution types and required parameters. The parameters that
are not used for specifying a distribution should be marked with a period (".") as a place holder.
Examples of Parameter Descriptions
Two examples of parameter descriptions are shown in Exhibit 4-28. These examples should
provide the user with a good idea of how the keywords and distribution definitions work.
In the first example, the MP# is 1. The air exchange rate (code AE) is defined for
microenvironment #1. The pollutant number is not used with air exchange rates, so this keyword
is missing. This MP is assigned to Sobol group 51. In this case, the parameter distribution is
only a function of two conditional variables, AvgTempCat, and /I CHome The parameter is not
resampled from the distribution every hour (ResampHours = NO) nor each day (ResampDays =
NO), although the parameter is resampled if the simulated person moves between home and
work (ResampWork = YES). Consequently, the conditional variable AvgTempCat has five
possible values (1-5) and AC Home has two possible values (1-2); these variables and their
values were defined in the Profile Functions file. Thus, after the line starting with 'Block' there
must be 5x2 =10 lines, one for each combination of the two conditional variables. The ten
distributions are lognormal in shape (although they have different parameters), and are listed in
orderfirst looping over the values of AvgTempCat and then A CHome
103
-------
MP#
= 1
Micro
number
= 1
Parameter Type
= AE
Condition #
1
= AvgTempCat
Condition #
2
= ACHome
ResampHours
= NO
ResampDays
= NO
ResampWork
= YES
Sobol
group
= 51
Block
DType
Season
Area
CI
C2
C3
Shape
Pari
Par2
Par3
Par4
LTrunc UTrunc
ResampOut
1
1
1
1
1
1
1
Lognormal
0 . 95
1.7
0
0
111 10.0
Y
1
1
1
1
2
1
1
Lognormal
0 . 65
1.7
0
0
111 10.0
Y
1
1
1
1
3
1
1
Lognormal
0 . 35
1.7
0
0
111 10.0
Y
1
1
1
1
4
1
1
Lognormal
0 . 33
1. 9
0
0
111 10.0
Y
1
1
1
1
5
1
1
Lognormal
0 . 33
1. 9
0
0
111 10.0
Y
1
1
1
1
1
2
1
Lognormal
0 . 50
2 . 0
0
0
111 10.0
Y
1
1
1
1
2
2
1
Lognormal
0 . 50
2 . 0
0
0
111 10.0
Y
1
1
1
1
3
2
1
Lognormal
0 . 60
2 . 0
0
0
111 10.0
Y
1
1
1
1
4
2
1
Lognormal
0 .80
2 . 0
0
0
111 10.0
Y
1
1
1
1
5
2
1
Lognormal
1.00
2 . 0
0
0
111 10.0
Y
MP#
= 25
Micro
number
= 12
Pollutant
= 3
Parameter Type
= PE
Hours
- Block
= 11
1 1
1 1
1 2
2 2 2 2 2 2
2 2 2 2
111
1 1
Weekday-DayType
= 12
2 2
2 2
1
Month-
-Season
= 11
2 2
2 3
3 3
4 4 4 1
Sobol
group
= 51
Block
DType
Season
Area
CI
C2
C3
Shape
Pari
Par2
Par3
Par4
LTrunc UTrunc
ResampOut
1
1
1
1
1
1
1
Point
1.0
2
1
1
1
1
1
1
Point
0 . 5
1
2
1
1
1
1
1
Point
0 . 9
2
2
1
1
1
1
1
Point
0 . 4
1
1
2
1
1
1
1
Point
0 . 8
2
1
2
1
1
1
1
Point
0 . 3
1
2
2
1
1
1
1
Point
1.0
2
2
2
1
1
1
1
Point
0 . 9
1
1
3
1
1
1
1
Point
0 . 8
2
1
3
1
1
1
1
Point
0 . 7
1
2
3
1
1
1
1
Point
0 . 6
2
2
3
1
1
1
1
Point
0 . 5
1
1
4
1
1
1
1
Point
0 . 5
2
1
4
1
1
1
1
Point
0 . 3
1
2
4
1
1
1
1
Point
0 . 2
2
2
4
1
1
1
1
Point
0 . 1
Exhibit 4-28. Example of a Parameter Descriptions Section of a Microenvironment
Descriptions File
In the second example, penetration factor (PE) is defined for Microenvironment #12. Here, the
distributions are not a function of any conditional variable, but rather different time blocks, day
types, and seasons. Distributions for PE must be defined for all possible combinations of these
time variables. The Hour - Block keyword line indicates a mapping of the hours of the day into
two different time blocks (1 and 2) roughly defining night and day. Thus a different parameter
distribution for PE will be used for these two time blocks. Similarly, the Weekday - DayType
mapping keyword line defines two different day types, "1" for Saturday and Sunday, and "2" for
the rest of the days of the week. Finally, the Month - Season mapping keyword line defines four
seasons, labeled 1-4, corresponding to winter, spring, summer, and autumn. The distributions
follow (again looping first over block, then day type, then season), and in this example the
parameter is defined as a single point value in all cases. Hence 2x2x4 = 16 distributions are
needed.
104
-------
It is clear that these methods allow the user a great deal of flexibility in defining different
distributions for the microenvironmental parameters. In most cases, many of the features of
these descriptions will not be used, but in some cases the user may wish to define a large number
of distributions for a single parameter. There is no limit in APEX on the number of distributions
that can be defined for a microenvironment parameter.
Control of the resampling is distinct from the number of distributions. APEX produces random
values in two steps. First, a uniform random number between zero and one is produced for every
hour of the simulation (separately for each person and each variable). The same uniform sample
may be used for multiple hours, depending on the resampling options. If all resampling is "NO"
(the default), then the same random sample applies to all hours of the simulation, for the given
person. The distributions determine how these samples are transformed. The uniform value
indicates the percentile of the CDF of the final distribution that is to be assigned. When several
distributions use the same uniform value, the same percentile is drawn from each. The value for
an MP may change whenever the underlying uniform sample changes, or the distribution
changes, or both.
4.25 Prevalence File
The Prevalence file is an optional APEX input file for modeling a subpopulation of persons with
a particular disease or condition. The Prevalence file is only required when the setting
DISEASE is set in the Control Options file. APEX uses the prevalence rates to assign a
YES/NO value to a physiological profile variable, ILL, and to produce output exposure summary
tables for persons with ILL = YES. If DISEASE is not set, then the Prevalence file is not
required and no summary tables for ill persons will be printed.
The Prevalence file must contain prevalence rates (probabilities) for all age and gender cohorts
from ages 0-99. The Gender, MinAge, and MaxAge lines define the bins. Each line of the
prevalence file contains a sector ID, followed by the values for each of the bins. The values in
the Prevalence file may be separated by one or more spaces. A portion of an example
Prevalence file is shown in Exhibit 4-29.
! asthma prevalence
1
Gender=
M
M
M
F
F
F
MinAge=
0
20
60
0
20
60
MaxAge=
19
59
99
19
59
99
06071001110
0.09744
0.22000
0. 12258
0. 13636
0.18462
0.
14524
06071001210
0.05056
0. 12857
0. 09230
0. 15200
0.24553
0.
19099
06071001220
0.14921
0.29310
0. 13249
0.11818
0.11339
0.
05856
06071001300
0.17143
0.18889
0.21798
0.16503
0.06861
0.
14479
Exhibit 4-29. First Part of an Example Prevalence File
105
-------
CHAPTER 5. APEX OUTPUT FILES
APEX can produce the output files listed below.
Log
Hourly
Timestep
Daily
Profile Summary (Persons)
Microenvironmental Summary
Microenvironmental Results
Output Tables
Sites
Events
MultiPollutant
Diary Clustering
Sobol Results
All of these output files are ASCII files which can be opened and reviewed using a text editor or
other software (e.g., spreadsheet, database, statistical analysis, or graphics). A brief summary of
these files is given in Table 5-1. Details of each file are provided in Sections 5.1 to 5.13 below.
If the clustering method of longitudinal diary assembly is used, then APEX may record an
additional output file, ChadClust.dat. This file is created for APEX internal use only and is not
intended as output for the user (see Volume //for details).
All output files (except for ChadClust.dat, which is a binary file) contain the same set of header
records, allowing files generated from the same run to be identified, and for audit trail
requirements. This header section consists of six lines followed by a blank line, as indicated
below.
Line 1: Type of output file
Line 2: APEX version, date and time of start of run
Line 3: Location description (from Control Options file)
Line 4: Scenario description (from Control Options file)
Line 5: Echoes first line of Control Options file
Line 6: List of the Pollutants (as given in Control Options file)
Next N lines: Echo the first line of the each of the Air Quality Data files for the N
pollutants in the simulation. If the output file is pollutant-specific, then only the line from
its corresponding Air Quality Data file is echoed.
The Location, Pollutant, and Scenario descriptions echo what the user provided for those
keywords in the Control Options file. In the first line of the Control Options file the user
typically gives general identifying information for the simulation. Similarly, the first lines of the
Air Quality Data files can identify the contents of the files.
106
-------
Table 5-1. APEX Output Files
Output File
Description
Log
Contains the record of the APEX model simulation as it progresses. If the simulation
completes successfully, the log file indicates the input files and parameter settings used
for the simulation and reports on a number of different factors. If the simulation ends
prematurely, the log file contains error messages describing the critical errors that
caused the simulation to end.
Hourly
Provides an hour-by-hour time series of exposures, doses, and other variables for each
modeled profile.
Daily
Provides a day-by-day time series of exposures, doses, and other variables for each
modeled profile.
Profile Summary
Provides a summary of each profile modeled in the simulation. Each line lists the
person's age, gender and race, in addition to a number of other personal profile
variables that the model uses to simulate exposure.
Microenvironmental
Summary
Provides a summary of the time and exposure by microenvironment for each profile
modeled in the simulation.
Microenvironmental
Results
Provides an hour-by-hour time series of microenvironment concentrations and
parameters for a pollutant for each modeled profile for each location (e.g., "Home,"
"Work," and "Other"). A Microenvironmental Results file is generated for each
pollutant.
Output Tables
Contains a series of tables summarizing the exposure (and dose, if calculated) results of
the simulation for a pollutant. The percentiles and exposure/dose cut-off points used in
these tables are defined in the Control Options file. A Tables file is generated for each
pollutant.
Sites
Lists the sectors, air districts, and zones in the study area, and identifies the mapping
between them.
Events
Contains event-level information (including MET, exposure, ventilation, and dose) for
individuals in the simulation. Settings in the Control Options file allow the user to
write this information for all persons, every Nth person, or for a set of specified profile
IDs.
Timestep
Has the same format as the Hourly file, except that it reports variables on every
timestep.
Multipollutant
Contains one row for each combination of micro, clock hour (0-23), and level for each
pollutant.
Sobol
Is produced only if the user requests a Sobol sensitivity analysis run. In that case, most
of the other output files are suppressed. The output consists of main and total indices
for both the average day and maximum day, for selected exposure metrics.
5.1 Log File
The Log file records the information listed below as a model run progresses.
