United States
 Environmental Protection
 Agency
 Environmental Monitoring
 Systems Laboratory
 Research Triangle Park NC 27711
 Research and Development
 EPA/600/S4-85/036 Apr. 1987
 Project  Summary
 A Sensitivity Analysis  of  the
 Enhanced Simulation  of  Human
 Air  Pollution  Exposure (SHAPE)
 Model

 Jacob Thomas, David Mage, Lance Wallace, and Wayne Ott
  A sensitivity analysis was undertaken
of the Simulation of Human Air Pollution
Exposure (SHAPE) model, which in-
corporates an enhanced version of the
Coburn-Forster-Kane (CFK) physiologi-
cal model for predicting a person's blood
carboxyhemoglobin (COHb) as a func-
tion  of  time in  response to  carbon
monoxide (CO) exposure. The SHAPE
model simulates the physical activities
of a sample of people in an urban area,
exposing them to pollutant concentra-
tions from appropriate microenviron-
ments as they move through time and
space in a 24-hour period.  The CFK
model dynamically calculates their blood
COHb from their CO exposures, and
their ventilation rate which is altered in
response to their level of activity. To
conduct this  sensitivity analysis,  the
SHAPE program  was run many times
using different combinations of values
for its parameters,  thus allowing  the
contribution to COHb of each of many
variables to be examined. The following
phenomena were found to have signifi-
cant effect, over and above the influence
of the ambient CO on the  predicted
frequency distribution of the maximum
COHb levels of the population: (1)  CO
exposure contributed by the highway
microenvironment; (2) altitude of  the
city; (3) the CFK  physiological  param-
eters (e.g., Haldane constant and  en-
dogenous CO production). In contrast,
if it is assumed that a person using an
indoor parking garage spends less than
10 minutes there, then use of a parking
garage has very little effect on the COHb
frequency distribution of the population.
For low-level CO exposure, use of either
the linear or nonlinear form of the CFK
model  yields  essentially the  same
results.
  This Prelect Summary was developed
by ERA'S Environmental Monitoring
Systems Laboratory, Research Triangle
Park, NC, to announce key findings of
the research project that Is fully docu-
mented In a separate report of the same
title (see Project Report ordering In-
formation at back).

Introduction
  The Simulation of Human Air Pollution
Exposure (SHAPE) model estimates
human exposure to carbon monoxide (CO).
It develops an "activity pattern profile"
for an individual by allocating each minute
of the day into any of 14 microenviron-
ments (home, car, office, parking garage,
etc.). By knowing where the person is and
approximately what he or she is doing,
another part of the program "exposes"
the person to CO concentrations appro-
priate for that microenvironment. Random
number generators provide minute-by-
minute concentration levels roughly
equivalent to those found in the actual
microenvironment, based on empirical
field  studies. For each  person,  these
sequential microenvironmental exposures
are brought together to form an "exposure
profile" providing CO concentration and
estimates of the  carboxyhemoglobin
(COHb) of the blood as functions of time
over the 24-hour period.
  The previous  version of the SHAPE
program used a simplified Coburn CO-
COHb model, a  first-order linear differ-

-------
ential equation with three constant coef-
ficients. This formulation did not permit.
physiological  and other  subcomponent
parameters (for example, blood volume,
ventilation rate, altitude) to be varied by
themselves, because the latter param-
eters were lumped together into the three
constants.
  The research summarized in the full
report  covers enhancements to  the
SHAPE model and a sensitivity analysis
of the enhanced model.
  The goal of this analysis is to determine
how changes in a variety of fundamental
parameters (highway CO concentrations,
CO concentrations inside a parking ga-
rage, time patterns of ambient CO con-
centrations, and various Coburn physio-
logical  parameters) affect the  following
three  dependent  variables  for  the
population:

  • Maximum 1 -hour average exposure
    (M1AE)
  • Maximum moving  8-hour average
    exposure (MM8AE)
  • Maximum 1 -hour  blood carboxy-
    hemoglobin(MICOHb)

In each case, we are interested  in the
population's frequency  distributions of
these variables and the impact of a change
in each parameter on these frequency
distributions.  In general,  we  examined
changes in frequency distributions that
result while one parameter is allowed to
vary and all the others are held constant.
In some cases,  two parameters were
varied at the same time.
  Our  simulation sample consisted of
400 non-smoking people employed  in a
large urban area. Their simulated com-
mute habits are based on data from the
U.S. Census Bureau's "Nationwide Per-
sonal Transportation Study", (1973).

