Pull request #333: SSP GDP, population, labor force update

Merge in JGCRI/gcam-core from xz/feature/SSPv3_2024 to master Squashed commit of the following: commit 731c683d384dcf5a423af5022e6a54eef804933e Author: Pralit Patel <[email protected]> Date: Fri Aug 23 13:45:07 2024 -0400 Given the change in socio-economics we should commit an updated set of calibrated macro TFP commit 3efcc7749c83d822210b2347919b1bded9e86e18 Author: Pralit Patel <[email protected]> Date: Fri Aug 23 13:44:37 2024 -0400 Bump version number for public release commit 6ad07affea665a370351a38474474dc96f77c794 Author: Pralit Patel <[email protected]> Date: Fri Aug 23 13:43:04 2024 -0400 Increase max tax for SSP4_2p6 as it is close to reaching the old one and can cause the target finder trouble if it tries one that would exceed it. commit 4b9001b9b6ab6791eb889f57bb72a50462edaa6d Author: Pralit Patel <[email protected]> Date: Thu Aug 22 23:37:32 2024 -0400 Fix some precursor statements and update the GCAM_DATA_MAP commit f842c57a7d99932168763447720e44a8ef1dab12 Author: Pralit Patel <[email protected]> Date: Thu Aug 22 23:35:04 2024 -0400 Include Matthew's fix to keep gas trade calibration flows from going negative which can throw of historical natural gas calibration commit 41fa0c1651935d735a3ee4c4b94e6f7fc9037359 Author: Zhao, Xin <[email protected]> Date: Thu Aug 22 12:00:14 2024 -0400 fix pop folder name case commit 8aa3f3c44326a9c0455acbd21e1dae9c09654d8b Author: Zhao, Xin <[email protected]> Date: Thu Aug 22 11:21:25 2024 -0400 improve the use of constants; minor code updates commit 2439d1b3a6c457cf5b066e9ced99cf2a6c614b35 Author: Zhao, Xin <[email protected]> Date: Wed Aug 21 17:14:32 2024 -0400 clean up data and code for old SSPs commit cea41330268eacc5f79953e86c88cef1bcf90623 Author: Pralit Patel <[email protected]> Date: Tue Aug 20 14:21:02 2024 -0400 Fix energy balance issue which was stemming from the fact that industrial EFW calculations correctly calculated coefficients based off of industrial self supply (L132.water_km3_ctry_ind_Yh) + desal. However, the total water flow was in the water_td_ind sector was calculated from just self supply. The latter was corrected to calculate coefficients from self supply + desal thus ensuring all water and energy flows are calculated as expected. Note: with this fix we can drop the temporary workaround in zwater_xml_electricity_water.R for adjusting 1990 Central Asia coal to avoid calibration errors. commit 57f4357c97b50570d31891c3df6685247e903d0e Author: Zhao, Xin <[email protected]> Date: Mon Aug 5 11:56:31 2024 -0400 temp fix of 1990 Central Asia coal calibration commit a46ab4cc3edfc62fdbfcfa75ab2e886fa3805d16 Author: Zhao, Xin <[email protected]> Date: Sat Aug 3 23:23:31 2024 -0400 update TFP per labor force data updates commit bba6416cbd181abd8843afb5cde04d7106a53c0e Author: Zhao, Xin <[email protected]> Date: Fri Aug 2 17:33:30 2024 -0400 update labor force and carbon price ceiling commit 3de4b0a95afca1a49a9d6188601d641da9305d79 Author: Zhao, Xin <[email protected]> Date: Wed Jul 31 09:08:44 2024 -0400 update tfp and filter years >1990 in national accounts queries commit 99b15e02e5c27eeeae55eb3cc3c09c24a8b98a18 Author: Zhao, Xin <[email protected]> Date: Tue Jul 30 16:56:14 2024 -0400 fix twn and south amer North commit 5f0785285b410b8ef1aa079acaabfcd77f991489 Author: dsheng1026 <[email protected]> Date: Mon Jul 29 10:36:56 2024 -0400 update TFP across SSP for open mode commit 03e7aecc4c7177ace54067b01480d20664b1b5df Author: Di Sheng <[email protected]> Date: Wed Jul 24 14:12:19 2024 -0700 hector version 3.2.0 commit 6d5b498dc8b2bf43f67c75337cd2dd56f0714e74 Author: Zhao, Xin <[email protected]> Date: Wed Jul 24 15:58:59 2024 -0400 Update historical GDP to use FAOSTAT (1971-2020 or -2023) and SSP database (future) commit 495afea8c4465917fc946dee143fb9ccb2d22839 Author: Zhao, Xin <[email protected]> Date: Tue Jul 23 14:37:41 2024 -0400 git data and code ready for hist GDP and SSP database updates commit 2618067f74bb85025b1b91cba1727157270c999d Author: Zhao, Xin <[email protected]> Date: Sat Jul 20 17:03:41 2024 -0400 adding SSP_Database_Verion options to constants.R commit 522a443164474fa48f130f2e9365e8868e448520 Author: Di Sheng <[email protected]> Date: Mon Jul 15 10:56:57 2024 -0700 manually update hector version ... and 7 more commits
JGCRI · Aug 23, 2024 · bdcab96 · bdcab96
1 parent dc374cd
commit bdcab96
Show file tree

