R/calcIniFoodPrice.R

In pik-piam/mrcommons: MadRat commons Input Data Library

#' @title calcIniFoodPrice
#'
#' @description provides global prices from the IMPACT model projections for MAgPIE commodities
#' for the initialization of the flexible demand model.
#'
#' @note The IMPACT projections start in 2005 and the prices are taken from that year.
#'
#' @param datasource The datasource specification. Currently available `FAO` and `IMPACT3.2.2World_Price`.
#' @param year Specifies the year for the initialization of prices in MAgPIE. Default is `y2005`.
#' @param products subselection of products to be returned
#'
#' @return List with a magpie object with commodity prices on global level in $17/tDM.
#' @author Mishko Stevanovic, Benjamin Leon Bodirsky
#' @seealso
#' [readIMPACT3.2.2World_Price()]
#' @examples
#' \dontrun{
#' calcOutput("IniFoodPrice", aggregate = FALSE)
#' }
#'
#' @importFrom stringr str_sub
#' @importFrom magpiesets findset

calcIniFoodPrice <- function(datasource = "FAO", year = "y2005", products = "kfo") {
  # dry matter factors
  dm <- 1 / readSource("ProductAttributes", "Products")[, , "wm"]
  dm <- collapseNames(dm)

  if (datasource == "IMPACT3.2.2World_Price") {
    out <- readSource("IMPACT3.2.2World_Price") # in 2017 prices now
    # select scenario and initial year
    out <- collapseNames(out[, year, "SSP2-NoCC-NoCC-379"])
    out <- setYears(out, NULL)
    # get the mapping
    mapping <- read.csv(toolGetMapping(type = "sectoral",
                                       name = "impact2magpie.csv",
                                       returnPathOnly = TRUE, where = "mappingfolder"),
                        na.strings = "")
    mapping <- mapping[1:nlevels(mapping$MAgPIE), ]
    out <- toolAggregate(out, rel = mapping, from = "IMPACT", to = "MAgPIE", dim = 3.1, partrel = TRUE)

    out <- add_columns(x = out, dim = 3.1, addnm = setdiff(findset("kall"), getNames(out)))

    # manually add prices for sugr_cane, sugr_beet, alcohol, fish, oilpalm and fodder from exteral sources

    out[, , "sugr_cane"]        <- 13.07939764 # the FAO producer price in Brazil in 2005 (biggest production)
    out[, , "sugr_beet"]        <- 55.311522 # the FAO producer price in France in 2005 (biggest production)
    out[, , "alcohol"]          <- 300 # a guess that a liter of beer is 1$
    # BB: on alocohol this is luxury. I would go for 30 cents like Sternie
    #    (also beacuse we are interested in the pirmary products going into alcohol, not the marketed product).
    # out[,,"fish"]             <- 2067.626632 # OECD/FAO price of fish food traded in 2005 #nolint: keep fish price for future
    out[, , "fish"]             <- 1400 # FAO FIGIS database, export value divided by export volume in 2005, rounded
    out[, , "oilpalm"]          <- 100 # palmoilpoint.blogspot.de (527 malaysian ringgit = 118 USD rounded to 100)
    out[, , "oilcakes"]         <- 300 # https://www.indexmundi.com/commodities/?commodity=soybean-meal&months=180
    out[, , "cottn_pro"]        <- 700 # https://cotton.ces.ncsu.edu/wp-content/uploads/2018/02/Seed-Cotton-Summary_Feb-13-2018.pdf?fwd=no 0.3353*2.2*1000 #nolint
    out[, , "foddr"]            <- 200 # equivalent to cereals
    out[, , "brans"]            <- 100 # quick google search
    out[, , "distillers_grain"] <- 200 # quick google search
    out[, , "ethanol"]          <- 900 # quick google search
    out[, , "fibres"]           <- 1500 # https://cotton.ces.ncsu.edu/wp-content/uploads/2018/02/Seed-Cotton-Summary_Feb-13-2018.pdf?fwd=no 0.690*2.2*1000 #nolint
    out[, , "molasses"]         <- 50 # quick google search
    out[, , "wood"]             <- 150 # quick google search
    out[, , "woodfuel"]         <- 100 # quick google search
    out[, , "begr"]             <- 50 # quick google search
    out[, , "betr"]             <- 50 # quick google search
    out[, , "res_cereals"]      <- 50 # quick google search
    out[, , "res_fibrous"]      <- 50 # quick google search
    out[, , "res_nonfibrous"]   <- 50 # quick google search
    out[, , "pasture"]          <- 50 # quick google search
    out[, , "scp"]              <- 1500 # using upper end prices similar to fishmeal

