World map overview with geom_world()

1. Introduction

geom_world() provides a convenient global base map for ggplot2. It comes bundled with country polygons, coastlines, and political/administrative boundaries.

Key features include:

• Automatic CRS transformation: Seamlessly projects data to your desired Coordinate Reference System.
• Antimeridian splitting: Handles the “Pacific wrap-around” issue automatically when changing central meridians.
• Layer control: Toggles for ocean background and administrative boundaries.

2. Basic usage

2.1 Default WGS84 map

By default, geom_world() plots the map using the WGS84 standard.

ggplot() +
  geom_world() +
  theme_void()

2.2 Explicit CRS specification

You can specify the CRS directly within the function.

ggplot() +
  geom_world(crs = 4326) +
  coord_sf(crs = 4326) +
  theme_void()

2.3 Hiding the ocean layer

For a cleaner look, you can remove the blue ocean background and change the land fill color.

ggplot() +
  geom_world(
    show_ocean   = FALSE,
    country_fill = "grey90"
  ) +
  theme_minimal()

2.4 Hiding administrative boundaries

If you only need continental landmasses without internal country borders, set show_admin_boundaries = FALSE.

ggplot() +
  geom_world(
    show_admin_boundaries = FALSE,
    country_fill          = "white"
  ) +
  theme_minimal()

Combining both options creates a minimalist silhouette map:

ggplot() +
  geom_world(
    show_ocean            = FALSE,
    show_admin_boundaries = FALSE
  ) +
  theme_minimal()

3. Projections

geom_world() shines when working with different map projections. It automatically projects the underlying polygons.

3.1 Robinson projection

crs_robin <- "+proj=robin +datum=WGS84"

ggplot() +
  geom_world(crs = crs_robin) +
  coord_sf(crs = crs_robin) +
  theme_void()

3.2 Robinson projection centred at 150°E

Changing the central meridian (centering the map on the Pacific) is often difficult in standard ggplot2. geom_world() handles the polygon splitting automatically.

crs_robin_150 <- "+proj=robin +lon_0=150 +datum=WGS84"

ggplot() +
  geom_world(crs = crs_robin_150) +
  coord_sf(crs = crs_robin_150) +
  theme_void()

3.3 Geographic CRS with shifted central meridian

crs_wgs84_150 <- "+proj=longlat +datum=WGS84 +lon_0=150"

ggplot() +
  geom_world(crs = crs_wgs84_150) +
  coord_sf(crs = crs_wgs84_150) +
  theme_void()

4. Axis labels and gridlines

A common issue with coord_sf() is that gridlines appear, but axis labels (coordinates) disappear. This often occurs when:

• expand = TRUE extends the map beyond ±180° or ±90°.
• The CRS lacks a geographic datum.
• Solid layers (like the ocean polygon) are drawn on top of the panel grid.

Recommended pattern for reliable axis labels: Use expand = FALSE inside coord_sf and set panel.ontop = TRUE in the theme.

ggplot() +
  geom_world() +
  coord_sf(
    crs    = 4326,
    expand = FALSE,
    datum  = sf::st_crs(4326)
  ) +
  theme_minimal() +
  theme(panel.ontop = TRUE)

5. Graticule annotation (meridians & parallels)

annotation_graticule() provides precise control over meridians and parallels. Unlike standard gridlines, these are annotation layers that:

• Are generated in WGS84 and transformed to your target CRS.
• Allow for custom line spacing (lon_step, lat_step) and label placement.

5.1 Global WGS84 map with graticules

ggplot() +
  geom_world() +
  annotation_graticule(
    lon_step     = 60,
    lat_step     = 30,
    label_offset = 5
  ) +
  coord_sf(
    crs    = 4326,
    expand = FALSE,
    datum  = sf::st_crs(4326)
  ) +
  theme_void() +
  theme(panel.ontop = TRUE)

5.2 Robinson projection

Note how the graticules curve naturally with the projection.

crs_robin <- "+proj=robin +datum=WGS84"

ggplot() +
  geom_world(crs = crs_robin) +
  annotation_graticule(
    crs          = crs_robin,
    lon_step     = 30,
    lat_step     = 15,
    label_offset = 3e5
  ) +
  coord_sf(crs = crs_robin) +
  theme_void()

5.3 Regional China map (clean axis labels)

For regional maps, the recommended pattern is to:

1. Use annotation_graticule() to draw the lines but hide its internal labels (label_color = NA).
2. Use standard labs() or coord_sf labels for the axes.
3. Keep the region exact with expand = FALSE.

cn_xlim <- c(70, 140)
cn_ylim <- c(0, 60)

ggplot() +
  geom_world() +
  annotation_graticule(
    xlim         = cn_xlim,
    ylim         = cn_ylim,
    crs          = 4326,
    lon_step     = 10,
    lat_step     = 10,
    label_color  = NA,
    label_offset = 1,
    label_size   = 3.5
  ) +
  coord_sf(
    xlim   = cn_xlim,
    ylim   = cn_ylim,
    expand = FALSE
  ) +
  labs(
    x = "Longitude",
    y = "Latitude"
  ) +
  theme_bw()

6. Highlighting selected countries

You can create “highlight” maps by layering geom_world() calls. The first call draws the base (e.g., white), and the second call filters for specific countries to color them.

