gcamdata: Build the GCAM Input Datasets

Documented in module_aglu_L203.ag_an_demand_input

# Copyright 2019 Battelle Memorial Institute; see the LICENSE file.

#' module_aglu_L203.ag_an_demand_input
#'
#' Builds agriculture demand inputs for the core and each SSP scenario.
#'
#' @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{L203.Supplysector_demand}, \code{L203.NestingSubsectorAll_demand_food},
#'   \code{L203.SubsectorAll_demand_food}, \code{L203.SubsectorAll_demand_nonfood},
#'   \code{L203.StubTech_demand_food}, \code{L203.StubTech_demand_nonfood}, \code{L203.SubregionalShares},
#'   \code{L203.DemandFunction_food}, \code{L203.DemandStapleParams}, \code{L203.DemandNonStapleParams},
#'   \code{L203.DemandStapleRegBias}, \code{L203.DemandNonStapleRegBias}, \code{L203.StapleBaseService},
#'   \code{L203.NonStapleBaseService}, \code{L203.GlobalTechCoef_demand}, \code{L203.GlobalTechShrwt_demand}, \code{L203.GlobalTechInterp_demand},
#'   \code{L203.StubTechProd_food}, \code{L203.StubTechProd_nonfood_crop}, \code{L203.StubTechProd_nonfood_meat},
#'   \code{L203.StubTechProd_For}, \code{L203.StubCalorieContent},
#'   \code{L203.PerCapitaBased}, \code{L203.BaseService}, \code{L203.IncomeElasticity}, \code{L203.PriceElasticity}. The
#'   corresponding file in the original data system was \code{L203.demand_input.R} (aglu level2).
#' @details This chunk specifies the input tables for agriculture demand: generic information for supply sector, subsector and technology,
#' food and non-food demand in calibration years, forest product demand, net exports and caloric contents in calibration and future years,
#' income elasticities for future years in core and SSP scenarios, as well as price elasticities.
#' @importFrom assertthat assert_that
#' @importFrom dplyr bind_rows filter if_else group_by lag left_join mutate select summarise
#' @importFrom tidyr gather replace_na spread
#' @author RC July 2017
module_aglu_L203.ag_an_demand_input <- function(command, ...) {
  if(command == driver.DECLARE_INPUTS) {
    return(c(FILE = "common/GCAM_region_names",
             FILE = "aglu/A_demand_food_staples",
             FILE = "aglu/A_demand_food_nonstaples",
             FILE = "aglu/A_demand_supplysector",
             FILE = "aglu/A_demand_nesting_subsector",
             FILE = "aglu/A_demand_subsector",
             FILE = "aglu/A_demand_technology",
             FILE = "aglu/A_fuelprefElasticity_ssp1",
             FILE = "aglu/A_diet_bias",
             "L101.ag_Food_Pcal_R_C_Y",
             "L101.ag_kcalg_R_C_Y",
             "L105.an_Food_Pcal_R_C_Y",
             "L105.an_kcalg_R_C_Y",
             "L109.ag_ALL_Mt_R_C_Y",
             "L109.an_ALL_Mt_R_C_Y",
             "L110.For_ALL_bm3_R_Y"
             ))
  } else if(command == driver.DECLARE_OUTPUTS) {
    return(c("L203.Supplysector_demand",
             "L203.NestingSubsectorAll_demand_food",
             "L203.SubsectorAll_demand_food",
             "L203.SubsectorAll_demand_nonfood",
             "L203.StubTech_demand_food",
             "L203.StubTech_demand_nonfood",
             "L203.SubregionalShares",
             "L203.DemandFunction_food",
             "L203.DemandStapleParams",
             "L203.DemandNonStapleParams",
             "L203.DemandStapleRegBias",
             "L203.DemandNonStapleRegBias",
             "L203.StapleBaseService",
             "L203.NonStapleBaseService",
             "L203.GlobalTechCoef_demand",
             "L203.GlobalTechShrwt_demand",
             "L203.StubTechProd_food",
             "L203.StubTechProd_nonfood_crop",
             "L203.StubTechProd_nonfood_meat",
             "L203.StubTechProd_For",
             "L203.StubCalorieContent",
             "L203.PerCapitaBased",
             "L203.BaseService",
             "L203.IncomeElasticity",
             "L203.PriceElasticity",
             "L203.FuelPrefElast_ssp1",
             "L203.GlobalTechInterp_demand"))
  } else if(command == driver.MAKE) {

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

    GCAM_commodity <- GCAM_region_ID <- element <- item <- value <- year <-
      region <- technology <- supplysector <- calOutputValue <- OtherUses_Mt <-
      NetExp_Mt <- fixedOutput <- Cons_bm3 <- efficiency <- energy.final.demand <-
      gcam.consumer <- nodeInput <- demand_type <- staples.food.demand.input <-
      non.staples.food.demand.input <- subsector <- NULL   # silence package check notes

