8  Choropleth Maps

Choropleth maps show themes based on predefined regions (e.g., countries, states, counties). Colors, or shades of a single color, are used to differentiate areas. The color of a region corresponds to a data characteristic or range of data values.

8.1 Initialize the Libraries and Data

There are a few things needed to setup the environment.

The examples will make use of the boundaries for the Hawaiian Islands. These have been stored in a (local) folder. The outlines are retrieved here.

Show the code 
## Set an option for the read_csv function
options(readr.show_col_types = FALSE) ## Suppress warning msg

## Initialize the Standard Libraries
library(readr)
library(ggplot2)
library(dplyr)
library(gt)

## Install the sitemaps package (uncomment; run once)
## library(devtools)
## install_github("kimbridges/sitemaps")
## install_github("kimbridges/plainmaps")

## Sitemaps
library(sitemaps)
library(plainmaps)

## Map Libraries
library(mapdata)
library(sf)

## Initialize defaults for Sitemaps
column <- site_styles()
hide   <- site_google_hides()

## Themes
plain_background_theme <- theme(
  panel.border     = element_blank(), 
  panel.grid       = element_blank(),
  panel.background = element_blank(),
  axis.text        = element_blank(),
  axis.ticks       = element_blank(),
  axis.title       = element_blank()) 

island_map_theme <- theme(
  panel.border = element_rect(color     = "black", 
                              fill      = NA, 
                              linewidth = 0.7),   
  panel.grid     = element_blank(),
  panel.background = element_rect(fill = "lightsteelblue1"),
  axis.text      = element_blank(), 
  axis.ticks     = element_blank(), 
  axis.title     = element_blank())


## Read stored border data
oahu_simple <- read_csv(col_names = TRUE, 
                    show_col_types = FALSE,
                    file = "hawaii_400/oahu_outline.csv")

hawaii_simple <- read_csv(col_names = TRUE, 
                    show_col_types = FALSE,
                    file = "hawaii_400/hawaii_outline.csv")

maui_simple <- read_csv(col_names = TRUE, 
                    show_col_types = FALSE,
                    file = "hawaii_400/maui_outline.csv")

kahoolawe_simple <- read_csv(col_names = TRUE, 
                    show_col_types = FALSE,
                    file = "hawaii_400/kahoolawe_outline.csv")

lanai_simple <- read_csv(col_names = TRUE, 
                    show_col_types = FALSE,
                    file = "hawaii_400/lanai_outline.csv")

molokai_simple <- read_csv(col_names = TRUE, 
                    show_col_types = FALSE,
                    file = "hawaii_400/molokai_outline.csv")

kauai_simple <- read_csv(col_names = TRUE, 
                    show_col_types = FALSE,
                    file = "hawaii_400/kauai_outline.csv")

niihau_simple <- read_csv(col_names = TRUE, 
                    show_col_types = FALSE,
                    file = "hawaii_400/niihau_outline.csv")

8.2 Choropleth Maps for Hawai`i

We can demonstrate a typical choropleth map using the political classification of the Hawaiian Islands.

Each island is a “predefined region.” The color code represents the county for each island.

Show the code 
## Assign islands to counties by using color.
fill <- NULL
fill$hawaii <- "darkseagreen3"
fill$maui   <- "goldenrod1" 
fill$oahu   <- "peachpuff3"
fill$kauai  <- "lightsalmon1"