Hide file tree

Showing 46 changed files with 31,613 additions and 198,812 deletions.
diff --git a/cvs/objects/util/base/include/version.h b/cvs/objects/util/base/include/version.h
@@ -41,10 +41,10 @@
  * NOTE: ADD 1 TO LATEST SUBVERSION REVISION NUMBER
  */
 //! The latest SVN revision number for identification of the build.
-#define __REVISION_NUMBER__ "gcam-v7.1"
+#define __REVISION_NUMBER__ "gcam-v7.2"
 /*****************************************************************************/
 
 //! GCAM model version.
-#define __ObjECTS_VER__ "7.1"
+#define __ObjECTS_VER__ "7.2"
 
 #endif // _VERSION_H_ 
diff --git a/input/gcamdata/R/constants.R b/input/gcamdata/R/constants.R
@@ -629,6 +629,12 @@ energy.FOOD_PROCESSING.ENERGY_INFILL_MIN_EJ_PCAL_COEF <- 0.000413 # minimum valu
 
 # Socioeconomics constants ======================================================================
 
+socioeconomics.SSP_DB_BASEYEAR <- 2020 # base year of SSP data base v3.0.1
+socioeconomics.GDP_Adj_Moving_Average_ISO <- c("ven")
+socioeconomics.GDP_Adj_Moving_Average_Duration <- 15 # used for smoothing GDP for South_America_North
+socioeconomics.GDP_Adj_No_Neg_Growth_ISO <- c("ven", "twn")
+socioeconomics.GDP_Adj_No_Neg_Growth_Year <- 2025 # used for adjusting GDP projection to avoid negative GDP growth after this year (for Taiwan and South_America_North)
+
 socioeconomics.CORE_GCAM_SCENARIO <- "gSSP2"
 
 # Population years - note that these sequences shouldn't have any overlap,
@@ -643,8 +649,6 @@ socioeconomics.PWT_CONSTANT_CURRENCY_YEAR <- 2011 # Currency base year in Penn W
 # we do the SSP scenarios if we update to UN 2015 population.)
 socioeconomics.FINAL_HIST_YEAR <- 2015
 
-# Sets the years during which the IMF projections are used over-riding the default (generally SSP) assumptions.
-socioeconomics.IMF_GDP_YEARS <- 2015:2024
 # There will be an imblance of trade by region historically which is implicitly balanced by
 # capital flows.  We can phase this out by the year assumed below (linearly).  Note, setting a value
 # beyond the final model year will hold the final historical net capital flows constant.  Either

diff --git a/input/gcamdata/R/utils-data.R b/input/gcamdata/R/utils-data.R
@@ -104,7 +104,7 @@ get_units <- function(x) { attr(x, ATTR_UNITS) }
 #' @param ... Names of precursor objects (character)
 #' @return \code{x} with units appended to any existing comments.
 add_precursors <- function(x, ...) {
-  pc <- as.character(list(...))
+  pc <- c(...)
   attr(x, ATTR_PRECURSORS) <- c(attr(x, ATTR_PRECURSORS), pc)
   x
 }

