R/simulate_ci.R

In mhesselbarth/rabmp: R agent-based model for plants

#' simulate_ci
#'
#' @description Update competition index
#'
#' @param data Dataframe with input data.
#' @param parameters List with all parameters.
#'
#' @details
#' The function calculated a compeition index using a kernel. Competition depends on
#' neighbouring trees (within max_dist). If \code{standardized = TRUE}, all values
#' are standardized to 0 <= CI <= 1.
#'
#' @return data.table
#'
#' @aliases simulate_ci
#' @rdname simulate_ci
#'
#' @references
#' Pommerening, A., Maleki, K., 2014. Differences between competition kernels and
#' traditional size-ratio based competition indices used in forest ecology. For. Ecol. Manage. 331, 135-143.
#'
#' @export
simulate_ci <- function(data, parameters){

  # get id of current living and no seedlings
  id <- data[type != "dead" & i == max(i), which = TRUE]

  # calculate CI (Pommerening et al. 2014 formula 6)
  # transformation of ci, which includes size of focal tree (Pommerening et al. 2014 formula 9)
  competition <- rcpp_calculate_ci(matrix = as.matrix(data[id, .(x, y, dbh)]),
                                   alpha = parameters$ci_alpha,
                                   beta = parameters$ci_beta,
                                   max_dist = parameters$ci_max_dist)

  # update tibble
  data[id, ci := competition]

  return(data)
}