R/zchunk_LA1321.Cement-gcam-korea.R

In rohmin9122/gcam-korea-release: Build the GCAM-Korea Input Datasets

#' module_gcam.korea_LA1321.Cement
#'
#' Allocate across the states national cement production, input-output cofficients, and energy inputs to cement production
#'
#' @param command API command to execute
#' @param ... other optional parameters, depending on command
#' @return Depends on \code{command}: either a vector of required inputs,
#' a vector of output names, or (if \code{command} is "MAKE") all
#' the generated outputs: \code{L1321.out_Mt_korea_cement_Yh}, \code{L1321.IO_GJkg_korea_cement_F_Yh}, \code{L1321.in_EJ_korea_cement_F_Y}. The corresponding file in the
#' original data system was \code{LA1321.Cement.R} (gcam-korea level1).
#' @details The tables for cement production, i.e., out, and energy inputs, i.e., in, were calculated by applying state shares to national data.
#' @details The state shares were determined by the states' relative values of cement shipments.
#' @details The input-out coefficients were downscaled to the states in proportation to the national data.
#' @importFrom assertthat assert_that
#' @importFrom dplyr filter mutate select
#' @importFrom tidyr gather spread
#' @author M. Roh
module_gcam.korea_LA1321.Cement <- function(command, ...) {
  if(command == driver.DECLARE_INPUTS) {
    return(c(FILE = "gcam-korea/Mining_Manufacturing_Survey",
             "L1321.out_Mt_R_cement_Yh",
             "L1321.IO_GJkg_R_cement_F_Yh",
             "L1321.in_EJ_R_cement_F_Y"))
  } else if(command == driver.DECLARE_OUTPUTS) {
    return(c("L1321.out_Mt_korea_cement_Yh",
             "L1321.IO_GJkg_korea_cement_F_Yh",
             "L1321.in_EJ_korea_cement_F_Y"))
  } else if(command == driver.MAKE) {

    all_data <- list(...)[[1]]

    . <- NULL   # silence package check notes

    # Load required inputs
    Mining_Manufacturing_Survey <- get_data(all_data, "gcam-korea/Mining_Manufacturing_Survey")
    L1321.out_Mt_R_cement_Yh <- get_data(all_data, "L1321.out_Mt_R_cement_Yh")
    L1321.IO_GJkg_R_cement_F_Yh <- get_data(all_data, "L1321.IO_GJkg_R_cement_F_Yh")
    L1321.in_EJ_R_cement_F_Y <- get_data(all_data, "L1321.in_EJ_R_cement_F_Y")

    # ===================================================

    variable <- value_state <- value_share <- value_national <- code <- state <-
      VoS <- sector <- fuel <- year <- value <- GCAM_region_ID <- NULL   # silence package check notes

    # Assigning national cement production to states on the basis of value of shipments (VoS) of
    # the North American Industry Classification System (NAICS) code 3273 by state
    # The NAICS code 3273 includes cement and concrete product manufacturing. Includes cement, ready-mix concrete,
    # concrete pipe, brick, and block, and other concrete products
    Mining_Manufacturing_Survey %>%
      # This is the code for Manufacture of Other Non-metallic Mineral Products
      # c23 contains cement, lime, plaster and ect.
      filter(code == "c23") %>%
      select(state, VoS) %>% # Selecting value of shipments
      gather(variable, value_state, -state) %>% # Converting to long form
      select(-variable) ->
      Cement_VoS_state

    # Here the relative state shares will be calculated. First the states will be summed to get the national total.
    # The national total is a single value
    Cement_VoS_state %>%
      select(-state) %>%
      summarise(value_national = sum(value_state)) %>%
      pull(value_national) ->
      Cement_VoS_national

    # The national total will be added as a separate column, so that the state share can be calculated
    Cement_VoS_state %>%
      # The national total is a single value, so left_join cannot be used
      repeat_add_columns(tibble(value_national = Cement_VoS_national)) %>%
      mutate(value_share = value_state / value_national) %>% # Calculating the state share
      select(state, value_share) ->
      Cement_share_state

