R/mapzen_dem.R

In neilcharles/geoviz: Elevation and GPS Data Visualisation

#' Gets Digital Elevation Model (DEM) data from 'mapzen' via 'Amazon Public Datasets'
#'
#' @param lat WGS84 latitude. Either a single point to use as the centre for a \code{square_km} sized raster, or a vector of track points
#' @param long WGS84 longitude. Either a single point to use as the centre for a \code{square_km} sized raster, or a vector of track points
#' @param square_km length of one edge the required square area, in km. Ignored if lat and long have length > 1
#' @param width_buffer If lat and long have length > 1, used as buffer distance around the provided points in km
#' @param max_tiles maximum number of map tiles to request. More tiles will give higher resolution scenes but take longer to download. Note that very small numbers of tiles may result in a scene that is not square.
#'
#' @return a raster with values corresponding to terrain height in metres
#'
#' @examples
#' lat = 54.4502651
#' long = -3.1767946
#' square_km = 2
#'
#' dem <- mapzen_dem(lat, long, square_km, max_tiles = 2)
#' @export
mapzen_dem <- function(lat, long, square_km, width_buffer = 1, max_tiles = 10){

  mapzen_terrain <-
    slippy_raster(
      lat,
      long,
      square_km,
      image_source = "mapzen",
      image_type = "dem",
      max_tiles = max_tiles
    )

  #(red * 256 + green + blue / 256) - 32768

  DEM <-
    (raster::raster(mapzen_terrain, layer = 1) * 256 +
      raster::raster(mapzen_terrain, layer = 2) +
      raster::raster(mapzen_terrain, layer = 3) / 256) - 32768

  return(DEM)

}