    # Load required inputs
    GCAM_region_names <- get_data(all_data, "common/GCAM_region_names", strip_attributes = TRUE)
    A_demand_supplysector <- get_data(all_data, "aglu/A_demand_supplysector", strip_attributes = TRUE)
    A_demand_nesting_subsector <- get_data(all_data, "aglu/A_demand_nesting_subsector", strip_attributes = TRUE)
    A_demand_subsector <- get_data(all_data, "aglu/A_demand_subsector", strip_attributes = TRUE)
    A_demand_technology <- get_data(all_data, "aglu/A_demand_technology", strip_attributes = TRUE)
    A_fuelprefElasticity_ssp1 <- get_data(all_data, "aglu/A_fuelprefElasticity_ssp1", strip_attributes = TRUE)
    A_demand_food_staples <- get_data(all_data, "aglu/A_demand_food_staples", strip_attributes = TRUE)
    A_demand_food_nonstaples <- get_data(all_data, "aglu/A_demand_food_nonstaples", strip_attributes = TRUE)
    A_diet_bias <- get_data(all_data, "aglu/A_diet_bias", strip_attributes = TRUE)
    L101.ag_Food_Pcal_R_C_Y <- get_data(all_data, "L101.ag_Food_Pcal_R_C_Y")
    L101.ag_kcalg_R_C_Y <- get_data(all_data, "L101.ag_kcalg_R_C_Y")
    L105.an_Food_Pcal_R_C_Y <- get_data(all_data, "L105.an_Food_Pcal_R_C_Y")
    L105.an_kcalg_R_C_Y <- get_data(all_data, "L105.an_kcalg_R_C_Y")
    L109.ag_ALL_Mt_R_C_Y <- get_data(all_data, "L109.ag_ALL_Mt_R_C_Y")
    L109.an_ALL_Mt_R_C_Y <- get_data(all_data, "L109.an_ALL_Mt_R_C_Y")
    L110.For_ALL_bm3_R_Y <- get_data(all_data, "L110.For_ALL_bm3_R_Y")

    # Build L203.Supplysector_demand: generic info for demand sectors by region
    A_demand_supplysector %>%
      write_to_all_regions(c(LEVEL2_DATA_NAMES[["Supplysector"]], LOGIT_TYPE_COLNAME), GCAM_region_names = GCAM_region_names) %>%
      filter(!region %in% aglu.NO_AGLU_REGIONS) -> # Remove any regions for which agriculture and land use are not modeled
      L203.Supplysector_demand

    # Build L203.NestingSubsectorAll_demand_food: generic info for food demand nesting subsectors by region
    # Filter for food demand since we only add extra nest to food sectors
    A_demand_nesting_subsector %>%
      write_to_all_regions(c(LEVEL2_DATA_NAMES[["SubsectorAll"]], LOGIT_TYPE_COLNAME), GCAM_region_names = GCAM_region_names) %>%
      filter(!region %in% aglu.NO_AGLU_REGIONS,
             !grepl("NonFood", supplysector)) -> # Remove any regions for which agriculture and land use are not modeled
      L203.NestingSubsectorAll_demand_food

    # Build L203.SubsectorAll_demand: generic info for demand subsectors by region
    A_demand_subsector %>%
      write_to_all_regions(c(c(LEVEL2_DATA_NAMES[["SubsectorAll"]], "subsector0"), LOGIT_TYPE_COLNAME), GCAM_region_names = GCAM_region_names) %>%
      filter(!region %in% aglu.NO_AGLU_REGIONS) -> # Remove any regions for which agriculture and land use are not modeled
      L203.SubsectorAll_demand

    # Split subsector table into food and non-food demand, because food gets extra nesting and non-food does not
    L203.SubsectorAll_demand %>%
      filter(grepl("NonFood", supplysector)) -> L203.SubsectorAll_demand_nonfood
    L203.SubsectorAll_demand %>%
      filter(!grepl("NonFood", supplysector)) -> L203.SubsectorAll_demand_food

    # Build L203.StubTech_demand: identification of stub technologies for demands by region
    A_demand_technology %>%
      write_to_all_regions(c(LEVEL2_DATA_NAMES[["Tech"]], "subsector0"), GCAM_region_names = GCAM_region_names) %>%
      rename(stub.technology = technology) %>%
      filter(!region %in% aglu.NO_AGLU_REGIONS) -> # Remove any regions for which agriculture and land use are not modeled
      L203.StubTech_demand

    # Split stub technology table into food and non-food demand, because food gets extra nesting and non-food does not
    L203.StubTech_demand %>%
      filter(grepl("NonFood", supplysector)) -> L203.StubTech_demand_nonfood
    L203.StubTech_demand %>%
      filter(!grepl("NonFood", supplysector)) -> L203.StubTech_demand_food

