Exercise 3_1: Mapping

Introduction to Geospatial Techniques for Social Scientists in R

Exercises

We’ve finally arrived at creating our first maps! This exercise focuses on ggplot2 as our primary tool — from a first look at the data all the way to a multi-layer customized map.

Note🏋 Exercise 1

Load attributes_cologne.shp using sf::read_sf(). Inspect the attribute table — which variables does the data contain? Choose two columns that interest you and create one choropleth map for each.

What do the two maps tell you? Can you spot any spatial patterns?

Caution💡 Tip

# inspect the attribute table
attributes_cologne |>
  sf::st_drop_geometry() |>
  dplyr::glimpse()

Note🏋 Exercise 2

Pick the variable from Exercise 1 you find most interesting. Change and add layers:

• Add the EV charging stations from charger_cologne.shp as a point layer on top
• Try different (colorblind-friendly) palettes.
• Change the shape of the charging station points.
• Map point size to charging power (column name powr_kw)
• Experiment with legend position
• Experiment with the thme
• Save the result as a PNG with ggsave()

Caution💡 Tip

• Palette options for scale_fill_viridis_c(): option = "plasma", "magma", "viridis", "cividis"
• Point shape: set shape = outside aes() with a number 0–25 (run ?pch for a visual overview)
• Size by power: aes(size = powr_kw) inside geom_sf(), then scale_size_continuous(name = "Power (kW)")
• Legend position: theme(legend.position = "bottom") — also try "left", "top", or c(0.1, 0.9)

Note🏋 Exercise 3

Display two of your maps next to each other for comparison. There are several ways to do this both within the ggplot argument and by using the map objects. Try at least one!

Caution💡 Tip

• facet_wrap: pivot first with tidyr::pivot_longer(), then facet_wrap(~ variable) — note: forces a shared color scale for both panels
• patchwork: library(patchwork) → map1 | map2
• cowplot: cowplot::plot_grid(map1, map2, ncol = 2, labels = c("A", "B"))

Solutions

Tip✅ Solution 1

library(sf)
library(dplyr)
library(ggplot2)
getwd()
attributes_cologne <- sf::read_sf("../../data/attributes_cologne.shp")

# inspect the attribute table
attributes_cologne |>
  sf::st_drop_geometry() |>
  dplyr::glimpse()

# two example maps — your variable choices will differ!
map_ecar <-
  ggplot() +
  geom_sf(
    data  = attributes_cologne,
    aes(fill = ecar),
    color = NA
  ) +
  scale_fill_viridis_c(
    option    = "plasma",
    direction = -1,
    name      = "E-Car Share"
  ) +
  theme_void() +
  labs(title = "Electric Cars")

map_greens <-
  ggplot() +
  geom_sf(
    data  = attributes_cologne,
    aes(fill = greens),
    color = NA
  ) +
  scale_fill_viridis_c(
    option    = "viridis",
    direction = -1,
    name      = "Vote Share"
  ) +
  theme_void() +
  labs(title = "Green Party Vote Share")

map_ecar
map_greens

Warning: Paket 'sf' wurde unter R Version 4.4.3 erstellt

Linking to GEOS 3.13.0, GDAL 3.10.1, PROJ 9.5.1; sf_use_s2() is TRUE

Warning: Paket 'dplyr' wurde unter R Version 4.4.3 erstellt

Attache Paket: 'dplyr'

Die folgenden Objekte sind maskiert von 'package:stats':

    filter, lag

Die folgenden Objekte sind maskiert von 'package:base':

    intersect, setdiff, setequal, union

Warning: Paket 'ggplot2' wurde unter R Version 4.4.3 erstellt

Tip✅ Solution 2

charger_cologne <- sf::read_sf("../../data/charger_cologne.shp")

my_map <-
  ggplot() +
  geom_sf(
    data  = attributes_cologne,
    aes(fill = ecar),
    color = NA
  ) +
  scale_fill_viridis_c(
    option    = "magma",
    direction = -1,
    name      = "E-Car Share"
  ) +
  geom_sf(
    data  = charger_cologne,
    # map size to charging power in kW
    aes(size = powr_kw),
    color = "black",
    shape = 17,    # filled triangle — try other numbers!
    alpha = .5
  ) +
  scale_size_continuous(name = "Power (kW)") +
  theme_void() +
  theme(legend.position = "bottom") +
  labs(
    title   = "Electric Cars & Charging Stations in Cologne",
    caption = "© Stadt Köln; Bundesnetzagentur"
  )

my_map

ggsave("my_map.png", my_map, dpi = 300)

Tip✅ Solution 3

# --- Option 1: patchwork ---
library(patchwork)

# combine maps with | (side by side) or / (stacked)
map_ecar | map_greens

# add a shared title
(map_ecar | map_greens) +
  patchwork::plot_annotation(
    title = "Electric Cars and Green Vote Share in Cologne"
  )

# --- Option 2: cowplot ---
library(cowplot)

cowplot::plot_grid(map_ecar, map_greens, ncol = 2, labels = c("A", "B"))

# --- Option 3: facet_wrap ---
library(tidyr)

# pivot two variables to long format first
cologne_long <-
  attributes_cologne |>
  tidyr::pivot_longer(
    cols      = c(cdu, greens),
    names_to  = "variable",
    values_to = "value"
  )

# facet_wrap uses the same color scale for all panels
ggplot() +
  geom_sf(
    data  = cologne_long,
    aes(fill = value),
    color = NA
  ) +
  scale_fill_viridis_c(option = "plasma", direction = -1) +
  facet_wrap(~ variable) +
  theme_void()

Warning: Paket 'patchwork' wurde unter R Version 4.4.3 erstellt