In-vehicle Air Pollution Exposure Measurement And Modeling

FINAL REPORTCalifornia Air Resources BoardCONTRACT NO. 07-310IN-VEHICLE AIR POLLUTION EXPOSURE MEASUREMENT AND MODELINGSubmitted byRalph J. Delfino, MD, PhD, and Jun Wu, PhDCo-Principal InvestigatorsDepartment of Epidemiology, School of Medicine, University of California, Irvine, 92697-7550Prepared for the California Air Resources Board andthe California Environmental Protection Agency.Co investigators:Scott Fruin, PhD, Keck School of Medicine, Environmental Health Division, University ofSouthern CaliforniaConstantinos Sioutas, ScD, Department of Civil & Environmental Engineering, University ofSouthern California.Lianfa Li, PhD, Program in Public Health, University of California, IrvineRufus Edwards, PhD, Department of Epidemiology, School of Medicine, University ofCalifornia, IrvineBeate Ritz, MD, PhD, Department of Epidemiology, UCLA School of Public HealthNorbert Staimer, PhD, Department of Epidemiology, School of Medicine, University ofCalifornia, IrvineJune 8, 2012i

CONTENTSFINAL REPORT .iCONTENTS.iiDisclaimer . viiAcknowledgements. viiLIST OF FIGURES. viiiLIST OF TABLES .xABSTRACT. xiiiEXECUTIVE SUMMARY. xivBODY OF REPORT . 161.Chapter One: Introduction. 161.1Background . 161.2. Scope and Purpose of the Project. 191.3. Tasks. 20Overview . 202. Chapter Two: A Predictive Model for Vehicle Air Exchange Rates based on aLarge, Representative Sample . 292.0 Introduction. 292.1Materials and Methods . 312.1.1 Vehicle selection . 312.1.2Instruments. . 322.1.3Air Exchange Rate Determinations. . 322.1.4Mathematical Equation and Assumptions. . 322.1.5Determination of Source Strength . 332.1.6Determination of Equilibrium Concentration . 332.1.7Speed. 342.1.8Data Analysis . 342.2.Results and Discussion . 352.2.1Vehicles Tested. 35ii

2.2.2Equilibrium Values and AERs Calculated . 352.2.3GEE Model Results. 362.3 Summary and Conclusions . 39References . 413.Chapter Three: Factors that Determine Ultrafine Particle Exposure in Vehicles. 433.0 Introduction. 433.1 Methods. 443.1.1 Vehicle Selection and Conditions Tested. 443.1.2 Particle Concentration Measurements . 453.1.3 Air Exchange Rate Measurements . 463.2 Results and Discussion . 483.2.1 Effect of Air Exchange Rate on I/O Ratios . 483.2.2 Effect of Vehicle Speed and Age on AER and I/O Ratios . 493.2.3 Effect of Particle Size on I/O Ratios . 513.2.4 Effect of Ventilation Fan Setting on I/O Ratios . 513.2.5 Effect of Cabin Air Filter and Loading on I/O Ratios. 523.3 Implications for In-Vehicle Particle Models. 553.4 implications for Exposure assessment . 563.5 Summary and Conclusions . 58References . 594. Chapter Four: Freeway Emission Rates and Vehicle Emission Factors of AirPollutants in Los Angeles. 614.0 Introduction. 614.1 Methods. 634.1.1 Mobile Measurement Platform (MMP) and continuous measurement instruments . 634.1.2 Sampling Routes. 644.1.3 Mathematical calculations and equations. 654.1.3.1 Emission Factor (EF). 654.1.3.2 Traffic Characterization . 68iii

4.1.3.3 Freeway emission rate calculations. 694.2 Results and Discussion . 694.2.1 Pollutant Concentrations . 694.2.2 LDV and HDV emission factors. 694.2.3 Fraction contribution of HDV to total emissions . 724.2.4 Freeway Pollutant Emission Rates . 734.2.4.1 Annual average emission rates . 734.2.4.2 Diurnal variation in freeway emission rates . 754.2.4.3 Freeway-to-freeway variability in emission rates . 764.3 Summary and Conclusions . 77References . 785. Chapter Five, Part I. Linking In-Vehicle Ultrafine Particle Exposures to On-RoadConcentrations . 815.0Introduction. 815.1Me th o d s . 825.1.1 Vehicle selection and ventilation conditions tested . 825.1.2 Speed and routes driven . 845.1.3 Particle concentration measurement, I/O and AER determination . 845.1.4 Predictive models. 855.2Results and Discussion . 865.2.1 In-vehicle-to-roadway concentration ratios. 865.2.2 Predictive model for ln(AER) at RC and OA setting . 875.2.3 Predictive model for logit(I/O) under RC and OA setting. 895.2.4. Fleet-wide distributions of AER and I/O . 935.2.5 Expected in-cabin concentrations for given roadway concentrations. 965.3. Summary and Conclusions . 97References . 98Chapter Five, Part II. Develop and validate the on-road exposure models for particlebounded PAH, PNC, PM2.5, NOx, and BC (based on Task 4: Develop and validate invehicle exposure models for BC, UFP number, PM2.5, particle-bounded PAH, and NOx.). 100iv

