move leap year handling into helper function

rules/build_electricity.smk

@@ -21,6 +21,7 @@ if config["enable"].get("prepare_links_p_nom", False):
 rule build_electricity_demand:
     params:
         snapshots=config_provider("snapshots"),
+        drop_leap_day=config_provider("enable", "drop_leap_day"),
         countries=config_provider("countries"),
         load=config_provider("load"),
     input:
@@ -68,6 +69,7 @@ rule base_network:
     params:
         countries=config_provider("countries"),
         snapshots=config_provider("snapshots"),
+        drop_leap_day=config_provider("enable", "drop_leap_day"),
         lines=config_provider("lines"),
         links=config_provider("links"),
         transformers=config_provider("transformers"),

scripts/_helpers.py

@@ -724,3 +724,15 @@ def validate_checksum(file_path, zenodo_url=None, checksum=None):
     assert (
         calculated_checksum == checksum
     ), "Checksum is invalid. This may be due to an incomplete download. Delete the file and re-execute the rule."
+
+
+def get_snapshots(snapshots, drop_leap_day=False, freq="h", **kwargs):
+    """
+    Returns pandas DateTimeIndex potentially without leap days.
+    """
+
+    time = pd.date_range(freq=freq, **snapshots, **kwargs)
+    if drop_leap_day and time.is_leap_year.any():
+        time = time[~((time.month == 2) & (time.day == 29))]
+
+    return time

scripts/add_electricity.py

@@ -93,7 +93,12 @@ import powerplantmatching as pm
 import pypsa
 import scipy.sparse as sparse
 import xarray as xr
-from _helpers import configure_logging, set_scenario_config, update_p_nom_max
+from _helpers import (
+    configure_logging,
+    get_snapshots,
+    set_scenario_config,
+    update_p_nom_max,
+)
 from powerplantmatching.export import map_country_bus
 from shapely.prepared import prep
 
@@ -760,15 +765,6 @@ def estimate_renewable_capacities(
             )
 
 
-def drop_leap_day(n):
-    if not n.snapshots.is_leap_year.any():
-        return
-    leap_days = (n.snapshots.day == 29) & (n.snapshots.month == 2)
-    n.set_snapshots(n.snapshots[~leap_days])
-    n.snapshot_weightings[:] = 8760 / len(n.snapshots)
-    logger.info("Dropped February 29 from leap year.")
-
-
 def attach_line_rating(
     n, rating, s_max_pu, correction_factor, max_voltage_difference, max_line_rating
 ):
@@ -805,7 +801,8 @@ if __name__ == "__main__":
 
     n = pypsa.Network(snakemake.input.base_network)
 
-    n.set_snapshots(pd.date_range(freq="h", **snakemake.params.snapshots))
+    time = get_snapshots(snakemake.params.snapshots, snakemake.params.drop_leap_day)
+    n.set_snapshots(time)
 
     Nyears = n.snapshot_weightings.objective.sum() / 8760.0
 
@@ -916,8 +913,5 @@ if __name__ == "__main__":
 
     sanitize_carriers(n, snakemake.config)
 
-    if snakemake.params.drop_leap_day:
-        drop_leap_day(n)
-
     n.meta = snakemake.config
     n.export_to_netcdf(snakemake.output[0])

scripts/base_network.py

@@ -75,7 +75,7 @@ import shapely
 import shapely.prepared
 import shapely.wkt
 import yaml
-from _helpers import configure_logging, set_scenario_config
+from _helpers import configure_logging, get_snapshots, set_scenario_config
 from packaging.version import Version, parse
 from scipy import spatial
 from scipy.sparse import csgraph
@@ -730,12 +730,12 @@ def base_network(
     transformers = _set_electrical_parameters_transformers(transformers, config)
     links = _set_electrical_parameters_links(links, config, links_p_nom)
     converters = _set_electrical_parameters_converters(converters, config)
-    snapshots = snakemake.params.snapshots
 
     n = pypsa.Network()
     n.name = "PyPSA-Eur"
 
-    n.set_snapshots(pd.date_range(freq="h", **snapshots))
+    time = get_snapshots(snakemake.params.snapshots, snakemake.params.drop_leap_day)
+    n.set_snapshots(time)
     n.madd("Carrier", ["AC", "DC"])
 
     n.import_components_from_dataframe(buses, "Bus")

scripts/build_daily_heat_demand.py

@@ -11,7 +11,7 @@ import geopandas as gpd
 import numpy as np
 import pandas as pd
 import xarray as xr
-from _helpers import set_scenario_config
+from _helpers import get_snapshots, set_scenario_config
 from dask.distributed import Client, LocalCluster
 
 if __name__ == "__main__":
@@ -32,11 +32,12 @@ if __name__ == "__main__":
 
     cutout_name = snakemake.input.cutout
 
-    time = pd.date_range(freq="h", **snakemake.params.snapshots)
+    time = get_snapshots(snakemake.params.snapshots, snakemake.params.drop_leap_day)
-    daily = pd.date_range(freq="D", **snakemake.params.snapshots)
+    daily = get_snapshots(
-    if snakemake.params.drop_leap_day:
+        snakemake.params.snapshots,
-        time = time[~((time.month == 2) & (time.day == 29))]
+        snakemake.params.drop_leap_day,
-        daily = daily[~((daily.month == 2) & (daily.day == 29))]
+        freq="D",
+    )
 
