Customize your maps

Richard Aubrey White, Chi Zhang

2021-01-15

library(csmaps)
library(ggplot2)
library(data.table)
library(magrittr)

The maps can be easily customized with color and labels.

Colored maps

Automatic coloring by location_code

pd <- copy(csmaps::nor_county_map_b2020_insert_oslo_dt)

q <- ggplot()
q <- q + geom_polygon(
  data = pd, 
  mapping = aes(
    x = long, 
    y = lat,
    group = group,
    fill = location_code
  ),  
  color="black",
  size=0.1
)
q <- q + annotate(
  "text",
  x = csmaps::nor_xxx_position_title_insert_oslo_b2020_insert_oslo_dt$long,
  y = csmaps::nor_xxx_position_title_insert_oslo_b2020_insert_oslo_dt$lat,
  label = "Oslo"
)
q <- q + theme_void()
q <- q + coord_quickmap()
q <- q + labs(title = "")
q

Customized coloring with external information

It is also possible to specify the color by user-defined groups. Here we show an example of assigning different (pseudo) risk level to each county.

pd <- copy(csmaps::nor_county_map_b2020_insert_oslo_dt)

# assign each location a random category for different colors
location_info <- unique(pd[,c("location_code")])
location_info[,category:=rep(
  c("Good","Normal","Neutral","Bad","Very Bad"),
  each=3)[1:.N]
]
location_info[,category:=factor(
  category,
  levels=c("Good","Normal","Neutral","Bad","Very Bad")
  )
]
print(location_info)
#>     location_code category
#>  1:  county_nor11     Good
#>  2:  county_nor15     Good
#>  3:  county_nor18     Good
#>  4:  county_nor03   Normal
#>  5:  county_nor30   Normal
#>  6:  county_nor34   Normal
#>  7:  county_nor38  Neutral
#>  8:  county_nor42  Neutral
#>  9:  county_nor46  Neutral
#> 10:  county_nor50      Bad
#> 11:  county_nor54      Bad

# join the map data.table
pd[location_info,on="location_code",category:=category]

q <- ggplot()
q <- q + geom_polygon(
  data = pd, 
  mapping = aes(
    x = long,
    y = lat,
    group = group,
    fill=category
  ), 
  color="black", 
  size=0.25
)
q <- q + annotate(
  "text",
  x = csmaps::nor_xxx_position_title_insert_oslo_b2020_insert_oslo_dt$long,
  y = csmaps::nor_xxx_position_title_insert_oslo_b2020_insert_oslo_dt$lat,
  label = "Oslo"
)
q <- q + coord_quickmap()
q <- q + labs(title="")
q <- q + theme_void()
q

Labeled maps

We can add labels of county index onto the maps. There are several options for adding texts on a graph in ggplot2. We recommend geom_label() to add the labels if no label overlap occurs, otherwise we recommend using ggrepel::geom_label_repel().

pd <- copy(csmaps::nor_county_map_b2020_insert_oslo_dt)
q <- ggplot()
q <- q + geom_polygon(
  data = pd, 
  mapping = aes(
    x = long,
    y = lat,
    group = group,
    fill = location_code
  ),  
  color="black",
  size=0.1
)
q <- q + annotate(
  "text",
  x = csmaps::nor_xxx_position_title_insert_oslo_b2020_insert_oslo_dt$long,
  y = csmaps::nor_xxx_position_title_insert_oslo_b2020_insert_oslo_dt$lat,
  label = "Oslo"
)
q <- q + geom_label(
  data = csmaps::nor_county_position_geolabels_b2020_default_dt,
  mapping = aes(x = long, y = lat, label = location_code)
  )
# ggrepel::geom_label_repel() for avoiding overlap
q <- q + theme_void()
q <- q + coord_quickmap()
q <- q + labs(title = "")
q

Labels can be easily added to other layouts, such as Oslo wards.

q <- ggplot(mapping = aes(x = long, y = lat))
q <- q + geom_polygon(
  data = csmaps::oslo_ward_map_b2020_default_dt,
  mapping = aes(group = group),
  color = "black",
  fill = "white",
  size = 0.2
)
q <- q + geom_label(
  data = csmaps::oslo_ward_position_geolabels_b2020_default_dt,
  mapping = aes(label = location_code),
  color = "red",
  size = 3,
  label.size = 0.1,
  label.r = grid::unit(0, "lines")
)
q <- q + theme_void()
q <- q + coord_quickmap()
q

Enrich plot with additional data

It is convenient to use csdata package to enrich Norway and Oslo maps with external information, such as location name and population. We illustrate how to do it here.

