Europe hexbin for cartogram

May 30, 2024 by Kinh Nguyen

There is not an existing hexbin for Europe at the time of writing, we can however manually create it as follows:

Load required packages

require(sf)
require(ggplot2)
require(dplyr)

Make an arbitrarily map extend, and assign a coordinate system.

x = st_sfc(st_polygon(list(rbind(c(0,0), c(10,0), c(10,10), c(0,0)))), crs = 4326)
g = st_make_grid(x, cellsize = 1, square = FALSE, flat_topped = TRUE)
g = st_as_sf(g)

Above we made 10 by 10 hexagon grid.

The coordinates should at least fall within the real world boundaries, says, coordinates 1000,1000 won’t exist.

Assign some IDs and identified, then plot them and start to pick values,

g$id = 1:nrow(g)
g$pick = 0
cmap <- c(
    DEU = 81,
    POL = 88,
    LTU = 95,
    LVA = 102,
    EST = 115,
    DNK = 94,
    SWE = 107,
    FIN = 114,
    NLD = 87,
    BEL = 80,
    LUX = 74,
    FRA = 67,
    CHE = 61,
    AUT = 68,
    CZE = 75,
    ITA = 48,
    ESP = 60,
    PRT = 53,
    IRL = 79,
    SVN = 55,
    SVK = 82, 
    HUN = 69,
    HRV = 62,
    MLT = 28,
    ROU = 63,
    BGR = 50,
    GRC = 43,
    CYP = 31
)
g$pick[g$id %in% cmap] <- 1
g$iso <- ''
g$iso[cmap] <- names(cmap)

Plot the full grid and assigned values so that you can arrange the countries, check the plot and add/edit the number above.

ggplot(g) + 
    geom_sf(aes(fill = factor(iso))) +
    theme(legend.position = 'none') +
    geom_sf_text(aes(label = id), color = 'white') +
    geom_sf_text(aes(label = iso), color = 'white', size = 3, nudge_y = -0.5)

Filter out selected countries

g <- g %>% filter(pick == 1)

and the end results can be. This can be further use for cartogram, for example, the WIP figure below (DOI to be updated soon)

Have fun!