Trees are often considered in urbanist circles to be a vital component of city planning. They provide vital shade (cooling areas down by an average of 3º F (1.6ºC) in comparison with urban non-green areas, according to the EPA) in turn decreasing demand for air conditioning and reducing energy use, provide habitat for wildlife, improve the aesthetic quality of neighborhoods, in addition to many more community benefits.

NYC’s Open Data site includes a dataset from the 2015 Street Tree Census, a volunteer and staff conducted count of all street tree in the 5 boroughs, documenting attributes such as species, diameter, perception of health, types of damage, and more.

When I first brought the dataset into tableau, I quickly realized the sheer number of trees was too high to discern any meaning from the map. So I narrowed the area down to just Brooklyn.

This narrowing showed that within Brooklyn there are 132 different catalogued species of street trees.

I also tried filtering the species count down to the top 10 most common and bottom 10 least common types of street trees. On first glance, my main takeaway from this data is that there are entirely too many Callery pear trees (a.k.a Bradford pear, the invasive ornamental tree notorious for its noxious odor often described as akin to rotting fish or bodily fluids. This odor attracts flies, the primary pollinator for the tree.)
There are virtually no benefits to the Callery pear as a street tree, as plenty of other white flowering trees exist that do not possess an odor that is near impossible to inhale without gagging. For instance, flowering dogwoods or cherry trees. Additionally, Callery pears are not edible (for humans, at least), adding to their uselessness.

Another dataset along similar lines I found interesting was the one from the Falling Fruit project, a user contributed map of edible, forageable fruit across the world. This one also had large amounts of data, so I’ve filtered it down here by season. The dataset also includes comments from contributors, which is often helpful.

I feel like there are more interesting analyses that can be made from both these datasets, and I hope to continue working to figure out what those might be.