Input files used
Settings for job control options
Number of diaries in total, and the number in each diary pool
Model execution time
Sectors in the study area
Air districts in the study area
Meteorology zones in the study area
Mappings of sectors to air districts and meteorology zones
Diary activities and counts of MET distributions
Descriptions of the microenvironment-specific parameters
107
-------
Statistical summaries of the simulated profiles
Output summary tables
If a model run stops abnormally, an error message will be written to the Log file. The user
should review the Log file after a model run to ensure that the simulation executed and
terminated normally and that the output results are valid. Output summary tables in this file (if
the LOGTABLES option is set to YES) are exactly the same as the tables in the Output Table
file. The level of detail of the information written to the Log file is controlled by the Control
Options file setting, DEBUGLEVEL. DEBUGLEVEL can have a value of 0, 1, 2, or 3; the
higher the level, the more information is written to the log. The Control Options file settings,
LOGDISTRICT, LOGPOPULATION, LOGPROFILES, LOG SECTORS, LOGTABLES, and
LOGZONES also control the writing of information to the Log file. See Table 4-4 for more
information on these settings.
If there are 40 or fewer profiles in the run, the person-level demographic variables are written to
the Log file for each person, as are the number of days that each person equals or exceeds the
exposure cut-points for the DM1HEXP (daily maximum 1-hour) table.
5.2 Hourly File
The Hourly file contains hourly time series of a number of APEX variables including
concentrations and doses for each simulated person or profile. Note: if the APEX timestep is
greater than 1 hour (TLMESTEPSPERDAY< 24), the Hourly file will not be written. In this
case, the Timestep file (see next section) provides the best summary of the exposure and dose
time series. The user can control which variables are written to the Hourly file via a list of
keywords using the Control Options file keyword HOURLYLLST. The variables and their
corresponding keywords are provided in Table 5-2.
Table 5-2. APEX Variables Written to the Hourly Output File
Variable
Description
Units
Control Options
File Keyword
Optional
Person
Simulated profile number
-
-
N
Hour
Hour # of the simulation
-
-
N
Ve
Ventilation
ml/min
VE
Y
Va
Alveolar ventilation
ml/min
VA
Y
EVR
Equivalent ventilation rate, VE
divided by body surface area
L/(min-m2)
EVR
Y
MET
Metabolic equivalent. Time-
averaged multiple of basal
energy expenditure for the hour.
MET
Y
EE
Energy expenditure
kcal/min
EE
Y
FEVEl term
Ozone-dependent variability
term in the %AFEV1 model
%2
FEVE1
Y
FEVE2 term
Non-ozone dependent variability
in %AFEV1
%2
FEVE2
Y
Hourly Max %AFEV1
Hourly maximum of the event-
level %AFEV1 values
%
DFEV1
Y
Micro Time
Time spent in microenvironment
N
min
TIME1 - TIMEN
Y
108
-------
Variable
Description
Units
Control Options
File Keyword
Optional
Micro Exposure
Exposure in microenvironment
N
OUTPUTUNITS
EXP1 -EXPN
Y
Ambient Concentration
Ambient pollutant
concentration, time averaged
over events
INPUTUNITS
AMB
Y
Ambient Concentration
(Home)
Ambient pollutant concentration
in the home district, time
averaged over events
INPUTUNITS
HOMEAMB
Y
Exposure
Time-averaged exposure for the
hour
OUTPUTUNITS
EXP
Y
Dose
Time-averaged dose for the
hour. Units of dose depend on
pollutant, see Volume II.
DOSE
Y
Intake Dose
PM pollutants only. Average
mass inhaled per minute
(includes mass not deposited)
during the hour
micrograms
/minute
(Hg/min)
INTAKEDOSE
Y
Deposited Dose
PM pollutants only. Total mass
deposited in the respiratory
system during the hour.
micrograms
(fig)
DEPDOSE
Y
Exposure Factor
The ratio of the hourly exposure
to the hourly ambient
concentration
EF
Y
Exposure Factor
Home)
The ratio of the hourly exposure
to the hourly ambient
concentration in the home
district
HOMEEF
Y
See Volume II for a description of the APEX ventilation algorithms and further information on
VE, VA, EVR, and EE. The variables VE, VA, EVR, MET, and EE, and the variables for
exposure, dose, and ambient concentration, are the time-weighted averages of the event values
for these variables. The ambient concentration is time-averaged over the events because the
simulated individual may move between home/work/other locations (and thus possibly between
air districts) in the course of an hour. Thus, the hourly ambient concentration may not be equal to
the home district AQ data for that hour.
The hourly exposure in microenvironment N is the portion of the total exposure for the hour
occurring in microenvironment N, equal to:
ZConcN * Duration
60
where ConcN is the concentration in microenvironment N for the event and Duration is the event
duration in minutes. A weighted average is used because it is possible for concentrations in a
given microenvironment to vary as the person moves between home/work/other locations during
the hour. ExpN is the total hourly exposure. The hourly exposure factor, EF, is simply the ratio
of the hourly exposure to the hourly ambient air concentration.
109
-------
The variables may be listed in any order in the Control Options file using the keyword
HOURLYLIST; however, they are printed in the output file in the order they appear in the above
table. The list can be on a single line or multiple lines, and may be comma or space-delimited.
The EXP, DOSE, EXPN, AMB, and EF keywords control the writing of that variable for all
pollutants in the simulation; the file headers for these variables will contain the pollutant name.
The dose variables will not be written for a pollutant if it has DODOSE = NO in the Control
Options file, even if a dose keyword is included in the HOURLYLIST.
An example regarding the use of the HOURLYLIST keyword would be:
HourlyList = EVR VE VA EE MET FEVE1 FEVE2 DFEV
HourlyList = AMB HOMEAMB EXP EF HOMEEF DOSE INTAKEDOSE DEPDOSE
An illustration of a portion of the resulting Hourly file for an example two-pollutant run (Poll
and Pol2) is shown in Exhibit 5-1.
APEX Hourly File
APEX Version 4.0 (dated February 19, 2007) Run Date = 20070321 Time = 101001
Location = Description of Location of the Study Area
Scenario = APEX Sensitivity Simulation
Simulation = ! APEX Sensitivity Simulation
Pollutant = Poll Pol2
Air Quality = ! Hourly Poll air quality data for an example metropolitan area
Air Quality = ! Hourly Pol2 air quality data for an example metropolitan area
p
Hour
Time
4 Time 12
Amb-Poll
Exp-Poll
EF-Pol1
Amb-Pol2
Exp-Pol2
EF-Pol2
1
1
60
0
1. 039E-02
5. 842E-03
0 .562
1.132E-02
6.351E-03
0 .561
1
2
60
0
9.000E-03
3.295E-03
0.366
9.000E-03
3.351E-03
0 . 372
1
3
60
0
7.000E-03
2.710E-03
0 .387
7.000E-03
2.756E-03
0.394
1
4
60
0
2 . 000E-03
9.189E-04
0 .459
2.000E-03
9.37 9E-04
0.469
1
5
25
0
7 . 000E-03
2.297E-03
0 . 328
7.000E-03
2.333E-03
0 . 333
1
6
26
0
2.300E-02
8.118E-03
0 . 353
2.300E-02
8.241E-03
0 . 358
1
7
60
0
2.100E-02
7.262E-03
0.346
2.100E-02
7.392E-03
0 . 352
1
8
60
0
1.800E-02
6.180E-03
0 .343
1.800E-02
6.291E-03
0.349
1
9
30
0
1. 800E-02
7.378E-03
0.410
1.800E-02
7.491E-03
0.416
1
10
12
30
1.900E-02
6.683E-03
0 . 352
1.900E-02
6.7 99E-03
0 . 358
1
11
60
0
2.026E-02
8.227E-03
0.406
2.044E-02
8.427E-03
0 . 412
Exhibit 5-1. First Part of an Example APEX Hourly Output File
Note that the hourly file could be very large if a large number of profiles are simulated. The
hourly file is only written if the HOURLYOUT parameter is to "YES" in the Control Options
file.
5.3 Timestep File
The Timestep file contains the timestep-level time series of a number of APEX variables,
including exposure and doses, for each simulated person or profile. The user can control which
variables are written to the Timestep file via a list of keywords using the Control Options file
keyword TIMESTEPLIST. The variables and their corresponding keywords are listed in Table
5-3.
110
-------
Table 5-3. APEX Variables Written to the Timeste
Variable
Description
Units
Control Options
File Keyword
Optional
Person
Simulated profile number
-
-
N
Hour
Hour # of the simulation
-
-
N
Timestep
Timestep # of the simulation
-
-
N
Ve
Ventilation
ml/min
VE
Y
Va
Alveolar ventilation
ml/min
VA
Y
EVR
Equivalent ventilation rate,
Ve, divided by body surface
area
L/(min-m2)
EVR
Y
MET
Metabolic equivalents. Time-
averaged multiple of basal
energy expenditure for the
timestep.
MET
Y
EE
Energy expenditure
kcal/min
EE
Y
Ambient Concentration
Ambient pollutant
concentration, time-averaged
over timestep
INPUTUNITS
AMB
Y
Exposure
Exposure, time-averaged over
events in the timestep
OUTPUTUNITS
EXP
Y
Dose
Time-averaged dose for the
hour. Units of dose depend on
pollutant, see Volume II.
DOSE
Y
Intake Dose
PM pollutants only. Average
mass inhalation rate (includes
mass not deposited) over
timestep
Hg/min
INTAKEDOSE
Y
Deposited Dose
PM pollutants only. Total
mass deposited in the
respiratory system during the
timestep
lig
DEPDOSE
Y
Exposure Factor
The ratio of the timestep
exposure to the timestep
ambient concentration
EF
Y
j Output File
See Volume II for a description of the APEX ventilation algorithms and further information on
VE, VA, EVR, and EE. The variables VE, VA, EVR, MET, and EE, and the variables for
exposure, dose, and ambient concentration, are the time-weighted averages of the event values
for these variables. The ambient concentration is time-averaged over the events because the
simulated individual may move between home/work/other locations (and thus possibly between
air districts) in the course of a timestep. Thus, the timestep ambient concentration may not be
equal to the home district AQ data for that timestep.
The variables may be listed in any order in the Control Options file using the keyword
TIMESTEPLISTbut they are printed in the output file in the order they appear in the above
table. The list can be on single or multiple lines and may be comma or space-delimited. The
EXP, DOSE, A MB, and EE keywords control the writing of that variable for all pollutants in the
simulation; the file headers for these variables will contain the pollutant name. However, the
dose variables will not be written for a pollutant if it has DODOSE = NO in the Control Options
file, even if a dose keyword is included in the TIMESTEPLIST.
Ill
-------
An example of the use of the TIMESTEPLIST keyword would be:
TimestepList = VE AMB EXP
An example of a portion of the resulting Timestep file for a one-pollutant run (ozone) is depicted
in Exhibit 5-2.
APEX Timestep File
APEX Version 4.0 (dated February 21, 2008) Run Date = 20080227 Time = 111351
Location = Description of Location of the Study Area
Scenario = APEX Sensitivity Simulation
Simulation = ! APEX Sensitivity Simulation
Pollutant = ozone
Air Quality = Name =0000200006
P Hour Timestep Ve Amb-ozone Exp-ozone
111 4858. 3.7 60E-03 3.760E-03
112 5951. 1.027E-02 1.027E-02
113 4156. 3.570E-03 3.570E-03
114 4949. 8.480E-03 8.480E-03
115 5060. 3.680E-03 3.680E-03
Exhibit 5-2. First Part of an Example APEX Timestep Output File
Note that the timestep file could be very large if a large number of profiles are simulated or if the
APEX timestep is very small. The timestep file is only written if the TIMESTEPOUT
parameter is to "YES" in the Control Options file. Also note that if the APEX timestep is equal
to the default (1 hour, or TIMESTEPSPERDA F=24), then the Timestep file in general would
contain the same information as the Hourly file, and thus in this case it is not written.
