The data that I’ve used in this visualization comes from two sources. One dataset is the NYC tree census from 2015 which includes the tree count and data collection information for each block of the city. The data was collected by NYC Parks & Recreation and partner organizations. The data is very detailed and includes each tree’s details such as species, size, and more but this data wasn’t conducive to the relationship I was trying to investigate through these visualizations. I think that a more detailed project on trees in NYC would find this data very useful and contribute to some questions of urban planning.
The second dataset includes the Heat Vulnerability Index (HVI) for each zip code of the 5 boroughs. The HVI of a given location refers to residents’ level of risk of dying during or after extreme heat. The model includes environmental factors like the surface temperature of the location, or how much heat people on the ground may be feeling in an episode of extremely hot weather. Other factors are access to air conditioning and the percentage of residents that are low-income. Income impacts accessibility to healthcare and other resources that can mitigate the impacts of heat-related medical crises.
I’m interested in this topic and wanted to use this data because of the direct link our choices in green space distribution and environmental protection have on the health of residents. This is a relationship that exists anywhere in the world, but it becomes specifically difficult in an urban environment like NYC because of constrictive zoning, air and ground pollution, and an unequal distribution of attention afforded to some communities over others despite their close proximity. There are other points to be made regarding large green spaces and their connection to mental health, recreation for children, and improvement of air quality. But what about homes and their surroundings?
The tree data used here is specifically counting the trees lining city blocks, or the trees around our homes that give us shade while we sit outside and walk around our neighborhoods. Through map visualizations, I wanted to point out a relationship between just trees and their impact on the immediate health of residents. The maps inadvertently highlight discrepancies that may be made up for by wealth and other factors. Each map is interactive with zooming in/out and tooltips but just their images are pasted here.
My first map shows the HVI rating for each zip code within NYC and is colored on scale accordingly. Across all 3 maps, red refers to HVI and green refers to trees. Each zipcode has a corresponding tooltip that lists the zipcode, the HVI, and the tree count sum for that entire area. There are some errors within the tree’s zipcode data, which appears as random points highlighting with a zipcode that is located across the city. It isn’t very impactful for the overall usage but should ideally be cleaned up. This map does a good job of visually showing the HVI distribution, but requires a closer look to see the tree counts and their possible correlation.
My second map is pretty simple and just visualizes the tree census data. I wanted to get a plain look at the visualization and from my knowledge of the city and its neighborhoods, it’s interesting to see the differences. It loses a bit of its impact in the fact that the color is only applied to the geographic data and the data only charts the outline of each block. I would prefer if the block itself was highlighted by the color because I think that all the empty space makes certain neighborhoods seem less green than they actually are. I also had to set limits on what got visualized, because certain green spaces were so tree-heavy that the rest of the map appeared barren. I thought it was more useful to disregard those green spaces as I mostly wanted to know about the city blocks. The tooltip simply shows the tree count for the block that the cursor is hovering on.
For my third map, I really wanted to highlight the discrepancies I noticed in the negative correlation relationship between tree coverage and HVI (more trees = lower HVI). The discrepancies, I wondered, may indicate higher levels of wealth and resources. As someone who is very familiar with the different NYC neighborhoods, I wanted to see if these discrepant communities aligned with the areas I knew to have generally higher income and more accessibility to city centers, which broadly aligns with accessibility to healthcare and resources for mitigating heat illness. I wanted to calculate the correlation between tree count and HVI for each zipcode and ultimately have a correlation coefficient, which could then determine the color coding of the map. In my thought process, the closer to -1 that the coefficient may be, the higher the likelihood that an area has greater income and resources. I unfortunately could not get this calculation to work and I wasn’t able to pinpoint why. Instead, I created a side-by-side comparison based on zipcode. With this visualization, it takes a bit more background knowledge and analysis to come to the conclusion that I was initially trying to display.
I started off with a vague knowledge of a relationship that I wanted to map out and through these visualizations I have only developed more questions and curiosities. In continuing through with this project, I would want to potentially incorporate more data that can add to the why behind the information being displayed. The why is largely outlined by the HVI statistical model itself, including the circumstantial and historical conditions that have shaped the communities in NYC. The risk factors within HVI like income, green space, and resources are deeply rooted in current and past racism. These racial factors need to have a substantial role in any conversation about community health in NYC. The care afforded to certain communities while others deteriorate over time indicates misuse of city money and resources, and I’m curious to see a comparison of similar data across other major cities in the US as well.