Detailed Emissions in Virginia (DEVA): Measuring Inequity Using the Atkinson Index

Author(s): Sara Alhassani

Mentor(s): Dr. Jenna Krall, Public Health

Abstract
Air pollution is a pressing issue, harming the environment and the people living in it. Even more concerning is that air pollution exposure disproportionally affects people of color and low-income populations. These groups are more likely to be exposed to pollution and experience harmful health effects. Particulate matter (PM) is known to cause health complications on the lungs and heart from exposure and inhalation. Policymakers and emission standard regulations need thorough data and research to guide the implementation of regulations to reduce the harm caused by air pollution. To help solve this issue, the “Detailed Emissions in Virginia: Identifying Air Pollution Sources for Environmental Justice” project seeks to provide data on pollutant and inequity in exposure for all counties in Virginia to be used in regulation development. In particular, the Atkinson Index is used to quantify inequity between racial and ethnic groups including Non-Hispanic or Latino White, Black or African American, Asian, and Hispanic or Latino. Combining 2011 NEI data on PM 2.5 emissions and 2010 Decennial Census data on race/ethnicity populations, the Atkinson Index is applied to determine the inequity for two values of epsilon, an inequality parameter. This subanalysis yielded that between the sectors of Wildfires, Coal Combustion in Industrial Broilers, and Diesel Heavy Duty Vehicles, inequity between the 4 racial/ethnic groups is shown to be at a midpoint for Wildfires and Coal Combustion. Diesel Trucks yield relatively low inequity. These results can be expanded and combined with data on other sectors and pollutants to be provided to policymakers as an all encompassing guide for reducing inequity between groups. This is crucial to protect the lives and health of vulnerable and susceptible populations.
Audio Transcript
My name is Sara Alhassani and I am a junior majoring in Applied Statistics with minors in Computational Data Sciences and Social Work. Over the past 2 semesters, I have been an undergraduate research assistant on the “Detailed Emissions in Virginia: Identifying Air Pollution Sources for Environmental Justice” project led by Dr. Henneman and Dr. Krall.

My primary focus for this project has been on measuring inequity in air pollution exposure using the Atkinson Index inequality measure.

With the alarming changes we’ve been seeing in the environment, it continues to be of utmost importance to study the environment. Researching air pollution, in particular, is crucial due to the detrimental impacts air pollution exposure has on the health of people, animals, and the earth. Increasing understanding of how air pollution impacts the world around us is used to guide and improve standards to protect everyone and everything at risk.

This presentation focuses on fine particulate matter, or PM 2.5, emissions. PM 2.5 are very small particles that can be inhaled deep into the lungs and pose serious health risks.

The objective of this subsection of the DEVA project is to evaluate the inequity in air pollution exposure between different race and ethnic groups. The upcoming analysis shows inequity between White, Black or African American, Asian, and Hispanic or Latino people.

Using past studies as guidance, we decided to use the Atkinson Index to measure inequity. This measure allows analysis to be done between groups. The Atkinson Index value ranges from 0 to 1 with 0 being complete equality and 1 being maximum inequality.

To determine inequity between races, I first obtained and organized data on all counties in Virginia from the 2010 Decennial Census to estimate inequity in 2011 PM 2.5 emissions using R programming.

Next, I combined that data with clean National Emissions Inventory PM 2.5 data provided by DEVA’s engineering team. I focused on 3 pollutant sectors: wildfires, coal combustion, and diesel trucks.

After this, I manipulated the data to obtain the variables needed to run the R Atkinson Index function for two values of epsilon, and obtained the results shown in the figures.

The bar plots show that when using an epsilon value of 0.75, the Atkinson Index values between groups for all 3 sectors is small and close to 0, not indicating a lot of inequality. However, when using epsilon value of 3, which accounts for sensitive analysis, we see much different results. While Diesel stays small, the Atkinson Index value for wildfires reaches almost 0.5, with coal combustion not too far below. The change in epsilon also appears to switch wildfires and coal combustion for which has greater inequality.

These results suggest that air pollution emission regulations need to be adjusted to lower air pollution exposure inequity caused by coal combustion in industrial broilers. While wildfires yielded similar results to coal combustion, their unpredictability makes it difficult to produce legislation to prevent them. This analysis can be enhanced by looking at inequity between different income levels and education attainment, but there is limited data on these factors at the county level.

Moving forward, DEVA hopes to quantify inequity for additional sectors and pollutants, to be used to support improvements in air pollution and exposure of susceptible and vulnerable populations.

I’d like to thank both Dr. Krall and Dr. Henneman for guiding me throughout this research, as well as the Jeffrees Memorial Trust for their support.

Thank you!

Leave a Reply