![[logo] US EPA](https://webarchive.library.unt.edu/eot2008/20090510232400im_/http://www.epa.gov/epafiles/images/logo_epaseal.gif){kind=logo}
Improving Estimates of Exposures in the Application of Human Exposure Models with Air Quality Model Output
Abstract:
Air pollution from ambient sources continues to adversely impact human health in the United States. A fundamental goal for EPA is to implement air quality standards and regulations that reduce health risks associated with exposures to criteria pollutants and air toxics. However, this is not an easy or straightforward task. The NRC has recommended that research should be conducted to identify those pollutants and sources that are responsible for the most significant risks in air sheds. A critical component of this research is to develop an understanding of how pollutants from sources impact ambient concentrations and, in turn, how these concentrations relate to exposures. This information provides a fundamental linkage for evaluating health impacts and developing effective mitigation strategies.
To help identify the air pollutant sources of greatest risk to humans, modeling tools can be applied to estimate air concentrations from sources and the resultant exposures for a population of interest. The appropriate combination of human exposure and air quality models for producing scientifically defensible estimates of the distribution of exposures will depend on many factors due to the complex nature of human contact with air pollutants. For a particular pollutant, air concentrations may vary significantly within an urban area, for example, and/or by time of day or season of the year. Human activities will also vary in space and time, and also by age, gender, and other demographic characteristics. The level of sophistication needed in both types of models will also depend on the research questions to be addressed. For example, a screening-level assessment to identify air pollutants that may pose an unacceptable health risk will require a different level of detail than an assessment aimed at quantifying the full range of exposures across a population and determining the relative importance of various exposure pathways.
The goal of this task is to apply population-based exposure models using modeled air pollutant concentrations to better understand the level of detail necessary for both urban- and local-scale assessments, and to ultimately provide improved estimates of exposures when pollutants are likely to have spatial gradients but measurements are limited. Approaches for linking air quality models to population-based exposure models for urban- and local-scale assessments will be developed and evaluated. Model applications will be conducted to evaluate the relationships between sources, ambient concentrations, and exposures, to identify exposure factors that influence these relationships, and to produce improved spatial and temporal population exposure estimates. This research is a collaborative effort with NERL's Atmospheric Modeling Division's (AMD) efforts to improve air quality modeling outputs for use in human exposure models (AMD task #15170). Comparisons will be conducted to evaluate the impact of improved urban-scale air pollutant concentrations from these efforts on population exposure estimates.
This research will have a broad impact on human air pollutant exposure assessments conducted by EPAs program offices (e.g., Office of Air and Radiation), EPAs Regions, and States by demonstrating where, when, and how there is value added with increased modeling complexity and scientific understanding of the impact of air pollutant sources on human exposure.
Objective:
To provide improved estimates of air pollutant exposures through combined air quality and exposure model applications:
- Apply human exposure models using emissions-based air quality modeling output
- Determine impact of air quality modeling approaches on modeled population distributions of exposure
- Identify factors contributing to higher exposures in modeling simulations
Relevance/Significance/Impact:
The research conducted under this task will demonstrate through selected applications the utility as well as the limitations in combining emissions-based air quality models with human exposure models for improving estimates of exposures for urban- and local-scale assessments. As monitoring networks are reduced in size and frequency due to their expense, exposure and risk assessments will have to rely more often on modeled concentrations as input to exposure models. This research will improve our understanding of how to best use these modeling tools for a variety of applications including risk assessment, accountability, development of mitigation strategies, and epidemiology.
Principal Investigator: Dr. Janet M. Burke