5.4 Daily File
The Daily Exposure file contains a daily time series of a large number of APEX variables for
each simulated person or profile. Writing of the file is controlled by the Control Options file
variable DAILYOUT. The user can control which variables are written to the file via a list of
keywords using the Control Options file keyword DAILYLIST. The variables and their
corresponding keywords are listed in Table 5-4.
"able 5-4. APEX Variables Written to the Daily Output File
Variable
Description
Units
Control Options
File Keyword
Optional
Person
Simulated profile number
-
-
N
Day
Day number of the simulation
-
-
N
Diary ID
ID of CHAD diary selected for the
current day for the profile
-
CHADID
Y
Diary Age
Age associated with the selected
CHAD diary (may be different from
the age of the simulated profile)
years
CHADAGE
Y
Diary
Employment
Employment status associated with
the selected CHAD diary
-
CHADEMP
Y
Diary
Occupation
The occupation of the CHAD diary
selected for that day.
-
CHADOCC
Y
112
-------
Variable
Description
Units
Control Options
File Keyword
Optional
Diary pool
Index of the APEX diary pool for
the current day (as determined by
profile functions file)
DIARYPOOL
Y
PAI
Physical activity index, the time-
averaged MET over the day for the
simulated person
PAI
Y
Key Diary
Variable
Daily value of the key diary variable
(statistic) used for the D&A
longitudinal diary assembly for the
simulated day for the profile (such
as time spent outdoor or in vehicles)
KEYVAR
Y
WindowRes
Conditional variable value indicating
whether residence windows are open
or closed (as determined by profile
functions file)
WINDOWRES
Y
WindowCar
Conditional variable value indicating
whether car windows are open or
closed (as determined by profile
functions file)
WINDOWCAR
Y
SpeedCat
Conditional variable value indicating
the speed at which a vehicle is
traveling (as determined by profile
functions file)
SPEEDCAT
Y
DailyCondl
Value of daily conditional variable 1
(as determined by profile functions
file)
DCOND1
Y
DailyCond2
Value of daily conditional variable 2
(as determined by profile functions
file)
DCOND2
Y
DailyCond3
Value of daily conditional variable 3
(as determined by profile functions
file)
DCOND3
Y
MaxTempCat
Conditional variable giving the
category for the maximum
temperature for the day (as
determined by profile functions file)
MAXTEMPCAT
Y
AvgTempCat
Conditional variable giving the
category for the average temperature
for the day (as determined by profile
functions file)
AVGTEMPCAT
Y
Maximum
Maximum hourly temperature for
Fahrenheit
MAXTEMP
Y
Temperature
the current day
Average
Temperature
Average of the hourly temperatures
for the current day
Fahrenheit
AVGTEMP
Y
Daily Max
%AFEV1
Daily maximum of the event-level
%AFEV1 calculations
%
DFEV1
Y
Average
Exposure
Time-averaged pollutant exposure
for the day.
OUTPUTUNITS
AVGEXP
Y
Max 1 Hour
Maximum 1-hour exposure on the
OUTPUTUNITS
MAX1EXP
Y
Exposure
given day; each hourly exposure
time-averaged over events.
Max 8 Hour
Maximum 8-hour exposure on the
OUTPUTUNITS
MAX8EXP
Y
Exposure
given day; each 8-hour exposure
time-averaged over events.
113
-------
Variable
Description
Units
Control Options
File Keyword
Optional
Max 8 Hour
Exposure Factor
(Home)
The ratio of the maximum 8-hour
exposure to the corresponding
average ambient concentration in the
home microenvironment
HOME8MAXEF
Y
Average Dose
Time-averaged pollutant dose for the
day. Units of dose depend on
pollutant, see Volume II.
AVGDOSE
Y
Intake Dose
PM pollutants only. Average mass
inhaled per minute (includes mass
not deposited) during the day
Hg/min
INTAKEDOSE
Y
Deposited Dose
PM pollutants only. Total mass
deposited in the respiratory system
during the day
M^g
DEPDOSE
Y
Max 1 Hour
Dose
Maximum 1-hour dose on the given
day; each hourly dose time-averaged
over events
MAX1DOSE
Y
Max 8 hour
Dose
Maximum 8-hour dose on the given
day; each 8-hour dose time-averaged
over events.
MAX8DOSE
Y
Max End-of-
Hour Dose
Maximum dose as calculated at the
end of each hour of the day.
MAX1FDOSE
Y
See Volume II for further information on the diary selection variables and conditional variables
on this list. The exposure and dose keywords will control printing for all pollutants in the
simulation; the file headers for these variables will contain the pollutant name.
Note that the Daily file could be very large if a large number of profiles or pollutants are
simulated. The daily file is only written if the DAILYOUT parameter is "YES" in the Control
Options file.
The keywords may be separated by either spaces or commas. An example of a DA1LYL1ST
would be:
DailyList = CHADID CHADAGE CHADEMP CHADOCC DIARYPOOL PAI KEYVAR WINDOWRES WINDOWCAR AVGEXP
An example portion of a Daily file created with the DAILYLIST example above for a theoretical
two-pollutant run (Poll and Pol2) is shown in Exhibit 5-3. Note that in the daily file, the values
may not fall directly under the corresponding label in the file header (in order to minimize file
size).
114
-------
APEX Daily File
APEX Version 4.0 (dated February 19, 2007) Run Date = 20070321 Time = 101001
Location = Description of Location of the Study Area
Scenario = APEX Sensitivity Simulation
Simulation = ! APEX Sensitivity Simulation
Pollutant = Poll Pol2
Air Quality = ! Hourly Poll air quality data for an example metropolitan area
Air Quality = ! Hourly Pol2 air quality data for an example metropolitan area
p
Day
CHADID
CHADAge
CHADEmp
DiaryPool
PAI
KeyVar
WindowRes
WindowCar
AvgExp-Poll
AvgExp-Pol2
1
1
NHW19167A
24
Works
2
2.20
540.00
0
0
7 . 833E-03
1.436E-02
1
2
CIN02759A
21
Works
2
2.41
49.00
0
0
7.74 8E-03
5 .4 5 6E-02
1
3
NHW10859A
20
Works
5
2 . 60
490.00
0
0
9.537E-03
7 .7 64E-02
1
4
NHA16047A
27
Works
5
1. 95
0.00
0
0
8.256E-03
8.37 9E-02
1
5
NHW13255A
24
Works
1
2.43
525.00
0
0
4.34 3E-03
5.7 4 7E-02
1
6
NHW15968A
21
Works
1
1.58
475.00
0
0
6 . 938E-03
7 . 345E-02
1
7
NHW12055A
20
Works
1
1.82
600.00
0
0
6 .196E-03
3 . 253E-02
1
8
WAS96832A
25
Works
1
2.24
15.00
0
0
4 . 37 2E-03
6.744E-02
1
9
DEN34716B
22
Works
1
3.63
11.00
0
0
6. 306E-03
9 .222E-02
1
10
CIN80040B
21
Works
4
2.70
390.00
0
0
5.712E-03
2 .54 3E-02
1
11
CIN00339B
24
Works
4
2.46
457.00
0
0
6. 08 9E-03
4 .334E-02
1
12
WAS6304 6A
24
Works
1
2.14
0.00
0
0
6 . 366E-03
6 . 4 35E-02
1
13
CIN61737C
26
Works
2
2 . 95
91.00
0
0
4 . 539E-03
6 .7 65E-02
1
14
CAA0 62 51A
21
Works
2
2.37
230.00
0
0
2 . 62 9E-03
6.27 9E-02
Exhibit 5-3. First Part of an Example Daily Output File
5.5 Profile Summary (Persons) File
Thq Profile Summary (Persons) file provides a summary of profile characteristics and
exposure/dose for each simulated person. Each record contains values for a number of variables
for each simulated individual. A small set of variables are written by default to the file, and
additional variables are only written if designated by the user in the Control Options file. The
variables are defined using the PSUMLIST keyword, followed by an equals (=) sign and a list of
variable-specific keywords. The available variables and their corresponding keywords are
provided in Table 5-5.
Table 5-5. APEX Variables Written to the Profile Summary File
Variable
Description
Control Options
File Keyword
Optional
Person
Sequential index number for simulated
individual
-
N
Home Sector
Sector in which the person lives (home)
-
N
Work Sector
Sector in which the person works (=home
sector for non-workers, or = 0 for those
working outside the study area)
N
Home District
Air district for the home sector
-
N
Work District
Air district for the work sector
-
N
Zone
Meteorology zone for the home sector
-
N
Age
Age of the simulated profile (years)
-
N
Gender
Male or female
-
N
Race
e.g., White, Black, Asian, Native American
(NatAm), Other (depending on pop. files)
-
N
Employment
Indicates employment outside the home
-
N
Height
Person height (inches)
-
N
Weight
Body mass (pounds)
-
N
Number of Diaries
Number of diaries used
Ul) I ARIES
Y
Group Number
Number of the Profile Factor subgroup
FGROUP
Y
Group Name
Name of the Profile Factor subgroup
GROUPNAME
Y
115
-------
Variable
Description
Control Options
File Keyword
Optional
Roadway District
Air district for the home roadway sector
ROADDIST
Y
Roadway Work District
Air district for the work roadway sector
RWDIST
Y
Commuting Distance
The distance (in km) from the home to the
work sector.
