Basic use cases

Source: vignettes/scotmaps.Rmd

Introduction

Use scotmaps to download map layers from www.spatialdata.gov.scot with single line of R code. Here we provide a few quick examples of how the package can be used.

Usage

See all the layers available:

install.packages("devtools")
devtools::install_github("fozy81/scotmaps")
library(scotmaps)
available_layers()

Download Layers

Either use get_layer("name of your layer") or name_of_layer(). For example you can down the marine areas using either of these functions:

ma <- get_layer("marine_areas")
# or...
ma <- marine_areas()

The first time you run a layer function, you will be prompted for permission to download that layer to your hard drive.

If you have already downloaded a layer you can see if it’s cached by using cached_layers().

Start Mapping!

To plot a map we’ll use the ggplot2 library. This package helps to format and present our data layers.

# install.packages("ggplot2")
library(ggplot2)

la <- local_authorities()

p <- ggplot(la) +
  geom_sf(data = la, aes(fill = local_authority), size = 0.05) +
  guides(fill = FALSE) +
  scale_fill_discrete() +
  theme_void() +
  theme(panel.grid.major = element_line(colour = "grey60", size = 0.1))

p

The geom_st recognises the simple feature or sf map downloaded layer and plots at a spatial object in the correct map projection.

Map all the Stats

Let’s download data from statistics.gov.scot and display it on a map.

# install.packages("devtools")
# install.packages("tidyverse")
# devtools::install_github("datasciencescotland/opendatascot")
library(opendatascot)
library(tidyverse)

# Immunisation data
immune <- ods_dataset("6-in-1-immunisation", geography = "la")
immune <- immune %>% filter(vaccinationUptake == "vaccinated-by-24-months",
                     measureType == "ratio",
                     refPeriod == 2018)

immune <- immune %>% mutate(value = as.numeric(value))

# Join to local authority spatial layer
la_immune <- inner_join(la, immune, by = c("code" = "refArea"))

p2 <- la_immune %>%
 ggplot() +
  geom_sf(aes(fill = value)) +
  scale_fill_viridis_c("Ratio") +
  theme_bw() +
  labs(title="6-in-1-immunisation by 24 months (2018)")

p2

Looks good but what about in 3D? Don’t worry, rayshader package has your back.

# devtools::install_github("tylermorganwall/rayshader")
library(rayshader)
plot_gg(p2, multicore = TRUE)
render_camera(zoom = 0.4, theta = -45, phi = 30)
Sys.sleep(0.2)
render_snapshot()

rgl::rgl.close()

Map your own stats

Let’s calculate bicycle parking capacity per person for each local authority in Scotland.

First, download the bicycle parking data from OpenStreetMap. OpenStreetMap is open data, licensed under the Open Data Commons Open Database License (ODbL) copyright OpenStreetMap contributors.

# install.packages("osmdata")
library(osmdata)
library(tidyverse)
scot <- getbb ("scotland", format_out = "polygon")
bicycles <- opq(scot[[3]][3]) %>%
  add_osm_feature(key = "amenity",
                  value = "bicycle_parking",
                  value_exact = TRUE) %>%
  osmdata_sf()

# Use osm point data
points <- bicycles$osm_points
# Tidy up missing capacity values
points$capacity <- as.numeric(points$capacity)
# Default if capacity not recorded
points$capacity[is.na(points$capacity)] <- median(points$capacity, na.rm = T)
# Convert to same crs as default scotmaps
points <- st_transform(points, crs = 27700)
plot(st_geometry(points), col = points$capacity)

Our OpenStreetMap query returned a wider area than Scotland, so let’s join points to associated Scottish local authority areas.

# devtools::install_github("fozy81/scotmaps")
# library(scotmaps)
la <- local_authorities()
points <- st_join(points, la, join = st_within)
# Cropped points with a local authority
plot(st_geometry(points[!is.na(points$local_authority), ]), col = points$capacity)

Now we can sum capacity available in each authority.

capacity <- points %>%
  group_by(code) %>%
  summarise(total = sum(capacity)) %>%
  st_set_geometry(NULL) # remove the geometry column
#> `summarise()` ungrouping output (override with `.groups` argument)

Next, download census population data using opendatascot package and calculate bicycle parking capacity per person.

# devtools::install_github("datasciencescotland/opendatascot")
library(opendatascot)
population <- ods_dataset("census-population-2001", geography = "la")
population <- population %>%
  filter(measureType == "count",
         age == "all",
         gender == "all")

total_points <- inner_join(capacity, population, by = c("code" = "refArea"))

# Calculate the pop / bicycle ratio
total_points$value <- as.numeric(total_points$value)
total_points <- total_points %>%  mutate(ratio = total / value)

# Join to authority areas
la_totals <- inner_join(la, total_points, by = c("code" = "code"))

Plot data on map:

# devtools::install_github("tylermorganwall/rayshader")
library(rayshader)
p3 <- ggplot(la_totals) +
  geom_sf(aes(fill = ratio), alpha = 0.5) +
  scale_colour_viridis_c()

plot_gg(p3, multicore = TRUE)
render_camera(zoom = 0.4, theta = -45, phi = 30)
Sys.sleep(0.2)
render_snapshot()

rgl::rgl.close()

Availability of Public Bicycle Parking Capacity

Alternatively, we could present the data as a league table:

total <- st_set_geometry(la_totals, NULL)
total <- mutate(total,  "per_100" = ratio * 100)

# make table
arrange(total, desc(per_100)) %>%
  mutate('No.' = seq_along(local_authority)) %>%
  select(No.,
         "Local Authority" = local_authority,
         "Capacity per 100 people" = per_100) %>%
  knitr::kable()
No. Local Authority Capacity per 100 people
1 City of Edinburgh 3.2452566
2 Dundee City 1.1251999
3 Fife 1.0079301
4 Glasgow City 0.9571373
5 Midlothian 0.9241299
6 East Lothian 0.8458396
7 Aberdeen City 0.7594579
8 Stirling 0.7539554
9 Highland 0.7318801
10 West Lothian 0.6659778
11 Orkney Islands 0.6443232
12 Perth and Kinross 0.4616559
13 Eilean Siar 0.4075164
14 North Lanarkshire 0.3821632
15 South Lanarkshire 0.3431321
16 West Dunbartonshire 0.3127075
17 Argyll and Bute 0.2978994
18 East Renfrewshire 0.2541680
19 East Dunbartonshire 0.2392764
20 Falkirk 0.2383068
21 South Ayrshire 0.2212370
22 Dumfries and Galloway 0.2206206
23 Scottish Borders 0.2098085
24 Renfrewshire 0.2024678
25 Inverclyde 0.1686401
26 Shetland Islands 0.1637257
27 Aberdeenshire 0.1617659
28 Moray 0.1230734
29 Clackmannanshire 0.1206398
30 North Ayrshire 0.0868816
31 Angus 0.0571956
32 East Ayrshire 0.0382584