## Plot each of the islands with its fill color.
ggplot() + 
  geom_polygon(data=hawaii_simple, aes(x=lon,y=lat),
               fill=fill$hawaii, 
               linewidth=0.4,
               color="black") +
  geom_polygon(data=oahu_simple, aes(x=lon,y=lat),
               fill=fill$oahu, 
               linewidth=0.4,
               color="black") +
  geom_polygon(data=lanai_simple, aes(x=lon,y=lat),
               fill=fill$maui, 
               linewidth=0.4,
               color="black") +
  geom_polygon(data=kauai_simple, aes(x=lon,y=lat),
               fill=fill$kauai, 
               linewidth=0.4,
               color="black") +
  geom_polygon(data=niihau_simple, aes(x=lon,y=lat),
               fill=fill$kauai, 
               linewidth=0.4,
               color="black") +
  geom_polygon(data=kahoolawe_simple, aes(x=lon,y=lat),
               fill=fill$maui, 
               linewidth=0.4,
               color="black") +
  geom_polygon(data=maui_simple, aes(x=lon,y=lat),
               fill=fill$maui, 
               linewidth=0.4,
               color="black") +
  geom_polygon(data=molokai_simple, aes(x=lon,y=lat),
               fill=fill$maui, 
               linewidth=0.4,
               color="black") +
  coord_fixed(1.1) +
  island_map_theme

8.3 Using Data to Assign Colors

Note that in this example, the colors are not assigned directly in the ggplot step. Instead, data values point to specific items in a color table.

This approach is especially valuable when the color_index values are computed. Here, these values need to be integers that correspond to the range of the items in the color table.

This turns the procedure into a more useful tool for displaying choropleth maps. Note the following characteristics:

  • The color are given with an index value (called quartile here). It is straightforward to change the colors in just this one place.

  • There are four quartile values. This is because a quantitative value (density) is being used as the basis of the colors. The division is into quartiles (four values based on an order list divided into 25% groups).

  • The population density is calculated.

  • The code for the data table, used for data confirmation, looks very long. However, there are formatting elements shown here for things like numeric formats and footnotes. It is a good habit to make complete (i.e., well annotated) data tables. Usually, it’s just a matter of copying the formatting code used in other tables.

  • A new variable (quartile) is calculate using Sitemap’s site_cuts function. You can get more information on this function in the Site Maps document.

  • The color for each island is merged into the basic data table by matching the common values for quartile.

  • The code used in the ggplot function is simplified by using the update_geom_defaults function to set values that will be used in each of the geom_polygon functions.

  • The color for each island’s fill value is retrieved by indexing rows the the island name.

  • The legend is added with a scale_fill_manual function. The values for this are initialized in the Data and Parameter Input section.

All of these steps make a generalizable procedure that can be used for other chloropleth plots.

The code is divided into large sections to help identify the overall visualization strategy.

Show the code 
## Data and Parameter Input ################################

## Color table: List of colors and their index values (quartile).
color_table <- 
  read_csv(file =
    "color,   quartile
     beige,   1 
     wheat2,  2
     wheat3,  3         
     wheat4,  4")

## Assign values for the legend.
ct_value <- color_table$quartile
ct_color <- color_table$color
legend_title <- "Population\nDensity\nQuartiles"

## Read the names and data values for the islands.
island <- read_csv(file =
  "name  , population, area
   Oahu,   953000,      597
   Hawaii, 185000,     4208
   Maui,   155000,     1173
   Kauai,   67000,      620
   Molokai,  7300,      260
   Lanai,    3100,      141
   Niihau,    130,       72
   Kahoolawe,   0,       45")

## Note: Island boundaries were set in the Initialization chunk.

## Data Manipulation #######################################

## Calculate the population density for each island
island <- island |>
  mutate(density = population/area) 

## Calculate an index for the population density
## This uses the Sitemaps site_cuts function to calculate quartiles.
island$quartile <- site_cuts(quant_var = island$density,
     cuttype = "quartiles4")