Modifications to SHAPE
  The following enhancements have been
made to the SHAPE computer program:

  • All  individual Coburn parameters
    appear explicitly.
  • Both linear and the more accurate
    nonlinear versions  of the Coburn
    model can be used to compute blood
    COHb.
  • Once the age and sex of the person
    have been  established,  height is
    determined by random sampling from
    distributions based on data from the
    National Health and  Nutrition Ex-
     amination Survey  (HANES) con-
    ducted by the  National Center for
     Health Statistics.
 • Weight is determined from empirical
   height-weight equations, also based
   on sex and age.
 • Blood  volume is  determined from
   height and weight, and hemoglobin
   content is randomly sampled from
   the HANES data, based on sex.
 • Some of the person's lung functions
   (alveolar ventilation and diffusion
   rates,  for example) are allowed to
   vary throughout  the  day  and are
   determined from the activities he or
   she performs, based  on metabolic
   rate assumptions  for each physical
   activity level.
 • For each person,  the program runs
   through an extra  24-hour exposure
   sequence  on the previous day in
   order to set the initial conditions for
   the day of interest.
  Once these enhancements were made,
the program was run many times with
different combinations of values. Ambient
data for three different dates were used,
each with relatively high  CO levels se-
lected from an earlier sensitivity analysis.
  In this study, only  subsamples of an
urban population were simulated for their
exposure experience  to CO. The  sub-
sample  consisted of   nonsmoking em-
ployed commuters during a week day. As
such, the distributions generated of maxi-
mum  1 hour CO  exposure, maximum  8
hour  exposure and  maximum 1  hour
COHb are seen to have small standard
deviations and less than 10% coefficients
of variation.
Conclusions

  The findings of the study are among
the following:

  • Changes in CO levels inside auto-
    mobiles exert a statistically signifi-
    cant effect on computed blood COHb
    of the population.
  • Altitude exerts a statistically signifi-
    cant effect on computed blood COHb.
  • Changes in the levels of CO in indoor
    parking garages have little effect on
    total CO exposure and blood COHb
    of the population due to the brief
    time (10 minutes)  assumed spent
    there.
  • For low-level CO exposure, the linear
    approximation of the solution to the
    Coburn equation does  not differ
    significantly  from the  nonlinear
    solution.

  This sensitivity analysis has identified
the relative significance of a  variety of
parameters in generating estimates of
blood COHb and has provided insight into
the manner in which  the  parameters
affect person-to-person variability of blood
COHb. It also has identified  the likely
contribution of various human activities
— driving a car, parking  in a garage — to
population  exposure distributions  and
human dose distributions.
  Jacob Thomas is with General Software Corporation.  Landover, MD 20785;
    the EPA authors,David Mage (also the EPA Project Officer, see below) and
    Lanco Wallace, are with Environmental Monitoring Systems  Laboratory,
    Research  Triangle Park, NC 27711; and the EPA author, Wayne Ott is with
    Office of Acid Deposition, Environmental Monitoring and Quality Assurance.
    Washington, DC 20460.
   Tho complete report, entitled "A Sensitivity Analysis of the Enhanced Simulation
    Lf Human Air Pollution Exposure (SHAPE) Model ." (Order No. PB 85-201
    101 /AS; Cost: $18.95, subject to change) will be available only from:
          National Technical Information Service
          5285 Port Royal Road
          Springfield, VA22161
          Telephone: 703-487-4650
   The EPA Project Officer can be contacted at:
          Environmental Monitoring Systems Laboratory
          U.S. Environmental Protection Agency
          Research Triangle Park, NC 27711

-------
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Official Business
Penalty for Private Use $300

EPA/600/S4-85/036
           0000329   PS

           U  S EHVIR PROTECTION AGENCY
           REGION 5  LIBRARY
           230 S  DEARBORN  STREET
           CHICAGO             It   60604

-------