Frequently Asked Questions

How can I plot WhiteboxRaster objects?

While there exists a built-in plot method for all WhiteboxRaster and WhiteboxVector objects, it is highly advisable to use {terra}’s plotting functionality by converting the Whitebox... objects to SpatRaster or SpatVector accordingly as follows:

library(wbw)
library(terra)

f <- system.file("extdata/dem.tif", package = "wbw")

wbw_read_raster(f) |>
  wbw_multidirectional_hillshade() |> 
  as_rast() |> 
  plot(legend = FALSE)

I’ve noticed that raster extent differs between WhiteboxRaster and terra/GDAL

Yes, that’s true! Since the original Python library Whitebox Workflows for Python is closed-source, the reason for this behavior is unknown. One possible explanation (i.e., my guess) is that GDAL typically defines the extent based on the outer edges of the corner pixels, while WbW may define the extent based on the centers of the corner pixels.

library(wbw)
library(terra)

f <- system.file("extdata/dem.tif", package = "wbw")

wbw_raster <- wbw_read_raster(f)
terra_raster <- rast(f)

# Compare extents
# Mind the difference in xmax (east) and ymin (south)
waldo::compare(
  as_vector(wbw_ext(wbw_raster)),
  as.vector(ext(terra_raster))
)
#> `names(old)`: "west" "east" "south" "north"
#> `names(new)`: "xmin" "xmax" "ymin"  "ymax" 
#> 
#> `old`: 1925449.3 1929446.2 5582091.5 5585716.6
#> `new`: 1925449.3 1929451.2 5582086.5 5585716.6

However, this issue is fixed when transitioning from WhiteboxRaster to SpatRaster via the as_rast() command:

converted_raster <- as_rast(wbw_raster)
waldo::compare(
  as.vector(ext(terra_raster)), 
  as.vector(ext(converted_raster))
  )
#> ✔ No differences