Thematic maps

A thematic map is a statistical map that focuses on the spatial variation of a particular distribution, topic or theme. The theme refers to the phenomena that is shown, which is often demographical, social, cultural, or economic. There are several types of thematic maps such as heat maps, cartogram, choropleths, and others.

Thematic maps are created by stacking layers, where per layer, data can be mapped to one or more aesthetics. It is also possible to generate small multiples.

Types

1.Choropleth Map

It is used to represent statistical values in different geographical scale, from global to local. It mainly display densities (ratios) of quantities using color.

2.Dot Distribution Map

It uses dots (variation of marks) to display the presence or absence of a feature.

3.Heat Maps

Heat maps display the density of points on a geographic map and can effectively visualise the intensity of the variable through a colour scale.

4.Cartogram

A cartogram is a thematic map type in which the size of an area is re-scaled to be proportional to the feature it represents.

Content Overview

Specifically, I will explain and demonstrate the usage of tmap package and I will focus on several major functions as per below:

• tm_shape
• qtm
• tm_tiles
• tm_facets
• tmap_mode
• tm_polygons
• tm_fill

Installation

The “World” DataFrame contains information about the population and the economy of countries all over the globe.

library(tmap)
data("World")

tmap is born

In order to create the map, tm_shape does the work. It accepts spatial data as an argument and in this case “World” is a SpatialPolygonsDataFrame.

However, in order to represent the data, we can use tm_polygons. The below countries are filtered by happiness index.

tm_shape(World) +
    tm_polygons("HPI")

tm_shape(World)+tm_polygons("HPI", palette=c("gray95", "gold"), contrast=.7, id="name", title="Happiness Index")

class(World)

## [1] "sf"         "data.frame"

Static vs Interactive

Every map can be either plotted as static image using plot or presented interactively using view modes. The mode can be changed using tmap_mode

tmap_mode("view")

tm_shape(World) +
    tm_polygons("HPI", palette= "-Blues")

Mix & Match

With tmap, we can combine different data together to the above map. What if we want to view the population of the metropolitan regions in the world, for example in Armenia?

We are presenting the information using bubbles tm_bubbles instead of polygons.

data (metro)

metro$growth <- (metro$pop2020 - metro$pop2010) / (metro$pop2010 * 10) * 100

    tm_shape(metro) +
    tm_bubbles("pop2010", col = "growth", 
 breaks=c(-Inf, seq(0, 6, by=2), Inf),
               palette="-RdYlBu", contrast=1, 
               title.size="Metro population", 
               title.col="Growth rate (%)", id="name") + 
    tm_format("World_wide")

Now, let’s join the above maps together!

data(World, metro)
metro$growth <- (metro$pop2020 - metro$pop2010) / (metro$pop2010 * 10) * 100

mapWorld <- tm_shape(World) +
    tm_polygons("income_grp", palette="-Blues", contrast=.7, id="name", title="Income group") +
    tm_shape(metro) +
    tm_bubbles("pop2010", col = "growth", 
               border.col = "black", border.alpha = .5, 
               style="fixed", breaks=c(-Inf, seq(0, 6, by=2), Inf),
               palette="-RdYlBu", contrast=1, 
               title.size="Metro population", 
               title.col="Growth rate (%)", id="name") + 
    tm_style("grey") +  tm_format("World_wide")

mapWorld

Let’s Build layers!

A tile layer is a way to visualize large data sets on the map.The function tm_basemap draws the tile layer as basemap (as bottom layer), whereas tm_tiles draws the tile layer as overlay layer. The stacking order corresponds to the order in which this layer is called.

The base map is “Stamen.Watercolor” and the overlay layer is the one having metro data. We can manipulate what layer we want to view by clicking on the icon on the map.

tmap_mode("view")
tm_basemap("Stamen.Watercolor") +
tm_shape(metro) + tm_bubbles(size = "pop2020", col = "red") +
tm_tiles("Stamen.TonerLabels")

Multiples

We can create small multiples of thematic maps using function tm_facets. This can be done in 3 different ways:

First by assigning multiple variable names to one aesthetic.

tmap_mode("view")
tm_shape(World) +
    tm_polygons(c("HPI", "economy")) +
    tm_facets(sync = TRUE, ncol = 2)

Second by splitting the spatial data with the by argument.

tmap_mode("plot")
## tmap mode set to plotting

data(NLD_muni)

NLD_muni$perc_men <- NLD_muni$pop_men / NLD_muni$population * 100

tm_shape(NLD_muni) +
    tm_polygons(col = "MAP_COLORS") +
    tm_facets(by = "province")

Third by using the tmap_arrange function:

The NLD_muni is the maps of Netherlands municipality level of 2013.

tmap_mode("plot")
## tmap mode set to plotting

data(NLD_muni)
tm1 <- tm_shape(NLD_muni) + tm_polygons("population", convert2density = TRUE, palette = "deepskyblue" )
tm2 <- tm_shape(NLD_muni) + tm_bubbles(size = "population")

tmap_arrange(tm1, tm2)

Quick Trick

We can create a thematic map with one function which is the quick thematic map (qtm).
The data frame used is rivers which shows 141 major rivers in North America, as compiled by the US Geological Survey.

data(rivers)

qtm (rivers)

And more.. World dataframe

qtm (World)

Map in Style

To switch styles for thematic maps, we can use tmap_style. Options include “white”, “gray”, “natural”, “classic”, “col_blind”, “albatross”, “beaver”, “bw”, and “watercolor”.

tmap_style("cobalt")
## tmap style set to "classic"
## other available styles are: "white", "gray", "natural", "cobalt", "col_blind", "albatross", "beaver", "bw", "watercolor"

qtm(World, fill = "life_exp")

qtm(World, "HPI", fill.palette = "-plasma")

Focus

What if we want to focus more or less on specific locations/features on the map? In this case, we use the function filter in which we can choose what we want to color & what we don’t. The below example focuses on coloring the continent Africa.

tmap_style("gray")

tm_shape(World, filter = World$continent=="Africa") +
    tm_polygons("HPI", id = "name", palette = "-viridis")