COMMDIST
Y
Commuting Time
The estimated time (in minutes) it takes to
travel from the home to work sector
COMMTIME
Y
Car AC type
Type of air conditioning in the car (depends of
Profile Functions file)
ACCAR
Y
Home AC type
Type of air conditioning in the residence
(depends of Profile Functions file)
ACHOM
Y
Disease status
Whether or not a profile is ill (depends on
Prevalence file)
DISEASE
Y
Gas Pilot
Indicates the presence of a gas pilot light in the
home (depends of Profile Functions file)
PILOT
Y
Gas Stove
Indicates the presence of a gas stove in the
home (depends of Profile Functions file)
STOVE
Y
ProJileConditionall
Value of profile conditional variable # 1 for the
person
PCOND1
Y
ProJileConditional2
Value of profile conditional variable # 2 for the
person
PCOND2
Y
ProJileConditional3
Value of profile conditional variable # 3 for the
person
PCOND3
Y
ProJileConditional4
Value of profile conditional variable # 4 for the
person
PCOND4
Y
ProfileConditional5
Value of profile conditional variable # 5 for the
person
PCOND5
Y
RegionalConditionall
Value of regional conditional variable # 1 for
the person
RCOND1
Y
RegionalConditional2
Value of regional conditional variable # 2 for
the person
RCOND2
Y
RegionalConditional3
Value of regional conditional variable # 3 for
the person
RCOND3
Y
RegionalConditional4
Value of regional conditional variable # 4 for
the person
RCOND4
Y
RegionalConditional5
Value of regional conditional variable # 5 for
the person
RCOND5
Y
Number of Events
Number of diary events covering the
simulation period for the person
EVENTS
Y
Blood Vol
The volume of blood in the body (ml)
BLOODVOL
Y
BSA
Body surface area (m2)
BSA
Y
Energy Conversion Factor
Energy conversion factor for person (L-
02/kcal)
ECF
Y
Lung Diffusivity
A lung diffusivity parameter used in the COHb
(CO dose) calculation (ml/min/torr)
DIFFUS
Y
Endogenous CO production
1
Endogenous CO production rate; only used for
calculating CO dose (ml/min)
ENDGN1
Y
Endogenous CO production
2
Endogenous CO production rate for women
between ages of 12 and 50 for half the
menstrual cycle; only used for calculating CO
dose (ml/min)
ENDGN2
Y
Hemoglobin
The amount of hemoglobin in the blood (g/ml)
HEMOGLOB
Y
MET max
Maximum obtainable MET level for the
person. (MET)
METMAX
Y
116
-------
Variable
Description
Control Options
File Keyword
Optional
Maximum Oxygen Uptake
Maximum obtainable oxygen uptake rate for
person (L-Ch/min)
V02MAX
Y
Maximum Oxygen Debt
Maximum obtainable oxygen debt for person
(ml-02/kg)
MOXD
Y
Physical Activity Index
Median of the daily PAI values (time-averaged
MET on each simulated day)
PAI
Y
Recovery Time
Time required to recover the maximum oxygen
debt (hours)
RECTIME
Y
Resting Metabolic Rate
Resting metabolic rate (kcal/min)
RMR
Y
VE Intercept
Regression parameter for the ventilation
routine
VEINTER
Y
VE Slope
Regression parameter for the ventilation
routine
VERES 11)
Y
VE Residual
Regression parameter for the ventilation
routine
VESLOPE
Y
%AFEV1 pl-p9
Model parameters for the %AFEV1 Ozone
calculations
B1-B9 (WRITE
EACH ONE
SEPARATELY)
Y
%AFEV1 Personal Variance
Model parameters for the %AFEV1 Ozone
calculations
FEVU
Y
%AFEV1 age slope
Regression parameters for the age fit for
%AFEV1 Ozone calculations
FEVSLP
Y
%AFEV1 age intercept
Regression parameters for the age fit for
%AFEV1 Ozone calculations
FEVINT
Y
BMI
Body-mass index (kg/m2)
BMI
Y
Average Exposure
Mean exposure concentration over the
simulation (ppm, ppb, or |.ig/ml as specified
by OUTPUTUNITS on Control Options file)
AVGEXP
Y
Maximum Exposure
Maximum exposure concentration on one
timestep over the simulation (ppm, ppb, or
M-g/m3, as specified in Control Options file)
MAXEXP
Y
Average Dose
Mean dose over the simulation. Units of dose
depend on pollutant, see Volume II.
AVGDOSE
Y
Maximum Dose
Maximum 1-hour dose on one timestep over
the simulation. Units of dose depend on
pollutant, see Volume II.
MAXDOSE
Y
Maximum DFEV1
Personal maximum of DFEV1 lung function
loss (for ozone only)
MAXDFEV
Y
Moderate EVR cutoff
Minimum event-level EVR to be considered as
moderate exertion
MO I) EVR
Y
Days with DFEV1>10%
Number of simulation days with more than
10% lung function loss on at least one event
(ozone runs only)
NDFEV10
Y
Days with DFEV1>15%
Number of simulation days with more than
15% lung function loss on at least one event
(ozone runs only)
NDFEV15
Y
Days with DFEV1>20%
Number of simulation days with more than
20% lung function loss on at least one event
(ozone runs only)
NDFEV20
Y
117
-------
The exposure and dose variables listed are written for all pollutants in a multiple-pollutant run.
An example portion of a Profile Summary file for a theoretical 2-pollutant (Poll and Pol2)
scenario is given in Exhibit 5-4. This file was created using the Control Options file command:
PSumList = PAI, AVGEXP, GROUPNAME
Note that each record in the file could be much longer, as many more variables could be printed.
APEX Diary Questionnaire File
APEX Version 4.0 (dated February 19, 2007) Run Date = 20070321 Time = 133813
Location = Description of Location of the Study Area
Scenario = APEX Sensitivity Simulation
Simulation = ! APEX Sensitivity Simulation
Pollutant = Poll Pol2
Air Quality = ! Hourly Poll air quality data for an example metropolitan area
Air Quality = ! Hourly Pol2 air quality data for an example metropolitan area
p
HSect
WSect
HDis
WDis
Zone
Age
Gender
Race
Empl
Height
Weight
PAI
AvgExp-Poll
AvgExp-Pol2
1
513
513
27
27
2
22
Male
White
Works
71.908
228.339
2.09
9.956E-03
1.005E-02
2
64
64
14
14
2
19
Male
Black
NoWrk
67 . 179
138.067
1.74
9.238E-03
9.413E-03
3
359
359
42
42
2
22
Female
Other
Works
61.018
173.609
1. 92
9.74 9E-03
9.415E-03
4
222
222
39
39
2
15
Male
Black
NoWrk
68.519
182.139
1.74
9.100E-03
9.131E-03
5
177
177
39
39
2
20
Female
Other
NoWrk
65.608
160.464
1. 65
8.906E-03
9.377E-03
6
287
287
49
49
2
32
Male
White
Works
65.658
155.154
2 .04
9.978E-03
1.059E-02
7
688
688
28
28
2
48
Female
Black
Works
66.264
183.261
1. 97
8.873E-03
9.257E-03
8
661
661
23
23
2
50
Female
White
Works
60.355
106.818
1. 93
8.7 65E-03
8.625E-03
9
280
280
55
55
2
39
Male
Black
NoWrk
69.081
209.165
1.76
9.120E-03
9.331E-03
10
793
793
17
17
2
32
Female
White
NoWrk
65.700
172.692
1. 87
1.020E-02
1.041E-02
Exhibit 5-4. First Part of an Example Profile Summary File
5.6 Microenvironmental Results File
The Microenvironmental Results file contains hourly values for a number of microenvironment
parameters and variables for all microenvironments, for all persons in a simulation. The file is
pollutant-specific, so one file will be created for each pollutant in the simulation. The variable
values are written for the "home," "work," "other," locations, along with the "road," "road
work," "near home," and "near work" locations if specified. There is a set of microenvironment
concentrations associated with each location for each profile. This file may be useful in
examining/testing the effects of conditional values on microenvironment concentrations.
The creation of the file for all pollutants is controlled by the Control Options file variable
MRESOUT. The files are written if MRESOUT= YES. The default is "NO", as these files are
very large, and writing them greatly affects the speed of the simulation! The printing of the
optional variables is dictated by the Control Options file keyword MRESLIST via a comma- or
space-separated list of variable keywords. The MRESLIST will control the writing of the
Microenvironmental Results file for all of the simulation pollutants. The variables that may be
written to the file and their corresponding keywords are provided in Table 5-6.
118
-------
Table 5-6. APEX Variables Written to the Microenvironmental Results File
Variable
Description
Control Options
File Keyword
Optional
Person
The number of the simulated profile
-
N
Hour #
Hour of the simulation. Hour ranges from -23 to
24 times the number of days in the simulation.
The hours -23 to 0 are included because APEX
extends the calculation of the microenvironment
concentrations to include the 24 hours prior to the
beginning of the simulation.
N
Micro #
Microenvironment number (See Section 4.18).
-
N
Location
APEX calculates concentrations for each
microenvironment including: home (1), work (2),
other (3), roadway (4), near home (5), near work
(6), road work (7), and near last (99) locations (see
Volume II). Any locations that are used are listed
in the file.
N
Proximity
Proximity factor: microenvironment parameter,
greater than or equal to 0.
PRX
Y
Penetration
Penetration factor: microenvironment parameter,
ranging from 0 to 1.
PEN
Y
CSum
Sum of concentration sources (CSource) terms
(INPUTUNITS)
CSUM
Y
Ambient
Pollutant concentration associated with the
AMB
Y
Concentration
location sector and hour as determined from the
Air Quality Data file (INPUTUNITS)
Micro Concentration
Pollutant concentration in the microenvironment
(.INPUTUNITS)
CONC
Y
ESum
Sum of emission sources (ESource) terms (M-g/hr)
ESUM
Y
Source Strength
Combined source strength for emission and
concentration sources (|ag/m3/hr)
SOURCE
Y
Micro Volume
Volume of microenvironment (m3)
VOL
Y
Air Exchange Rate
(AER)
Rate of air exchange in microenvironment (1/hr)
AER
Y
Removal (Decay)
Total removal rate of pollutant from
RR
Y
Rate
microenvironment (1/hr)
WindowRes
Conditional variable value indicating whether
residence windows are open or closed, as
determined by profile functions file
WINDOWRES
Y
WindowCar
Conditional variable value indicating whether car
windows are open or closed, as determined by
profile functions file
WINDOWCAR
Y
MaxTempCat
Daily maximum temperature category conditional
variable - will be same for all hours in a day, as
determined by profile functions file
MAXTEMPCAT
Y
AvgTempCat
Daily average temperature category conditional
variable (will be same for all hours in a day), as
determined by profile functions file
AVGTEMPCAT
Y
SpeedCat
Conditional variable value indicating the speed at
which a vehicle is traveling, as determined by
profile functions file
SPEEDCAT
Y
DailyConditionall
Value of daily conditional variable 1 for the hour
DCOND1
Y
DailyConditional2
Value of daily conditional variable 2 for the hour
DCOND2
Y
DailyConditional3
Value of daily conditional variable 3 for the hour
DCOND3
Y
TempCat
Hourly temperature category conditional variable
TEMPCAT
Y
119
-------
Variable
Description
Control Options
File Keyword
Optional
HumidCat
Hourly humidity category conditional variable
HUMIDCAT
Y
PrecipCat
Hourly precipitation category conditional variable
PRECIPCAT
Y
WindCat
Hourly wind speed category conditional variable
WINDCAT
Y
DirCat
Hourly wind direction category conditional
variable
DIRCAT
Y
Day
Day of the simulation
DAY
Y
Month
Month of the year
MONTH
Y
Air District
Air district person is in (i.e., the district
corresponding to the home/work/other location)
DISTRICT
Y
Day Week
Day of the week
DAYWEEK
Y
AQConditionall
Value of AQ conditional 1 for the time step
AQCOND1
Y
AQConditional2
Value of AQ conditional 2 for the time step
AQCOND2
Y
AQConditional3
Value of AQ conditional 3 for the time step
AQCOND3
Y
AQConditional4
Value of AQ conditional 4 for the time step
AQCOND4
Y
AQConditional5
Value of AQ conditional 5 for the time step
AQCOND5
Y
A number of the parameters in the file are undefined for a FACTORS microenvironment (see
Section 4.24.1). These parameters will be padded with 0 in that case.
An example of the use of MRESLIST in the Control Options file is:
MResList = AER, PRX, PEN, AMB, CONC, MAXTEMPCAT, AVGTEMPCAT, AQCOND1
The resulting example of aMicroenvironmentalResults file for a theoretical pollutant (Poll) is
given in Exhibit 5-5. Note that in the file the variable values may not fall directly under the
corresponding label in the file header (in order to minimize file size).
