In r-tmap/tmap: Thematic Maps

knitr::opts_chunk$set(
  collapse = TRUE,
  fig.width = 8,
  comment = "#>"
)
hook_output <- knitr::knit_hooks$get("output")
knitr::knit_hooks$set(output = function(x, options) {
   lines <- options$output.lines
   if (is.null(lines)) {
     return(hook_output(x, options))  # pass to default hook
   }
   x <- unlist(strsplit(x, "\n"))
   more <- "..."
   if (length(lines)==1) {        # first n lines
     if (length(x) > lines) {
       # truncate the output, but add ....
       x <- c(head(x, lines), more)
     }
   } else {
     x <- c(more, x[lines], more)
   }
   # paste these lines together
   x <- paste(c(x, ""), collapse = "\n")
   hook_output(x, options)
 })

library(tmap)
tmap_options(scale = 0.75)

Extension types

There are three types of tmap extensions:

1. New map layer types
2. New spatial data classes
3. New output modes

Rather than explaining how to extend tmap for each of these three types (which is rather complex), it is easier to demonstrate with proof-of-concept extension packages:

New map layer types

This type of extension requires:

• Specification of visual variables. For new visual variable, options need to be added.
• Spatial transformation function needs to be specified: e.g. should centroids be used?
• The mapping between visual variables and (standard) visual or transformation parameters.

See tmap.glyphs. A layer function tm_donuts is added. More glyph types layer functions will follow, e.g. tm_pies, or tm_radars.

library(tmap.glyphs)
ZH_muni = NLD_muni[NLD_muni$province == "Zuid-Holland", ]
ZH_muni$income_middle = 100 - ZH_muni$income_high - ZH_muni$income_low

tm_shape(ZH_muni) +
  tm_polygons() +
  tm_donuts(
    parts = tm_vars(c("income_low", "income_middle", "income_high"), multivariate = TRUE),
    fill.scale = tm_scale_categorical(values = "-pu_gn_div"),             
    size = "population",
    size.scale = tm_scale_continuous(ticks = c(50000, 100000, 250000, 500000))) 

New spatial data classes

This type of extension requires methods to obtain:

• the spatial geometries (cast to either vectorized objects (sfc) or a stars object with indices) and
• the data variables and its levels (if any)

See tmap.networks which supports sfnetworks

library(sfnetworks)
library(tmap.networks)

(sfn = as_sfnetwork(roxel))

Besides this new spatial data class "sfnetwork", this package also features new map layers, albeit very basic so far:

tm_shape(sfn) +
    tm_network()

tm_shape(sfn) +
    tm_edges(col = "type", lwd = 4) +
    tm_nodes()

New mode

This type of extension is by far the most difficult one. It requires:

• Initialization
• Loading of the used spatial object
• Plotting function for each map layer type
• A function for each map component
• Functions to preprocess and show the legends
• Run the plot
• Shiny integration functions

As a proof of concept we created tmap.deckgl. It remains a proof-of-concept because the underlying package deckgl hasn't been updated the last 2 years.

For this proof of concept, only one map layer is tm_polygons(). The shiny integration does not work yet.

library(tmap.deckgl)

tmap_mode("deck")

tm_shape(NLD_dist) +
  tm_polygons(
    fill = "employment_rate", 
    fill.scale = tm_scale_intervals(values = "scico.roma"),
    lwd = 0.1) +
tm_shape(NLD_muni) +
  tm_polygons(fill = NULL, lwd = 1) +
tm_deck(pitch = 75)

r-tmap/tmap documentation built on Feb. 28, 2025, 7:54 a.m.