6.1 Highlighting China

ggplot() +
  geom_world(
    country_fill = "white",
    show_frame   = TRUE
  ) +
  geom_world(
    filter_attribute = "SOC",
    filter           = "CHN",
    country_fill     = "red"
  ) +
  theme_void()

6.2 Highlighting multiple countries

Pass a vector of ISO codes to highlight multiple regions.

focus <- c("CHN", "JPN", "KOR")

ggplot() +
  geom_world(
    country_fill = "grey95",
    show_frame   = TRUE
  ) +
  geom_world(
    filter_attribute = "SOC",
    filter           = focus,
    country_fill     = "#f57f17"
  ) +
  theme_void()

7. Visualizing Custom Data

Users can also merge external datasets (e.g., GDP, population, or other metrics) with the map data to create choropleth maps. This requires accessing the underlying spatial data using check_geodata and load.

7.1 Merging external metrics

First, ensure the necessary geospatial data files are available and load them. Then, merge your custom data using the ISO country code (SOC).

# 1. Ensure data availability and GET FILE PATHS
map_files <- check_geodata(c("world_countries.rda", "world_coastlines.rda"))
#> extdata dir: D:/Program Files/R-4.3.3/library/ggmapcn/extdata (writable = TRUE)
#> cache   dir: C:\Users\Administrator\AppData\Roaming/R/data/R/ggmapcn (writable = TRUE)
#> Using existing cache file: C:/Users/Administrator/AppData/Roaming/R/data/R/ggmapcn/world_countries.rda
#> Using existing extdata file: D:/Program Files/R-4.3.3/library/ggmapcn/extdata/world_coastlines.rda

# 2. Load the world countries data (object name: 'countries')
load(map_files[1])

# 3. Create custom data: Real 2023 Population Estimates (Top 25+ major nations)
# Unit: Millions
custom_data <- data.frame(
  iso_code = c("CHN", "IND", "USA", "IDN", "PAK", "NGA", "BRA", "BGD", 
               "RUS", "MEX", "JPN", "ETH", "PHL", "EGY", "VNM", "COD", 
               "TUR", "IRN", "DEU", "THA", "GBR", "FRA", "ITA", "ZAF", 
               "KOR", "ESP", "COL", "CAN", "AUS", "SAU"),
  pop_mil  = c(1425.7, 1428.6, 339.9, 277.5, 240.5, 223.8, 216.4, 172.9, 
               144.4, 128.5, 123.3, 126.5, 117.3, 112.7, 98.9, 102.3, 
               85.8, 89.2, 83.2, 71.8, 67.7, 64.7, 58.9, 60.4, 
               51.7, 47.5, 52.1, 38.8, 26.6, 36.9)
)

# 4. Merge custom data with the 'countries' object
# Note: Use 'all.x = TRUE' to preserve the map geometry for all countries
merged_data <- merge(
  countries, 
  custom_data, 
  by.x  = "SOC", 
  by.y  = "iso_code", 
  all.x = TRUE
)

# 5. Plot with layering strategy
ggplot() +
  # Layer 1: Data Fill (No borders, just color)
  geom_sf(
    data  = merged_data, 
    aes(fill = pop_mil), 
    color = "transparent"
  ) +
  # Layer 2: World Boundaries (Transparent fill, standard borders)
  geom_world(
    country_fill = NA, 
    show_ocean   = FALSE
  ) +
  # Styling
  scale_fill_viridis_c(
    option    = "plasma", 
    na.value  = "grey95", 
    direction = -1,      # Reverse color scale so dark = high population
    name      = "Population (Millions)"
  ) +
  theme_void() +
  theme(legend.position = "bottom")

This workflow highlights a key design philosophy: by accessing raw spatial data via check_geodata() and processing it with geom_sf(), you gain complete flexibility to visualize custom datasets. At the same time, overlaying geom_world() ensures that the final map retains the consistent, high-quality basemap and administrative boundary styles provided by ggmapcn.