120
-------
APEX Microenvironmental Results File
APEX Version 4.0 (dated February 19, 2007) Run Date = 20070321 Time = 152320
Location = Description of Location of the Study Area
Scenario = APEX Sensitivity Simulation
Simulation = ! APEX Sensitivity Simulation
Pollutant = Poll
Air Quality = ! Hourly Poll air quality data for an example metropolitan area
Simulation Start Date = 20040101
Person
Micro
Loc
Hour
Prx
Pen
Amb
Cone
AER
WindowRes
MaxTempCat
AvgTempCat
TempCat
1
1
1
-23
1.0000
1.0000
8 . 000E-03
4 . 227E-04
0
4104
0
1
1
1
1
1
1
-22
1.0000
1.0000
9 . 000E-03
6. 518E-04
0
5042
0
1
1
1
1
1
1
-21
1.0000
1.0000
7 . 000E-03
7 . 04 4E-04
0
6962
0
1
1
1
1
1
1
-20
1.0000
1.0000
2.000E-03
2.335E-04
0
5010
0
1
1
1
1
1
1
-19
1.0000
1.0000
7.000E-03
1.64 5E-04
0
1511
0
1
1
1
1
1
1
-18
1.0000
1.0000
2.300E-02
1.712E-03
0
5822
0
1
1
1
1
1
1
-17
1.0000
1.0000
2.100E-02
1. 922E-03
0
6193
0
1
1
1
1
1
1
-16
1.0000
1.0000
1. 800E-02
1.398E-03
0
4841
0
1
1
1
1
1
1
-15
1.0000
1.0000
1. 800E-02
1.207E-03
0
4370
0
1
1
1
1
1
1
-14
1.0000
1.0000
1.900E-02
1.192E-03
0
4138
0
1
1
1
1
1
1
-13
1.0000
1.0000
2 .100E-02
1. 496E-03
0
4923
0
1
1
1
1
1
1
-12
1.0000
1.0000
2.4 00E-02
3 . 14 7E-03
1
0217
0
1
1
1
1
1
1
-11
1.0000
1.0000
2.800E-02
3.395E-03
0
8523
0
1
1
1
1
1
1
-10
1.0000
1.0000
3 . 000E-02
3.091E-03
0
6968
0
1
1
1
1
1
1
-9
1.0000
1.0000
3.200E-02
2 . 390E-03
0
4730
0
1
1
1
1
1
1
-8
1.0000
1.0000
3.000E-02
1.132E-03
0
1955
0
1
1
1
1
1
1
-7
1.0000
1.0000
2.800E-02
1.395E-03
0
3449
0
1
1
1
1
1
1
-6
1.0000
1.0000
2 . 600E-02
1. 4 87E-03
0
3753
0
1
1
1
1
1
1
-5
1.0000
1.0000
2.200E-02
1.44 0E-03
0
4299
0
1
1
1
1
1
1
-4
1.0000
1.0000
2.300E-02
1.960E-03
0
6052
0
1
1
1
1
1
1
-3
1.0000
1.0000
2 . 600E-02
2 . 610E-03
0
7176
0
1
1
1
1
1
1
-2
1.0000
1.0000
2.300E-02
1. 683E-03
0
4316
0
1
1
1
1
1
1
-1
1.0000
1.0000
2.300E-02
1.27 9E-03
0
3517
0
1
1
1
1
1
1
0
1.0000
1.0000
2 . 200E-02
1.22 6E-03
0
3640
0
1
1
1
1
1
1
1
1.0000
1.0000
8 . 000E-03
6.081E-04
0
4104
0
1
1
1
Exhibit 5-5. First Part of an Example Microenvironmental Results File
5.7 Microenvironmental Summary File
Thq Microenvironmental Summary file provides the amount of time spent, mean exposure
concentration, and maximum exposure concentration within each microenvironment during the
period of simulation, for each simulated person. The Microenvironmental Summary file is
pollutant-specific, and thus one is created for each pollutant in the simulation. After the six
header records and one blank record, there is one record labeling the columns of the subsequent
records in the file. These labels and descriptions of the values in the corresponding columns are
given in Table 5-7. The first part of an example Microenvironmental Summary file is shown in
Exhibit 5-6.
Table 5-7. Format of the APEX Microenvironmental Summary File
Column
Variable
Type
Description
1
Person
Num
Profile numbersequential index number for the simulated individual
2
Micro
Num
Microenvironment numberSequential index number for each
microenvironment (as designated in the Microenvironment Mapping
file)
3
Name
Char
Microenvironment name (as designated in the Microenvironment
Mapping file) (maximum of 40 characters)
4
Minutes
Num
Total time spent in the microenvironment by this individual (minutes)
121
-------
Column
Variable
Type
Description
5
Meanconc
Num
Average concentration during the time spent in the microenvironment
by this individual (ppm, ppb, or |.ig/ml as specified by INPUTUNITS
in the Control Options file)
6
Maxconc
Num
Maximum concentration during the time spent in the
microenvironment by this individual (ppm, ppb, or |.ig/m\ as specified
by INPUTUNITS in the Control Options file)
APEX
Microenvironmental Summary File
APEX Version
5.0 (dated July 25, 2017) Run Date = 20191028 Time = 152729
Location
Example Metropolitan
Area
Scenario
Example Metropolitan
Area, Jun-Aug 2010, 2010 pop, RandomSeed 4178348
Simulation
Example Metropolitan
Area
Pollutant
Ozone
Air
Quality
! 2010 base ozone AQ
data for Example Metropolitan Area
Person
Mic
Minutes MeanConc
MaxConc
1
0
0 0 . 000E+00
0.000E+00
1
1
70969 5.913E-03
3.262E-02
1
2
17738 3.17 3E-02
8.904E-02
1
3
25779 4.011E-02
9.600E-02
1
4
10541 3.759E-02
7.800E-02
1
5
7453 3.7 02E-02
9.600E-02
2
0
0 0.000E+00
0.000E+00
Exhibit 5-6. First Part of an Example Microenvironmental Summary File
5.8 Output Tables File
The Output Tables file can provide hundreds of summary tables, depending on the table
specifications in the Control Options file. The Tables file is pollutant-specific; thus one is
created for each pollutant in the simulation. APEX users should specify which tables to print by
using the TABLESLIST variable in the Control Options file. This parameter takes the
arguments listed below.
EXP1H, EXP8H, EXPTS, EXPAVG. prints the tables pertaining to the 1-hour, 8-hour,
time-step maxima, and average daily exposures (OUTPUTUNITS)
DOSE1H, DOSE8H, DOSE 1 EH DOSETS DOSEA VG prints the tables pertaining to
the 1-hour, 8-hour, time-step, 1-hour end-of-hour maxima, and average daily doses.
Dose units depend on the pollutant.
MICROTIME. prints the time spent in each microenvironment (minutes).
DOSETIME: prints the time spent in each dose category (minutes).
CHILDREN, prints additional tables that include only children.
ACTIVE, ACTCHILD: prints tables including active individuals, or the subset of active
children.
ILLNESS, ILLCHILD: prints tables that includes the subgroup of those with the user-
specified illness, for all simulated individuals or children only.
MOD, HEA VY: prints the tables for individuals with exposures in the moderate or heavy
EVR categories.
EMPLOYED, printed tables for all employed persons
122
-------
5.8.1 Exposure Summary Tables
APEX can write out well over 100 different exposure summary tables for each pollutant. There
are 11 different types of exposure tables, listed below.
1. Minutes in each exposure interval by microenvironment
2. Minutes at or above each exposure level by microenvironment
3. Person-days at or above each daily maximum 1-hour exposure level
4. Person-days at or above each daily maximum 8-hour exposure level
5. Person-days at or above each daily maximum timestep exposure level
6. Number of simulated persons with multiple exposures at or above each daily maximum 1-
hour exposure level
7. Number of simulated persons with multiple exposures at or above each daily maximum 8-
hour exposure level
8. Number of simulated persons with multiple exposures at or above each daily maximum
timestep exposure level.
9. Number of simulated persons with multiple exceedances (in the simulation) of the
threshold timestep exposure levels.
10. Person-days at or above each daily average exposure level
11. Number of persons at or above each overall average exposure level
Table types 1, 2, 10, and 11 are generated only once for the entire population. Table types 3 to 9
are generated for seven population subgroups, under three exertion levels. Tables may be
omitted if the subgroup contains no simulated persons.
The seven population subgroups are listed below.
1. All Persons. The table statistics are based on the entire population.
2. Children. The table statistics are based on the population of children, as defined by the age
range given by the Control Options file settings CHILDMIN and CHILDMAX.
3. Active Persons. The table statistics are based on the population of people having a median
PAI over the whole simulation period that exceeds the value designated by the Control
Options file setting ACTIVEPAI.
4. Active Children. The table statistics are based on the population of active children, as
determined by the Control Options file settings CHILDMIN, CHILDMAX, and
ACTIVEPAI
5. Ill Persons. The table statistics are based on the population of ill people. The population is
determined by the probabilities given in the Prevalence file. This population is only
considered if the input variable DISEASE is set in the Control Options file.
6. Ill Children. The table statistics are based on the population of ill people. The population
is determined by the probabilities given in the Prevalence file and the Control Options
file settings CHILDMIN and CHILDMAX. This population is only considered if the
input setting DISEASE is set in the Control Options file.
7. Employed Persons. The table statistics are based on the population of all employed
people.
The three exertion levels are listed below.
123
-------
1. All Exertion Conditions. The table statistics are based on exposures experienced by the
population subgroup under any ventilation conditions.
2. Moderate Exertion. The table statistics are based on exposures experienced by the
population subgroup only during periods in which their average EVR is in the
"moderate" range. The period of time during which EVR is averaged is either 1 hour or 8
hours, based on the table being generated. The "moderate" EVR ranges are defined by
the Control Options file settings MODEVR1 and HEA VYEVR1 (for 1-hour exposures)
and MODEVR8 and HEA VYEVR8 (for 8-hour exposures). An individual's EVR is in
the moderate range if it is greater than or equal to the MODEVR# setting and less than
the HEA VYEVR# setting for the exposure period.
3. Heavy Exertion. The table statistics are based on exposures experienced by the population
subgroup only during periods in which their average EVR is in the "heavy" range. The
period of time during which EVR is averaged is either 1 hour or 8 hours, based on the
table being generated. The "heavy" EVR ranges are defined by the Control Options file
settings HEAVYEVR1 (for 1-hour exposures) and HEAVYEVR8 (for 8-hour exposures).
An individual's EVR is in the heavy range if it is greater than or equal to the
HEAVYEVR# setting for the exposure period.
For each table that is generated, APEX prints out a label that identifies the table uniquely. For
example, a table of type #1, for all people under all exertion conditions, has the identifier TIME,
WITHIN, ALL, ALL. Users can reference these identifier labels in custom programs that read in
and process the APEX Tables file.
Exposure Table Type #1: Minutes in each exposure interval by microenvironment
This table lists the total minutes spent by all simulated persons in each microenvironment when
exposure concentration is within various ranges. The bounds of a range are specified at the top
of each column and the top of the next column to the right (Exhibit 5-7). For each
microenvironment, the table provides three rows of data for the three variables defined below.
Minutes. The number of person-minutes summed over all the simulated persons that are
spent in the specified microenvironment and that fall within the exposure concentration
range bounded by the values indicated at the top of the column and the top of the next
column to the right;
Row_%. The percent of the minutes spent in the specified microenvironment that fall
within the exposure concentration range; and
Tot_%. the percent of the total minutes that are spent in the microenvironment and that
fall within the exposure concentration range.