    # Build L203.GlobalTechCoef_demand: input names of demand technologies
    A_demand_technology %>%
      repeat_add_columns(tibble(year = MODEL_YEARS)) %>%
      rename(sector.name = supplysector, subsector.name = subsector) %>%
      select(c(LEVEL2_DATA_NAMES[["GlobalTechCoef"]])) ->
      L203.GlobalTechCoef_demand

    # Build L203.GlobalTechShrwt_demand: shareweights of demand technologies
    L203.GlobalTechCoef_demand %>%
      select(LEVEL2_DATA_NAMES[["GlobalTechYr"]]) %>%
      mutate(share.weight = 1) ->
      L203.GlobalTechShrwt_demand

    # Build L203.GlobalTechInterp_demand: Interpolation rule to fix initial shareweights
     A_demand_technology %>%
      filter(subsector != technology) %>%
      mutate(from.year = MODEL_FINAL_BASE_YEAR,
             to.year = max(MODEL_YEARS)) %>%
      repeat_add_columns(tibble(year = MODEL_YEARS)) %>%
      rename(sector.name = supplysector, subsector.name = subsector) %>%
      select(LEVEL2_DATA_NAMES[["GlobalTechInterp"]])-> L203.GlobalTechInterp_demand

    # Calibrated staple and non-staple demands of crops and meat
    # Create table of regions, technologies and all base years
    # NOTE: Easiest if the model base years are subsetted from a full table as a last step in the construction of each of these tables
    A_demand_technology %>%
      write_to_all_regions(c(LEVEL2_DATA_NAMES[["Tech"]], "minicam.energy.input", "market.name", "subsector0"), GCAM_region_names = GCAM_region_names) %>%
      mutate(market.name = region, stub.technology = technology) ->
      A_demand_technology_R
    # Add all base years
    A_demand_technology_R %>%
      repeat_add_columns(tibble(year = MODEL_BASE_YEARS)) ->
      A_demand_technology_R_Yh
    # Add all model years
    A_demand_technology_R %>%
      repeat_add_columns(tibble(year = MODEL_YEARS)) ->
      A_demand_technology_R_Y

    # Build L203.StubTechProd_food: crop and meat food supply by technology and region
    L101.ag_Food_Pcal_R_C_Y %>%
      # Combine crop and meat food demand in Pcal
      bind_rows(L105.an_Food_Pcal_R_C_Y) %>%
      filter(year %in% MODEL_BASE_YEARS) %>%
      left_join_error_no_match(GCAM_region_names, by = "GCAM_region_ID") ->
      L203.ag_an_Food_Pcal_R_C_Y

    A_demand_technology_R_Yh %>%
      # Select food demand
      filter(supplysector %in% c("FoodDemand_Staples", "FoodDemand_NonStaples")) %>%
      # Map in food demand by region / commodity / year
      left_join_error_no_match(L203.ag_an_Food_Pcal_R_C_Y, by = c("region", "technology" = "GCAM_commodity", "year")) %>%
      mutate(calOutputValue = round(value, aglu.DIGITS_CALOUTPUT),
             share.weight.year = year,
             # Subsector and technology shareweights (subsector requires the year as well)
             subs0.share.weight = if_else(calOutputValue > 0, 1, 0), # NESTING SUBSECTOR SWS??????
             subs.share.weight = if_else(calOutputValue > 0, 1, 0),
             tech.share.weight = if_else(calOutputValue > 0, 1, 0)) %>%
      select(c(LEVEL2_DATA_NAMES[["StubTechProd"]]), subsector0, subs0.share.weight) %>%
      filter(!region %in% aglu.NO_AGLU_REGIONS) %>%           # Remove any regions for which agriculture and land use are not modeled
      filter(year %in% MODEL_BASE_YEARS) ->                         # Also subset the calibration tables to only the model base years
      L203.StubTechProd_food

    # NonFoodDemands
    L109.ag_ALL_Mt_R_C_Y %>%
      # Combine the balance tables of crop and meat in Mt
      bind_rows(L109.an_ALL_Mt_R_C_Y) %>%
      filter(year %in% MODEL_BASE_YEARS) %>%
      left_join_error_no_match(GCAM_region_names, by = "GCAM_region_ID") %>%
      select(GCAM_region_ID, region, GCAM_commodity, year, OtherUses_Mt, NetExp_Mt) -> # only nonfood and net exports data will be used
      L203.ag_an_ALL_Mt_R_C_Y

    A_demand_technology_R_Yh %>%
      # Select nonfood demand
      filter(supplysector %in% c("NonFoodDemand_Crops", "NonFoodDemand_Meat")) %>%
      # Map in nonfood demand by region / commodity / year
      left_join_error_no_match(L203.ag_an_ALL_Mt_R_C_Y, by = c("region", "technology" = "GCAM_commodity", "year")) %>%
      mutate(calOutputValue = round(OtherUses_Mt, aglu.DIGITS_CALOUTPUT),
             share.weight.year = year,
             # Subsector and technology shareweights (subsector requires the year as well)
             subs.share.weight = if_else(calOutputValue > 0, 1, 0),
             tech.share.weight = if_else(calOutputValue > 0, 1, 0)) %>%
      select(LEVEL2_DATA_NAMES[["StubTechProd"]]) %>%
      filter(!region %in% aglu.NO_AGLU_REGIONS) %>%           # Remove any regions for which agriculture and land use are not modeled
      filter(year %in% MODEL_BASE_YEARS) ->                         # Also subset the calibration tables to only the model base years
      L203.StubTechProd_nonfood