5.4. In tro d u c tio n . 1005.5. Ma te ria ls . 1015.5.1 Mobile Measurement Platform and Concentrations Measured . 1015.5.2 Road and Traffic Classification . 1025.5.3 Meteorological Parameters . 1025.5.4 Independent and Dependent Variables. 1035.6. Me th o d s . 1055.6.1 Exploratory Data Analysis . 1055.6.2 Selection of Predictor variables. 1065.6.3 General Linear and Non-Linear Models with Inclusion of Factor Variables. 1075.6.3.1 Basic model: linear regression with factor variables . 1075.6.3.2 Non-linear model: generalized additive model with factor variables . 1075.6.4 Time series model with temporal autocorrelation and factor variables. 1095.6.5 Model validation . 1105.6.5.1 Holdout validation as an independent test and validation. 1105.6.5.2 3x3 cross-validation. 1115.6.5.3 Measurement criteria. 1115.7. Re s u lts a n d Dis c u s s io n . 1125.7.1 Dependent variable concentrations. 1125.7.2 Transformation and correlation analysis . 1145.7.3 Grouping Comparison . 1195.7.3.1 Roadway types . 1205.7.3.2 Time of day. 1215.7.3.3. Atmospheric Stability . 1245.7.4 Regression models for prediction. 1265.7.4.1 PAH modeling . 1265.7.4.2 PNC modeling . 1285.7.4.3 PM2.5 modeling . 1315.7.4.4 NOX modeling. 133v

5.7.4.5 BC modeling. 1355.7.5 Time series analysis. 1385.7.6 Discussion. 1405.7.6.1 Correlation analysis and scatter plots. 1405.7.6.2 Influence of roadway types. 1415.7.6.3 Influence of time of day . 1415.7.6.4 Influence of traffic variables. 1415.7.6.5 Influence of meteorological factors . 1425.7.6.6 Linear vs. non-linear models . 1425.7.6.7 Validation of predictive models. 1435.7.6.8 Consideration of temporal autocorrelation. 1435.8 S u mm a ry a n d Co n c lu s io n s . 1445.9 Re fe re n c e s . 1456. Ch a p te r Six. Ta s k 5: Va lid a te th e in -ve h ic le e xp os u re m o d e l fo r P AH a g a in s tm e a s u re m e n ts fro m re p re s e n ta tive s u b je c ts . . 1466.1. Ma te ria ls a n d Me th o d s . 1466.2.Re s u lts a n d Dis c u s s io n . 1496.3 S u m m a ry a n d Co n c lu s io n s . 1526.4 References. 1547. Chapter 7. Study Limitations . 1548. Chapter 8. Overall Summary and Conclusions. 156References. 1609. Chapter 9. Re c o mm e n d a tio n s . 16210. LIST OF PUBLICATIONS PRODUCED. 163vi

DISCLAIMERThe statements and conclusions in this report are those of the University and not necessarilythose of the California Air Resources Board. The mention of commercial products, theirsource, or their use in connection with material reported herein is not construed as actual orimplied endorsement of such products.ACKNOWLEDGEMENTSWe thank Neelakshi Hudda, graduate student research assistant at the Department of Civil &Environmental Engineering, University of Southern California, for her diligent and superbwork on this project. As she was lead or co-author of several papers resulting from thisresearch, with her permission, some of the chapters in this report are partly adapted from herPh.D. thesis. We also thank James Liacos, Winnie Kam, Evangelia Kostenidou, Sandrah P.Eckel at the Department of Civil & Environmental Engineering, University of SouthernCalifornia, and Luke D. Knibbs at Queensland University of Technology, Brisbane, Australia.We thank Thomas Tjoa, Department of Epidemiology, UCI, for his work in constructingdatasets and help in programming the data analysis.This Report was submitted in fulfillment of California Air Resources Board contract no. 07310 by the University of California, Irvine under the sponsorship of the California AirResources Board. Work was completed as of June 28, 2012.vii

LIST OF FIGURESFigure 2.1: Typical Time-series plot for runs conducted at Cemetery along with the initial buildup and freeway run. Average speed during Freeway run was 89 10 km hr-1 forstable portion highlighted in black). The second black highlight corresponds tostable values during 51.1 9.4 km hr-1 and 31.3 5.5 km hr-1 speed runs.Figure 2.2: AER results for all 59 vehicles tested.Figure 2.3: Model-predicted AER increase with age and speed for median age study vehicle.Figure 2.4: Model predictions versus actual measurements, and the normality of theresiduals. Each data point represents a measured AER used to populate thepredictive model.Figure 2.5:. Comparison of model predictions and results from Knibbs et al., 2009.Figure 3.1: I/O ratio dependence on AER for 2

Ralph J. Delfino, MD, PhD, and Jun Wu, PhD . Co-Principal Investigators . 32 Determination of Source Strength . The dashed lines join values from the same vehicle. Figure 3.4: Comparison of I/O ratios at different speeds and fan settings. Figure 3.5: I/O ratios by filter condition or absence under OA conditions in a 2010 Toyota .