     cutout = atlite.Cutout(cutout_name).sel(time=time)
 

scripts/build_electricity_demand.py

@@ -39,7 +39,7 @@ import logging
 
 import numpy as np
 import pandas as pd
-from _helpers import configure_logging, set_scenario_config
+from _helpers import configure_logging, get_snapshots, set_scenario_config
 from pandas import Timedelta as Delta
 
 logger = logging.getLogger(__name__)
@@ -263,7 +263,9 @@ if __name__ == "__main__":
     configure_logging(snakemake)
     set_scenario_config(snakemake)
 
-    snapshots = pd.date_range(freq="h", **snakemake.params.snapshots)
+    snapshots = get_snapshots(
+        snakemake.params.snapshots, snakemake.params.drop_leap_day
+    )
 
     fixed_year = snakemake.config["load"].get("fixed_year", False)
     years = (

scripts/build_hourly_heat_demand.py

@@ -10,7 +10,7 @@ from itertools import product
 
 import pandas as pd
 import xarray as xr
-from _helpers import generate_periodic_profiles, set_scenario_config
+from _helpers import generate_periodic_profiles, get_snapshots, set_scenario_config
 
 if __name__ == "__main__":
     if "snakemake" not in globals():
@@ -24,9 +24,9 @@ if __name__ == "__main__":
         )
     set_scenario_config(snakemake)
 
-    snapshots = pd.date_range(freq="h", **snakemake.params.snapshots)
+    snapshots = get_snapshots(
-    if snakemake.params.drop_leap_day:
+        snakemake.params.snapshots, snakemake.params.drop_leap_day
-        snapshots = snapshots[~((snapshots.month == 2) & (snapshots.day == 29))]
+    )
 
     daily_space_heat_demand = (
         xr.open_dataarray(snakemake.input.heat_demand)

scripts/build_hydro_profile.py

@@ -65,7 +65,7 @@ import atlite
 import country_converter as coco
 import geopandas as gpd
 import pandas as pd
-from _helpers import configure_logging, set_scenario_config
+from _helpers import configure_logging, get_snapshots, set_scenario_config
 from numpy.polynomial import Polynomial
 
 cc = coco.CountryConverter()
@@ -73,9 +73,7 @@ cc = coco.CountryConverter()
 
 def get_eia_annual_hydro_generation(fn, countries, capacities=False):
     # in billion kWh/a = TWh/a
-    df = pd.read_csv(
+    df = pd.read_csv(fn, skiprows=2, index_col=1, na_values=[" ", "--"]).iloc[1:, 1:]
-        fn, skiprows=2, index_col=1, na_values=[" ", "--"]
-    ).iloc[1:, 1:]
     df.index = df.index.str.strip()
     df.columns = df.columns.astype(int)
 
@@ -175,9 +173,7 @@ if __name__ == "__main__":
 
     params_hydro = snakemake.params.hydro
 
-    time = pd.date_range(freq="h", **snakemake.params.snapshots)
+    time = get_snapshots(snakemake.params.snapshots, snakemake.params.drop_leap_day)
-    if snakemake.params.drop_leap_day:
-        time = time[~((time.month == 2) & (time.day == 29))]
 
     cutout = atlite.Cutout(snakemake.input.cutout).sel(time=time)
 

scripts/build_line_rating.py

@@ -58,7 +58,7 @@ import numpy as np
 import pandas as pd
 import pypsa
 import xarray as xr
-from _helpers import configure_logging, set_scenario_config
+from _helpers import configure_logging, get_snapshots, set_scenario_config
 from shapely.geometry import LineString as Line
 from shapely.geometry import Point
 
@@ -147,9 +147,7 @@ if __name__ == "__main__":
     set_scenario_config(snakemake)
 
     n = pypsa.Network(snakemake.input.base_network)
-    time = pd.date_range(freq="h", **snakemake.params.snapshots)
+    time = get_snapshots(snakemake.params.snapshots, snakemake.params.drop_leap_day)
-    if snakemake.params.drop_leap_day:
-        time = time[~((time.month == 2) & (time.day == 29))]
 
     cutout = atlite.Cutout(snakemake.input.cutout).sel(time=time)
 

scripts/build_population_weighted_energy_totals.py

@@ -15,7 +15,7 @@ if __name__ == "__main__":
 
         snakemake = mock_snakemake(
             "build_population_weighted_energy_totals",
-            kind='energy',
+            kind="energy",
             simpl="",
             clusters=60,
         )

scripts/build_renewable_profiles.py

@@ -186,9 +186,8 @@ import time
 import atlite
 import geopandas as gpd
 import numpy as np
-import pandas as pd
 import xarray as xr
-from _helpers import configure_logging, set_scenario_config
+from _helpers import configure_logging, get_snapshots, set_scenario_config
 from dask.distributed import Client
 from pypsa.geo import haversine
 from shapely.geometry import LineString
@@ -227,11 +226,9 @@ if __name__ == "__main__":
     else:
         client = None
 
