Skip to Main Content
 

Global Search Box

 
 
 
 

Files

File List


19684.pdf (21.13 MB)

ETD Abstract Container

Abstract Header

Land Use Random Forests for Estimation of Exposure to Elemental Components of Particulate Matter

Brokamp, Richard C
http://orcid.org/0000-0002-0289-3151
http://rave.ohiolink.edu/etdc/view?acc_num=ucin1463130851

Abstract Details

2016, PhD, University of Cincinnati, Medicine: Biostatistics (Environmental Health).
Particulate matter (PM) has long been known to have a negative effect on public health. Epidemiological studies associating air pollution and other sources of PM often rely on land use modeling for exposure assessment. This approach relies on the association of characteristics of the surrounding land with PM concentrations. Land use regression (LUR) is the most commonly implemented land use model and has several drawbacks, including model instability due to correlated predictors and an inability to capture non-linear relationships and complex interactions. Here, I utilize the machine learning random forest model within a land use framework to generate a novel land use random forest (LURF) model. Using ambient air sampling data from the Cincinnati Childhood Allergy and Air Pollution (CCAAPS) study, I developed LURF and LUR models for eleven elemental components of particulate matter, including Al, Cu, Fe, K, Mn, Ni, Pb, S, Si, V, Zn. We show that LURF models utilized a higher number and more diverse selection of land use predictors than the LUR models. Furthermore, the LURF models were more accurate and precise predictors of all elemental PM concentrations, except for Fe, Mn, and Ni. To extend the usability of the LURF models, I utilized the recent application of the infinitesimal jackknife (IJ) to the random forest model in order to estimate the prediction variance. The IJ theorems were originally verified under the assumptions of traditional random forest framework, namely using CART trees and bootstrap resampling. Alternatives to the traditional random forest, such as subsampling instead of bootstrap resampling and conditional inference trees instead of CART trees have been shown to increase the accuracy of the random forest algorithm and eliminate its variable selection bias. Here, I conduct simulation experiments to show that the IJ performs well when using these random forest variations. Specifically, using the conditional inference tree instead of the CART tree and subsampling instead of bootstrap resampling results in increasing the accuracy and precision of the IJ estimator of random forest prediction variance. To associate the exposure of elemental components of PM with respiratory health, I applied the novel LURF model to the CCAAPS cohort. The exposures of children in this ongoing, prospective birth cohort located in Cincinnati, Ohio were calculated using their residential address history. Comparison of estimated elemental exposures with total PM2.5 estimated exposure showed that they were not correlated. Lung function and asthma testing was conducted on all children at age seven. We found that Al, Fe, Pb, Si, Zn, and total PM2.5 were associated with decreased lung function on their own, but after unconfounding the effect of exposure to PM with neighborhood level effects, the associations generally disappeared. Finally, I discuss the current limitation of two-stage models that associate spatial pollutants with health effects, namely omitting the uncertainty from the exposure assessment stage. When incorporating this uncertainty in the future, the increased accuracy and precision of LURF models compared to LUR models should allow for more precise estimation of the health effects of these pollutants.
Patrick Ryan, Ph.D. (Committee Chair)
Roman A. Jandarov, PH.D. (Committee Member)
Marepalli Rao, Ph.D. (Committee Member)
124 p.
Biostatistics
particulate matter; random forest; land use model; respiratory health; children; infinitesimal jackknife

Recommended Citations

Citations

  • Brokamp, R. C. (2016). Land Use Random Forests for Estimation of Exposure to Elemental Components of Particulate Matter [Doctoral dissertation, University of Cincinnati]. OhioLINK Electronic Theses and Dissertations Center. http://rave.ohiolink.edu/etdc/view?acc_num=ucin1463130851

    APA Style (7th edition)

  • Brokamp, Richard. Land Use Random Forests for Estimation of Exposure to Elemental Components of Particulate Matter. 2016. University of Cincinnati, Doctoral dissertation. OhioLINK Electronic Theses and Dissertations Center, http://rave.ohiolink.edu/etdc/view?acc_num=ucin1463130851.

    MLA Style (8th edition)

  • Brokamp, Richard. "Land Use Random Forests for Estimation of Exposure to Elemental Components of Particulate Matter." Doctoral dissertation, University of Cincinnati, 2016. http://rave.ohiolink.edu/etdc/view?acc_num=ucin1463130851

    Chicago Manual of Style (17th edition)

ucin1463130851
320
© , some rights reserved.Creative Commons License Land Use Random Forests for Estimation of Exposure to Elemental Components of Particulate Matter by Richard C Brokamp is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported License. Based on a work at etd.ohiolink.edu.
This open access ETD is published by University of Cincinnati and OhioLINK.