124
-------
TIME,WITHIN,ALL,ALL,ALL
Exposure: Minutes in each Exposure interval ( ppm ), by microenvironment, for N = 100 Profiles
Micro
Level:
0.0000
0.0100
0.0200
0.0300
0.0400
0.0500
0.0600
0
Minutes
8041.
0 .
0 .
0 .
0 .
0 .
0 .
0
Row %
100.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0
Tot~%
0.0607
0.0000
0.0000
0.0000
0.0000
0.0000
0 .0000
1
Minutes
6815807.
1525776.
397884.
137206.
55886.
20271.
4 674 .
1
Row %
76.0842
17.0321
4.4415
1.5316
0.6239
0.2263
0.0522
1
Tot~%
51.4478
11.5170
3.0034
1.0357
0.4218
0.1530
0.0353
2
Minutes
116515.
281146.
456343.
559532.
419900.
282998.
170243.
2
Row %
4.8170
11.6232
18.8662
23.1323
17.3596
11.6998
7.0382
2
Tot~%
0.8795
2.1222
3.4446
4.2235
3.1695
2.1362
1.2850
3
Minutes
30586.
57073.
108273.
158511.
177594.
162320.
118771.
3
Row %
3.2347
6.0358
11.4506
16.7636
18.7817
17.1664
12.5608
3
Tot~%
0 . 2309
0.4308
0.8173
1.1965
1.3405
1.2252
0.8965
4
Minutes
7837 .
11971.
20586.
29189.
26537.
29647.
21117.
4
Row %
4.7583
7.2683
12.4990
17.7224
16.1122
18.0005
12.8214
4
Tot~%
0.0592
0.0904
0.1554
0.2203
0.2003
0.2238
0.1594
5
Minutes
53028.
69107 .
87928.
90228.
106943.
136538.
106956.
5
Row %
7.0458
9.1822
11.6830
11.9886
14.2095
18.1417
14.2112
5
Tot~%
0.4003
0.5216
0.6637
0.6811
0.8072
1.0306
0.8073
Exhibit 5-7. Example of Exposure Table Type #1 in the Output Tables File
Exposure Table Type #2: Minutes in each exposure interval by microenvironment
This table is similar to Table Type #1, except that it reports the cumulative person-minutes that
are spent in a microenvironment with an exposure concentration that equals or exceeds the value
indicated at the top of the column.
Exposure Table Type #3: Person-days at or above each daily maximum 1-hour exposure
level
This table provides a statistical summary of the cumulative person-days, for both simulated
persons and the population in the study area, with a daily maximum 1-hour (hourly) average
exposure concentration that equals or exceeds the value indicated at the top of the column
(Exhibit 5-8). The interpretations of the variables in Table Type #3 (and other "person-days"
tables) are provided in Table 5-8.
125
-------
PERSONDAYS,DM1H,ALL,ALL,ALL
Exposure: Person-Days at or above each Daily Maximum 1-Hour Exposure Level ( ppm ), for N = 100
Profiles. Area Population = 16519458
Group: All People
Level: 0.00000 0.01000 0.02000 0.03000 0.04000 0.05000 0.06000 0.07000
Counts(Pop)
1.52 0E+0 9
1.491E+09
1.450E+09
1.
3 97E + 0 9
1.28 0E + 0 9
1.019E+09
6.533E+08
3.507E+0
#Meet (Pop)
16519458
16519458
16519458
16519458
16519458
16519458
16519458
16519458
%Meet (Pop)
100.000
100.000
100.000
100.000
100
.000
100.000
100
.000
100.000
Mean
92.000
90.280
87.780
84
.590
77 .
510
61.690
39.
550
21.230
Std.Dev.
0 .000
3.364
5.809
8 .
468
11.
053
14.730
15.
981
13.357
CV
0 .000
0 . 037
0.066
0 .
100
0 . 143
0.239
0 . 4
04
0 . 629
Minimum
92.000
70.000
62.000
55
.000
45.
000
26.000
14 .
000
3 .000
10.0 %ile
92.000
87.100
79.100
72
.000
59.
400
39.300
19.
000
8 .000
25.0 %ile
92.000
90.000
86.250
83
.000
74 .
000
53.000
29.
000
12.250
50.0 %ile
92.000
91.500
90.000
87
.000
80 .
000
62.500
38 .
000
17.500
75.0 %ile
92.000
92.000
92.000
90
.000
86.
000
72.750
49.
000
27.750
90.0 %ile
92.000
92.000
92.000
91
. 900
89.
000
81.900
65.
000
40.000
95.0 %ile
92.000
92.000
92.000
92
.000
90 .
000
85.000
12.
000
52.900
99.0 %ile
92.000
92.000
92.000
92
.000
91.
990
90.970
85.
950
67.900
Maximum
92.000
92.000
92.000
92
.000
92 .
000
91.000
86.
000
68.000
Mean (%)
100.000
98.130
95.413
91
. 946
84 .
250
67.054
42 .
989
23.076
Min (%)
100.000
76.087
67.391
59
.783
48 .
913
28.261
15.
217
3.261
Max (%)
100.000
100.000
100.000
100.000
100
.000
98.913
93 .
478
73.913
Counts (Sim)
9.200E+03
9.028E+03
8.778E+03
8 .
459E+03
7.751E+03
6.169E+03
3.955E+03
2.123E+0
#Meet (Sim)
100
100
100
100
100
100
100
100
Exhibit 5-8. Example of Exposure Table Type #3 in the Output Tables File
Table 5-8. Interpretation of the Variables in Exposure Table Type #3 and Other "Person-
Days" Based Tables
Table Entry
Interpretation
Counts (Pop)
Total number of person-days at or above the level specified at the top of each column for the
population (of the subgroup) in the study area (while at this exertion).
#Meet (Pop)
Number of persons (in the subgroup) in the study area population who have at least one
exposure at or above the level specified at the top of each column (while at this exertion).
NOTE: For exertion level tables, the 0.0 level count will not necessarily be equal to the
population of the subgroup, since some persons may have no events at the given exertion
level.
%Meet (Pop)
Percentage of people (in the subgroup) in the population who have at least one exposure at or
above the level specified at the top of each column (while at this exertion). NOTE: For
exertion level tables this may not be 100% at the 0.0 level, since some persons may have no
events at the exertion level.
Mean
Mean number of days per person (in the subgroup) during which an exposure at or above the
level specified at the top of each column is experienced (while at this exertion).
Std. Dev.
Standard deviation across persons (in the subgroup) in the number of days during which an
exposure at or above the level specified at the top of each column is experienced (while at this
exertion).
CV
Coefficient of variation across persons (in the subgroup) in the number of days during which
an exposure at or above the level specified at the top of each column is experienced (while at
this exertion).
Minimum
The lowest total number of days across persons (in the subgroup) during which an exposure at
or above the level specified at the top of each column is experienced (while at this exertion).
Percentiles
The Nth percentile of number of days across persons (in the subgroup) during which an
exposure at or above the level specified at the top of each column is experienced (while at this
exertion).
Maximum
The highest total number of days across persons (in the subgroup) during which an exposure
at or above the level specified at the top of each column is experienced (while at this exertion).
126
-------
Table Entry
Interpretation
Mean (%)
Mean number of days per person (in the subgroup) during which an exposure at or above the
level specified at the top of each column (while at this exertion) is experienced, as percentage
of possible days.
Min (%)
The lowest total number of days across persons (in the subgroup) during which an exposure at
or above the level specified at the top of each column is experienced (while at this exertion),
as percentage of possible days.
Max (%)
The highest total number of days across persons (in the subgroup) during which an exposure
at or above the level specified at the top of each column is experienced (while at this exertion),
as percentage of possible days.
Counts (Sim)
Total number of simulated person-days (in the subgroup) during which an exposure at or
above the level specified at the top of each column is experienced (while at this exertion).
NOTE: At the 0.0 level in the exertion-dependent tables, Counts (Sim) might not necessarily
be equal to #Meet (Sim)*NumDays, since some persons may have no events at the exertion
level.
#Meet (Sim)
The total number of simulated persons (in the subgroup) who experience at least one exposure
at or above the level specified at the top of each column (while at this exertion).
Exposure Table Type #4: Person-days at or above each daily maximum 8-hour exposure
level
This table provides a statistical summary of the cumulative person-days, for both simulated
persons and the population in the study area, with a daily maximum 8-hour average exposure
concentration that equals or exceeds specified levels. The table and its interpretation are the
same as Table Type #3 (Exhibit 5-8) except that the exposure metric is the daily max 8-hour
average exposure concentration.
Exposure Table Type #5: Person-days at or above each daily maximum timestep exposure
level
This table provides a statistical summary of the cumulative person-days, for both simulated
persons and the population in the study area, with a daily maximum timestep average exposure
concentration that equals or exceeds specified levels. The table and its interpretation are the
same as Table Type #3 (Exhibit 5-8) except that the exposure metric is the daily max timestep
average exposure concentration. This table is not written if the timestep is equal to one hour.
Exposure Table Type #6: Number of simulated persons with multiple exposures at or
above each daily maximum 1-hour exposure level
This table simply provides a count of the number of simulated persons who have at least 1 (2, 3,
4, 5, 6) days in the simulation during which they have experienced an exposure above each of the
daily maximum 1-hour exposure levels. An example is shown in Exhibit 5-9.
127
-------
MULTIPLE
DM1H,ALL
ALL,ALL
Exposure
Number
of
Simulated Persons with Multiple Exposures
at or above
each Daily Maximum 1-
Hour
Exposure Level
( ppm ), for
N = 100
Profiles.
Group: All People
At least
1
At least 2
At least 3
At least 4
At least 5
At least 6
Exposure
Exposures
Exposures
Exposures
Exposures
Exposures
Levei
0
000
100
100
100
100
100
100
0
010
100
100
100
100
100
100
0
020
100
100
100
100
100
100
0
030
100
100
100
100
100
100
0
040
100
100
100
100
100
100
0
050
100
100
100
100
100
100
0
060
100
100
100
100
100
100
0
070
100
100
100
97
97
95
0
080
97
95
84
75
67
61
0
090
85
6 6
50
37
30
26
0
100
54
33
19
12
7
5
0
110
38
15
2
1
0
0
0
120
4
1
0
0
0
0
0
130
1
0
0
0
0
0
0
140
0
0
0
0
0
0
Exhibit 5-9. Example of Exposure Table Type #6 in the Output Tables File
Exposure Table Type #7: Number of simulated persons with multiple exposures at or
above each daily maximum 8-hour exposure level
This table simply provides a count of the number of simulated persons who have at least 1 (2, 3,
4, 5, 6) days in the simulation during which they have experienced an exposure above each of the
daily maximum 8-hour exposure levels. The table is the same as Table Type #6 (Exhibit 5-9)
except that the exposure metric is the daily max 8-hour average exposure concentration.
Exposure Table Type #8: Number of simulated persons with multiple exposures at or
above each daily maximum timestep exposure level. This table provides a count of the
number of simulated persons who have at least 1 (or 2, 3, 6, etc.) days in the simulation during
which they have experienced an exposure above each of the daily maximum timestep exposure
levels. The table is the same as Table Type #6 (Exhibit 5-9) except that the exposure metric is
the daily max timestep average exposure concentration. This table is not written if the timestep
is equal to one hour.