diff --git a/input/gcamdata/R/zaglu_L162.ag_prodchange_R_C_Y_GLU_irr.R b/input/gcamdata/R/zaglu_L162.ag_prodchange_R_C_Y_GLU_irr.R
@@ -112,6 +112,7 @@ module_aglu_L162.ag_prodchange_R_C_Y_GLU_irr <- function(command, ...) {
       mutate(Irr_Rfd = "RFD") %>%
       rename(yield_kgHa = Yield_kgHa_rainfed) %>%
       bind_rows(L162.ag_irrYield_kgHa_Rcrs_Ccrs_Y) %>%
+      mutate(yield_kgHa = as.numeric(yield_kgHa)) %>%
       group_by(CROSIT_ctry, CROSIT_crop, Irr_Rfd) %>%
       mutate(tag1 = if_else(yield_kgHa[year == 2030] < yield_kgHa[year == 2005], 1, 0), # if 2030 < 2005, then AgProdChange1 = 0
              yield_kgHa = if_else(tag1 == 1 & year == 2030, yield_kgHa[year == 2005], yield_kgHa),

diff --git a/input/gcamdata/R/zaglu_L202.an_input.R b/input/gcamdata/R/zaglu_L202.an_input.R
@@ -425,7 +425,7 @@ module_aglu_L202.an_input <- function(command, ...) {
       left_join(L110.IO_Coefs_pulp%>% left_join_error_no_match(GCAM_region_names, by = c("GCAM_region_ID")), by = c("region","year")) %>%
       mutate(coefficient = IO,
              coefficient=if_else(stub.technology == "woodpulp",aglu.FOREST_PULP_CONVERSION,coefficient)) %>%
-      group_by(GCAM_region_ID,stub.technology) %>%
+      group_by(region, stub.technology) %>%
       mutate(coefficient= if_else(is.na(coefficient),approx_fun(year, coefficient, rule = 1),coefficient)) %>%
       ungroup() %>%
       select(colnames(L202.StubTechCoef_an))->L202.StubTechCoef_an_Forest

diff --git a/input/gcamdata/R/zenergy_L1327.paper.R b/input/gcamdata/R/zenergy_L1327.paper.R
@@ -17,7 +17,7 @@
 #' @author MMC July 2022
 module_energy_L1327.paper <- function(command, ...) {
   if(command == driver.DECLARE_INPUTS) {
-    return(c("L1011.en_bal_EJ_R_Si_Fi_Yh",
+    return(c("L1012.en_bal_EJ_R_Si_Fi_Yh",
              FILE = "energy/A_regions",
              FILE = "energy/mappings/enduse_fuel_aggregation",
              "L1326.in_EJ_R_indenergy_F_Yh",
@@ -50,7 +50,7 @@ module_energy_L1327.paper <- function(command, ...) {
     all_data <- list(...)[[1]]
 
     # Load required inputs
-    L1011.en_bal_EJ_R_Si_Fi_Yh <- get_data(all_data, "L1011.en_bal_EJ_R_Si_Fi_Yh", strip_attributes = TRUE)
+    L1012.en_bal_EJ_R_Si_Fi_Yh <- get_data(all_data, "L1012.en_bal_EJ_R_Si_Fi_Yh", strip_attributes = TRUE)
     A_regions <- get_data(all_data, "energy/A_regions",strip_attributes = TRUE)
     enduse_fuel_aggregation <- get_data(all_data, "energy/mappings/enduse_fuel_aggregation", strip_attributes = TRUE)
     L1326.in_EJ_R_indenergy_F_Yh <- get_data(all_data, "L1326.in_EJ_R_indenergy_F_Yh", strip_attributes = TRUE)
@@ -85,7 +85,7 @@ module_energy_L1327.paper <- function(command, ...) {
 
 
     # Get pulp and paper energy use from IEA energy balances
-    L1011.en_bal_EJ_R_Si_Fi_Yh %>%
+    L1012.en_bal_EJ_R_Si_Fi_Yh %>%
       filter(grepl("paper", sector)) ->
       L1327.in_EJ_R_paper_F_Yh
 
