Street Trees in San Francisco, CA

Library imports:

library(ggplot2)
library(readr)
library(knitr)
library(magrittr)
library(lubridate)
library(dplyr)

Data imports:

library(readr)
df <- read_csv("Street_Tree_List.csv", 
    col_types = cols(PlantDate = col_datetime(format = "%m/%d/%Y %H:%M:%S AM")))

Let’s take a quick look at the data, isolating the variables we will be using:

simpledf <- df %>% select(PlantType, qSpecies, DBH, PlantDate)
kable(head(simpledf))

Question 1: What are the most common tree species on San Francisco streets and in what quantity?

First we will see how many total species are cataloged in this dataset.

length(unique(simpledf$qSpecies))

## [1] 565

That’s a lot of species!

Next, we can take a glimpse at the frequency of each planted tree species in SF streets.

x <- as.data.frame(table(simpledf$qSpecies), na.rm = TRUE)
kable(head(x))

Looking at the data in x we can see that the most common species is Sycamore: London Plane with 11,489 tree individuals - impressive. Next, we will import the SF_Top_Ten Dataset which soley includes the top ten species, their common names, and frequency counts.

SF_Top_Ten <- read_csv("SF_Top_Ten.csv")
kable(head(SF_Top_Ten))

Using this we can construct a barplot of the top ten tree species and their frequency.

ggplot(data= SF_Top_Ten, aes(x= reorder(Species, -Frequency), y = Frequency)) +
   geom_bar(stat="identity", fill="steelblue")+
  theme_minimal()+
  labs(x= "Tree species", y= "Number of tree individuals", title= "Top ten tree species planted on the streets of San Francisco") + theme(axis.text.x = element_text(angle = 90, hjust = 1)) 

Looking at this graph, we see that there are nearly 12,000 London Plane Sycamore trees planted in SF streets. There is a significant drop in quantity moving to the second most frequent tree, the New Zealand Christmas tree.

Question 2: How do the DBH (diameter at breast height) values of Juniperus chinensis species change with age?

We are going to filter the simpledf data to isolate only the Juniperus chinesis records. We will also remove any NA values from the DBH and PlantDate column.

Juniperdf <- simpledf %>% filter(qSpecies == "Juniperus chinensis :: Juniper")
Juniperdf %>% na.omit()

From here we will find the elapsed time between planting date and today’s date.

Juniperdf$age <- today() - as.Date(Juniperdf$PlantDate)

ggplot() +
  geom_point(data= Juniperdf, mapping=aes(x=age, y=DBH), color="darkgreen")+
  labs(y= "DBH in cm", x="Age in days", title= "Change in DBH as Juniper ages") +
  ylim(0,30) 

It’s important to note that 5,000 days is roughly 13.5 years old and 15,000 days is around 42 years old. We can see that there is a slight upwards trend in DBH as Juniper ages, but most trees don’t grow over 15 cm in diameter at breast height.

Next, we can also look at the distribution of DBH values for our Juniper species to see where most diameters fall.

ggplot()+
  geom_histogram(data=Juniperdf, mapping= aes(x= DBH), color="darkblue", fill="lightblue")+
  labs(title= "Diameter at Breast Height (cm) distribution for Juniper", y= "Count", x="DBH (cm)")

This graph shows us that the majority of planted Juniper trees are roughly 17cm diameter or less, but there are some outliers extending to above 60cm in diameter.

Question 3: How do the DBH values compare for the top three most prevalent species planted?

To do this analysis, we must first filter our data so we isolate the top three species - London Plane Sycamore, New Zealand Christmas tree, and Brisbane Box. From there we can create a facet grid, comparing the DBH values of these species.

topthree <- simpledf %>% filter(qSpecies== "Platanus x hispanica :: Sycamore: London Plane" | qSpecies== "Metrosideros excelsa :: New Zealand Xmas Tree"| qSpecies=="Lophostemon confertus :: Brisbane Box")

ggplot()+
  geom_histogram(data= topthree, 
                 mapping = aes(x=DBH, fill=qSpecies), bins= 20) +
  facet_grid(. ~ qSpecies) +
  scale_x_log10() +
  labs(title= "Distribution of DBHs for the top three most common species in SF streets", y="Tree individuals", x= "DBH in cm")+
  scale_fill_discrete(name = "Species")