Week 1: Discussion Section

Rapa Nui (Easter Island or Isla de Pascua) in Chile

In this section, we’ll kick-off our mapmaking skills using the tmap package. Your task will be to make a map of Rapa Nui (Easter Island or Isla de Pascua). Rapa Nui is a volcanic island in Polynesia and territory of Chile. It is widely known for monumental statues called moai, but it is also a place of geoscientific inquiry. Several research programs about the island’s volcanic history, petrology, and human-environment interactions exist.

Source Materials

The following materials are modified from the tmap book.

Learning Objectives

• Read in spatial objects with st_read() and read_stars()
• Create map with single spatial object
• Create map with multiple spatial objects
• Use different types of tmap plotting formats (e.g. tm_polygons(), tm_fill(), tm_dots(), etc.)
• Include essential map elements (e.g. scale bar & north arrow or graticules)
• Create an interactive map
• Challenge: Reproduce map using ggplot2 instead of tmap

1. Get Started

• Create a version-controlled R Project
• Add (at least) a subfolder to your R project: data
• Create a Quarto document

Let’s load all necessary packages:

library(here)
library(tidyverse)
library(sf)
library(stars)
library(tmap)

Next, let’s download some data from Rapa Nui. Unzip and move this to your version-controlled R Project’s data folder.

• Read in the spatial objects for Rapa Nui:
  • ei_points: file contains several points on the island
    • Use the pipe operator to filter() output for points representing volcanoes
  • ei_elev: raster with elevation data
  • ei_borders: polygon with the island outline
  • ei_roads: lines contains a road network for the island

ei_points <- st_read(here("data", "easter_island", "ei_points.gpkg")) |> 
  filter(type == "volcano")
ei_elev <- read_stars(here("data", "easter_island", "ei_elev.tif"))
ei_borders <- st_read(here("data", "easter_island", "ei_border.gpkg"))
ei_roads <- st_read(here("data", "easter_island", "ei_roads.gpkg"))

2. Create a single-object map with tmap

• Create a map of the outline Rapa Nui using tm_shape() + tm_borders()

Solution

# Add Easter Island vector in default gray
tm_shape(ei_borders) +
  tm_borders() +
# Set general map layout options
  tm_layout(main.title = "Easter Island")

3. Create a multiple-object map with tmap

Expand on your single-object map to create a map of Rapa Nui and…

• …denote the island’s borders and continuous elevation
• …denote the island’s volcanoes and roads
• …play with the color palette and essential map elements

Solution

ei_map <- tm_shape(ei_elev) +
  tm_graticules() +
  tm_raster(col.scale = tm_scale_continuous(values = "-rd_yl_gn"),
            col.legend = tm_legend("Elevation (m asl)")) +
  tm_shape(ei_borders) + 
    tm_borders() +
  tm_shape(ei_roads) +
    tm_lines() +
  tm_shape(ei_points) +
    tm_symbols(shape = 24, 
               size = "elevation", 
               size.legend = tm_legend("Volcanoes (m asl)")) +
  tm_compass(position = c("right", "top")) +
  tm_scalebar() +
  tm_title("Easter Island")

ei_map

4. Create an interactive map with tmap

• Switch from the default "plot" mode to "view"

Solution

tmap_mode("view") # Convert default static map to interactive map 

ℹ tmap mode set to "view".

ei_map

5. Create a multiple-object map with ggplot2

More map making in R

There are MANY different ways to make maps in R, all with different pros/cons. Check out this resource for examples of what map making looks like in ggplot2, leaflet, and more!

As a bonus challenge, reproduce your multiple-object map from earlier using ggplot2!

Solution

# extract lat & long from geom column 
volcanoes_point <- ei_points %>%
  mutate(lon = unlist(map(ei_points$geom,1)), # longitude = first component (x)
         lat = unlist(map(ei_points$geom,2))) # latitude = second component (y)

ggplot() +
  # first add Easter Isand borders
  geom_sf(data = ei_borders, color = "#212529") +
  # add elevation raster
  geom_stars(data = ei_elev) +
  scale_fill_distiller(name = "Elevation (m asl)", # legend title with units
                       palette = "RdYlGn",         # red-yellow-green palette (no need to reverse)
                       na.value = "lightblue") +  # set NA color (background ocean)
  # add road vector
  geom_sf(data = ei_roads, color = "#343a40") +
  # add volcano vector
  geom_point(data = volcanoes_point, 
             aes(x = lon, y = lat, 
                 size = elevation), # point size depends on attribute value
             shape = 17,            # triangle
             color = "#22577a") +
  scale_size_continuous(name = "Volcanoes (m asl)") +       # legend title with units
  ggspatial::annotation_north_arrow(location = "br",        # bottom right
                                    which_north = "true") + # point to north pole
  ggspatial::annotation_scale(location = "bl",              # bottom left
                              width_hint = 0.5) +           # proportion of map area the scalebar should occupy
  labs(title = "Easter Island") +
  theme_minimal()