R: how to turn raster image / matrix values into series in order to obtain boxplot for multiple raster images?

Question

I have a series of tiff files representing temperature data (~40 files) for each of which I would like to get a simple boxplot of the value distribution. I know how to boxplot the raster files directly; however, I would like to use ggplot (requires dataframe base) to arrange individual plots in a specific manner.

Ideally a solution would provide a dataframe where each raster image's values are represented by a column as the x-y-position of the data is unimportant, but I am not sure of what the best solution is here?

Answer 1

You can use the following code

library(tidyverse)
library(raster)

#Make a list of the files
files <- list.files(path="E:\\...", #Provide the path containing the tif files
                    pattern="tif", full.names=TRUE, recursive=TRUE)

#Stack rasters
Stack <- stack(files)

#Convert the rasters into data frame
df <- as.data.frame(Stack, xy=TRUE) %>% 
  drop_na()

#Create boxplot for each raster layer
df %>% 
  pivot_longer(-c(x, y)) %>% 
  ggplot(aes(x = name, y = value)) +
  geom_boxplot()