    # [DC] At some point probably better to do another 'quick google search',
    # for now convert these to 2017 values using US inflation, also for consistency
    conv <- c("sugr_cane", "sugr_beet", "alcohol", "fish", "oilpalm", "oilcakes", "cottn_pro", "foddr",
              "brans", "distillers_grain", "ethanol", "fibres", "molasses", "wood", "woodfuel", "begr",
              "betr", "res_cereals", "res_fibrous", "pasture", "scp")
    getItems(out, dim = 1) <- "USA"
    out[, , conv] <- GDPuc::toolConvertGDP(out[, , conv], unit_in = "constant 2005 US$MER",
                                           unit_out = "constant 2017 US$MER",
                                           replace_NAs = "no_conversion")
    getItems(out, dim = 1) <- "GLO"

    # correct the prices for dry matter values
    out <- out[, , ] / dm[, , ]

    description <- paste0("Prices from the ", datasource,
                          " model projections plus crude estimates for missing commodities for year ", year, ".")

  } else if (datasource == "FAO") {
    out <- calcOutput(type = "PriceAgriculture", datasource = "FAO", aggregate = "GLO")
    out <- out[, year, ]
    out <- collapseNames(out)

    missingCommodities <- setdiff(findset("kall"), getNames(out))
    out <- add_columns(x = out, dim = 3.1, addnm = missingCommodities)

    # out[,,"fish"]             <- 2067.626632 # OECD/FAO price of fish food traded in 2005 #nolint
    out[, , "fish"]             <- 1400 # FAO FIGIS database, export value divided by export volume in 2005, rounded
    # out[, , "oilpalm"]          <- 100 # palmoilpoint.blogspot.de (527 malaysian ringgit   = 118 USD rounded to 100) #nolint
    out[, , "oilcakes"]         <- 300 # https://www.indexmundi.com/commodities/?commodity=soybean-meal&months=180
    out[, , "brans"]            <- 100 # quick google search
    out[, , "distillers_grain"] <- 200 # quick google search
    out[, , "ethanol"]          <- 900 # quick google search
    out[, , "alcohol"]          <- 300 # a guess that a liter of beer is 1$
    # BB: this is luxury. I would go for 30 cents like Sternie
    #     (also beacuse we are interested in the pirmary products going into alcohol, not the marketed product).
    out[, , "molasses"]         <- 50 # quick google search
    out[, , "wood"]             <- 150 # quick google search
    out[, , "woodfuel"]         <- 100 # quick google search
    out[, , "begr"]             <- 50 # quick google search
    out[, , "betr"]             <- 50 # quick google search
    out[, , "res_cereals"]      <- 50 # quick google search
    out[, , "res_fibrous"]      <- 50 # quick google search
    out[, , "res_nonfibrous"]   <- 50 # quick google search
    out[, , "pasture"]          <- 50 # quick google search
    out[, , "scp"]              <- 1500 # using upper end prices similar to fishmeal
    out[, , "foddr"]            <- 50 # quick google search
    out[, , "sugar"]            <- 250 # quick google search

    conv <- c("fish", "oilcakes", "brans", "distillers_grain", "ethanol", "alcohol",
              "molasses", "wood", "woodfuel", "begr",
              "betr", "res_cereals", "res_fibrous", "pasture", "scp",
              "foddr", "sugar")
    getItems(out, dim = 1) <- "USA"
    out[, , conv] <- GDPuc::toolConvertGDP(out[, , conv], unit_in = "constant 2005 US$MER",
                                           unit_out = "constant 2017 US$MER",
                                           replace_NAs = "no_conversion")
    getItems(out, dim = 1) <- "GLO"

    # correct the prices from online sources for dry matter values
    out[, , missingCommodities] <- out[, , missingCommodities] / dm[, , missingCommodities]

    # set years to NULL
    out <- setYears(out, NULL)

    description <- paste0("Prices from the ", datasource,
                          " model projections plus crude estimates for missing commodities for year ", year, ".")
  }

  out <- out[, , findset(products, noset = "original")]

  return(list(x = out,
              weight = NULL,
              unit = "US$2017/tDM",
              description = description,
              isocountries = FALSE))
}