Plotting with ggplot2

Objectives

• Be able to create simple plots with ggplot2 and tweak them

• Know how to look for help

• Reading data into R from disk or web

• Know how to tweak example plots from a gallery for your own purpose

• We will build up an equivalent notebook to the original Python notebook

Repeatability/reproducibility

From Claus O. Wilke: “Fundamentals of Data Visualization”:

One thing I have learned over the years is that automation is your friend. I think figures should be autogenerated as part of the data analysis pipeline (which should also be automated), and they should come out of the pipeline ready to be sent to the printer, no manual post-processing needed.

• Try to minimize manual post-processing. Automated visualization prevents last-minute regeneration panic.

• No single library or language is perfect for every use-case.

• In R, common visualization libraries include:

  • ggplot2: grammar of graphics

  • lattice: grid-based graphics

  • plotly: interactive visualizations

  • highcharter: interactive visualizations based on Highcharts

  • leaflet: interactive maps

  • shiny: interactive web apps

  • base R: traditional graphics system

• Two main visualization paradigms: procedural (base R) and declarative (ggplot2).

Why are we starting with ggplot2?

• Concise, powerful, and consistent grammar of graphics

• Intuitive mapping of visual aesthetics to data

• Easily integrates with data manipulation libraries like dplyr

• Well-documented, widely adopted, and open source

• Simple to export to multiple formats (png, svg, pdf)

Example data: Weather data from two Norwegian cities

We will use weather data from Norsk KlimaServiceSenter, Meteorologisk institutt (MET) (CC BY 4.0). The data is in CSV format for Oslo and Tromsø:

• View data repository

We’ll load CSV files using read.csv() or directly from the web with read.csv(url).

Dataframes in R are excellent for tabular data and are directly usable with ggplot2.

Reading data into a dataframe

Let’s combine monthly data from two CSV files:

url_prefix <- "https://raw.githubusercontent.com/coderefinery/data-visualization-python/main/data/"

data_tromso <- read.csv(paste0(url_prefix, "tromso-monthly.csv"))
data_oslo <- read.csv(paste0(url_prefix, "oslo-monthly.csv"))

data_monthly <- rbind(data_tromso, data_oslo)

# view combined data
data_monthly

Fixing date format:

data_monthly$date <- as.Date(paste0("01.", data_monthly$date), format="%d.%m.%Y")

What is in a dataframe?

Dataframes contain rows and columns:

Explore interactively in R:

head(data_monthly)
tail(data_monthly)
names(data_monthly)
str(data_monthly)
dim(data_monthly)
data_monthly$max.temperature
summary(data_monthly$max.temperature)
max(data_monthly$max.temperature)
subset(data_monthly, max.temperature > 20)

Where to learn more about dataframes in R

• R for Data Science

• Quick-R

• Data Carpentry R lessons

Plotting the data

Simple plot:

library(ggplot2)

ggplot(data_monthly, aes(x = date, y = precipitation, fill = name)) +
  geom_bar(stat = "identity", position = "dodge")

Improve temporal axis:

ggplot(data_monthly, aes(x = date, y = precipitation, fill = name)) +
  geom_bar(stat = "identity", position = "dodge") +
  scale_x_date(date_labels = "%Y-%m")

Plot side-by-side panels:

ggplot(data_monthly, aes(x = date, y = precipitation, fill = name)) +
  geom_bar(stat = "identity") +
  facet_wrap(~name)

Stacked bars:

ggplot(data_monthly, aes(x = date, y = precipitation, fill = name)) +
  geom_bar(stat = "identity")

Temperature range area plot:

ggplot(data_monthly, aes(x = date, ymin = min.temperature, ymax = max.temperature, fill = name)) +
  geom_ribbon(alpha = 0.5)

Exercise: Using visual channels to re-arrange plots

• Switch axes:

ggplot(data_monthly, aes(y = date, x = precipitation, fill = name)) +
  geom_bar(stat = "identity") + facet_wrap(~name)

• Change axis title:

ggplot(data_monthly, aes(y = date, x = precipitation, fill = name)) +
  geom_bar(stat = "identity") + facet_wrap(~name) +
  labs(x = "Precipitation (mm)")

• Side-by-side temperature plots:

ggplot(data_monthly, aes(x = date, ymin = min.temperature, ymax = max.temperature)) +
  geom_ribbon(alpha = 0.5) + facet_wrap(~name)

More fun with visual channels

Read daily data:

data_tromso_daily <- read.csv(paste0(url_prefix, "tromso-daily.csv"))
data_oslo_daily <- read.csv(paste0(url_prefix, "oslo-daily.csv"))
data_daily <- rbind(data_tromso_daily, data_oslo_daily)
data_daily$date <- as.Date(data_daily$date, format="%d.%m.%Y")
data_daily <- subset(data_daily, date > "2022-12-01" & date < "2023-05-01")

Snow depth plot:

ggplot(data_daily, aes(x = date, y = snow.depth, fill = name)) +
  geom_bar(stat = "identity") + facet_wrap(~name)

Color by temperature:

ggplot(data_daily, aes(x = date, y = snow.depth, color = max.temperature)) +
  geom_point() + facet_wrap(~name)

Change color scheme:

ggplot(data_daily, aes(x = date, y = snow.depth, color = max.temperature)) +
  geom_point() + facet_wrap(~name) + scale_color_viridis_c(option="plasma")

Themes

# dark theme
theme_set(theme_dark())

Other themes:

• theme_classic()

• theme_minimal()

• theme_bw()

Exercise: Anscombe’s quartet

data <- read.csv("example.csv")

ggplot(data, aes(x=x, y=y, color=dataset)) +
  geom_point() + facet_wrap(~dataset)

---

Keypoints

• Browse example galleries for inspiration.

• Minimize manual post-processing.

• CSV files are good for storing data for visualization.

• Load data into R dataframes and plot using ggplot2.