Assessing Geographic Variability in Key Indicators of Water Quality and Air Quality: A Comparison of Urban vs. Rural Communities

Research center:
Lead researcher:
Project funded:
September 2021
Anticipated completion date:
April 2023

This project aims to explore urban-rural and regional differences in potential exposure to contaminated waters and/or polluted air. In addition to identifying such differences, we will explore potential covariates that help examine the observed spatial patterns.

Our team will use a combination of spatial analytical techniques and regression modeling to determine whether statistically significant differences exist in potential exposure to contaminated waters and polluted air among rural and urban areas (defined at the Census tract level using 2013 Rural Urban Commuting Area codes), as well as U.S. Environmental Protection Agency defined regions, with findings prepared in a policy brief format. Sociodemographic composition variables (e.g., percent non-White, percent below 200% Federal Poverty Limit, etc. at the Census tract level) will be tested for their individual relationship with air and water quality outcomes, as well as acting as a potential moderator between urban-rural status and each respective outcome.

We propose a tiered level of spatial analyses with the first level being at the Census tract level and the second an inter-regional comparison using the Environmental Protection Agencies administrative regions. Communities disproportionately impacted by environmental harms are defined as "environmental justice communities" by the U.S. Environmental Protection Agency. Factors that go into this designation include social vulnerability, socioeconomic factors, as well as exposure to/spatial proximity to various environmental hazards. Using the Environmental Justice Screen Mapping Tool, stakeholders can visualize and interact with this data at the block group level (i.e., typically 600-3,000 persons per unit). We will aggregate this data to the Census tract level to help our team examine urban-rural differences in our outcomes of interest, where point-level data are not available or sparse.