    # Build L203.StubTechProd_For: Forest product demand by technology and region
    L110.For_ALL_bm3_R_Y %>%
      unique() %>%
      filter(year %in% MODEL_BASE_YEARS) %>%
      left_join_error_no_match(GCAM_region_names, by = "GCAM_region_ID") %>%
      ungroup() %>%
      select(GCAM_region_ID, region, year, Cons_bm3) -> # Select forest demand
      L203.For_ALL_bm3_R_Y

    A_demand_technology_R_Yh %>%
      filter(supplysector == "NonFoodDemand_Forest") %>%
      # Map in forest product demand in bm3
      left_join_error_no_match(L203.For_ALL_bm3_R_Y, by = c("region", "year")) %>%
      mutate(calOutputValue = round(Cons_bm3, aglu.DIGITS_CALOUTPUT),
             share.weight.year = year,
             # Subsector and technology shareweights (subsector requires the year as well)
             subs.share.weight = if_else(calOutputValue > 0, 1, 0),
             tech.share.weight = if_else(calOutputValue > 0, 1, 0)) %>%
      select(LEVEL2_DATA_NAMES[["StubTechProd"]]) %>%
      filter(!region %in% aglu.NO_AGLU_REGIONS) %>%           # Remove any regions for which agriculture and land use are not modeled
      filter(year %in% MODEL_BASE_YEARS) ->                         # Also subset the calibration tables to only the model base years
      L203.StubTechProd_For

    # Build L203.StubCalorieContent:
    # calorie content of food crops (incl secondary products) and meat commodities
    L101.ag_kcalg_R_C_Y %>%
      # Combine the weigted average caloric content of crop and meat products
      bind_rows(L105.an_kcalg_R_C_Y) %>%
      filter(year %in% MODEL_BASE_YEARS) %>%
      left_join(GCAM_region_names, by = "GCAM_region_ID") ->
      L203.ag_an_kcalg_R_C_Y

    A_demand_technology_R_Y %>%
      filter(supplysector %in% c("FoodDemand_Staples", "FoodDemand_NonStaples")) %>%
      # Create NAs for future years, use left_join instead
      left_join(L203.ag_an_kcalg_R_C_Y, by = c("region", "technology" = "GCAM_commodity", "year")) %>%
      mutate(efficiency = round(value, aglu.DIGITS_CALOUTPUT)) %>%
      # For each region / commodity,
      group_by(region, subsector0, subsector, technology) %>%
      # Calorie content are held constant in the future, so set value for future years at the final base year value
      mutate(efficiency = replace(efficiency, year > max(MODEL_BASE_YEARS), efficiency[year == max(MODEL_BASE_YEARS)])) %>%
      ungroup() %>%
      select(c(LEVEL2_DATA_NAMES[["StubTechCalorieContent"]], "subsector0")) %>%
      filter(!region %in% aglu.NO_AGLU_REGIONS) ->          # Remove any regions for which agriculture and land use are not modeled
      L203.StubCalorieContent

    # FINAL DEMANDS
    # Build L203.PerCapitaBased: per-capita final demand attributes that do not vary by time period
    A_demand_supplysector %>%
      filter(!is.na(energy.final.demand)) %>%
      write_to_all_regions(LEVEL2_DATA_NAMES[["PerCapitaBased"]], GCAM_region_names = GCAM_region_names) %>%
      filter(!region %in% aglu.NO_AGLU_REGIONS) ->          # Remove any regions for which agriculture and land use are not modeled
      L203.PerCapitaBased

    A_demand_supplysector %>%
      filter(!is.na(energy.final.demand)) %>%
      repeat_add_columns(tibble(year = MODEL_FUTURE_YEARS)) %>%
      write_to_all_regions(LEVEL2_DATA_NAMES[["IncomeElasticity"]], GCAM_region_names = GCAM_region_names) %>%
      filter(!region %in% aglu.NO_AGLU_REGIONS) ->          # Remove any regions for which agriculture and land use are not modeled
      L203.IncomeElasticity

    A_demand_supplysector %>%
      filter(!is.na(energy.final.demand)) %>%
      repeat_add_columns(tibble(year = MODEL_FUTURE_YEARS)) %>%
      write_to_all_regions(LEVEL2_DATA_NAMES[["PriceElasticity"]], GCAM_region_names = GCAM_region_names) %>%
      filter(!region %in% aglu.NO_AGLU_REGIONS) ->          # Remove any regions for which agriculture and land use are not modeled
      L203.PriceElasticity

    # Fuel preference elasticity
    # Build L203.FuelPrefElast_ssp1: Fuel preference elasticities for meat in SSP1
    #Keep the nesting subsector
    names_FuelPrefElast_nest <- c("region", "supplysector", "subsector0", "subsector",  "year.fillout", "fuelprefElasticity")
    A_fuelprefElasticity_ssp1 %>%
      mutate(year.fillout = min(MODEL_BASE_YEARS)) %>%
      write_to_all_regions(names_FuelPrefElast_nest, GCAM_region_names = GCAM_region_names) %>%
      filter(!region %in% aglu.NO_AGLU_REGIONS) ->           # Remove any regions for which agriculture and land use are not modeled
      L203.FuelPrefElast_ssp1


    # Build L203.BaseService: base service of (standard) final demands
    # This excludes food demands, which have a different demand formulation
    Prod_colnames <- c("region", "supplysector", "year", "calOutputValue")
    L203.StubTechProd_nonfood[Prod_colnames]%>%
      # Combine all food and nonfood demand
      bind_rows(L203.StubTechProd_For[Prod_colnames]) %>%
      group_by(region, supplysector, year) %>%
      # Sum the total of all commodities by region and supply sector
      summarise(calOutputValue = sum(calOutputValue)) %>%
      ungroup() %>%
      rename(energy.final.demand = supplysector, base.service = calOutputValue) %>%
      filter(!region %in% aglu.NO_AGLU_REGIONS) %>%           # Remove any regions for which agriculture and land use are not modeled
      filter(year %in% MODEL_BASE_YEARS) ->                   # also subset the calibration tables to only the model base years
      L203.BaseService

    # FOOD DEMAND MODEL
    # Sub-regional population and income shares can be used to sub-divide a region's consumers into classes (e.g., by income
    # group, urban/rural, etc). Shares are required to be read in, but as no sub-divisions are currently used, the shares
    # in all regions are 1.
    L203.SubregionalShares <- A_demand_food_staples %>%
      select(gcam.consumer, nodeInput) %>%
      mutate(pop.year.fillout = min(MODEL_BASE_YEARS),
             inc.year.fillout = min(MODEL_BASE_YEARS),
             subregional.population.share = 1,
             subregional.income.share = 1) %>%
      write_to_all_regions(LEVEL2_DATA_NAMES[["SubregionalShares"]], GCAM_region_names = GCAM_region_names) %>%
      filter(!region %in% aglu.NO_AGLU_REGIONS)

    L203.DemandFunction_food <- A_demand_food_staples %>%
      write_to_all_regions(LEVEL2_DATA_NAMES[["DemandFunction_food"]], GCAM_region_names = GCAM_region_names) %>%
      filter(!region %in% aglu.NO_AGLU_REGIONS)

    L203.DemandStapleParams <- A_demand_food_staples %>%
      write_to_all_regions(LEVEL2_DATA_NAMES[["DemandStapleParams"]], GCAM_region_names = GCAM_region_names) %>%
      filter(!region %in% aglu.NO_AGLU_REGIONS)

    L203.DemandNonStapleParams <- A_demand_food_nonstaples %>%
      write_to_all_regions(LEVEL2_DATA_NAMES[["DemandNonStapleParams"]], GCAM_region_names = GCAM_region_names) %>%
      filter(!region %in% aglu.NO_AGLU_REGIONS)

    if(nrow(A_diet_bias) > 0) {
    L203.DemandStapleRegBias <- select(L203.DemandStapleParams, region, gcam.consumer, nodeInput, staples.food.demand.input) %>%
      left_join_error_no_match(A_diet_bias,
                               by = c(staples.food.demand.input = "demand_type")) %>%
      select(LEVEL2_DATA_NAMES[["DemandStapleRegBias"]])
    } else {
      # no convergence values, just return an empty tibble
      L203.DemandStapleRegBias <- as_tibble(sapply(LEVEL2_DATA_NAMES[['DemandStapleRegBias']], function(d) {character()}))
    }


    if(nrow(A_diet_bias) > 0) {
    L203.DemandNonStapleRegBias <- select(L203.DemandNonStapleParams, region, gcam.consumer, nodeInput, non.staples.food.demand.input) %>%
      left_join_error_no_match(A_diet_bias,
                               by = c(non.staples.food.demand.input = "demand_type")) %>%
      select(LEVEL2_DATA_NAMES[["DemandNonStapleRegBias"]])
      #select(c("region", "gcam.consumer", "nodeInput", "non.staples.food.demand.input", "regional.bias.year", "regional.bias"))
    } else {
      # no convergence values, just return an empty tibble
      L203.DemandNonStapleRegBias <- as_tibble(sapply(LEVEL2_DATA_NAMES[['DemandNonStapleRegBias']], function(d) {character()}))
    }

    L203.Demand <- L203.StubTechProd_food %>%
      group_by(region, supplysector, year) %>%
      summarise(base.service = sum(calOutputValue)) %>%
      ungroup()

    L203.StapleBaseService <- filter(L203.Demand, supplysector %in% A_demand_food_staples$staples.food.demand.input) %>%
      rename(staples.food.demand.input = supplysector) %>%
      left_join_error_no_match(select(L203.DemandStapleParams, region, gcam.consumer, nodeInput, staples.food.demand.input),
                             by = c("region", "staples.food.demand.input")) %>%
      select(LEVEL2_DATA_NAMES[["StapleBaseService"]])

    L203.NonStapleBaseService <- filter(L203.Demand, supplysector %in% A_demand_food_nonstaples$non.staples.food.demand.input) %>%
      rename(non.staples.food.demand.input = supplysector) %>%
      left_join_error_no_match(select(L203.DemandNonStapleParams, region, gcam.consumer, nodeInput, non.staples.food.demand.input),
                               by = c("region", "non.staples.food.demand.input")) %>%
      select(LEVEL2_DATA_NAMES[["NonStapleBaseService"]])

    #FINAL OUTPUT
    L203.Supplysector_demand %>%
      add_title("Generic information for agriculture demand sectors") %>%
      add_units("Unitless") %>%
      add_comments("Specify generic info for demand sectors") %>%
      add_comments("Remove any regions for which agriculture and land use are not modeled") %>%
      add_legacy_name("L203.Supplysector_demand") %>%
      add_precursors("common/GCAM_region_names",
                     "aglu/A_demand_supplysector") ->
      L203.Supplysector_demand

    L203.NestingSubsectorAll_demand_food %>%
      add_title("Generic information for agriculture food demand sectors") %>%
      add_units("Unitless") %>%
      add_comments("Specify generic info for food demand sectors") %>%
      add_comments("Remove any regions for which agriculture and land use are not modeled") %>%
      add_legacy_name("L203.Supplysector_demand") %>%
      add_precursors("common/GCAM_region_names",
                     "aglu/A_demand_supplysector",
                     "aglu/A_demand_nesting_subsector") ->
      L203.NestingSubsectorAll_demand_food

    L203.SubsectorAll_demand_food %>%
      add_title("Generic information for agriculture food demand subsectors") %>%
      add_units("Unitless") %>%
      add_comments("Specify generic info for food demand subsectors") %>%
      add_comments("Remove any regions for which agriculture and land use are not modeled") %>%
      add_legacy_name("L203.SubsectorAll_demand") %>%
      add_precursors("common/GCAM_region_names",
                     "aglu/A_demand_subsector") ->
      L203.SubsectorAll_demand_food

    L203.SubsectorAll_demand_nonfood %>%
      add_title("Generic information for agriculture non-food demand subsectors") %>%
      add_units("Unitless") %>%
      add_comments("Specify generic info for non-food demand subsectors") %>%
      add_comments("Remove any regions for which agriculture and land use are not modeled") %>%
      add_legacy_name("L203.SubsectorAll_demand") %>%
      add_precursors("common/GCAM_region_names",
                     "aglu/A_demand_subsector") ->
      L203.SubsectorAll_demand_nonfood

    L203.StubTech_demand_food %>%
      add_title("Identification for stub technologies for agriculture food demands") %>%
      add_units("Unitless") %>%
      add_comments("Specify identification of stub technologies for food demands") %>%
      add_comments("Remove any regions for which agriculture and land use are not modeled") %>%
      add_legacy_name("L203.StubTech_demand") %>%
      add_precursors("common/GCAM_region_names",
                     "aglu/A_demand_technology") ->
      L203.StubTech_demand_food

    L203.StubTech_demand_nonfood %>%
      add_title("Identification for stub technologies for agriculture non-food demands") %>%
      add_units("Unitless") %>%
      add_comments("Specify identification of stub technologies for non-food demands") %>%
      add_comments("Remove any regions for which agriculture and land use are not modeled") %>%
      add_legacy_name("L203.StubTech_demand") %>%
      add_precursors("common/GCAM_region_names",
                     "aglu/A_demand_technology") ->
      L203.StubTech_demand_nonfood

    L203.GlobalTechCoef_demand %>%
      add_title("Input names of agriculture demand technologies") %>%
      add_units("Unitless") %>%
      add_comments("Specify input names of demand technologies") %>%
      add_legacy_name("L203.GlobalTechCoef_demand") %>%
      add_precursors("aglu/A_demand_technology") ->
      L203.GlobalTechCoef_demand

    L203.GlobalTechShrwt_demand %>%
      add_title("Shareweights of agriculture demand technologies") %>%
      add_units("Unitless") %>%
      add_comments("Specify shareweights of agriculture demand technologies") %>%
      add_legacy_name("L203.GlobalTechShrwt_demand") %>%
      same_precursors_as(L203.GlobalTechCoef_demand) ->
      L203.GlobalTechShrwt_demand

    L203.GlobalTechInterp_demand %>%
      add_title("Interpolation rule for demand technologies") %>%
      add_units("Unitless") %>%
      add_comments("Specify interpolation rule for agriculture demand technologies") %>%
      add_legacy_name("L203.GlobalTechInterp_demand") %>%
      same_precursors_as(L203.GlobalTechShrwt_demand) ->
      L203.GlobalTechInterp_demand

    L203.StubTechProd_food %>%
      add_title("Food supply (staple and non-staple) by technology and region") %>%
      add_units("Pcal") %>%
      add_comments("Crop and meat food supply in calibration years by region / commodity") %>%
      add_comments("Remove any regions for which agriculture and land use are not modeled") %>%
      add_legacy_name("L203.StubTechProd_food_crop") %>%
      add_precursors("common/GCAM_region_names",
                     "aglu/A_demand_technology",
                     "L101.ag_Food_Pcal_R_C_Y",
                     "L105.an_Food_Pcal_R_C_Y") ->
      L203.StubTechProd_food

    L203.StubTechProd_nonfood %>%
      filter(supplysector == "NonFoodDemand_Crops") %>%
      add_title("Crop non-food demand by technology and region") %>%
      add_units("Mt") %>%
      add_comments("Map in crop non-food demand in calibration years by region / commodity") %>%
      add_comments("Remove any regions for which agriculture and land use are not modeled") %>%
      add_legacy_name("L203.StubTechProd_nonfood_crop") %>%
      add_precursors("common/GCAM_region_names",
                     "aglu/A_demand_technology",
                     "L109.ag_ALL_Mt_R_C_Y") ->
      L203.StubTechProd_nonfood_crop

    L203.StubTechProd_nonfood %>%
      filter(supplysector == "NonFoodDemand_Meat") %>%
      add_title("Meat non-food demand by technology and region") %>%
      add_units("Mt") %>%
      add_comments("Map in meat non-food demand in calibration years by region / commodity") %>%
      add_comments("Remove any regions for which agriculture and land use are not modeled") %>%
      add_legacy_name("L203.StubTechProd_nonfood_meat") %>%
      add_precursors("common/GCAM_region_names",
                     "aglu/A_demand_technology",
                     "L109.an_ALL_Mt_R_C_Y") ->
      L203.StubTechProd_nonfood_meat

    L203.StubTechProd_For %>%
      add_title("Forest product demand by technology and region") %>%
      add_units("bm3") %>%
      add_comments("Map in forest demand in calibration years by region / commodity") %>%
      add_comments("Remove any regions for which agriculture and land use are not modeled") %>%
      add_legacy_name("L203.StubTechProd_For") %>%
      add_precursors("common/GCAM_region_names",
                     "aglu/A_demand_technology",
                     "L110.For_ALL_bm3_R_Y") ->
      L203.StubTechProd_For


    L203.StubCalorieContent %>%
      add_title("Caloric content of food crops") %>%
      add_units("kcal/g") %>%
      add_comments("Map in weighted average of caloric content in calibration years by region / commodity") %>%
      add_comments("Future years are set to the final base year value") %>%
      add_comments("Remove any regions for which agriculture and land use are not modeled") %>%
      add_legacy_name("L203.StubCalorieContent_crop") %>%
      add_precursors("common/GCAM_region_names",
                     "aglu/A_demand_technology",
                     "L101.ag_kcalg_R_C_Y",
                     "L105.an_kcalg_R_C_Y") ->
      L203.StubCalorieContent

    L203.PerCapitaBased %>%
      add_title("Per-capita final agriculture demand attributes") %>%
      add_units("Unitless") %>%
      add_comments("Attributes do not vary by time period") %>%
      add_comments("Remove any regions for which agriculture and land use are not modeled") %>%
      add_legacy_name("L203.PerCapitaBased") %>%
      add_precursors("common/GCAM_region_names",
                     "aglu/A_demand_supplysector") ->
      L203.PerCapitaBased

    L203.BaseService %>%
      add_title("Base service of final demands") %>%
      add_units("Pcal/Mt/bm3") %>%
      add_comments("Calculate the total final demands by supply sector in each region") %>%
      add_comments("Remove any regions for which agriculture and land use are not modeled") %>%
      add_legacy_name("L203.BaseService") %>%
      add_precursors("common/GCAM_region_names",
                     "aglu/A_demand_supplysector",
                     "aglu/A_demand_technology",
                     "L101.ag_Food_Pcal_R_C_Y",
                     "L105.an_Food_Pcal_R_C_Y",
                     "L109.ag_ALL_Mt_R_C_Y",
                     "L109.an_ALL_Mt_R_C_Y",
                     "L110.For_ALL_bm3_R_Y") ->
      L203.BaseService