@@ -414,7 +414,7 @@ module_energy_L1327.paper <- function(command, ...) {
       add_units("EJ") %>%
       add_comments("PAPERPRO sector from IEA energy balances aggregated to GCAM regions") %>%
       add_legacy_name("L1327.in_EJ_R_paper_F_Yh") %>%
-      add_precursors("L1011.en_bal_EJ_R_Si_Fi_Yh", "energy/A_regions",
+      add_precursors("L1012.en_bal_EJ_R_Si_Fi_Yh", "energy/A_regions",
                      "common/iso_GCAM_regID", "energy/mappings/enduse_fuel_aggregation",
                      "L123.in_EJ_R_indchp_F_Yh", "energy/paper_food_TFE", "energy/A327.china_biomass") ->
       L1327.in_EJ_R_paper_F_Yh
@@ -432,7 +432,7 @@ module_energy_L1327.paper <- function(command, ...) {
       add_units("GJ/kg paper") %>%
       add_legacy_name("L1327.IO_GJkg_R_paper_F_Yh") %>%
       add_comments("IO coefficients for heat energy and electricity are calculated from IEA energy consumption and FAO paper production data") %>%
-      add_precursors("L1011.en_bal_EJ_R_Si_Fi_Yh", "aglu/FAO/FAO_Paper_Prod_t_FORESTAT", "aglu/AGLU_ctry", "energy/A_regions",
+      add_precursors("L1012.en_bal_EJ_R_Si_Fi_Yh", "aglu/FAO/FAO_Paper_Prod_t_FORESTAT", "aglu/AGLU_ctry", "energy/A_regions",
                      "common/iso_GCAM_regID","energy/mappings/enduse_fuel_aggregation", "L123.in_EJ_R_indchp_F_Yh",
                      "energy/paper_food_TFE", "energy/A327.globaltech_coef", "energy/A327.china_biomass") ->
       L1327.IO_GJkg_R_paper_F_Yh
@@ -443,7 +443,7 @@ module_energy_L1327.paper <- function(command, ...) {
       add_comments("Subtracted pulp and paper energy use from industrial energy use values in L1326.in_EJ_R_indenergy_F_Yh") %>%
       add_comments("To determine adjusted input energy for industrial energy use") %>%
       add_legacy_name("L1327.in_EJ_R_indenergy_F_Yh") %>%
-      add_precursors("L1011.en_bal_EJ_R_Si_Fi_Yh", "L1326.in_EJ_R_indenergy_F_Yh", "aglu/FAO/FAO_Paper_Prod_t_FORESTAT",
+      add_precursors("L1012.en_bal_EJ_R_Si_Fi_Yh", "L1326.in_EJ_R_indenergy_F_Yh", "aglu/FAO/FAO_Paper_Prod_t_FORESTAT",
                      "aglu/AGLU_ctry", "energy/A_regions", "common/iso_GCAM_regID","energy/mappings/enduse_fuel_aggregation") ->
       L1327.in_EJ_R_indenergy_F_Yh
 
@@ -452,7 +452,7 @@ module_energy_L1327.paper <- function(command, ...) {
       add_units("None") %>%
       add_comments("Subsector share weights for process heat technologies will be reset from 0 to 1 for these regions") %>%
       add_legacy_name("L1327.elec_noheat_adj_shwt_R") %>%
-      add_precursors("L1011.en_bal_EJ_R_Si_Fi_Yh", "energy/A_regions",
+      add_precursors("L1012.en_bal_EJ_R_Si_Fi_Yh", "energy/A_regions",
                      "common/iso_GCAM_regID","energy/mappings/enduse_fuel_aggregation") ->
       L1327.elec_noheat_adj_shwt_R
 
