Convert reduced representation to full raster

Convert reduced representation to full raster

rasterizeReduced(
  reduced,
  fullRaster,
  newRasterCols,
  mapcode = names(fullRaster),
  ...
)

Arguments

reduced

data.frame or data.table that has at least one column of codes that are represented in the fullRaster.

fullRaster

RasterLayer/SpatRaster of codes used in reduced that represents a spatial representation of the data. Note that if fullRaster is a factor SpatRaster, the active category level values are used, not the IDs (see terra::activeCat and terra::cats)

newRasterCols

Character vector, length 1 or more, with the name(s) of the column(s) in reduced whose value will be returned as a RasterLayer/SpatRaster or list of RasterLayer/SpatRasters.

mapcode

a character, length 1, with the name of the column in reduced that is represented in fullRaster.

...

Other arguments. None used yet.

Value

A RasterLayer/SpatRaster or list of RasterLayer/SpatRaster of with same dimensions as fullRaster representing newRasterCols spatially, according to the join between the mapcode contained within reduced and fullRaster

See also

terra::rast()

Author

Eliot McIntire

Examples

library(data.table)
library(terra)

origDTThreads <- data.table::setDTthreads(2L)
origNcpus <- options(Ncpus = 2L)

ras <- rast(ext(0, 15, 0, 15), res = 1)
fullRas <- randomPolygons(ras, numTypes = 2)
names(fullRas) <- "mapcodeAll"
uniqueComms <- unique(fullRas)
reducedDT <- data.table(uniqueComms,
                        communities = sample(1:1000, length(uniqueComms)),
                        biomass = rnbinom(length(uniqueComms), mu = 4000, 0.4))
biomass <- rasterizeReduced(reducedDT, fullRas, "biomass")

# The default key is the layer name of the fullRas, so rekey incase of miskey
setkey(reducedDT, biomass)

communities <- rasterizeReduced(reducedDT, fullRas, "communities")
coltab(communities) <- c("blue", "orange", "red")
if (interactive()) {
  terra::plot(c(biomass, communities, fullRas))
}

## with a factor SpatRaster, the mapcode should correspond to the
## active category (not the ids)
cls <- data.frame(id = sort(unique(as.vector(fullRas[]))))
cls$Bclass <- LETTERS[cls$id]
levels(fullRas) <- cls
is.factor(fullRas)
#> [1] TRUE

clsDT <- as.data.table(cls)
reducedDT <- reducedDT[clsDT, on = "mapcodeAll==id"]
reducedDT[, mapcodeAll := Bclass]
#>    mapcodeAll communities biomass Bclass
#>        <char>       <int>   <num> <char>
#> 1:          A         860     148      A
#> 2:          B         860     148      B

biomass2 <- rasterizeReduced(reducedDT, fullRas, "biomass")

# clean up
data.table::setDTthreads(origDTThreads)
options(Ncpus = origNcpus)