Skip to content

Instantly share code, notes, and snippets.

@oscarperpinan
Last active December 20, 2019 10:43
Show Gist options
  • Save oscarperpinan/63f6e4ab95ca90e31a350e4392663b12 to your computer and use it in GitHub Desktop.
Save oscarperpinan/63f6e4ab95ca90e31a350e4392663b12 to your computer and use it in GitHub Desktop.
beautiful-thematic-maps: alternative code without using ggplot2
## https://timogrossenbacher.ch/2016/12/beautiful-thematic-maps-with-ggplot2-only/
library(raster)
library(sp)
library(rgdal)
library(viridisLite)
library(rasterVis)
## Read age data
data <- read.csv("input/avg_age_15.csv", stringsAsFactors = F)
## Read shapefile
gde_15 <- readOGR("input/geodata/gde-1-1-15.shp", layer = "gde-1-1-15")
## Join data with shapefile
mapData <- merge(gde_15, data, by.x = 'BFS_ID', by.y = 'bfs_id')
## Import the raster file
relief <- raster("input/geodata/02-relief-georef-clipped-resampled.tif")
## Number of classes for the age data
no_classes <- 6
## Colors using the magma palette
pal <- rev(
magma(n = no_classes,
## Limit the first and last colors
begin = 0.1,
end = 0.8))
## Compute breaks with quantiles
quantiles <- quantile(mapData$avg_age_15,
probs = seq(0, 1,
length.out = no_classes + 1))
## Not needed here, but a better alternative to original code (a "for"
## loop)
## labels <- paste(signif(quantiles[-7], 4),
## signif(quantiles[-1], 4),
## sep = '-')
## Create a trellis object using sp::spplot to display the
## SpatialPolygons object.
psp <- spplot(mapData["avg_age_15"],
col.regions = pal,
## define the points where the colors change
at = quantiles,
## set the border color and width
col = 'lightgray', lwd = 0.2,
par.settings = list(axis.line=list(col="transparent")),
## adjust the legend
colorkey =
list(space = 'bottom',
height = 0.7,
labels = list(
at = quantiles,
labels = signif(quantiles, 3),
rot = 45)
)
)
## Create a trellis object to display the relief using
## rasterVis::levelplot
plevel <- levelplot(relief,
maxpixels = 2e6,
margin = FALSE,
par.settings = GrTheme)
## Ok, ready to go: overlay these trellis objects to display the
## result
psp + as.layer(plevel, under = TRUE)
@oscarperpinan
Copy link
Author

map

@mmaechler
Copy link

Very very admirable! Finally someone proving that using ggplot2 is not a must at all for high quality graphics in R. Thank you!

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment