8  Mountains

Munros are the hills in Scotland over 3000 feet high. Many Scottish people are obsessed with Munro bagging — trying to see how many of the 282 Munros you can climb. I’ve lived in Scotland since 2003, but I still haven’t been up a Munro!

So I thought for the mountains prompt, I’d map all of the Munros in Scotland by height. Then someday, when I bag a munro, I can mark it on the map.

Setup

library(tidyverse)     # for data wrangling
library(sf)            # for maps
library(rnaturalearth) # for map coordinates
library(ggthemes)      # for the map theme
library(plotly)        # for interactive plots
library(showtext)      # for fonts

# install a good Scottish font
# https://www.fontspace.com/hill-house-font-f40002
font_add(family = "Hill House",
         regular = "fonts/Hill_House.otf")
showtext_auto()

8.1 Data

The Database of British and Irish Hills v17.3 has a table of the Munros, with columns for many years (I guess which hills are classified as Munros changes over time). Let’s get just the current munros and fix some of the names.

Code

munros <- read_csv("data/munrotab_v8.0.1.csv",
                   show_col_types = FALSE) %>%
  filter(`2021` == "MUN") %>%
  select(-c(`1891`:`2021`)) %>% # get rid of the year columns
  rename(height_m = "Height (m)", 
         height_ft = "Height\n(ft)")

Make a quick histogram of their heights to get an overview of the data. I’d usually use the metric system, but since Munros are defined as hills over 3000 feet, I’ll use feet.

Code

ggplot(munros, aes(x = height_ft)) +
  geom_histogram(binwidth = 100, boundary = 0, 
                 color = "black", fill = "white")

8.2 Map of Scotland

Now I need a map of Scotland.

Code

scotland_sf <- ne_states(geounit = "Scotland", 
                         returnclass = "sf")

The munro table locates the peaks using grid coordinates, and the map uses latitude and longitude. So I translated the grid coordinates to latitude and longitude using Stackoverflow code from hrbrmstr.

Code

latlon <- map2(munros$xcoord, munros$ycoord, os.grid.to.lat.lon)

munros$lat <- map_dbl(latlon, `[[`, 1)
munros$lon <- map_dbl(latlon, `[[`, 2)

Then plot the latitude and longitude coordinates on the map, colored by height.

Code

ggplot() + 
  geom_sf(data = scotland_sf,
          mapping = aes(),
          color = "black", 
          fill = "grey",
          size = .1) +
  coord_sf(xlim = c(-8, 0), ylim = c(54, 61)) +
  geom_point(aes(x = lon, y = lat, color = height_ft), munros) +
  theme_map()

8.3 Make it prettier

There’s no munros on the Northern Isles, so I’ve cropped them out of the map. I also made the colours better. I added a label to geom_point(), which will produce a warning that it isn’t used, but you’ll see why in the next step.

The version of Hill House I have doesn’t have lowercase letters, so I’m using all uppercase for the title.

Code

munro_plot <- ggplot() + 
  geom_sf(data = scotland_sf,
          mapping = aes(),
          color = "chocolate4", 
          fill = "tan",
          size = .1) +
  coord_sf(xlim = c(-7.4, -2), ylim = c(54.8, 58.5)) +
  geom_point(aes(label = Name, color = height_ft, y = lat, x = lon), 
             data = arrange(munros, height_ft),
             size = 0.5) +
  scale_color_viridis_c(name = "Height (ft)",
                        limits = c(3000, 4500)) +
  labs(x = NULL, y = NULL,
       title = "THE MUNROS OF SCOTLAND",
       caption = "Data from The Database of British and Irish Hills | Plot by @lisadebruine") +
  theme_map() +
  theme(legend.position = c(0, 0),
        legend.background = element_blank(),
        panel.background = element_rect(fill = "transparent", color = "transparent"),
        plot.background = element_rect(fill = "lightskyblue1", color = "transparent"),
        plot.title = element_text(family = "Hill House", size = 26, hjust = .5)
  )

munro_plot

The location and height of Scotland’s Munros.

8.4 Make it interactive

The plotly package makes it pretty easy to make a ggplot interactive. Select an are to get a closer look, or hover over an individual munro to see why we added the unused label argument to geom_point() above.

I had to add some css to get the hover bar to stop acting weird, but it’s still not quite right.

Code

.js-plotly-plot .plotly .modebar { position: relative; top: -2em; }
.plotly.plot-container { padding-top: 2em; background-color: #B0E1FE;}

Code

plotly::ggplotly(
  p = munro_plot, 
  tooltip = c("label", "colour")
)