R/readForestLossDrivers.R

In pik-piam/mrland: MadRaT land data package

#' Read ForestLossDrivers
#'
#' Read-in an Forest loss data (range 2001-2015 but only single annual number her)
#' (Source:DOI: 10.1126/science.aau3445 Table 1).
#'
#'
#' @return magpie object of the Curtis et al., 2018 Data
#' @author Abhijeet Mishra
#' @seealso \code{\link{readSource}}
#' @examples
#' \dontrun{
#' a <- readSource("ForestLossDrivers")
#' }
#'
#' @importFrom magclass as.magpie
#' @importFrom madrat toolSubtypeSelect
#' @import readxl
#' @import countrycode
#' @importFrom stats complete.cases

readForestLossDrivers <- function() {
  ## Mapping file
  mapping <- read.csv("mapping.csv", header = TRUE, sep = ";")

  ## Magpie standard
  isoCountry <- toolGetMapping(type = "regional", name = "regionmappingH12.csv", where = "madrat")

  ## Merge Mappings
  fullMapping <- merge(mapping, isoCountry, by = "CountryCode")[, c(-2, -4)]
  colnames(fullMapping) <- c("CountryCode", "RegionCodeSource", "RegionCodeMAgPIE")

  ## Original Data
  file <- "forest_loss.csv"
  # This is Tree Cover loss in Mha at this stage, all drivers are in percentage
  df <- read.csv(file = file, header = TRUE, sep = ",")
  # Division by 100 because we change from percentage to proportion. Division by 15 to get annual data
  df[, -c(1:3)] <- (df$treecoverloss_01_15 * (df[, -c(1:3)] / 100)) / 15

  dfMag <- as.magpie(df[, -c(2, 3)], temporal = NULL, spatial = "region") ## This is Tree Cover loss in Mha

  ## FRA Forest Area
  # Convert is set to TRUE For Mha
  a <- collapseNames(readSource("FRA2020", "forest_area", convert = TRUE)[, , "naturallyRegeneratingForest"])

  ## Create iso level data based on forest data as weight
  forestLoss <- toolAggregate(x = dfMag, weight = setYears(a[, "y2015", ], NULL), rel = fullMapping,
                              from = "RegionCodeSource", to = "CountryCode")

  return(forestLoss)
}