Integrating Uncertainty and Interindividual Variability in Environmental Risk Assessment

205

188

A population exposure model for particulate matter ( PM ) , called the Stochastic Human Exposure and Dose Simulation ( SHEDS -PM ) model, has been developed and applied in a case study of daily PM 2.5 exposures for the population living in Philadelphia, PA. SHEDS -PM is a probabilistic model that estimates the population distribution of total PM exposures by randomly sampling from various input distributions. A mass balance equation is used to calculate indoor PM concentrations for the residential microenvironment from ambient outdoor PM concentrations and physical factor data ( e.g., air exchange, penetration, deposition ) , as well as emission strengths for indoor PM sources ( e.g., smoking, cooking ) . PM concentrations in nonresidential microenvironments are calculated using equations developed from regression analysis of available indoor and outdoor measurement data for vehicles, offices, schools, stores, and restaurants / bars. Additional model inputs include demographic data for the population being modeled and human activity pattern data from EPA's Consolidated Human Activity Database ( CHAD ) . Model outputs include distributions of daily total PM exposures in various microenvironments ( indoors, in vehicles, outdoors ) , and the contribution from PM of ambient origin to daily total PM exposures in these microenvironments. SHEDS -PM has been applied to the population of Philadelphia using spatially and temporally interpolated ambient PM 2.5 measurements from 1992 -1993 and 1990 US Census data for each census tract in Philadelphia. The resulting distributions showed substantial variability in daily total PM 2.5 exposures for the population of Philadelphia ( median = 20 g / m 3 ; 90th percentile = 59 g / m 3 ) . Variability in human activities, and the presence of indoor -residential sources in particular, contributed to the observed variability in total PM 2.5 exposures. The uncertainty in the estimated population distribution for total PM 2.5 exposures was highest at the upper end of the distribution and revealed the importance of including estimates of input uncertainty in population exposure models. The distributions of daily microenvironmental PM 2.5 exposures ( exposures due to time spent in various microenvironments ) indicated that indoor -residential PM 2.5 exposures ( median = 13 g / m 3 ) had the greatest influence on total PM 2.5 exposures compared to the other microenvironments. The distribution of daily exposures to PM 2.5 of ambient origin was less variable across the population than the distribution of daily total PM 2.5 exposures ( median = 7 g / m 3 ; 90th percentile =18 g / m 3 ) and similar to the distribution of ambient outdoor PM 2.5 concentrations. This result suggests that human activity patterns did not have as strong an influence on ambient PM 2.5 exposures as was observed for exposure to other PM 2.5 sources. For most of the simulated population, exposure to PM 2.5 of ambient origin contributed a significant percent of the daily total PM 2.5 exposures ( median = 37.5% ) , espec...

Section: Methodsmentioning

confidence: 99%

A population exposure model for particulate matter: case study results for PM2.5 in Philadelphia, PA

Burke

Zufall

Özkaynak

2001

205

188

“…This time ±activity pattern diary is considered the simulated individual's time ± activity pattern from which percent of time spent indoors, outdoors, or at work is derived. Two -stage Monte Carlo simulation (Bogen and Spear, 1987;IAEA, 1989;MacIntosh et al, 1995b) is used to derive estimates of both inter-individual variability in the population and uncertainty in the estimated empirical chlorpyrifos absorbed dose distributions in the NHEXAS study populations. After age and gender are randomly sampled from U.S. Census demographic distributions for AZ or MSP, the model randomly samples from uncertainty probability distributions for each parameter in pathwayspecific equations to estimate absorbed dose for a sampled individual.…”

Section: Model Structure and Methodsmentioning

confidence: 99%

“…Explicit consideration of uncertainty in environmental assessments is important for understanding the range and likelihood of potential outcomes, and the relative influence of different assumptions, decisions, knowledge gaps, and stochastic variability in inputs on these outcomes ( Bogen and Spear, 1987;Iman and Helton, 1988;IAEA, 1989;Morgan and Henrion, 1990;Frey, 1992;U.S. EPA, 1992c;Taylor, 1993) .…”

Section: Evaluation Of Uncertainty In Population Distributionsmentioning

confidence: 99%

Modeled estimates of chlorpyrifos exposure and dose for the Minnesota and Arizona NHEXAS populations