Exposure Table Type #9: Number of simulated persons with multiple exposures at or
above some threshold timestep exposure level. This table provides a count of the number of
simulated persons who have at least 1 (or 2, 30, or 300, for example) timesteps in the entire
simulation during which they have experienced an exceedance of each timestep threshold
exposure level. The different number of exceedances to include in the table are listed in the
Control Options file using the keyword TSMULTILEVELS. The threshold exposures are listed
using the keyword TSEXP. This table is not written if the timestep is equal to one hour.
Exposure Table Type #10: Person-days at or above each daily average exposure level
This table provides a statistical summary of the cumulative person-days, for both simulated
persons and the population in the study area, with a daily average exposure concentration that
128
-------
equals or exceeds specified levels. The table and its interpretation are the same as Table Type #3
(Exhibit 5-8) except that the exposure metric is the daily average exposure concentration.
Exposure Table Type #11: Persons at or above each overall average exposure level
This table provides a statistical summary of cumulative numbers of both simulated persons and
people in the study area whose overall average exposure concentrations equal or exceed
specified levels. The overall average exposure concentration is the average of hourly exposure
concentrations over the whole period of simulation. An example of this table is provided in
Exhibit 5-10.
PERSONDAYS,SAVG
,ALL,ALL
Exposure: Persons at or above each Overall Average Exposure Level
(ppm) , for
N = 100
Profiles. Area
Population = 3976069
Level:
0.000 0.500
1.000
2 .000
3 .000
4 .000
5.000
Counts (Pop)
: 0.3 98E + 07 0.398E + 07
0.392E+07
0.38 6E + 0 6
0.000E+00
0.000E+00
0.000E+00
#Meet (Pop):
3976069 3976069
3916428
385679
0
0
0
%Meet (Pop):
100.000 100.000
98.500
9.700
0 .000
0 .000
0 .000
Counts (Sim)
: 0.100E+04 0.100E+04
0.985E+03
0.97 0E + 02
0.000E+00
0.000E+00
0.000E+00
# Meet (Sim)
: 1000 1000
985
97
0
0
0
Exhibit 5-10. Example of Exposure Table Type #11 in the Output Tables File
5.8.2 Dose Summary Tables
APEX can write out over 100 different dose summary tables for each pollutant. There are 10
different types of dose summary tables. The contents of each table type are described in detail
below. Table Types #7-10 are generated only once, for the entire population. Table Types #1-6
are each generated for six population subgroups, under three exertion levels. See the previous
section on exposure tables for the definition of population subgroups and exertion levels. For the
pollutant CO, dose is blood dose (%COHb), and for any PM pollutant the dose is the rate of mass
deposited in the respiratory system in |ig/min (See Volume II). For all other pollutants dose is
simply exposure*ventilation.
Dose Table Type #1 Person-days at or above each daily max end-of-hour dose level
This table provides a statistical summary of the cumulative person-days for both simulated
persons and the population in the study area, for which the daily maximum end-of-hour dose is
equal to or exceeds specified levels. The format of the table is the same as Exposure Table Type
#3 (Exhibit 5-8).
Dose Table Type #2 Person-days at or above each daily max 1-hour dose level
This table provides a statistical summary of the cumulative person-days, for both simulated
persons and the population in the study area, for which the daily maximum 1-hour average dose
is equal to or exceeds specified levels. The format of the table is the same as Exposure Table
Type #3 (Exhibit 5-8). The definitions of the variables in this table can be found in Table 5-8.
129
-------
Dose Table Type #3Person-days at or above each daily max 8-hour dose level
This table provides a statistical summary of the cumulative person-days, for both simulated
persons and the population in the study area, for which the daily maximum 8-hour average dose
is equal to or exceeds specified levels. The format of the table is the same as Exposure Table
Type #3 (Exhibit 5-8). The definitions of the variables in this table can be found in Table 5-8.
Dose Table Type #4: Person-days at or above each daily maximum timestep dose level
This table provides a statistical summary of the cumulative person-days, for both simulated
persons and the population in the study area, with a daily maximum timestep average dose that
equals or exceeds specified levels. The table and its interpretation are the same as Exposure
Table Type #3 (Exhibit 5-8), except that the exposure metric is the daily maximum timestep
average exposure concentration. This table is not written if the timestep is equal to one hour.
The definitions of the variables in this table can be found in Table 5-8.
Dose Table Type #5: Number of simulated persons with multiple timestep doses at or above
some threshold timestep dose level.
This table provides a count of the number of simulated persons who have at least 1 (or 2, 30, or
300, for example) timesteps in the entire simulation during which they have experienced an
exceedance of each timestep threshold dose level. The different number of exceedances to
include in the table are listed in the Control Options file using the keyword TSMULTILEVELS.
The threshold exposures are listed using the keyword TSDOSE. This table is not written if the
timestep is equal to one hour.
Dose Table #6 Person-days at or above each daily average dose level
This table provides a statistical summary of the cumulative person-days, for both simulated
persons and the population in the study area, for which the daily average dose is equal to or
exceeds specified levels. The format of the table is the same as Exposure Table Type #3 (Exhibit
5-8). The definitions of the variables in this table can be found in Table 5-8.
Dose Table #7 Persons at or above each overall average dose level
This table provides a statistical summary of cumulative numbers of both simulated persons and
the people in the study area whose overall average doses are equal to or exceed a specified level.
The overall average dose is the average of hourly dose levels during the entire simulation period.
Dose Table #8Person-hours at or above each end-of-hour dose level
This table provides a statistical summary of the number of person-hours, for both simulated
persons and the population in the study area, for which each end-of-hour dose level is equal to or
exceeds specified levels. The format of the table is the same as Exposure Table Type #3 (Exhibit
5-8), except that the time units are hours rather than days. The definitions of the variables in this
table can be found in Table 5-8.
130
-------
Dose Table #9
Minutes in each dose interval
This table provides a statistical summary of cumulative person-minutes, for both simulated
persons and the population in the study area, for which the dose (for example, blood %COHb
level) is within a specified range. The bounds of the dose range are specified by the levels at the
top of each column and the top of the next column to the right. The definitions of the variables
in this table are similar to those found in Table 5-8, except that the time units are in minutes
rather than days.
Dose Table #10 Minutes at or above each dose level
This table provides a statistical summary of cumulative person-minutes spent by both simulated
persons and the population in the study area, for which the dose (for example, blood %COHb
level) is equal to or exceeds specified levels. The definitions of the variables in this table are
similar to those found in Table 5-8, except that the time units are in minutes rather than days.
5.8.3 ResponseProb Summary Tables
This table type provides the number of simulated persons estimated to experience the risk
endpoint, e.g., lung function decrements over selected thresholds, at various pollutant exposure
levels. The exposure statistic that the exposure-response function is applied to is indicated by the
second item in the first line of the table header; in Exhibit 5-11 it is the daily maximum 8-hour
(DM8H) level.
ResponseProb,DM8H,ALL,MOD,ALL
FEV1 Decrement: Number of persons affected in each exposure bin, for N= 1000 profiles. Area
population = 4919330
Group: All Ages, Moderate Exertion (EVR 13.00-99.00)
Level:
0 .
0000
0
0100
0 .
0200
#Sim
Frac.Sim
#Pop
FEV1>10,
2 . 5
pctl
0 .
0000
0
0000
0 .
0000
1.510
0.1510E-02
7428 .
FEV1>10,
50 . 0
pctl
0 .
0129
0
1416
0 .
3730
3.561
0 . 3561E-02
0.17 52E + 05
FEV1>10,
97 . 5
pctl
0 .
0323
0
3234
0 .
774 6
6.548
0.654 8E-02
0.3221E+05
FEV1>15,
2 . 5
pctl
0 .
0000
0
0000
0 .
0000
0.196
0.1960E-03
964 . 4
FEV1>15,
50 . 0
pctl
0 .
0024
0
0281
0 .
0812
1.008
0.1008E-02
4 961.
FEV1>15,
97 . 5
pctl
0 .
0089
0
0931
0 .
2304
1.983
0.1983E-02
9756.
FEV1>2 0,
2 . 5
pctl
0 .
0000
0
0000
0 .
0000
0.0561
0.5612E-04
27 6. 1
FEV1>2 0,
50 . 0
pctl
0 .
536E-
04
0
853E-03
0 .
0350
0.1682
0.1682E-03
827 . 3
FEV1>2 0,
97 . 5
pctl
0 .
869E-
03
0
0105
0 .
0308
0.4756
0.4756E-03
2340 .
Bin counts
36
.00
86
.00
80
.00
314.0
0.3140
0.1545E+07
Exhibit 5-11. Portion of ResponseProb Table
Columns for higher ozone levels in Exhibit 5-11 have been removed for clarity, as indicated by
ellipses. The label before the colon in each line is user-supplied. Each value is the product of a
probability and a person count. The last row labeled "Bin counts" has all probabilities set to one
to show the number of persons in each column. The three rows for each FEV1 cut-point
correspond to different estimates of the associated probability. For example, of the 38 persons in
the first column, 0.0129 are expected to suffer a loss of over 10% in FEV1 lung function, using
the median probability. If the ozone levels went high enough, the entries would all become zero
because the person counts would be zero. That is, even if the risk factor were high, if no one
131
-------
experiences that particular ozone level, then no one can have lung function loss due to ozone at
that level.
The "#Sim" column is the sum of the previous columns, representing the number of affected
persons in the simulation. In the above example there were 314 persons in the moderate exertion
category, of whom 3.561 would be expected to lose over 10% lung function using the median
probabilities. The final column extrapolates these results to the total population of the study
area. In this case, 1.545 million persons would be in the moderate exertion category, and 17,520
would lose over 10% lung function using the median probabilities.
5.9 Sites File
The Sites output file lists the sectors, air districts, and zones in the study area, and identifies the
mapping between them. Thus, each record contains the information shown below.
Sector#: sector ID
Latitude, sector latitude (decimal degrees)
Longitude, sector longitude (decimal degrees)
SectorName. sector name
Air#: air district ID
AirDistance: distance from air district to sector (km)
AirLatitude. air district latitude (decimal degrees)
AirLongitude. air district longitude (decimal degrees)
Air Name air district name
Met#: meteorology zone ID
MetDistance: distance from meteorology zone to sector (km)
MetLatitude: meteorology zone latitude (decimal degrees)
MetLongitude: meteorology zone longitude (decimal degrees)
MetName: meteorology zone name
5.10 Events File
The Events file contains a summary of the activity diary, with accompanying exposure and dose,
at the diary event level. The variables printed in this file include those listed below.
Person : the profile number of the simulated individual
Seq: the event number for the profile
Day : the day number of the simulation, incremented from Day 1 of simulation
Year: the year of the event (4-digit)
Mn : the month of the event (1 to 12)
Dy: the day of the week the event (1 to 7)
Hr : the hour of the event (1 to 24)
Dur : the duration of the event (integer minutes)
Act: the MET distribution code for the event activity
Mic: the microenvironment code for the event
HW 1 = event in home sector, 2 = event in work sector, 3 = elsewhere
132
-------
Exposure: exposure level (concentration in the microenvironment) during the event (ppm,
ppb or ng/m3, as indicated by OUTPUTUNITS)
Optionally, the user can ask APEX to include the variables (bulleted below) by using the
keyword EVENTSLIST, and then listing each of the variable names to be included (shown
below).
