In April 2021, I co-presented a workshop on working with spatial data in R to the NMSU Fish, Wildlife and Conservation Ecology Graduate Student Organization (GSO). One of the topics I presented on was how to make a simple inset map with R.

Packages

library(sf)
library(tidyverse)
library(rgdal)
library(maps) ##not used, but has some world map data in it
library(mxmaps)
library(cowplot)
library(ggspatial)

Building the Focal Map

First we can load the data by directly reading a shapefile (saved in the working directory) and bringing it into R as a SF object. The shapefile includes the states of Mexico. We’ll want to filter to only include the Northwestern states, then “zoom” into our focal area.

estados <- read_sf("dest_2010gw/dest_2010gw.shp")

## filter out the data so we can plot a focal area (NW Mexican states) 
area <- estados %>% filter(ENTIDAD == "SONORA" | ENTIDAD == "SINALOA" | 
                             ENTIDAD == "BAJA CALIFORNIA" | ENTIDAD == "BAJA CALIFORNIA SUR")

## since the data is an SF object, we can easily make a simple map of it
ggplot(area) + geom_sf(aes()) + theme_linedraw()

Lets “zoom” into the focal area of the Midriff Islands Region, by setting limits on our axes using lat/long coordinates.

ggplot(area) + geom_sf(aes()) + theme_linedraw() + coord_sf(xlim = c(-114, -112), 
                                                            ylim = c(28, 30), 
                                                            expand = TRUE)

Since ggplot gives us nice axes/gridlines that functions as a clean graticule, we won’t incldue a scale bar. But lets add a north arrow.

ggplot(area) + geom_sf(aes()) + theme_linedraw() + coord_sf(xlim = c(-114, -112), 
                                                            ylim = c(28, 30), 
                                                            expand = TRUE) +
  annotation_north_arrow()

This doesn’t look too aesthetically pleasing, but luckily we can fix that by playing with the the style and the location of the north arrow. Note that the location can be specified as top left (“tl”), bottom left (“bl”), etc.

ggplot(area) + geom_sf(aes()) + theme_linedraw() + coord_sf(xlim = c(-114, -112), 
                                                            ylim = c(28, 30), 
                                                            expand = TRUE) +
  annotation_north_arrow(location = "tl", style = north_arrow_fancy_orienteering()) 

## this is looking pretty good so lets save this as an object so that we can add the inset
p1 <- ggplot(area) + geom_sf(aes()) + theme_linedraw() + coord_sf(xlim = c(-114, -112), 
                                                                  ylim = c(28, 30), 
                                                                  expand = TRUE) +
  annotation_north_arrow(location = "tl", style = north_arrow_fancy_orienteering())

Building the inset map

There are lots of packages that come with different spatial data. Rather than use the same shapefile, we can use the package mxmaps that has some simple map data of Mexico to create our map in a different way.

ggplot(mxstate.map, aes(long, lat, group=group)) + geom_polygon()

We can now begin to change the aesthetics such as the fill, outline thickness and theme of the map. The void theme is the best choice for our inset map, as we can also easily add layers of aesthetics that we want or don’t want.

ggplot(mxstate.map, aes(long, lat, group=group)) +
  geom_polygon(fill = "white", color = "black", size = .2) + theme_void()

We don’t want the whole map of mexico, we just want to show the northwest area of the country. Once again we set set limits to the x and y axes using coodinates to “zoom” into an area.

ggplot(mxstate.map, aes(long, lat, group=group)) +
  geom_polygon(fill = "white", color = "black", size = .2) + theme_void() + 
  coord_map(ylim=c(22.5,32.5), xlim=c(-117.5, -106))

To draw a box over the map area to highlight the area of the main map, we will add a rectangle using the annotate function, specifying the limits using the coordinates we used above. We can also specify the color of the box- here we’ll use red so that it pops.

ggplot(mxstate.map, aes(long, lat, group=group)) +
  geom_polygon(fill = "white", color = "black", size = .2) + theme_void() + 
  coord_map(ylim=c(22.5,32.5), xlim=c(-117.5, -106)) + 
  annotate(geom = "rect", ymax = 30, ymin = 28, xmax = -114, xmin = -112, 
           colour = "red", fill = NA)

Finally lets add a black outline to this map with a blank background.

ggplot(mxstate.map, aes(long, lat, group=group)) +
  geom_polygon(fill = "white", color = "black", size = .2) + theme_void() + 
  coord_map(ylim=c(22.5,32.5), xlim=c(-117.5, -106)) + 
  annotate(geom = "rect", ymax = 30, ymin = 28, xmax = -114, xmin = -112, 
           colour = "red", fill = NA)  +
  theme(panel.background = element_rect(fill = "white"), panel.border = 
          element_rect(color = "black", fill=NA, size = 1.5)) 

## this looks pretty good so we'll save it as a 2nd object
p2 <- ggplot(mxstate.map, aes(long, lat, group=group)) +
  geom_polygon(fill = "white", color = "black", size = .2) + theme_void() + 
  coord_map(ylim=c(22.5,32.5), xlim=c(-117.5, -106)) + 
  annotate(geom = "rect", ymax = 30, ymin = 28, xmax = -114, xmin = -112, 
           colour = "red", fill = NA) + labs("") +
  theme(panel.background = element_rect(fill = "white"), panel.border = 
          element_rect(color = "black", fill=NA, size = 1.5))

We’ll use the package cowplot to put the pieces together, specifying where to place the inset.

ggdraw() + draw_plot(p1) + draw_plot(p2, x = 0.5, y = 0.65, width = 0.3, height = 0.3)

You’ll maybe want to play with where everything lines up/fits to minimize white space.

ggdraw() + draw_plot(p1) + draw_plot(p2, x = 0.725, y = 0.65, width = 0.3, height = 0.3)

And with that, we have a simple map with an inset!!

Don’t forget to save/copy your ggplot however you want!