Mapping Urban Trees and Income in Manhattan

Caitlin Dempsey

Updated:

The collective inventory of trees in a city is known as the urban forest.  The urban forest contains trees found on private property, park locations, and trees that line the streets (known as street trees).  

Benefits of trees in cities

The presence of trees in an urban environment provide innumerable benefits (Dwyer et. al, 1992).  

Urban trees provide shade and serve as a windbreak, reducing heating costs in the winter and cooling costs in the summer.  Trees slow storm water runoff, allowing valuable rainfall to percolate down and replenish groundwater sources.  Trees help to pull in air pollution and serve as a sink for carbon dioxide.  

Trees also provide many therapeutic benefits; the presence of trees has been demonstrated to reduce stress.  

Trees improve property values, slow traffic (making it safer), and reduce noise.  



Free weekly newsletter

Fill out your e-mail address to receive our newsletter!
Email:  

By entering your email address you agree to receive our newsletter and agree with our privacy policy.
You may unsubscribe at any time.



Trees are an investment

Despite the many benefits of a well-developed urban canopy, trees are also an investment requiring time and money and many cities have disparities in the distribution of trees that can be tied to the relative wealth of the neighborhoods.

A study by Zhu and Zhang (2008) that surveyed urban areas across the United States found that for every 1 percent increase in per capita income, demand for forest cover increased by 1.76 percent.  

This article takes a look at the density of street trees in Manhattan in New York City.

Open tree data for New York City

Every ten years, the city of New York inventories all of the street trees across the five boroughs (Figure 1) that make up the city: Manhattan, Staten Island, the Bronx, Brooklyn, and Queens. 

Screenshot of the data portal for the 2015 tree census done by NYC.
Data from the 2015 Street Tree Census is available from the NYC Open Data portal.

This database is a census only of street trees and therefore doesn’t contain tree information for park trees or trees on private property.  Over 2,600 citizen volunteers worked with the city’s Park staff to collect street tree locations along with information about each tree’s health, size, species, and numerous other attributes about the tree. Also included with each tree record are the geographic coordinates.   

This database of over 646,000 trees is made freely available to the public via its OpenNYC data portal.  This article looks at the street trees for Manhattan, a total of 65,423 trees.  

Map with the NYC borough of Manhattan in red with the other four Burroughs in medium yellow.  The surrounding areas are in light gray and the ocean is blue.
Map showing Manhattan in red. Map: Caitlin Dempsey.

Using QGIS, an open source GIS software program, a heat map of the trees in Manhattan was created. This map shows the density of street trees per 60 x 60 foot cells in the borough of Manhattan.  

Since the database doesn’t cover private and park trees, there are several deceptively low tree densities on the map.  For example, the white area in the middle of Manhattan is Central Park.  

Overall, the heat map shows a strong street tree coverage across Manhattan, with pockets of high density along the streets bordering Central Park and along the major route (Westside Highway) that parallels the Hudson River (wider streets allow for a high frequency of street trees due to medians).  

Heat map showing densities of trees in Manhattan in shades of green.  The darker green is a higher density of trees. The surrounding areas are light gray for New Jersey and medium yellow for other areas of NYC.
Heat map showing the density of street trees across Manhattan from NYC’s 2015 tree census. Map: Caitlin Dempsey.

Mapping Manhattan’s median income using US Census Bureau data

For income data was mapped using 2015 median household income data for Manhattan at the blockgroup level from the U.S. Census Bureau.  The blockgroup is the smallest geography that the U.S. Census offers income data for (they are restricted due to privacy concerns from offering income at the smallest census geography which is the block).  

The map shows the disparity of incomes with the Upper East side (surrounding Central Park), Upper West Side, and Greenwich Village as some of the highest areas of income.  

A choropleth map with shades of blue for median income divided by quartiles for Manhattan.
Map showing Median Household Income (in US $) categorized by quartiles. Data from U.S. Census, 2015. Map: Caitlin Dempsey.

Analyzing the relationship between tree density and median income

Since both datasets contain geolocation information, a spatial join was performed using QGIS to assign the median household income value to each tree.

A figure showing two tables with red arrows pointing from a map to the two tables.
The attributes from both the tree and income datasets were spatially joined. Image: Caitlin Dempsey.

After joining, the data was filtered in order to examine potential correlations between tree species, density, size, health, stewardship, and income.

To simplify the analysis and to be able to compare tree characteristics between the highest and lowest income levels, the median household income levels were divided into quartiles.

