R/lsm_l_ed.R

In r-spatialecology/landscapemetrics: Landscape Metrics for Categorical Map Patterns

#' ED (landscape level)
#'
#' @description Edge Density (Area and Edge metric)
#'
#' @param landscape A categorical raster object: SpatRaster; Raster* Layer, Stack, Brick; stars or a list of SpatRasters.
#' @param count_boundary Count landscape boundary as edge
#' @param directions The number of directions in which patches should be connected: 4 (rook's case) or 8 (queen's case).
#'
#' @details
#' \deqn{ED = \frac{E} {A} * 10000}
#' where \eqn{E} is the total landscape edge in meters and \eqn{A} is the total
#' landscape area in square meters.
#'
#' ED is an 'Area and Edge metric'. The edge density equals all edges in the landscape
#' in relation to the landscape area. The boundary of the landscape is only included in the
#' corresponding total class edge length if \code{count_boundary = TRUE}.
#' The metric describes the configuration of the landscape, e.g. because an overall aggregation
#' of  classes will result in a low edge density. The metric is standardized to the
#' total landscape area, and therefore comparisons among landscapes with different total
#' areas are possible.

#' \subsection{Units}{Meters per hectare}
#' \subsection{Range}{ED >= 0}
#' \subsection{Behaviour}{Equals ED = 0 if only one patch is present (and the landscape
#' boundary is not included) and increases, without limit, as the landscapes becomes more
#' patchy}
#'
#' @seealso
#' \code{\link{lsm_l_te}},
#' \code{\link{lsm_l_ta}}, \cr
#' \code{\link{lsm_c_ed}}
#'
#' @return tibble
#'
#' @examples
#' landscape <- terra::rast(landscapemetrics::landscape)
#' lsm_l_ed(landscape)
#'
#' @references
#' McGarigal K., SA Cushman, and E Ene. 2023. FRAGSTATS v4: Spatial Pattern Analysis
#' Program for Categorical Maps. Computer software program produced by the authors;
#' available at the following web site: https://www.fragstats.org
#'
#' @export
lsm_l_ed <- function(landscape,
                          count_boundary = FALSE, directions = 8) {
    landscape <- landscape_as_list(landscape)

    result <- lapply(X = landscape,
                     FUN = lsm_l_ed_calc,
                     count_boundary = count_boundary,
                     directions = directions)

    layer <- rep(seq_along(result),
                 vapply(result, nrow, FUN.VALUE = integer(1)))

    result <- do.call(rbind, result)

    tibble::add_column(result, layer, .before = TRUE)
}

lsm_l_ed_calc <- function(landscape, count_boundary, directions, resolution, extras = NULL) {

    if (missing(resolution)) resolution <- terra::res(landscape)

    if (is.null(extras)){
        metrics <- "lsm_l_ed"
        landscape <- terra::as.matrix(landscape, wide = TRUE)
        extras <- prepare_extras(metrics, landscape_mat = landscape,
                                            directions = directions, neighbourhood = 4, resolution = resolution)
    }

    # all values NA
    if (all(is.na(landscape))) {
        return(tibble::new_tibble(list(level = "landscape",
                              class = as.integer(NA),
                              id = as.integer(NA),
                              metric = "ed",
                              value = as.double(NA))))
    }

    # get patch area
    area_patch <- lsm_p_area_calc(landscape,
                                  directions = directions,
                                  resolution = resolution,
                                  extras = extras)

    # summarise to total area
    area_total <- sum(area_patch$value)

    # get total edge
    edge_landscape <- lsm_l_te_calc(landscape,
                                    count_boundary = count_boundary,
                                    resolution = resolution,
                                    extras = extras)

    # relative edge density
    ed <- edge_landscape$value / area_total

    return(tibble::new_tibble(list(level = rep("landscape", length(ed)),
                 class = rep(as.integer(NA), length(ed)),
                 id = rep(as.integer(NA), length(ed)),
                 metric = rep("ed", length(ed)),
                 value = as.double(ed))))
}