Mask a raster based on a vector

Masks a raster file or object on the basis of a vector file or object. Pixels not covered by the vector features are set to NoData. If the input raster is multi-band, the mask is automatically applied to all bands. An optional buffer can be applied to the input vector to allow a more "lenient" masking, or to remove also the borders of the vector.

mask_rast(in_rast, mask, mask_value = 0, crop = FALSE, buffer = NULL,
  out_type = "rastobject", out_file = NULL, out_dtype = NULL,
  out_nodata = NULL, compress = "None", overwrite = FALSE,
  parallel = FALSE, cores = parallel::detectCores() - 2,
  verbose = TRUE)

Arguments

in_rast	Raster file or object inheriting class raster to be masked
mask	Vector file or object of class *sf or sp to be used as a mask 2. Raster file or object of class *sf or sp to be used as a mask NOTE: If mask is a raster object or file, the function checks if it has the same number of pixels and extent of in_rast. If this is true, then mask is used directly to mask the input. Otherwise, it is first of all vectorized to a vector mask. The vector mask is then used in the processing.
mask_value	integer if mask is a raster object/file, value corresponding to the areas that should be removed from the ouptut, Default: 0 (Ignored if the provided mask is a vector!)
crop	logical if TRUE, in_rast is also cropped on the extent of mask, Default: FALSE
buffer	numeric if not NULL, width of a buffer to be applied to mask before masking in_rast. If negative, mask is "reduced" prior to masking (see examples), Default: NULL
out_type	character if == "rastobj", return a Raster object; if == "filename", return the filename of the masked layer (GTiff or gdal vrt format depending on other arguments - see below) Default: "rastobj" (If an invalid string is provided, defaults to rastobj)
out_file	character filename where the masked raster should be saved If NULL, the masked raster is saved on a temporary file in the R temporary folder, named <basename(in_rast)>_sprawlmask.tif. The file is saved in GTiff format, with compress compression.
out_dtype	character data type of the output masked files, according to gdal specifications for GTiff files ("Byte", "UInt16", "Int16", "UInt32", "Int32", "Float32", "Float64", "CInt16", "CInt32", "CFloat32" and "CFloat64"). If NULL, the data type is retrieved from the input, Default: NULL
out_nodata	numeric value to be assigned to areas outside the mask, Default: NULL
compress	logical allow verbose output from foreach for debugging purposes, Default: FALSE
overwrite	logical if TRUE, and out_file is set and existing, existing files are overwritten, Default: FALSE
parallel	logical if TRUE, use ClusterR to implement multicore processing. This speeds up execution for large rasters, Default: FALSE
cores	numeric Number of cores to use in case of parallel processing. If not provided, it defaults to parallel::detectCores()-2
verbose	logical if TRUE, extended processing information is sent to the console in the form of messages

Value

object of class raster (if out_type == rastobj), or character string corresponding to the filename of the created raster (if out_type == rastfile)

Examples

library(sprawl)
library(sprawl.data)
library(raster)
in_polys <- read_vect(system.file("extdata/shapes","lc_polys.shp",
                                  package = "sprawl.data"),
                       stringsAsFactors = TRUE)
in_rast  <- raster::stack(system.file("extdata/MODIS_test", "EVIts_test.tif",
                       package = "sprawl.data"))[[1]]
in_polys <- sf::st_transform(in_polys, proj4string(in_rast))
masked   <- mask_rast(in_rast, in_polys, verbose = FALSE)
plot_rast(in_rast, in_poly = in_polys)
plot_rast(masked, in_poly = in_polys)