Skip to contents

Grid dataset

Source: R/gisco-get-grid.R
gisco_get_grid.Rd

These datasets contain grid cells covering the EU and neighbouring countries at resolutions from 1 km to 100 km. Population figures are available for selected reference years.

Usage

gisco_get_grid(
  resolution = c(100, 50, 20, 10, 5, 2, 1),
  spatialtype = c("REGION", "POINT"),
  cache_dir = NULL,
  update_cache = FALSE,
  verbose = FALSE
)

Source

https://ec.europa.eu/eurostat/web/gisco/geodata/grids.

Arguments

resolution

The grid cell resolution in km. Available values are "1", "2", "5", "10", "20", "50" and "100". See Details.

spatialtype

A character string selecting "REGION" or "POINT".

cache_dir

A character string with a path to a cache directory. See Caching strategies section in gisco_set_cache_dir().

update_cache

A logical value indicating whether to refresh the cached file. Defaults to FALSE. When set to TRUE, it forces a new download.

verbose

A logical value indicating whether to display informational messages.

Value

A sf object.

Details

gisco_get_grid() downloads the GeoPackage representation of the grid as polygon cells (spatialtype = "REGION") or cell-centre points (spatialtype = "POINT"). The official distribution also provides tabular CSV and Parquet files, which this function does not download.

All grid geometries use EPSG:3035. Population columns are named TOT_P_YYYY, where YYYY is the reference year. To calculate population density, divide a population value by the cell area in square kilometres (resolution^2).

The file sizes range from 428 KB (resolution = 100) to 1.7 GB (resolution = 1).

Population variables (TOT_P_*) have year- and country-specific licensing conditions. Other grid elements are covered by the general Eurostat copyright provisions. Review the licensing table and metadata on the official grid page before redistributing or publishing the data.

Examples

grid <- gisco_get_grid(resolution = 20)

# Proceed if downloaded correctly.
if (!is.null(grid)) {
  library(dplyr)

  grid <- grid |>
    mutate(popdens = TOT_P_2021 / 20^2)

  breaks <- c(0, 1, 10, 25, 50, 100, 250, 500, 1000, Inf)

  # Cut groups.
  grid <- grid |>
    mutate(popdens_cut = cut(popdens,
      breaks = breaks,
      include.lowest = TRUE
    ))

  cut_labs <- prettyNum(breaks, big.mark = " ")[-1]
  cut_labs[1] <- "0"
  cut_labs[9] <- "> 1000"

  pal <- c(
    "black",
    hcl.colors(length(breaks) - 2, palette = "Spectral", alpha = 0.9)
  )

  library(ggplot2)

  ggplot(grid) +
    geom_sf(aes(fill = popdens_cut), color = NA, linewidth = 0) +
    coord_sf(
      xlim = c(2500000, 7000000),
      ylim = c(1500000, 5200000)
    ) +
    scale_fill_manual(
      values = pal, na.value = "black",
      name = "",
      labels = cut_labs
    ) +
    theme_void() +
    labs(
      title = "Population density in Europe (2021)",
      subtitle = "Grid: 20 km. People by square km.",
      caption = paste(
        "Source: Eurostat GISCO grid dataset.\n",
        "Review the applicable population-data licence."
      )
    ) +
    theme(
      text = element_text(colour = "white"),
      plot.background = element_rect(fill = "grey2"),
      plot.title = element_text(hjust = 0.5),
      plot.subtitle = element_text(hjust = 0.5, face = "bold"),
      plot.caption = element_text(
        color = "grey60", hjust = 0.5, vjust = 0,
        margin = margin(t = 5, b = 10)
      ),
      legend.key.height = unit(0.5, "lines"),
      legend.key.width = unit(1, "lines")
    )
}