Makeover – Unlocking the Power of Data Visualization with R

Map showing healthcare accessibility in Spain

Overview

This map will focus on improving the visualization of healthcare accessibility across Spain.

Makeover of Day 7: Accessibility — #30DayMapChallenge 2025

Data: OSM hospitals + worldpop population + simplified travel time model

# Load libraries
library(sf)
library(tmap)
library(terra)
library(osmdata)
library(eurostat)
library(dplyr)
library(rnaturalearth)
library(rnaturalearthdata)

# 1. Get country boundaries (Europe)
europe <- ne_countries(continent = "Europe", returnclass = "sf")

# 2. Select a country
spain <- europe %>% filter(admin %in% c("Spain"))

# 3. Download hospitals from OpenStreetMap
spain %>%
  st_bbox() %>%
  st_as_sfc() %>%
  st_transform(crs = 4326) -> spain_bbox

q <- opq(bbox = spain_bbox) %>%
  add_osm_feature(key = "amenity", value = "hospital")

hospitals <- readRDS("../day7_accessibility/data/hospitals_spain.rds")

# Use a metric CRS (Web Mercator)
spain_3857 <- st_transform(spain, 3857)
spain_v <- vect(spain_3857)

hospitals <- st_transform(hospitals, 3857)
hospitals_v <- vect(hospitals)

# --------------------------
# Create raster template
# --------------------------
# Resolution in meters (e.g., 1000 = 1 km per pixel)
r_template <- rast(spain_3857, res = 1000)
# Raster must contain values before mask()
values(r_template) <- 1

# Clip raster to Spain
r_template <- crop(r_template, spain_v) 
r_template <- mask(r_template, spain_v)
r_smooth <- focal(r_template, w = 3, fun = mean, 
                  na.policy = "omit")

# --------------------------
# Compute distance (meters -> km)
# --------------------------
# distance() computes Euclidean distance to the nearest point
dist_rast_m <- distance(r_template, y = hospitals_v)  # meters
dist_rast_km <- dist_rast_m / 1000                    # convert to km

dist_rast_spain <- mask(crop(dist_rast_km, spain_v), spain_v)

# --------------------------
# TEST Plot
# --------------------------
plot(dist_rast_km, main = "Simulated Distance to Nearest Hospital (km)")
plot(spain_v, add = TRUE)
points(hospitals_v, pch = 20, cex = 0.4)

# Create the map
tmap_mode("plot")

map_access <- tm_shape(dist_rast_spain) +
tm_raster(col.legend = tm_legend(title="Distance (km)"),
          col.scale = tm_scale(
    values = c(  "#D55E00",  # strong orange/red
                  "#E69F00",  # warm yellow
                  "#F0E442",  # soft lemon
                  "#56B4E9",  # light sky blue
                  "#0072B2",  # mid blue
                  "#2C3E50"   # deep navy/blue-black
                )))+
  tm_shape(spain_v) +
  tm_borders(col = "grey40",lwd = 2) +
  tm_shape(hospitals_v) +
  tm_dots(size = 0.05, 
          shape = 10, 
          col = "grey20") +
  tm_title("Accessibility\nHealthcare Facilities in Spain",
           fontface="bold") +
  tm_layout(legend.outside = TRUE,
            panel.show = F,
            legend.frame = FALSE,
            legend.width = 7.5,
            frame = FALSE,
            main.title.size = 1.1) +
  tm_credits("#30DayMapChallenge 2025\nDay 30: Accessibility - Makeover\nData: OSM, simulated population\n",
             size = 0.6, 
             position =  tm_pos_out(pos.h = 0, pos.v = 0.1))

map_access

Save the map

tmap_save(
  tm = map_access,
  filename = "day30_makeover.png",
  width = 6,      
  height = 4,    
  dpi = 320        
)

# Optional: also view interactively
# tmap_mode("view")
# map_access