    L203.IncomeElasticity %>%
      add_title("Future income elasticity of non-food demands by region") %>%
      add_units("Unitless") %>%
      add_comments("Values copied from assumptions") %>%
      add_legacy_name("L203.IncomeElasticity") %>%
      add_precursors("common/GCAM_region_names", "aglu/A_demand_supplysector") ->
      L203.IncomeElasticity

    L203.PriceElasticity %>%
      add_title("Price elasticities of non-food demands by supply sector") %>%
      add_units("Unitless") %>%
      add_comments("Values copied from assumptions") %>%
      add_legacy_name("L203.PriceElasticity") %>%
      add_precursors("common/GCAM_region_names", "aglu/A_demand_supplysector") ->
      L203.PriceElasticity

    L203.FuelPrefElast_ssp1 %>%
      add_title("Fuel preference elasticities for meat in SSP1") %>%
      add_units("Unitless") %>%
      add_comments("Specify the minimum base year value") %>%
      add_comments("Remove any regions for which agriculture and land use are not modeled") %>%
      add_legacy_name("L203.FuelPrefElast_ssp1") %>%
      add_precursors("aglu/A_fuelprefElasticity_ssp1") ->
      L203.FuelPrefElast_ssp1

    L203.SubregionalShares %>%
      add_title("Subregional population and income shares for food demand") %>%
      add_units("Unitless") %>%
      add_comments("Names copied from assumptions") %>%
      add_precursors("aglu/A_demand_food_staples") ->
      L203.SubregionalShares

    L203.DemandFunction_food %>%
      add_title("Demand function for food demand") %>%
      add_units("Unitless") %>%
      add_comments("Names copied from assumptions") %>%
      add_precursors("aglu/A_demand_food_staples") ->
      L203.DemandFunction_food

    L203.DemandStapleParams %>%
      add_title("Food demand function parameters for staples") %>%
      add_units("Unitless") %>%
      add_comments("Values copied from assumptions to all regions") %>%
      add_precursors("aglu/A_demand_food_staples") ->
      L203.DemandStapleParams

    L203.DemandNonStapleParams %>%
      add_title("Food demand function parameters for non-staples") %>%
      add_units("Unitless") %>%
      add_comments("Values copied from assumptions to all regions") %>%
      add_precursors("aglu/A_demand_food_nonstaples") ->
      L203.DemandNonStapleParams

    L203.DemandStapleRegBias %>%
      add_title("Food demand function regional bias parameters for staples") %>%
      add_units("Unitless") %>%
      add_comments("Values taken from assumptions; computed offline in ancillary analysis") %>%
      add_precursors("aglu/A_demand_food_staples", "aglu/A_diet_bias") ->
      L203.DemandStapleRegBias

    L203.DemandNonStapleRegBias %>%
      add_title("Food demand function regional bias parameters for non-staples") %>%
      add_units("Unitless") %>%
      add_comments("Values taken from assumptions; computed offline in ancillary analysis") %>%
      add_precursors("aglu/A_demand_food_nonstaples", "aglu/A_diet_bias") ->
      L203.DemandNonStapleRegBias

    L203.StapleBaseService %>%
      add_title("Calibrated food demand quantities by year (staples)") %>%
      add_units("PCal/yr") %>%
      add_comments("FAO demand quantities multiplied by caloric content and added by demand category") %>%
      same_precursors_as(L203.BaseService) %>%
      add_precursors("aglu/A_demand_food_staples") ->
      L203.StapleBaseService

    L203.NonStapleBaseService %>%
      add_title("Calibrated food demand quantities by year (non-staples)") %>%
      add_units("PCal/yr") %>%
      add_comments("FAO demand quantities multiplied by caloric content and added by demand category") %>%
      same_precursors_as(L203.BaseService) %>%
      add_precursors("aglu/A_demand_food_nonstaples") ->
      L203.NonStapleBaseService

    return_data(L203.Supplysector_demand, L203.NestingSubsectorAll_demand_food, L203.SubsectorAll_demand_food,
                L203.SubsectorAll_demand_nonfood, L203.StubTech_demand_food, L203.StubTech_demand_nonfood,
                L203.GlobalTechCoef_demand, L203.GlobalTechShrwt_demand, L203.GlobalTechInterp_demand, L203.StubTechProd_food,
                L203.StubTechProd_nonfood_crop, L203.StubTechProd_nonfood_meat, L203.StubTechProd_For,
                L203.StubCalorieContent, L203.PerCapitaBased, L203.BaseService,
                L203.IncomeElasticity, L203.PriceElasticity, L203.FuelPrefElast_ssp1,
                L203.SubregionalShares, L203.DemandFunction_food, L203.DemandStapleParams, L203.DemandNonStapleParams,
                L203.DemandStapleRegBias, L203.DemandNonStapleRegBias, L203.StapleBaseService, L203.NonStapleBaseService)
  } else {
    stop("Unknown command")
  }
}