R/toolBiomassSupplyAggregate.R

In pik-piam/mrremind: MadRat REMIND Input Data Package

#' toolBiomassSupplyAggregate
#' The function aggregates biomass supply curves to regionmapping different from H12.
#' It only works if all regions are subregions of H12 regions. The offset parameter (a) is taken from the H12 region.
#' The slope parameter (b) is multiplied by a weight. The weight is the inverse of the share of agricultural area of the subregion in the H12 region.
#'
#' @return return: returns region aggregated biomass supply curve data
#'
#' @param x magclass object that should be aggregated
#' @param rel relation matrix containing a region mapping.
#' @param weight aggregation weight
#' @author Felix Schreyer
#' @export
#' @importFrom magclass as.magpie dimReduce
#' @importFrom dplyr %>% mutate select rename left_join
#' @importFrom quitte as.quitte

toolBiomassSupplyAggregate <- function(x, rel=NULL, weight = calcOutput("FAOLand", aggregate = F)[,,"6610",pmatch=TRUE][,"y2010",]){

  # variable definitions needed for data.frame operations below
  region <- NULL
  value <- NULL
  Total <- NULL
  country <- NULL

  # FS: aggregate from iso-countries to regions
  # inverse of the disaggregation described in convertMAgPIE.R (subytpe = "supplyCurve_magpie_40")
  # copy offset parameter a from iso-country to region
  # multiply agricultural land share of iso-country in region with iso-country slope parameter b for
  # one iso-country to obtain slope parameter b of region

  # calculate share in agricultural land area of countries relative to MAgPIE regions
  # get agricultural land for iso-countries in 2010 from FAO
  AgrLandIso <- weight
  # aggregate agricultural land to regions in regionmapping
  AgrLandReg <- toolAggregate(AgrLandIso, rel)

  # calculate share of agricultural land area for each iso-country relative to the MAgPIE region it is in
  AgrLandShare <- as.quitte(AgrLandIso) %>%
    select(region, value) %>%
    rename(country = region) %>%
    left_join(rel) %>%
    left_join((as.quitte(AgrLandReg) %>%
                 select(region, value) %>%
                 rename(Total = value))) %>%
    mutate(value = value / Total) %>%
    # if no agricultural area at all -> assume very low share of 1e-5
    mutate( value = ifelse(value == 0, 1e-5, value)) %>%
    select(country, value) %>%
    data.frame %>%
    as.magpie(spatial = 1, datacol=2) %>%
    dimReduce()

  # multiply slope parameter with agricultural land share
  x <- x
  x[,,"b"] <- x[,,"b"] * AgrLandShare

  # select one country per region (first country in alphabet) to set weight = 1,
  # rest of countries weight 0
  rel.sort <- rel %>%
                    arrange(region)
  iso.sel <- rel.sort$country[cumsum(table(rel.sort$region))]

  weight <- new.magpie(getRegions(x), fill = 0)
  weight[iso.sel,,] <- 1

  # aggregate to regionmapping, take a and b value only of one country within the region,
  # other weights are 0
  y <- toolAggregate(x, rel, weight = weight)

  return(y)
}