San Francisco is always changing. A formerly wind-swept, grass and sand covered peninsula has completely transformed into a wind-swept, tree and sand covered peninsula.
While San Francisco’s trees are varied, captivating, and omnipresent, they’re a relatively new resident of this fine dune.
Urban Forest Map is undertaking the immense task of collecting data on every tree in San Francisco. They’ve released fresh data as of a couple days ago for the city’s western neighborhoods. The dataset already includes over 15,000 trees and will eventually cover the entire city.
I took a quick look at the most common trees in their database alongside another dataset of plant species native to San Francisco.
T(h)ree things are immediately obvious.
1: Tree names are amazing. Once the hard work of producing a scientific name is out of the way it’s all fun and games.
2: Very few of the trees in this part of San Francisco are native. I assume this will hold true as the data collection expands.
3: New Zealand Christmas trees! What a name. What abundance. These are those trees that produce beautiful red bristly flower crowns. Not to be confused with bottle brush trees. In addition to being nice looking they are apparently a misguided 1980’s beautification scheme turned sidewalk and sewer destroyer.
I really love this data and wish I had more time to play with it. Luckily, Urban Forest Map provides a ton of great information about the trees that can be easily dug in to. Don’t miss the actual map for zeroing in on specific trees.
Notes on the data
A quick way to make sense of a big dataset is to “group” on interesting columns. That will give you a count of how many occurences of each unique value occur in that column, which is the bulk of how I produced the included chart. I did that in a Numeracy project here. You can use this project as a starting point for your own analysis.
Whether a species is native or not is a nuanced scientific topic. Calscape publishes the most reasonable dataset I could find. I'd love to hear from you if you have thoughts about this choice. It’d be very easy to swap out that dataset for trees native to California, the west coast, or the United States if you wanted a broader view.
The common names used in this analysis are from the Urban Forest Map dataset. I’ve noticed seemingly minor differences between datasets but don't have an educated opinion about the most agreed-upon names.