Buck

Özkaynak

Xue³

et al. 2001

This paper presents a probabilistic, multimedia, multipathway exposure model and assessment for chlorpyrifos developed as part of the National Human Exposure Assessment Survey ( NHEXAS) . The model was constructed using available information prior to completion of the NHEXAS study. It simulates the distribution of daily aggregate and pathway -specific chlorpyrifos absorbed dose in the general population of the State of Arizona ( AZ) and in children aged 3 ± 12 years residing in Minneapolis ± St. Paul, Minnesota ( MSP) . Pathways included were inhalation of indoor and outdoor air, dietary ingestion, non -dietary ingestion of dust and soil, and dermal contact with dust and soil. Probability distributions for model input parameters were derived from the available literature, and input values were chosen to represent chlorpyrifos concentrations and demographics in AZ and MSP to the extent possible. When the NHEXAS AZ and MSP data become available, they can be compared to the distributions derived in this and other prototype modeling assessments to test the adequacy of this pre -NHEXAS model assessment. Although pathway -specific absorbed dose estimates differed between AZ and MSP due to differences in model inputs between simulated adults and children, the aggregate model results and general findings for simulated AZ and MSP populations were similar. The major route of chlorpyrifos intake was food ingestion, followed by indoor air inhalation. Two -stage Monte Carlo simulation was used to derive estimates of both interindividual variability and uncertainty in the estimated distributions. The variability in the model results reflects the difference in activity patterns, exposure factors, and concentrations contacted by individuals during their daily activities. Based on the coefficient of variation, indoor air inhalation and dust ingestion were most variable relative to the mean, primarily because of variability in concentrations due to use or no -use of pesticides. Uncertainty analyses indicated a factor of 10 ± 30 for uncertainty of model predictions of 10th, 50th, and 90th percentiles. The greatest source of uncertainty in the model stems from the definition of no household pesticide use as no use in the past year. Because chlorpyrifos persists in the residential environment for longer than a year, the modeled estimates are likely to be low. More information on pesticide usage and environmental concentrations measured at different post -application times is needed to refine and evaluate this and other pesticide exposure models.

“…However, uncertainty and variability are two distinct phenomena with different interpretations and implications for risk managers (Bogen and Spear, 1987;Frey, 1992;NRC, 1994;Kelly and Campbell, 2000). Interindividual variability determines the population fraction at risk, whereas uncertainty determines the reliability of model predictions.…”

Section: Introductionmentioning

confidence: 99%

Separation of uncertainty and interindividual variability in human exposure modeling

Ragas

Brouwer

Büchner

et al. 2008

The NORMTOX model predicts the lifetime-averaged exposure to contaminants through multiple environmental media, that is, food, air, soil, drinking and surface water. The model was developed to test the coherence of Dutch environmental quality objectives (EQOs). A set of EQOs is called coherent if simultaneous exposure to different environmental media that are all polluted up to their respective EQOs does not result in exceeding the acceptable or tolerable daily intake (ADI or TDI). Aim of the present study is to separate the impact of uncertainty and interindividual variability in coherence predictions with the NORMTOX model. The method is illustrated in a case study for chlorfenvinphos, mercury and nitrate. First, ANOVA was used to calculate interindividual variability in input parameters. Second, nested Monte Carlo simulation was used to propagate uncertainty and interindividual variability separately. Lifetime-averaged exposure to chlorfenvinphos, mercury and nitrate was modeled for the Dutch population. Output distributions specified the population fraction at risk, due to a particular exposure, and the reliability of this risk. From the case study, it was obtained that at lifelong exposure to all media polluted up to their standard, 100% of the Dutch population exceeds the ADI for chlorfenvinphos, 15% for mercury and 0% for nitrate. Variance in exposure to chlorfenvinphos, mercury and nitrate is mostly caused by interindividual variability instead of true uncertainty. It is concluded that the likelihood that ADIs of chlorfenvinphos and mercury will be exceeded should be further explored. If exceeding is likely, decision makers should focus on identification of high-risk subpopulations, rather than on additional research to obtain more accurate estimates for particular parameters.