remove data/hydro_capacities.csv with untraceable source

config/config.default.yaml

@@ -258,7 +258,7 @@ renewable:
     cutout: europe-2013-sarah3-era5
     carriers: [ror, PHS, hydro]
     PHS_max_hours: 6
-    hydro_max_hours: "energy_capacity_totals_by_country" # one of energy_capacity_totals_by_country, estimate_by_large_installations or a float
+    hydro_max_hours_quantile: 0.1
     flatten_dispatch: false
     flatten_dispatch_buffer: 0.2
     clip_min_inflow: 1.0

doc/configtables/hydro.csv

@@ -2,7 +2,7 @@
 cutout,--,Must be 'europe-2013-sarah3-era5',Specifies the directory where the relevant weather data ist stored.
 carriers,--,"Any subset of {'ror', 'PHS', 'hydro'}","Specifies the types of hydro power plants to build per-unit availability time series for. 'ror' stands for run-of-river plants, 'PHS' represents pumped-hydro storage, and 'hydro' stands for hydroelectric dams."
 PHS_max_hours,h,float,Maximum state of charge capacity of the pumped-hydro storage (PHS) in terms of hours at full output capacity ``p_nom``. Cf. `PyPSA documentation <https://pypsa.readthedocs.io/en/latest/components.html#storage-unit>`_.
-hydro_max_hours,h,"Any of {float, 'energy_capacity_totals_by_country', 'estimate_by_large_installations'}",Maximum state of charge capacity of the pumped-hydro storage (PHS) in terms of hours at full output capacity ``p_nom`` or heuristically determined. Cf. `PyPSA documentation <https://pypsa.readthedocs.io/en/latest/components.html#storage-unit>`_.
+hydro_max_hours_quantile,-,"[0, 1]","Quantile of hydro reservoir storage durations with data per country to use as a fill-value for missing datapoints. Defaults to 10% quantile, assuming reservoirs with large capacities have data. If no hydro reservoir has a storage duration given, a duration of 24 hours is assumed."
 flatten_dispatch,bool,"{true, false}",Consider an upper limit for the hydro dispatch. The limit is given by the average capacity factor plus the buffer given in  ``flatten_dispatch_buffer``
 flatten_dispatch_buffer,--,float,"If ``flatten_dispatch`` is true, specify the value added above the average capacity factor."
 clip_min_inflow,MW,float,"To avoid too small values in the inflow time series, values below this threshold are set to zero."

doc/release_notes.rst

@@ -21,6 +21,14 @@ Release Notes
   rules. This simplifies the use of ``mock_snakemake`` and places downloaded
   data more transparently into the ``data`` directory.
 
+* Removed use of ``data/hydro_capacities.csv`` for hydro energy storage
+  capacities because its source is untraceable. Many larger hydro reservoirs
+  from ``powerplantmatching`` have storage capacities given. The storage
+  capacity of hydro-electric plants with missing storage capacities is now
+  approximated through a configurable quantile of the storage durations of hydro
+  reservoirs with data, seperately for each country. The default is the 10%
+  quantile, assuming that missing data points are rather smaller reservoirs.
+
 PyPSA-Eur 0.12.0 (30th August 2024)
 ===================================
 

rules/build_electricity.smk

@@ -468,7 +468,6 @@ rule add_electricity:
         ),
         regions=resources("regions_onshore.geojson"),
         powerplants=resources("powerplants.csv"),
-        hydro_capacities=ancient("data/hydro_capacities.csv"),
         unit_commitment="data/unit_commitment.csv",
         fuel_price=lambda w: (
             resources("monthly_fuel_price.csv")

scripts/add_electricity.py

@@ -616,41 +616,22 @@ def attach_hydro(n, costs, ppl, profile_hydro, hydro_capacities, carriers, **par
         )
 
     if "hydro" in carriers and not hydro.empty:
-        hydro_max_hours = params.get("hydro_max_hours")
+        hydro_max_hours_quantile = params.get("hydro_max_hours_quantile", 0.1)
 
-        assert hydro_max_hours is not None, "No path for hydro capacities given."
-
-        hydro_stats = pd.read_csv(
-            hydro_capacities, comment="#", na_values="-", index_col=0
+        max_hours_country = hydro.groupby("country").max_hours.quantile(
+            hydro_max_hours_quantile
         )
-        e_target = hydro_stats["E_store[TWh]"].clip(lower=0.2) * 1e6
-        e_installed = hydro.eval("p_nom * max_hours").groupby(hydro.country).sum()
-        e_missing = e_target - e_installed
-        missing_mh_i = hydro.query("max_hours.isnull()").index
-
-        if hydro_max_hours == "energy_capacity_totals_by_country":
-            # watch out some p_nom values like IE's are totally underrepresented
-            max_hours_country = (
-                e_missing / hydro.loc[missing_mh_i].groupby("country").p_nom.sum()
-            )
-
-        elif hydro_max_hours == "estimate_by_large_installations":
-            max_hours_country = (
-                hydro_stats["E_store[TWh]"] * 1e3 / hydro_stats["p_nom_discharge[GW]"]
-            )
-
-        max_hours_country.clip(0, inplace=True)
 
         missing_countries = pd.Index(hydro["country"].unique()).difference(
             max_hours_country.dropna().index
         )
         if not missing_countries.empty:
             logger.warning(
-                f'Assuming max_hours=6 for hydro reservoirs in the countries: {", ".join(missing_countries)}'
+                f'Assuming max_hours=24 for hydro reservoirs in the countries: {", ".join(missing_countries)}'
             )
         hydro_max_hours = hydro.max_hours.where(
             hydro.max_hours > 0, hydro.country.map(max_hours_country)
-        ).fillna(6)
+        ).fillna(24)
 
         if params.get("flatten_dispatch", False):
             buffer = params.get("flatten_dispatch_buffer", 0.2)