How can I automize raster processing in R - r

Question

I have 90 rasters that I need to clip and ressample to match my template raster. Is there a way for me to "automize this"? (i.e., only have to run the process once, so that I can, for instance, leave it running overnight)
I know that the answer is probably a "loop", but I'm not sure how to do it because each original file has a different name and the final product needs to have that same name, but end in "_final.asc"
This is the code I've been using:
library(raster)
#load original world raster (the one to be processed)
mau <- raster("PATH/name_of_the_raster.asc")
#load bounding box raster made by me (to clip the original and therefore make the next steps easier)
rastergrande <- raster("PATH")
#clip original raster by bounding box raster
intermedio <- crop(mau, rastergrande)
#load my template raster (the one that has the extent and cell size I want)
template <- raster("PATH")
#resample raster to template parameters
novoraster <- resample(intermedio, template, "bilinear")
#assign NA's to where template is NA
values(novoraster)[is.na(values(template))] <- NA
#Export to ascii
writeRaster(novoraster, "PATH/name_of_the_raster_final.asc", overwrite=TRUE)

Answer 1

Variations of this question have been asked a number of times. Here is one example
The general approach can be to get a vector with all your filenames, typically with list.files, and use that to make a vector of output filenames by changing the names or the path. Something like this:
input <- list.files(path=".", pattern="asc$", full.names=TRUE)
output <- gsub(".tif", "_out.tif", input)
Or change the folder
output <- gsub("this/path", "that/path", input)
And then use a loop
for (i in 1:length(input)) {
# read input[i]
# write output[i]
}
A working minimal example (using a single file):
library(raster)
input <- system.file("external/test.grd", package="raster")
# write to current working directory, but with asc extensions
output <- gsub(".grd", ".asc", basename(input))
for (i in 1:length(input)) {
r <- raster(input[i])
writeRaster(r, output[i], overwrite=TRUE)
}
Also, not directly related to your question, but this:
values(novoraster)[is.na(values(template))] <- NA
Should be
novoraster <- mask(novoraster, template)

Answer 2

You can use lapply. You would need a vector of file names, maybe with
fn_list <- list.files("PATH", pattern="asc$")
Your workhorse function could look like
one_file <- function(fn, template, rastergrande) {
mau <- raster(fn)
intermedio <- crop(mau, rastergrande)
novoraster <- resample(intermedio, template, "bilinear")
values(novoraster)[is.na(values(template))] <- NA
fn_out <- gsub("asc$", "_final.asc", fn)
writeRaster(novoraster, fn_out)
}
You can then apply one_file on fn_list:
library(raster)
library(usdm)
library(ggplot2)
library(rgdal)
rastergrande <- raster("PATH")
template <- raster("PATH")
lapply(fn_list, one_file, template=templ, rastergrande=rastergrande)
By the way, take a look at the terra package: https://www.youtube.com/watch?v=O8V4QJ13Gjc

Answer 3

Here you can work with a for loop or apply (lapply and mapply)
file_names <- list.files("PATH/", pattern = ".tif")
r_list <- lapply(file_names,raster)
rastergrande <- raster("PATH")
template <- raster("PATH")
intermedio <- list()
novoraster <- list()
Creating a vector with rasters names
rasternames <- c("name1","name2" ...)
For loop to process all rasters
for(i in 1:length(file_names){
intermedio[[i]] <- crop(r_list[[i]], rastergrande)
novoraster[[i]] <- resample(intermedio[[i]], template, "bilinear")
values(novoraster[[i]])[is.na(values(template))] <- NA
writeRaster(novoraster[[i]], paste0("PATH/", rasternames[i], ".asc"), overwrite=TRUE)
}