Median household income ranges – division of quartiles:

  • Up to $27,471 – Income group 1
  • Between $27,472 and $63,556 – Income group 2
  • Between $63,557 and $107,065 – Income group 3
  • Between $107,066 and $340,545 – Income group 4

Nine years ago, New York City had embarked on an ambitious MillionTreesNYC campaign which is an effort to infill both private and public spots within the city with trees in order to increase the urban canopy.  The campaign recently announced that the 1 millionth tree was recently planted.  

Plots of both tree density and tree size against income have demonstrated that NYC officials have been very successful at creating an equitable distribution of trees across the city.  In this first graph below, the R2 value demonstrates almost no statistical correlation between tree distribution and income.  

Graph plotting the density of trees against household income in Manhattan.  The background of the graph is light gray, the plots are blue dots, and the lines are black.
Plot of tree density against income by Census block group in Manhattan. The R2 value of 0.06626 indicates that the correlation between income and tree density is very low. Graph: Caitlin Dempsey.

This lack of correlation is supported by the pie chart below which shows the distribution of street levels across the income quartile groups.

Pie chart showing number of trees is equal for each median income quartile in Manhattan.  The colors on the pie chart are dark yellow, medium blue, medium gray, and dark orange.
Number of street trees by income group quartile in Manhattan. Graph: Caitlin Dempsey.

Number of tree species by median income in Manhattan

Where disparities among the income groups start to emerge is when looking at tree species.  

The borough of Manhattan has over 150 unique tree species. The species diversity across the income groups is fairly equitable:

  • First quartile has 109 unique species.  
  • Second quartile has 116 unique species.  
  • Third quartile has 101 unique species  
  • Fourth quartile has 102 unique species.

 However, a look at the dominant species within each income group starts to reveal discrepancies.  This graph visualizes the 16 most numerous tree species based on the proportion that falls within each income group.

Dark yellow, grey, orange, and blue bar chart for the dominant tree species within each income group.
Tree species dominance within each of the four income groups. Graph: Caitlin Dempsey.

While some of the top species have a fairly equitable distribution (such as Littleleaf Linden and American Linden) other species have wide variations in distribution.  For example, almost half of the American elm trees (45%) in Manhattan are located in the lowest income areas.  

Similarly, 40% of Northern red oak trees are located in the lowest income areas.  Gingko is mostly found in the upper income areas with 35% of those trees found in the highest income areas and an additional 28% located in the second highest income areas.  

The most populous tree in Manhattan is the Honey Locust.  This tree is popular thanks to its brilliant fall foliage, fast growth, ability to withstand full sun, and can grow in a range of soils.  This tree is the dominant species regardless of income quartile, making up 16% of all trees in the lowest income quartile and 20% of trees in the highest income quartile.  

Discrepancies in dominant tree species start to emerge when looking at the rest of the top five trees between the highest and lowest income groups.  

Graphic showing the top five tree species in the highest and lowest income groups in Manhattan.  The graphic is designed to look like two trees.
The top five most populous trees for Manhattan’s lowest income areas (L) and high income areas (R). Graphic: Caitlin Dempsey

Native to Vietnam and China, Callery Pear trees make up the second most dominant tree in higher income areas of Manhattan.  These trees produce a beautiful although unfortunate smelling white blossoms in the spring.  The Callery Pear has a very short life span, requiring it to be replaced every 15-20 years.

The pin oak is the second most dominant tree in lower income areas.  A fast growing tree, pin oaks are valued for their pollution tolerance and ease of transplant, making them hardy urban trees.  

The third dominant tree in higher income neighborhoods is the Ginkgo tree.  Native to China, this tree is valued for its incredibly long life (some live over 2,000 years) and resistance to disease and insects.  The erratic branches of this tree mean that it requires more care than other species.   

The third most populous tree in lower income is the London Planetree.  This tree is favored for its ability to absorb air pollution.  

The fourth most dominant tree in high income is the Japanese pagodatree which is highly tolerant of urban pollution.  Native to Europe, the Little Leaf Linden tree is the fifth most populous trees in the higher income areas.  This medium sized tree is tolerant to urban pollution and can live to 1,000 years.  

The reason for the differences in dominant tree species between high and low income areas is unclear.  All of these tree species represent a mix of native and non-native species as well as a range of life spans.

Geography of stewardship of trees in Manhattan

The sharpest differences between low and high income areas when it comes to trees lies in the care of the trees.  Income was a stronger predictor of care within the neighborhood. 

The NYC Parks and Recreation department relies on a citizen-based tree-care program to help care for its street trees.  Stewards are volunteers that tend to the health of street maps, ensuring their longevity. Stewards are relied upon to help water, beautify, and nurture street trees to promote maximum longevity and health of those trees.  