Add county name and population to Norway map

# enrich with population and location name
dpop_2020 <- csdata::nor_population_by_age_cats()[calyear==2020]

# join, create label
labels <- copy(csmaps::nor_county_position_geolabels_b2020_insert_oslo_dt)
labels[
  dpop_2020, 
  on = "location_code",
  pop_total := pop_jan1_n
]
labels[
  csdata::nor_locations_names(), 
  on = "location_code",
  location_name := location_name
]
labels[, label := paste0(location_name, '\n', pop_total)]
print(head(labels))
#>    location_code     long      lat pop_total        location_name
#> 1:  county_nor30  8.85000 60.60000   1241165                Viken
#> 2:  county_nor03 10.72028 59.98000    693494                 Oslo
#> 3:  county_nor34 11.00000 61.86886    371385            Innlandet
#> 4:  county_nor38  8.50000 59.32481    419396 Vestfold og Telemark
#> 5:  county_nor42  7.80000 58.30000    307231                Agder
#> 6:  county_nor11  6.10000 58.70000    479892             Rogaland
#>                           label
#> 1:               Viken\n1241165
#> 2:                 Oslo\n693494
#> 3:            Innlandet\n371385
#> 4: Vestfold og Telemark\n419396
#> 5:                Agder\n307231
#> 6:             Rogaland\n479892

# plot
pd <- copy(csmaps::nor_county_map_b2020_insert_oslo_dt)
q <- ggplot()
q <- q + geom_polygon(
  data = pd, 
  mapping = aes(
    x = long,
    y = lat,
    group = group,
    fill = location_code
  ),
  color="black",
  size=0.1
)
q <- q + annotate(
  "text",
  x = csmaps::nor_xxx_position_title_insert_oslo_b2020_insert_oslo_dt$long,
  y = csmaps::nor_xxx_position_title_insert_oslo_b2020_insert_oslo_dt$lat,
  label = "Oslo"
)
q <- q + ggrepel::geom_label_repel(
  data = labels,
  mapping = aes(x = long, y = lat, label = label)
)
q <- q + theme_void()
q <- q + coord_quickmap()
q <- q + labs(title = "")
q

Add location name for ward and population for Oslo map

# enrich with population and location name
dpop_2020 <- csdata::nor_population_by_age_cats()[calyear==2020]

# join, create label
labels <- copy(csmaps::oslo_ward_position_geolabels_b2020_default_dt)
labels[
  dpop_2020, 
  on = "location_code",
  pop_total := pop_jan1_n
]
labels[
  csdata::nor_locations_names(), 
  on = "location_code",
  location_name := location_name
]
labels[, label := paste0(location_name, '\n', pop_total)]
print(head(labels))
#>         location_code     long      lat pop_total  location_name
#> 1: wardoslo_nor030101 10.79760 59.91010     58671     Gamle Oslo
#> 2: wardoslo_nor030102 10.78000 59.92567     62423    Grünerløkka
#> 3: wardoslo_nor030103 10.76683 59.93981     45089         Sagene
#> 4: wardoslo_nor030104 10.73555 59.91230     38945 St. Hanshaugen
#> 5: wardoslo_nor030105 10.66500 59.89925     59269        Frogner
#> 6: wardoslo_nor030106 10.65000 59.92500     34569         Ullern
#>                    label
#> 1:     Gamle Oslo\n58671
#> 2:    Grünerløkka\n62423
#> 3:         Sagene\n45089
#> 4: St. Hanshaugen\n38945
#> 5:        Frogner\n59269
#> 6:         Ullern\n34569

q <- ggplot(mapping = aes(x = long, y = lat))
q <- q + geom_polygon(
  data = csmaps::oslo_ward_map_b2020_default_dt,
  mapping = aes(group = group),
  color = "black",
  fill = "white",
  size = 0.2
)
q <- q + geom_label(
  data = labels,
  mapping = aes(label = label),
  color = "red",
  size = 3,
  label.size = 0.1,
  label.r = grid::unit(0, "lines")
)
q <- q + theme_void()
q <- q + coord_quickmap()
q