MET. MET level for the event (unitless)
UMET: unmodified MET level for the event (unitless)
VA: alveolar ventilation during the event (ml/min)
VE: ventilation during the event (ml/min)
EVR: equivalent ventilation rate
DFEV1: end-of-event percent decrement in the forced expiratory volume in 1 sec
FEVX: end of event X term used in the %AFEV1 calculation
Deficit, oxygen debt, percent of nominal
AMB. (or "ambient") the ambient concentration during the event (ppm, ppb or |ag/m\ as
indicated by OUTPUTUNITS)
HomeAmb. the ambient concentration in the home district during the event (ppm, ppb or
Hg/m3, as indicated by OUTPUTUNITS)
FEVE1: the intra-individual variability term associated with the %AFEV1 model
FEVE2: the E2 error term associated with the %AFEV1 model
If DODOSE = Yes in the Control Options file, then two variables related to dose will be printed
for all cases, and an additional two will be printed for PM exposure (shown below).
Dose, average dose over the event
FDose: final dose for the event
DepDose. deposited mass dose for PM events
IntakeDose: intake dose rate for the PM event
The distinction between average dose and final dose is only relevant for pollutants with dose
modeled as a continuous function of time, such as carbon monoxide (CO). The dose for CO is
measured by the carboxyhemoglobin concentration in the blood (%COHb). While the CO
exposure in the lungs is assumed to be constant over one event, the blood %COHb changes
continuously with time, much like the air concentration in the mass balance model. The Dose is
the average over the event, while FDose is the value at the end of the event (and therefore
becomes the initial value for the next event). For many pollutants, no such dose model has been
built into APEX yet.
An example of the EVENTSLIST keyword would be:
EventsList = UMET VA VE EVR MET DEFICIT
133
-------
APEX Events File
APEX Version 5.0 (dated July 25, 2017) Run Date = 20191029 Time = 121815
Location = Los Angeles Area
Scenario = Los Angeles (CSA 348), Jun-Aug 2010 Example case, 2010 pop, RandomSeed 4178348
Simulation = COF_LA_Short_Course.txt modified from APEX Intro Document Section 3.3; Example
APEX model run for the ISES-ISEE APEX short course.
Pollutant = Ozone
Air Quality = ! 2010 base ozone air quality data for CSA 348 : Los Angeles-Long Beach-
Riverside, CA
Person Seq Day Year Mn Dy Hr Dur Act Mic HW IJMET MET VA VE Deficit EVR Exp_Ozone
5 11 2010 6 1 1 60 5 1 1 0.999 0.999 3530. 7042. 0.00000 3.764 3.878E-03
5 2 1 2010 6 1 2 60 5 1 1 0.908 0.908 3206. 6513. 0.00000 3.482 3.963E-03
5 3 1 2010 6 1 3 60 5 1 1 0.868 0.868 3068. 6283. 0.00000 3.359 3.965E-03
5 4 1 2010 6 1 4 60 5 1 1 0.951 0.951 3359. 6764. 0.00000 3.616 4.030E-03
5 5 1 2010 6 1 5 60 5 1 1 0.887 0.887 3134. 6394. 0.00000 3.418 3.965E-03
5 6 1 2010 6 1 6 60 5 1 1 0.953 0.953 3367. 6778. 0.00000 3.623 3.850E-03
5 7 1 2010 6 1 7 60 5 1 1 0.838 0.838 2961. 6104. 0.00000 3.263 3.735E-03
5 8 1 2010 6 1 8 60 19 1 1 9.074 5.103 18029. 29385. 0.95311 15.708 3.642E-03
5 9 1 2010 6 1 9 60 19 1 1 5.628 3.516 12423. 20042. 0.99555 10.713 3.491E-03
5 10 1 2010 6 1 10 20 100 5 99 3.548 3.218 11369. 18520. 0.99861 9.900 4.000E-02
Exhibit 5-12. Portion of an Events File
A portion of an example Events file is shown in Exhibit 5-12 above. This file can become very
largeabout 1.4 MB per person-year if all variables are written to the file. For this reason, the
user is given the option of writing the events for only a fraction of the simulated persons. This is
controlled by the Control Options file settings EVENTSAMPLE and CUSTOMSAMPLE. See
Section 4.2.3 for more information on these keywords.
5.11 Multipollutant File
This file can only be generated in a run containing at least two pollutants. It reports the amount
of time (averaged across persons) in each combination of pollutant levels, for each combination
of clock hour and microenvironment. The time is reported in hours, and cannot exceed the
number of days in the simulation (because there is at most 1 hour per person, per day, in that
bin). The divisor is the number of profiles with a non-zero amount of time in that bin. Also, the
fraction of profiles that "visited" that bin is reported.
If there are N1 cut-points (listed on the Control Options file) for pollutant #1 and N2 cut-points
for pollutant #2, then there will be 24*M*(N1+1)*(N2+1) rows of data on this file, where M is
the number of micros. For each pollutant, the number of "bins" is one more than the number of
cut-points. For example, if two cut-points are specified at 10 and 20 (in the OUTPUTUNITS for
that pollutant), then the 3 bins are "below 10", "10 - 20", and "above 20". If an exposure falls
exactly on a cut-point, the time is assigned to the higher bin.
Similar to other output files, the results are tallied over the entire population if COMMUTING =
NO or if KEEPLEA VERS = YES. If commuting is on and KEEPLEA VERS = NO, then the
population under consideration is the set of profiles who both live and work in the study area,
which will generally be fewer than the number requested on the Control Options file.
134
-------
5.12 Diary Clustering Files
There are up to three diary clustering output files, available when the CLUSTERDIARY option
is used for longitudinal diary assembly. The file names are specified on the Control Options file
using the keywords CLUSTER, CWEIGHTand TRANSITIONS. The file names are separate
from the control switches with yes/no settings; the keywords for the switches are
CLUSTEROUT; CWEIGHTOUT, and TRANSOUT
The "Cluster" file contains one record per diary day in the input diary database. The pool and
cluster assignments are listed, along with the score along each cluster axis. The score has two
forms: a time in minutes, and a ratio of this diary time to the average time across all diaries. The
cluster assignment is the largest of these ratio scores. If two ratios are exactly equal, the cluster
is the smaller of the axis numbers. For writing out the values, the maximum ratio is limited to
999.
The "CWeight" file has one record for each unique combination of "from" pool, "to" pool,
"from" cluster, "to" cluster, and age bin. Each record on this file has a statistical weight that is
used when selecting the cluster for the next (or "to") day.
The "Transitions" file has one record per example of a transition in the diary database, with the
"from" pool, "to" pool, "from" cluster, "to" cluster, pool bin, and age bin, along with the
CHADID of the "to" day, the diary age and whether it is part of an "adjacent" transition.
5.13 Sobol Results File
This file is generated only when a Sobol sensitivity analysis run is performed, which requires
setting SOBOLRUN = Yes on the Control Options file (see CHAPTER 11 of Volume II). In this
case, many of the other output files are suppressed. The Sobol file consists of two sections.
First, each random variable is listed along with Sobol group number. This allows the user to
identify which variables belong to a given group. The second section has two tables for each
specified exposure metric, average day or maximum day.
The exposure metrics are listed in Table 11-3 of Volume //, and may be one or more of the
following: AvgExp, MaxlExp, Max8Exp, MaxTSExp, MuxSEC. AvgDose, Max!Dose,
MuxSDose, MaxTSdose, MaxlFDose, Intake, or Dep. The dose metrics are available only if
APEX is asked to perform dose calculations (using DODOSE = YES on the Control Options
file). The last two {Intake and Dep) are available only for particulate matter. Each selected
metric is evaluated on each simulation day, for each profile. The "average day" is the average
for each person, over all the simulated days. The "maximum day" is the worst daily value for
that metric, for each person.
Sobol analyses measures the sensitivity of the selected exposure metric(s) to changes in the
random input variables. Within each table on the output file, there is one row for each Sobol
group number. There is one column for main effects, and a second for total effects.
Each main or total effect should be between zero and one. This range may occasionally be
exceeded when the stochastic noise is extremely large, due to small sample size. In that case, the
indices are not reliable. Otherwise, the total effect should be at least as large as the main effect,
135
-------
and may be substantially larger. The main effect is a measure of the fractional importance of one
specified group of variables, by themselves (including interactions among sets of variables
entirely within that group). The total effect includes the main effect plus interactions between
the specified group and variables in other groups.
The Sobol analysis provides sampled estimates for each "term" in a variance decomposition of
the selected exposure metric. If the estimates were without error (which would only happen in
the limit of infinite sample size), then the sum of all the main effects for all Sobol groups, plus
each unique interaction, would equal one. The sum of all main effects (across all the Sobol
groups) should never exceed one, and may be less than one because interactions between groups
are left out. The sum of all total effects should be at least one, because each interaction between
groups will be counted multiple times (once for each group that is part of the interaction). Larger
indices reflect greater importance for the variables in the corresponding Sobol group. See
Saltelli et al. (2004) for a detailed description of this method. Mokhtari et al. (2006) present an
application of the Sobol method to an exposure model. The paper Glen and Isaacs (2012)
provides guidance on using the Sobol method with stochastic models like APEX.
136
-------
REFERENCES
Glen W.G. and K. Isaacs (2012). Estimating Sobol sensitivity indices using correlations.
Environmental Modeling and Software 37: 157-166.
Graham S.E. and T. McCurdy (2005). Revised ventilation rate (Ve) equations for use in
inhalation-oriented exposure models. EPA/600/X-05/008.
McCurdy, T., G. Glen, L. Smith, and Y. Lakkadi (2000). The National Exposure Research
Laboratory's Consolidated Human Activity Database, Journal of Exposure Analysis and
Environmental Epidemiology 10: 566-578 (2000).
Mokhtari, A., H.C. Frey, and J. Zheng (2006). Evaluation and recommendation of sensitivity
analysis methods for application to Stochastic Human Exposure and Dose Simulation models,
Journal of Exposure Analysis and Environmental Epidemiology (2006) 1-16.
National Research Council (1991). Human exposure assessment for airborne pollutants:
advances and opportunities. Washington, DC: National Academy of Sciences.
Saltelli, A., S. Tarantola, F. Campolongo, and M. Ratto (2004). Sensitivity Analysis in Practice
A Guide to Assessing Scientific Models. John Wiley & Sons, Ltd, Chichester, England.
U.S. Environmental Protection Agency (1999).
https://www.epa.gov/fera
U.S. Environmental Protection Agency (2017).
https://www.epa.gov/fera/apex-user-guides and
Total Ri sk Integrated Methodol ogy. Web site:
An Introduction to APEX. Available at:
in the APEX installer.
U.S. Environmental Protection Agency (2019). The Consolidated Human Activity Database
(CHAD) Documentation and User's Guide, EPA-452/B-19-001. Available at:
https://www.epa.gov/fera/human-exposure-modeling-databases-support-exposure-modeling
137
-------
United States Office of Air Quality Planning and Standards Publication No. EPA-452/R-23-009a
Environmental Protection Health and Environmental Impacts Division June 2023
Agency Research Triangle Park, NC
------- |