Those households with the lowest median incomes were the most likely to have no stewards assigned to care for the street trees in their neighborhoods.   Trees in higher income are significantly more likely to benefit from multiple stewards than trees in lower income areas.

Graph showing the number of stewards per tree broken down by income group in Manhattan.  Blue is the highest income, orange is the middle income and grey is the lowest income level.
Number of stewards per tree segmented by income group. Graph: Caitlin Dempsey.

Being a stewards require a commitment not only of time but also of money.  Stewards are expected to provide all of the mulch, fertilizer, water, and equipment needed to tend to trees. This expected financial investment would be a barrier for low-income families seeking to join the tree adoption program. 

A list of requirements for being a tree steward in NYC.
List of needed items to care for street trees. Source: MillionTreesNYC Tree Care Guide.

Presence of tree guards in Manhattan

Income was also a strong predictor for the presence of tree guards.  Tree guards are important to protect the tree against damage from cars, pedestrians walking on roots, and bicyclists attaching their bikes to tree trunks.

Neighborhoods at the lower end of the income range were the least likely to have tree guards with over half of Manhattan’s trees without them.  

The average income for neighborhoods without tree guards was $62,645.732 while the average income for neighborhoods with tree guards was $95,509.238.  

As with other forms of stewardship, tree guard purchase and installation is done at the investment of the area residents.  The bar graph below shows that income is a stronger predictor for whether a tree will have a tree guard.

A graph showing in blue if there is a tree guard and orange if there isn't a tree guard by income quartiles.
Distribution of tree guards segmented by income group. Graph: Caitlin Dempsey.

Conclusion

This analysis showed that while the NYC Park’s department has achieved a fairly equitable distribution of trees across its city, there is a great need for a more equitable strategy for the care of those trees.   

The adequate care of its street streets is an important long term strategy, particularly in light of NYC’s aggressive tree planting program.  The OneMillionTreeNYC program relies heavily on volunteer contributions to plant and care for the growing urban forestry in the city.  Funding and more investment by the city is needed to develop stewards that will care for trees across the income groups.  

As indicated by this pie chart, lower income areas already have a higher proportion of dead trees compared to the higher income areas. 

Pie chart showing the break down of dead trees in Manhattan by income group.
Distribution of dead trees by income group in Manhattan. Graph: Caitlin Dempsey.

Closing remarks about trees and income in Manhattan

This article was a cursory look at street trees in Manhattan and median income. A complete dataset of all the trees found within Manhattan (including all park trees and trees on private property) is not available.  

It would be interesting to see if there are indeed differences in the urban tree canopy between higher and lower income neighborhoods when private trees are also factored in to the study.  This results of that analysis could be used by NYC Parks in order to site more street trees in locations where private property and park trees are sparse.  

All trees are not valued equally and the analysis in this article does not address that disparity.  Smaller trees, poorly placed trees, and diseased trees don’t offer the same benefits as a well located, leafy, large tree.  

This analysis was limited to the information that could be easily extracted and quantified from the NYC street tree database.  More complex studies employ tree valuation and tree rating calculations to determine a tree’s value to an area (NYC tree valuation protocol, 2012).  

Given the disparity in available stewards for lower income areas, an analysis that look at the siting of trees that require less maintenance, are hardier, and longer-lived in areas lacking adequate stewards would be beneficial.  

References

Dwyer, J. F., McPherson, E. G., Schroeder, H. W., & Rowntree, R. A. (1992). Assessing the benefits and costs of the urban forest. Journal of Arboriculture18, 227-227. Retrieved from http://joa.isa-arbor.com/request.asp?JournalID=1&ArticleID=2515&Type=2

NYC tree valuation protocol.  (2012, March).  Retrieved from https://www.nycgovparks.org/pagefiles/187/New-York-City-Tree-Valuation-Protocol-2023__64270276d0a05.pdf

Segel, E., & Heer, J. (2010). Narrative visualization: Telling stories with data. IEEE Transactions on Visualization and Computer Graphics16(6), 1139-1148. Retrieved from http://vis.stanford.edu/files/2010-Narrative-InfoVis.pdf

Zhu, P., & Zhang, Y. (2008). Demand for urban forests in United States cities. Landscape and Urban Planning, 84 (3-4), 293-300 DOI: 10.1016/j.landurbplan.2007.09.005

Photo of author
About the author
Caitlin Dempsey
Caitlin Dempsey is the editor of Geography Realm and holds a master's degree in Geography from UCLA as well as a Master of Library and Information Science (MLIS) from SJSU.