## Merge the colors using the quartile values as an index
island <- merge(island, color_table, by="quartile")

## Data Check #############################################

## Show a well-formatted data table
gt(island) |>
  ## Number formats
  fmt_number(
    columns  = c(population, area),  
    use_seps = TRUE,
    decimals = 0) |> 
  fmt_number(
    columns = density,
    use_seps = TRUE,
    decimals = 2) |>
  ## Footnotes
  tab_footnote(
    footnote = "miles\U00B2", ## Symbol for squared
    locations = cells_column_labels(columns=area)) |> 
  tab_footnote(
    footnote = "people/mile\U00B2",
    locations = cells_column_labels(columns=density)) |> 
  ## Data source
  tab_source_note(source_note="Source: Gemini AI query")
quartile name population area1 density2 color
1 Niihau 130 72 1.81 beige
1 Kahoolawe 0 45 0.00 beige
2 Molokai 7,300 260 28.08 wheat2
2 Lanai 3,100 141 21.99 wheat2
3 Hawaii 185,000 4,208 43.96 wheat3
3 Kauai 67,000 620 108.06 wheat3
4 Oahu 953,000 597 1,596.31 wheat4
4 Maui 155,000 1,173 132.14 wheat4
Source: Gemini AI query
1 miles²
2 people/mile²
Show the code 
## Plot Section ################################################
 
## Make a set of default values for the geom_polygon parameters.
update_geom_defaults("polygon", 
                     list(color = "black", linewidth = 0.4))

## Plot each of the islands using the color as fill.
island_map <- ggplot() + 
  geom_polygon(data=hawaii_simple, aes(x=lon,y=lat,
      fill=island$color[island$name=="Hawaii"])) +
  geom_polygon(data=maui_simple, aes(x=lon,y=lat, 
      fill=island$color[island$name=="Maui"])) +
  geom_polygon(data=molokai_simple, aes(x=lon,y=lat,
      fill=island$color[island$name=="Molokai"])) +
  geom_polygon(data=kahoolawe_simple, aes(x=lon,y=lat, 
      fill=island$color[island$name=="Kahoolawe"]))+
  geom_polygon(data=lanai_simple, aes(x=lon,y=lat, 
      fill=island$color[island$name=="Lanai"])) +
  geom_polygon(data=oahu_simple, aes(x=lon,y=lat,
      fill=island$color[island$name=="Oahu"])) +
  geom_polygon(data=kauai_simple, aes(x=lon,y=lat,
      fill=island$color[island$name=="Kauai"])) + 
  geom_polygon(data=niihau_simple, aes(x=lon,y=lat,
      fill=island$color[island$name=="Niihau"])) + 
  coord_fixed(1.1) +
  island_map_theme

## Create the map with a legend
choro_map <- island_map +
  scale_fill_manual(legend_title,
                    labels = ct_value,
                    values = ct_color)

## Plot the map and legend
choro_map

8.4 Annotation

The functions in Sitemaps can be used to add text annotations to the choropleth map.

Text position coordinates can be copied from Google Maps by right-clicking on a location and copying the coordinates from the pop-up box by left-clicking on the coordinates. You get a message that the coordinates have been copied to the clipboard.

Recall that the text is centered on the lat, lon coordinates unless you change the justify value to left or right. [NOTE: JUSTIFY DOESN’T SEEM TO BE WORKING RIGHT IN SITEMAPS]

The example below uses the Population Density map from the previous chunk.

Show the code 
## Create a file of names for the annotation.
annotation <- read_csv(file =
      "text,   lat,       lon,         fontface, name_text_size
  Ni`ihau,     21.67626, -160.09480,   plain,    4
  Kaua`i,      22.11367, -158.96844,   plain,    4
  O`ahu,       21.35112, -158.41976,   plain,    4
  Moloka`i,    21.35607, -156.68232,   plain,    4
  Lāna`i,      20.78355, -157.30721,   plain,    4
  Maui,        20.97766, -155.93282,   plain,    4
  Kaho`olawe,  20.40621, -157.01449,   plain,    4
  Hawai`i,     19.55664, -156.38691,   plain,    4
  Pacific Ocean, 19.74548, -158.75696, italic,   5")

## Plot the map with the annotation.
choro_map +
  site_names(datatable = annotation)