R/mask_data.R

In Doi90/bushfireSOS: Fit Species Distribution Models In Response To Australian Bushfire Crisis

#' Mask environmental data
#'
#' @param region
#' @param env_data
#'
#' @return
#' @export
#'
#' @examples
#'
mask_data <- function(env_data,
                      region,
                      crop = FALSE){

  ## Don't forget the ACT
  ## Add ACT to NSW
  if("NSW" %in% region){
    if(!"ACT" %in% region)
      region <- c(region, "ACT")
  }

  ## Convert region character

  region <- gsub("VIC", "Victoria", region)
  region <- gsub("NSW", "New South Wales", region)
  region <- gsub("QLD", "Queensland", region)
  region <- gsub("SA", "South Australia", region)
  region <- gsub("WA", "Western Australia", region)
  region <- gsub("NT", "Northern Territory", region)
  region <- gsub("TAS", "Tasmania", region)
  region <- gsub("ACT", "Australian Capital Territory", region)


# Get state polygon data and make it sf

  AUS.shapes <- rnaturalearth::ne_states("australia",
                                         returnclass = "sf")

  AUS.shapes <- sf::st_transform(AUS.shapes,
                                 crs = raster::crs(env_data))

  AUS.shapes <- AUS.shapes[AUS.shapes$name %in% region, ]

  ## crop the raster
  if(crop){
    env_data <- raster::crop(env_data, AUS.shapes)
  }

  env_data <- raster::mask(env_data,
                             mask = AUS.shapes)

  return(env_data)

}