Merge pull request #1018 from koen-vg/existing-caps-from-pm

Use powerplantmatching IRENASTAT for renewable capacities in `add_existing_baseyear`

config/config.default.yaml

@@ -72,7 +72,6 @@ enable:
   retrieve_sector_databundle: true
   retrieve_cost_data: true
   build_cutout: false
-  retrieve_irena: false
   retrieve_cutout: true
   build_natura_raster: false
   retrieve_natura_raster: true

doc/configtables/enable.csv

@@ -5,7 +5,6 @@ retrieve_databundle,bool,"{true, false}","Switch to retrieve databundle from zen
 retrieve_sector_databundle,bool,"{true, false}","Switch to retrieve sector databundle from zenodo via the rule :mod:`retrieve_sector_databundle` or whether to keep a custom databundle located in the corresponding folder."
 retrieve_cost_data,bool,"{true, false}","Switch to retrieve technology cost data from `technology-data repository <https://github.com/PyPSA/technology-data>`_."
 build_cutout,bool,"{true, false}","Switch to enable the building of cutouts via the rule :mod:`build_cutout`."
-retrieve_irena,bool,"{true, false}",Switch to enable the retrieval of ``existing_capacities`` from IRENASTAT with :mod:`retrieve_irena`.
 retrieve_cutout,bool,"{true, false}","Switch to enable the retrieval of cutouts from zenodo with :mod:`retrieve_cutout`."
 build_natura_raster,bool,"{true, false}","Switch to enable the creation of the raster ``natura.tiff`` via the rule :mod:`build_natura_raster`."
 retrieve_natura_raster,bool,"{true, false}","Switch to enable the retrieval of ``natura.tiff`` from zenodo with :mod:`retrieve_natura_raster`."

doc/release_notes.rst

@@ -184,6 +184,8 @@ Upcoming Release
 
 * Fix custom busmap read in `cluster_network`.
 
+* Data on existing renewable capacities is now consistently taken from powerplantmatching (instead of being retrieved separately); the dataset has also been updated to include 2023 values.
+
 * Added shapes to .nc file for different stages of the network object in `base_network`, `build_bus_regions`, and `cluster_network`.
 
 * Fix p_nom_min of renewables generators for myopic approach and add check of existing capacities in `add_land_use_constraint_m`.

doc/retrieve.rst

@@ -118,11 +118,6 @@ This rule downloads techno-economic assumptions from the `technology-data reposi
 
 - ``resources/costs.csv``
 
-Rule ``retrieve_irena``
-================================
-
-.. automodule:: retrieve_irena
-
 Rule ``retrieve_ship_raster``
 ================================
 

envs/environment.yaml

@@ -25,7 +25,7 @@ dependencies:
 - yaml
 - pytables
 - lxml
-- powerplantmatching>=0.5.5,!=0.5.9
+- powerplantmatching>=0.5.13
 - numpy
 - pandas>=2.1
 - geopandas>=0.11.0

rules/retrieve.smk

@@ -42,24 +42,6 @@ if config["enable"]["retrieve"] and config["enable"].get("retrieve_databundle",
             "../scripts/retrieve_databundle.py"
 
 
-if config["enable"].get("retrieve_irena"):
-
-    rule retrieve_irena:
-        output:
-            offwind="data/existing_infrastructure/offwind_capacity_IRENA.csv",
-            onwind="data/existing_infrastructure/onwind_capacity_IRENA.csv",
-            solar="data/existing_infrastructure/solar_capacity_IRENA.csv",
-        log:
-            "logs/retrieve_irena.log",
-        resources:
-            mem_mb=1000,
-        retries: 2
-        conda:
-            "../envs/retrieve.yaml"
-        script:
-            "../scripts/retrieve_irena.py"
-
-
 if config["enable"]["retrieve"] and config["enable"].get("retrieve_cutout", True):
 
     rule retrieve_cutout:

rules/solve_myopic.smk

@@ -26,9 +26,6 @@ rule add_existing_baseyear:
         existing_heating_distribution=resources(
             "existing_heating_distribution_elec_s{simpl}_{clusters}_{planning_horizons}.csv"
         ),
-        existing_solar="data/existing_infrastructure/solar_capacity_IRENA.csv",
-        existing_onwind="data/existing_infrastructure/onwind_capacity_IRENA.csv",
-        existing_offwind="data/existing_infrastructure/offwind_capacity_IRENA.csv",
     output:
         RESULTS
         + "prenetworks-brownfield/elec_s{simpl}_{clusters}_l{ll}_{opts}_{sector_opts}_{planning_horizons}.nc",

rules/solve_perfect.smk

@@ -25,9 +25,6 @@ rule add_existing_baseyear:
             "existing_heating_distribution_elec_s{simpl}_{clusters}_{planning_horizons}.csv"
         ),
         existing_heating="data/existing_infrastructure/existing_heating_raw.csv",
-        existing_solar="data/existing_infrastructure/solar_capacity_IRENA.csv",
-        existing_onwind="data/existing_infrastructure/onwind_capacity_IRENA.csv",
-        existing_offwind="data/existing_infrastructure/offwind_capacity_IRENA.csv",
     output:
         RESULTS
         + "prenetworks-brownfield/elec_s{simpl}_{clusters}_l{ll}_{opts}_{sector_opts}_{planning_horizons}.nc",

scripts/add_existing_baseyear.py

@@ -13,6 +13,7 @@
 import country_converter as coco
 import numpy as np
 import pandas as pd
+import powerplantmatching as pm
 import pypsa
 import xarray as xr
 from _helpers import (
@@ -60,14 +61,22 @@ def add_existing_renewables(df_agg, costs):
     Append existing renewables to the df_agg pd.DataFrame with the conventional
     power plants.
     """
-    carriers = {"solar": "solar", "onwind": "onwind", "offwind": "offwind-ac"}
+    tech_map = {"solar": "PV", "onwind": "Onshore", "offwind": "Offshore"}
 
-    for tech in ["solar", "onwind", "offwind"]:
-        carrier = carriers[tech]
+    countries = snakemake.config["countries"]
+    irena = pm.data.IRENASTAT().powerplant.convert_country_to_alpha2()
+    irena = irena.query("Country in @countries")
+    irena = irena.groupby(["Technology", "Country", "Year"]).Capacity.sum()
 
-        df = pd.read_csv(snakemake.input[f"existing_{tech}"], index_col=0).fillna(0.0)
+    irena = irena.unstack().reset_index()
+
+    for carrier, tech in tech_map.items():
+        df = (
+            irena[irena.Technology.str.contains(tech)]
+            .drop(columns=["Technology"])
+            .set_index("Country")
+        )
         df.columns = df.columns.astype(int)
-        df.index = cc.convert(df.index, to="iso2")
 
         # calculate yearly differences
         df.insert(loc=0, value=0.0, column="1999")
@@ -97,14 +106,16 @@ def add_existing_renewables(df_agg, costs):
 
         for year in nodal_df.columns:
             for node in nodal_df.index:
-                name = f"{node}-{tech}-{year}"
+                name = f"{node}-{carrier}-{year}"
                 capacity = nodal_df.loc[node, year]
                 if capacity > 0.0:
-                    df_agg.at[name, "Fueltype"] = tech
+                    df_agg.at[name, "Fueltype"] = carrier
                     df_agg.at[name, "Capacity"] = capacity
                     df_agg.at[name, "DateIn"] = year
-                    df_agg.at[name, "lifetime"] = costs.at[tech, "lifetime"]
-                    df_agg.at[name, "DateOut"] = year + costs.at[tech, "lifetime"] - 1
+                    df_agg.at[name, "lifetime"] = costs.at[carrier, "lifetime"]
+                    df_agg.at[name, "DateOut"] = (
+                        year + costs.at[carrier, "lifetime"] - 1
+                    )
                     df_agg.at[name, "cluster_bus"] = node