Mireille Vargas: Investigating Correlations between Urban Tree Cover and the Disadvantage Community Index

Mireille Vargas; Dr. Chris Fields; Dr. Christa Anderson

The investigation of the distribution of urban tree canopy and its correlations with socioeconomic factors have become an important topic in environmental justice. Likewise, the Office of Environmental Health Hazard Assessment (OEHHA) created a screening tool that visualizes the distributional equity of pollution burden by census tract in the state of California. This study continues this discussion by examining the distributional equity of urban tree cover (UTC) in cities in the Bay Area using high spatial land cover data and comparing it with CalEnviroScreen 3.0 Disadvantage Community Index along with certain individual indicators that make up the calculated index. Data was analyzed using ordinary least squares regression (OLS) and spatial autoregressive models (SAR). Bivariate regression showed a negative relationship between the disadvantage community index score and UTC in some cities and multivariable regressions of individual indicators showed strong negative relationships of certain indicators and UTC. Residuals of the OLS and SAR models showed that most cities had residuals clustered at zero for the SAR models than the OLS model suggesting that spatial autocorrelation is an important feature in the data. These findings suggest the importance of taking UTC into consideration for pinpointing disadvantaged communities due to the strong relationships among UTC and disadvantage communities’ indicators.

Mireille’s Presentation about her Research

5 Comments on “Mireille Vargas: Investigating Correlations between Urban Tree Cover and the Disadvantage Community Index

  1. Hi Mireille,

    Great job! You did a nice job explaining your methods and results in a easy to understand way. Were there any indicator results that you were surprised by, or did it all pan out as expected?

    Do you think you would have found similar findings in other parts of California, or even other parts of the country? I wonder how UTC would correlate with some of these variables in other geographic areas.

    Good job!

    • Hey Bianca,

      I very much found it interesting that in my results of all census tracts in the Bay area ozone was negatively significant with UTC but wasn’t significant in the data with only urban cities. I didn’t find it too surprising that there was a significant negative relationship between UTC and housing burden in my urban cities data since that was prevalent in San Francisco and Oakland (the larger two of my dataset).

      In the last few weeks of SESUR, I started to investigate other places in California such as Los Angeles and Orange county. From just the few analysis I did on these datasets, there were a few similarities in which indicators had a significant relationship with UTC.

  2. Hi Mireille! It was great working with you this summer! I loved hearing about your project and all of its interesting findings. I was wondering if you could talk a little bit about why you became interested in this project and what your favorite part of the research process was.

    Great job!

  3. Hi Mireille-

    Nice job, I loved your explanation of your methods and findings. You have a great understanding of regression and explained your work so clearly.

    I found it fascinating to think about the interplay of all these factors. You would think a simple solution would be to go plant more trees in disadvantaged communities but then realize there may be lack of green spaces and other areas to support tree cover and a lack of resources to take care of those trees, among other issues.

    What do you think could be next steps in this project?

  4. Hi Mireille,

    You did a great job presenting each of the steps of your research. It was easy to follow your process.

Leave a Reply to Alison Bowers Cancel reply

Your email address will not be published. Required fields are marked *

css.php