Faster operations on rasters

This alternative to raster::mask is included here.

fastMask(
  x,
  y,
  cores = NULL,
  useGDAL = getOption("reproducible.useGDAL", TRUE),
  verbose = getOption("reproducible.verbose", 1),
  ...
)

Arguments

x	A Raster* object.
y	A SpatialPolygons object. If it is not in the same projection as x, it will be reprojected on the fly to that of x
cores	An integer* or 'AUTO'. This will be used if gdalwarp is triggered. 'AUTO' will calculate 90 number of cores in the system, while an integer or rounded float will be passed as the exact number of cores to be used.
useGDAL	Logical or "force". Defaults to getOption("reproducible.useGDAL" = TRUE). If TRUE, then this function will use gdalwarp only when not small enough to fit in memory (i.e., if the operation fails the raster::canProcessInMemory(x, 3) test). Using gdalwarp will usually be faster than raster::projectRaster, the function used if this is FALSE. Since since the two options use different algorithms, there may be different projection results. "force" will cause it to use GDAL regardless of the memory test described here.
verbose	Numeric, -1 silent (where possible), 0 being very quiet, 1 showing more messaging, 2 being more messaging, etc. Default is 1. Above 3 will output much more information about the internals of Caching, which may help diagnose Caching challenges. Can set globally with an option, e.g., options('reproducible.verbose' = 0) to reduce to minimal
...	Currently unused.

Value

A Raster* object, masked (i.e., smaller extent and/or several pixels converted to NA)

Examples

library(sp)
library(raster)

Sr1 <- Polygon(cbind(c(2, 4, 4, 0.9, 2), c(2, 3, 5, 4, 2)))
Sr2 <- Polygon(cbind(c(5, 4, 2, 5), c(2, 3, 2, 2)))
Sr3 <- Polygon(cbind(c(4, 4, 5, 10, 4), c(5, 3, 2, 5, 5)))

Srs1 <- Polygons(list(Sr1), "s1")
Srs2 <- Polygons(list(Sr2), "s2")
Srs3 <- Polygons(list(Sr3), "s3")
shp <- SpatialPolygons(list(Srs1, Srs2, Srs3), 1:3)
d <- data.frame(vals = 1:3, other = letters[3:1], stringsAsFactors = FALSE)
row.names(d) <- names(shp)
shp <- SpatialPolygonsDataFrame(shp, data = d)
poly <- list()
poly[[1]] <- raster(raster::extent(shp), vals = 0, res = c(1, 1))
poly[[2]] <- raster(raster::extent(shp), vals = 1, res = c(1, 1))
origStack <- stack(poly)
# original mask function in raster
newStack1 <- mask(x= origStack, mask = shp)
newStack2 <- fastMask(x = origStack, y = shp)
#>       useGDAL is TRUE, but problem is small enough for RAM; skipping GDAL; useGDAL = 'force' to override
#> This function is using raster::mask
#>  because fastMask doesn't have a specific method for these RasterStack or RasterBrick yet

# test all equal
all.equal(newStack1, newStack2)
#> [1] TRUE

newStack1 <- stack(newStack1)
newStack2 <- stack(newStack2)

if (interactive()) {
  plot(newStack2[[1]])
  plot(shp, add = TRUE)
}