Across the globe, large cities strive to advance equity. Cities are mainly concerned with providing their residents with an equal opportunity to live their healthiest lives. Equity doesn’t just promote inclusivity, but it also results in more resilient cities. Consequently, cities want to ensure that all residents have equal access to amenities like public transportation, parks, and quality streets irrespective of income, zip code, or demographics. Cities spend an immense amount of time and resources attempting to compare neighborhoods’ food access, transportation options, and environmental efforts. All these efforts are invested to create an equal distribution of resources in different areas of the city.
Among all the cities' physical characteristics, streets are perhaps the most important and one of the most challenging in contemporary city planning. Streets are where people meet their neighbors. Streets are also an extension of people's front yard and the largest urban public space in most cities. Most notably, streets are how people get around from point A to point B. The aesthetics and appearance of a street can greatly influence the safety and perception.
Street trees are important because they help protect pedestrians. This is not only because they contribute to the perception of safety, but also because they create a physical boundary between the sidewalk and cars (UC Berkeley study). Additionally, the presence of trees are believed to encourage people to choose more active modes of transport like running, cycling and walking. That is because trees can enhance a street's public realm (Center for Active Design).
Because trees play such a key role in street design, and due to the inequality observed in tree distribution between different NYC neighborhoods, a metric needed to be developed to capture tree canopy coverage. The aim of this metric is to make it easy and straightforward for planners to visually and quantitatively compare tree canopy distribution across different areas of the city. I used NYC's street tree census and NYC street centerline from the NYC Open Data platform to produce a street level calculation of tree canopies.
Before diving into the details of the data analysis here are a few examples from real streets:
1. South Portland Street is a very highly covered street. As such, trees provide an abundance of shade and make it pleasant to walk along on a sunny day
2. Fulton Street consists of a medium tree coverage with some trees, but not as much shade or “sense of place” as South Portland Street.
3. Finally Lafayette Street right of the BAM has a very low tree coverage.
HOW THE STREET TREE COVERAGE WAS MEASURED:
Ideally, to understand the city tree canopy we’d need to measure trees by their crown diameter. However, this data is not immediately available.The 2015 tree census released by NYC parks, provides a comprehensive account of each and every street tree in NYC. Among other tree features, like exact location and tree species, NYC Parks also provide users with the Diameter at Breast Height (DBH) for each street tree in the city. The DBH is a common measurement of a tree’s diameter measured at approximately 4.5 feet height.
A few academic studies previously found that there is a positive statistical relationship between a tree diameter at breast height and its crown diameter. For example, in their 1994 paper, Gering and May found that this relationship holds in four tree species groups. Their findings were quite simple. The bigger the DBH is, the bigger the crown diameter is expected to be. For this reason, using the DBH as a proxy for tree crown size is how I decided to measure urban tree canopies.
In non-data urbanist language: based on the size of the tree’s trunk, I am closely estimating the size of the foliage canopy.
For the analysis, each of NYC’s 683,788 street trees needed to be associated with their respective street segments for more than 119K street segments in NYC. After associating each tree with its respective street, a tree canopy coverage score was computed for each street segment. This was done by summing over each tree’s diameter at breast height on a given street segment and then dividing by the street segment’s length. This process resulted in a measurement I call STC, short for Street Tree Coverage. Finally, I divided the STC into five equal probability groups. These groups provide a relative measurement of the street tree coverage across the city. In this way, the top 20% of STC were rated as Very High Street Tree Coverage. In contrast, the lowest 20% of streets were grouped as Very Low STC. Streets without trees were categorized as No Trees.
Here’s a zoomed-in map of Fort Greene, Brooklyn. The three streets discussed at the beginning of this post are marked on the map as: 1) for South Portland Street with Very High Coverage, 2) Fulton Street with Medium Coverage. And 3) for Lafayette Street’s Low Coverage.
The benefit of this metric is that it can provide a fine-grained understanding of a city’s tree canopy distribution. In some cities this metric can be calculated solely relying on Open Data that is already collected. This means large scale neighborhood tree canopy coverage studies can be conducted without investing large funding and without compromising accuracy.
As many cities struggle with limited budgets and lack the data science expertise, this metric and others can improve many cities’ decision making process. These sort of metrics can make cities more efficient without having them spend substantial amounts of money or time. Metrics such as these can help cities stimulate economic activity and raise real estate values. Finally, cities can, with the aid of data science, can better allocate capital investments based on which neighborhoods are underserved for specific city amenities.
About the Author:
Avigail Vantu is an urban data scientist. In 2016, she received her MS from New York University. Avigail holds a BA in Economics and Political Science from the Hebrew University of Jerusalem. Currently Avigail is working on her Public and Urban Policy doctorate at the New School and is an adjunct faculty at the New York University Tandon School of Engineering. There she teaches a computing for sustainable urban environments class. Avigail also works on several applied urban data science and analytics projects with start-up companies and academic groups.
Avigail is eager to form new collaborations with city agencies and start-ups to solve complicated urban data questions firstname.lastname@example.org
- This project was initially developed to follow a book chapter co-written by Dr. Kristen Day and the author titled “Using Big Data to Support Public Space Research”. The chapter is currently under review and is planned to be published by Routledge Companions in 2020 as part of the textbook Companion to Public Spaces (editors: Mehta, V., & Palazzo,. D).