@@ -470,7 +470,7 @@ module_energy_L1327.paper <- function(command, ...) {
       add_units("GJ/kg") %>%
       add_comments("Woodpulp in Mt divided by paper sector biomass energy in EJ") %>%
       add_legacy_name("L1327.IO_woodpulp_energy") %>%
-      add_precursors("L1011.en_bal_EJ_R_Si_Fi_Yh", "energy/A_regions", "common/GCAM_region_names",
+      add_precursors("L1012.en_bal_EJ_R_Si_Fi_Yh", "energy/A_regions", "common/GCAM_region_names",
                      "common/iso_GCAM_regID", "energy/mappings/enduse_fuel_aggregation",
                      "L123.in_EJ_R_indchp_F_Yh", "energy/paper_food_TFE", "energy/A327.china_biomass",
                      "L110.For_ALL_bm3_R_Y") ->

diff --git a/input/gcamdata/R/zenergy_L2391.gas_trade_flows.R b/input/gcamdata/R/zenergy_L2391.gas_trade_flows.R
@@ -326,14 +326,30 @@ module_energy_L2391.gas_trade_flows <- function(command, ...) {
       mutate(value = abs(value)) -> L2391.NG_import_calOutput_statdiff
 
     # Divide export difference for each region between LNG & pipelines according to historical shares,
-    # and then adjust the corresponding calOutputValues accordingly.  Thiss will keep any region's
+    # and then adjust the corresponding calOutputValues accordingly.  This will keep any region's
     # exports from going negative and keep LNG / pipeline shares relatively constant.
+    # First, recalculate historical shares from adjusted data.
+    L2391.gas_flow_balances %>%
+      select(region, year, calExport_pipe, calExport_LNG) %>%
+      tidyr::gather(GCAM_Commodity_traded, value, -c(region, year)) %>%
+      mutate(GCAM_Commodity = "natural gas") %>%
+      mutate(GCAM_Commodity_traded = gsub("calExport_", "", GCAM_Commodity_traded),
+             GCAM_Commodity_traded = if_else(GCAM_Commodity_traded == "pipe", "gas pipeline", GCAM_Commodity_traded)) %>%
+      group_by(region, year, GCAM_Commodity, GCAM_Commodity_traded) %>%
+      summarise(value = sum(value)) %>%
+      ungroup() %>%
+      group_by(region, year) %>%
+      mutate(share = value / sum(value)) %>%
+      ungroup() %>%
+      mutate(share = if_else(is.nan(share), 0, share)) %>%
+      select(-value) -> L2391.NG_export_shares_adj
+
     L2391.NG_import_calOutput_statdiff %>%
       left_join_error_no_match(GCAM_region_pipeline_bloc_export, by = c("region" = "origin.region")) %>%
       # L2391.NG_export_shares contains shares for LNG and pipeline
       # join will duplicate rows by each carrier
       # LJENM will error, left_join() is used
-      left_join(L2391.NG_export_shares, by = c("region", "year")) %>%
+      left_join(L2391.NG_export_shares_adj, by = c("region", "year")) %>%
       mutate(value = value * share) %>%
       select(-share) -> L2391.NG_import_calOutput_adj
 

diff --git a/input/gcamdata/R/zsocio_L100.GDP_hist.R b/input/gcamdata/R/zsocio_L100.GDP_hist.R
@@ -19,43 +19,73 @@
 #' @importFrom dplyr filter mutate select
 #' @author BBL February 2017
 module_socio_L100.GDP_hist <- function(command, ...) {
+
+  MODULE_INPUTS <-
+    c(FILE = "aglu/AGLU_ctry",
+      FILE = "socioeconomics/GDP/GCAMFAOSTAT_GDP",
+      FILE = "socioeconomics/GDP/GDP_twn")
+
+  MODULE_OUTPUTS <-
+    c("L100.gdp_mil90usd_ctry_Yh")
+
   if(command == driver.DECLARE_INPUTS) {
-    return(c(FILE = "socioeconomics/USDA_GDP_MER",
-             FILE = "socioeconomics/WB_ExtraCountries_GDP_MER"))
+    return(MODULE_INPUTS)
   } else if(command == driver.DECLARE_OUTPUTS) {
-    return(c("L100.gdp_mil90usd_ctry_Yh"))
+    return(MODULE_OUTPUTS)
   } else if(command == driver.MAKE) {
 
     Country <- year <- value <- iso <- NULL # silence package checks.
 
     all_data <- list(...)[[1]]
 
