Merge pull request #219 from iiasa/costs/ssp-fix

Fix naming of GDP and population columns in SSP data aggregation within `tools.costs`
iiasa · Sep 2, 2024 · 5fe36ca · 5fe36ca
2 parents cf687da + 33aa238
commit 5fe36ca
Show file tree

Hide file tree

Showing 3 changed files with 22 additions and 14 deletions.
diff --git a/doc/whatsnew.rst b/doc/whatsnew.rst
@@ -1,8 +1,11 @@
 What's new
 **********
 
-.. Next release
-.. ============
+Next release
+============
+
+- Fix and update :doc:`/api/tools-costs` (:pull:`219`)
+  - Fix naming of GDP and population columns in SSP data aggregation (:pull:`219`).
 
 v2024.8.6
 =========

diff --git a/message_ix_models/tools/costs/gdp.py b/message_ix_models/tools/costs/gdp.py
@@ -22,14 +22,15 @@ def process_raw_ssp_data(context: Context, config: Config) -> pd.DataFrame:
     pandas.DataFrame
         with the columns:
 
-        - scenario_version
-        - scenario
-        - region
-        - year
-        - total_gdp
-        - total_population
-        - gdp_ppp_per_capita
-        - gdp_ratio_reg_to_reference
+        - scenario_version: version of SSP scenario data
+        - scenario: scenario (SSP1-5, LED)
+        - region: region name
+        - year: year of data
+        - total_population: total population aggregated to the regional level
+        - total_gdp: total GDP aggregated to the regional level
+        - gdp_ppp_per_capita: regional GDP per capita in PPP
+        - gdp_ratio_reg_to_reference: ratio of regional GDP per capita to \
+            reference region's GDP per capita
     """
     from collections import defaultdict
 
@@ -97,8 +98,8 @@ def merge(*dfs: pd.DataFrame) -> pd.DataFrame:
         return (
             pd.concat(
                 [
-                    dfs[0].to_series().rename("total_gdp"),
-                    dfs[1].to_series().rename("total_population"),
+                    dfs[0].to_series().rename("total_population"),
+                    dfs[1].to_series().rename("total_gdp"),
                     dfs[2].to_series().rename("gdp_ppp_per_capita"),
                     dfs[3].to_series().rename("gdp_ratio_reg_to_reference"),
                 ],

diff --git a/message_ix_models/tools/wb.py b/message_ix_models/tools/wb.py
@@ -67,8 +67,12 @@ def get_weight(code: "sdmx.model.common.Code") -> float:
             return 1.0
 
     elif method == "population":
+        # Work around khaeru/sdmx#191: ensure the HTTPS URL is used
+        client = sdmx.Client("WB_WDI")
+        client.source.url = client.source.url.replace("http://", "https://")
+
         # Retrieve WB_WDI data for SERIES=SP_POP_TOTAL (Population, total)
-        dm = sdmx.Client("WB_WDI").data(
+        dm = client.data(
             "WDI", key="A.SP_POP_TOTL.", params=dict(startPeriod=2020, endPeriod=2020)
         )
 
@@ -154,7 +158,7 @@ def fetch_codelist(id: str) -> "sdmx.model.common.Codelist":
     import sdmx
 
     file = pooch.retrieve(
-        url="http://api.worldbank.org/v2/sdmx/rest/codelist/WB/", known_hash=None
+        url="https://api.worldbank.org/v2/sdmx/rest/codelist/WB/", known_hash=None
     )
     # Read the retrieved SDMX StructureMessage and extract the code list
     sm = sdmx.read_sdmx(file)