Skip to contents

These data sets are inherited from the certegis package.

Usage

geo_gemeenten

geo_ggdregios

geo_nuts3

geo_postcodes2

geo_postcodes3

geo_postcodes4

geo_postcodes6

geo_provincies

postcodes

postcodes4_afstanden

Examples

geo_provincies
#> Simple feature collection with 12 features and 3 fields
#> Geometry type: MULTIPOLYGON
#> Dimension:     XY
#> Bounding box:  xmin: 13565.4 ymin: 306846.2 xmax: 278026.1 ymax: 619352.4
#> Projected CRS: Amersfoort / RD New
#> First 10 features:
#>        provincie inwoners oppervlakte_km2                       geometry
#> 1        Drenthe   494705        2680.368 MULTIPOLYGON (((228930.1 57...
#> 2      Flevoland   428210        1466.570 MULTIPOLYGON (((164912.9 51...
#> 3      Friesland   651430        3531.269 MULTIPOLYGON (((154047.9 56...
#> 4     Gelderland  2096460        5118.931 MULTIPOLYGON (((169898.4 48...
#> 5      Groningen   586920        2392.983 MULTIPOLYGON (((276560.6 58...
#> 6        Limburg  1115835        2210.001 MULTIPOLYGON (((192680.3 41...
#> 7  Noord-Brabant  2573810        5054.037 MULTIPOLYGON (((100303.1 41...
#> 8  Noord-Holland  2887460        2864.383 MULTIPOLYGON (((150846.9 47...
#> 9     Overijssel  1166335        3408.845 MULTIPOLYGON (((182721.3 51...
#> 10       Utrecht  1360955        1553.980 MULTIPOLYGON (((130059 4794...

geo_provincies |>
  plot2(title = "Inwoners in Nederlandse provincies")
#>  Assuming datalabels.centroid = TRUE. Set to FALSE for a point-on-surface
#> placing of datalabels.
#>  Using category = inwoners
#>  Using datalabels = provincie


geo_provincies |>
  plot2(colour_fill = "umcgoranje")
#>  Adding white to the category scale - set two colours to colour_fill to
#> prevent this.
#>  Assuming datalabels.centroid = TRUE. Set to FALSE for a point-on-surface
#> placing of datalabels.
#>  Using category = inwoners
#>  Using datalabels = provincie


geo_gemeenten |>
  filter_geolocation(provincie == "Groningen") |>
  plot2(category = inwoners / oppervlakte_km2,
        category.title = "Inwoners per km^2",
        title = "Populatiedichtheid in Groningse gemeenten")
#>  Assuming datalabels.centroid = TRUE. Set to FALSE for a point-on-surface
#> placing of datalabels.
#>  Using datalabels = gemeente


geo_gemeenten |>
  filter_geolocation(provincie == "Groningen") |>
  plot2(category = inwoners / oppervlakte_km2,
        category.title = "Inwoners per km^2",
        title = "Populatiedichtheid in Groningse gemeenten",
        subtitle = "Met postcode-4 als overlay",
        datalabels = FALSE,
        colour = "umcgblauw",
        linewidth = 1) |>
  add_sf(geo_postcodes4 |> filter_geolocation(provincie == "Groningen"),
         colour = "umcgoranje",
         colour_fill = NA,
         linewidth = 0.25)
#> ! In 'sf' plots, colour will set the borders, not the areas - did you mean
#> colour_fill = "umcgblauw"?