-    # Load required inputs
-    usda_gdp_mer <- get_data(all_data, "socioeconomics/USDA_GDP_MER")
-    WB_ExtraCountries_GDP_MER <- get_data(all_data, "socioeconomics/WB_ExtraCountries_GDP_MER")
-    assert_that(tibble::is_tibble(usda_gdp_mer))
-    assert_that(tibble::is_tibble(WB_ExtraCountries_GDP_MER))
-
-    # bind qatar's GDP data to the USDA's nation-level data
-    usda_gdp_mer <- bind_rows(usda_gdp_mer, WB_ExtraCountries_GDP_MER) %>%
-      arrange(Country)
-
-    # Convert to long form, filter to historical years, convert units
-    usda_gdp_mer %>%
-      select(-Country) %>%
-      gather_years %>%
-      filter(!is.na(value), !is.na(iso)) %>%
-      mutate(value = value * CONV_BIL_MIL * gdp_deflator(1990, base_year = 2010),
-             year = as.integer(year)) %>%
-      add_title("Historical GDP downscaled to country (iso)") %>%
-      add_comments("Units converted to constant 1990 USD") %>%
-      add_precursors("socioeconomics/USDA_GDP_MER",
-                     "socioeconomics/WB_ExtraCountries_GDP_MER") %>%
+    # Load required inputs ----
+    get_data_list(all_data, MODULE_INPUTS, strip_attributes = TRUE)
+
+    # Note that GCAMFAOSTAT_GDP has data in 2015$ so regional GDP deflators are not used
+    # But values are brought back to 1990$ using USA GDP deflators
+    # FAO_GDP_Deflators is added to this chunk for BYU uses
+    # see module_aglu_L100.regional_ag_an_for_prices for examples
+    # Note that historical regional values have been fixed using FAO_AREA_DISAGGREGATE_HIST_DISSOLUTION_ALL in gcamfaostat
+    GCAMFAOSTAT_GDP %>%
+      gather_years() %>%
+      filter(element_code == 6184, # 2015$
+             !is.na(value)) %>%
+      # ToDo: GCAMFAOSTAT_GDP is more recent than other GCAMFAOSTAT files
+      # FAOSTAT recent changed name of 2 countries
+      # next GCAMFAOSTAT update (ready already) will use area_code to avoid dealing with country name changes
+      mutate(area = if_else(grepl("Yemen ", area), "Yemen", area),
+             area = if_else(area == "Netherlands (Kingdom of the)", "Netherlands", area),
+             area = if_else(area == "Sint Maarten (Dutch part)", "Sint Maarten (Dutch Part)", area)) %>%
+      left_join_error_no_match(AGLU_ctry %>% select(area = FAO_country, iso), by = "area") %>%
+      select(iso, year, value) %>%
+      # agg two Yeman
+      group_by(iso, year) %>%
+      summarize(value = sum(value), .groups = "drop") %>%
+      mutate(value = value * gdp_deflator(1990, base_year = 2015) ) ->
+      L100.gdp_mil90usd_ctry_Yh_0
+
+    # Taiwan from pwt but also extend to recent years using Taiwan statistics
+    GDP_twn %>%
+      select(iso, year, value = gdp_2015USD) %>%
+      mutate(value = value * gdp_deflator(1990, base_year = 2015)) ->
+      L100.gdp_mil90usd_ctry_Yh_0_TWN
+
+    L100.gdp_mil90usd_ctry_Yh_0 %>%
+      bind_rows(
+        L100.gdp_mil90usd_ctry_Yh_0_TWN
+      ) %>%
+      add_title("Historical GDP by country (iso) since 1970 per FAOSTAT") %>%
+      add_comments("Units 2015$ converted to constant 1990 USD using USA GDP deflators") %>%
+      add_precursors("socioeconomics/GDP/GCAMFAOSTAT_GDP",
+                     "aglu/AGLU_ctry",
+                     "socioeconomics/GDP/GDP_twn") %>%
       add_units("Million 1990 USD") %>%
       add_legacy_name("L100.gdp_mil90usd_ctry_Yh") ->
       L100.gdp_mil90usd_ctry_Yh
 
-    return_data(L100.gdp_mil90usd_ctry_Yh)
+
+    return_data(MODULE_OUTPUTS)
+
   } else {
     stop("Unknown command")
   }