    # Creating a list of states
    # This will be helpful when expanding the national table into state-level
    # Note that not every state has cement manufacturing
    state_list <- unique(Cement_share_state$state)

    # This section is calculating state-level data by multiplying the state share by the component in the global data
    # This will generate an output table: Cement production by state / historical year
    L1321.out_Mt_R_cement_Yh %>%
      filter(GCAM_region_ID == gcamkorea.RegionNum) %>% # Filtering for the component
      repeat_add_columns(tibble(state = state_list)) %>% # Expanding the table to the state-level
      left_join_error_no_match(Cement_share_state, by = "state") %>% # Adding the state share we calculated above
      mutate(value = value * value_share) %>% # Multiplying the national amount with the state share
      select(state, sector, year, value) ->
      L1321.out_Mt_korea_cement_Yh

    # This section is downscaling the national input/output (IO) coefficients to the state level
    # Assuming all states have the same IO coefficients for heat, electricity, and limestone
    # This will generate an output table: Input-output coefficients of cement production by state / fuel / historical year
    L1321.IO_GJkg_R_cement_F_Yh %>%
      filter(GCAM_region_ID == gcamkorea.RegionNum) %>%
      repeat_add_columns(tibble(state = state_list)) %>%
      select(state, sector, fuel, year, value) ->
      L1321.IO_GJkg_korea_cement_F_Yh

    # Calculating energy inputs to cement production by state
    # Again, this section is downscaling the national data to the state level, using the state share calculated above
    # Note that this assumes the same fuel blend in all states
    # This will generate an output table: Energy inputs to cement production by state / fuel / historical year
    L1321.in_EJ_R_cement_F_Y %>%
      filter(GCAM_region_ID == gcamkorea.RegionNum) %>%
      repeat_add_columns(tibble(state = state_list)) %>%
      left_join_error_no_match(Cement_share_state, by = "state") %>%
      mutate(value = value_share * value) %>%
      select(state, sector, fuel, year, value) ->
      L1321.in_EJ_korea_cement_F_Y

    # ===================================================

    L1321.out_Mt_korea_cement_Yh %>%
      add_title("Cement production by state / historical year ") %>%
      add_units("Mt") %>%
      add_comments("downscaling national data using state shares") %>%
      add_comments("these state shares were calculated to be proportional to the their values of cement shipments") %>%
      add_legacy_name("L1321.out_Mt_korea_cement_Yh") %>%
      add_precursors("gcam-korea/Mining_Manufacturing_Survey", "L1321.out_Mt_R_cement_Yh") ->
      L1321.out_Mt_korea_cement_Yh

    L1321.IO_GJkg_korea_cement_F_Yh %>%
      add_title("Input-output coefficients of cement production by state / fuel / historical year") %>%
      add_units("GJ/kg and kg/kg") %>%
      add_comments("downscaling national data assuming the same IO coefficients for each respective fuel") %>%
      add_legacy_name("L1321.IO_GJkg_korea_cement_F_Yh") %>%
      add_precursors("L1321.IO_GJkg_R_cement_F_Yh") ->
      L1321.IO_GJkg_korea_cement_F_Yh

    L1321.in_EJ_korea_cement_F_Y %>%
      add_title("Energy inputs to cement production by state / fuel / historical year") %>%
      add_units("EJ") %>%
      add_comments("downscaling national data using state shares") %>%
      add_comments("these state shares were calculated to be proportional to the their values of cement shipments") %>%
      add_legacy_name("L1321.in_EJ_korea_cement_F_Y") %>%
      add_precursors("gcam-korea/Mining_Manufacturing_Survey", "L1321.in_EJ_R_cement_F_Y") ->
      L1321.in_EJ_korea_cement_F_Y

    return_data(L1321.out_Mt_korea_cement_Yh, L1321.IO_GJkg_korea_cement_F_Yh, L1321.in_EJ_korea_cement_F_Y)
  } else {
    stop("Unknown command")
  }
}