-    time = pd.date_range(freq="h", **snakemake.params.snapshots)
+    sns = get_snapshots(snakemake.params.snapshots, snakemake.params.drop_leap_day)
-    if snakemake.params.drop_leap_day:
-        time = time[~((time.month == 2) & (time.day == 29))]
 
-    cutout = atlite.Cutout(snakemake.input.cutout).sel(time=time)
+    cutout = atlite.Cutout(snakemake.input.cutout).sel(time=sns)
     regions = gpd.read_file(snakemake.input.regions)
     assert not regions.empty, (
         f"List of regions in {snakemake.input.regions} is empty, please "

scripts/build_solar_thermal_profiles.py

@@ -9,9 +9,8 @@ Build solar thermal collector time series.
 import atlite
 import geopandas as gpd
 import numpy as np
-import pandas as pd
 import xarray as xr
-from _helpers import set_scenario_config
+from _helpers import get_snapshots, set_scenario_config
 from dask.distributed import Client, LocalCluster
 
 if __name__ == "__main__":
@@ -32,9 +31,7 @@ if __name__ == "__main__":
     config = snakemake.params.solar_thermal
     config.pop("cutout", None)
 
-    time = pd.date_range(freq="h", **snakemake.params.snapshots)
+    time = get_snapshots(snakemake.params.snapshots, snakemake.params.drop_leap_day)
-    if snakemake.params.drop_leap_day:
-        time = time[~((time.month == 2) & (time.day == 29))]
 
     cutout = atlite.Cutout(snakemake.input.cutout).sel(time=time)
 

scripts/build_temperature_profiles.py

@@ -9,9 +9,8 @@ Build time series for air and soil temperatures per clustered model region.
 import atlite
 import geopandas as gpd
 import numpy as np
-import pandas as pd
 import xarray as xr
-from _helpers import set_scenario_config
+from _helpers import get_snapshots, set_scenario_config
 from dask.distributed import Client, LocalCluster
 
 if __name__ == "__main__":
@@ -29,9 +28,7 @@ if __name__ == "__main__":
     cluster = LocalCluster(n_workers=nprocesses, threads_per_worker=1)
     client = Client(cluster, asynchronous=True)
 
-    time = pd.date_range(freq="h", **snakemake.params.snapshots)
+    time = get_snapshots(snakemake.params.snapshots, snakemake.params.drop_leap_day)
-    if snakemake.params.drop_leap_day:
-        time = time[~((time.month == 2) & (time.day == 29))]
 
     cutout = atlite.Cutout(snakemake.input.cutout).sel(time=time)
 

scripts/build_transport_demand.py

@@ -13,7 +13,12 @@ import logging
 import numpy as np
 import pandas as pd
 import xarray as xr
-from _helpers import configure_logging, generate_periodic_profiles, set_scenario_config
+from _helpers import (
+    configure_logging,
+    generate_periodic_profiles,
+    get_snapshots,
+    set_scenario_config,
+)
 
 logger = logging.getLogger(__name__)
 
@@ -183,10 +188,9 @@ if __name__ == "__main__":
 
     options = snakemake.params.sector
 
-    snapshots = pd.date_range(freq="h", **snakemake.params.snapshots, tz="UTC")
+    snapshots = get_snapshots(
-    if snakemake.params.drop_leap_day:
+        snakemake.params.snapshots, snakemake.params.drop_leap_day, tz="UTC"
-        leap_day = (snapshots.month == 2) & (snapshots.day == 29)
+    )
-        snapshots = snapshots[~leap_day]
 
     nyears = len(snapshots) / 8760
 

scripts/make_summary.py

@@ -13,7 +13,7 @@ import sys
 import numpy as np
 import pandas as pd
 import pypsa
-from _helpers import configure_logging, set_scenario_config
+from _helpers import configure_logging, get_snapshots, set_scenario_config
 from prepare_sector_network import prepare_costs
 
 idx = pd.IndexSlice
@@ -690,7 +690,8 @@ if __name__ == "__main__":
         for planning_horizon in snakemake.params.scenario["planning_horizons"]
     }
 
-    Nyears = len(pd.date_range(freq="h", **snakemake.params.snapshots)) / 8760
+    time = get_snapshots(snakemake.params.snapshots, snakemake.params.drop_leap_day)
+    Nyears = len(time) / 8760
 
     costs_db = prepare_costs(
         snakemake.input.costs,

scripts/prepare_sector_network.py

@@ -3507,10 +3507,6 @@ def set_temporal_aggregation(n, resolution, solver_name, drop_leap_day=False):
         logger.info("Aggregate to frequency %s", resolution)
         n = average_every_nhours(n, resolution)
 
-    if drop_leap_day:
-        sns = n.snapshots[~((n.snapshots.month == 2) & (n.snapshots.day == 29))]
-        n.set_snapshots(sns)
-
     return n