Merge pull request #245 from OSeMOSYS/summaries-updates

Updates to result-summaries

workflow/rules/postprocess.smk

@@ -19,16 +19,20 @@ RESULT_SUMMARIES = [
     "AnnualTotalTradeFlowsNode",
     "AnnualTotalTradeFlowsCountry",
     "AnnualEmissionIntensity",
+    "AnnualEmissionIntensityGlobal",
     "PowerCapacityNode",
     "TransmissionCapacityNode",
     "PowerCapacityCountry",
     "TransmissionCapacityCountry",
     "GenerationSharesNode",
     "GenerationSharesCountry",
+    "GenerationSharesGlobal",
     "PowerCostNode",
     "TotalCostNode",
     "PowerCostCountry",
     "TotalCostCountry",
+    "PowerCostGlobal",
+    "TotalCostGlobal",
     "Metrics"
 ]
 
@@ -113,6 +117,7 @@ rule calculate_carbon_intensity:
         annual_emissions = "results/{scenario}/results/AnnualEmissions.csv",
     output:
         emission_intensity = "results/{scenario}/result_summaries/AnnualEmissionIntensity.csv",
+        emission_intensity_global = "results/{scenario}/result_summaries/AnnualEmissionIntensityGlobal.csv",
     log:
         log = 'results/{scenario}/logs/carbon_intensity.log'
     script: 
@@ -127,8 +132,10 @@ rule calculate_costs:
     output:
         node_pwr_cost = "results/{scenario}/result_summaries/PowerCostNode.csv",
         country_pwr_cost = "results/{scenario}/result_summaries/PowerCostCountry.csv",
+        global_pwr_cost = "results/{scenario}/result_summaries/PowerCostGlobal.csv",
         node_cost = "results/{scenario}/result_summaries/TotalCostNode.csv",
         country_cost = "results/{scenario}/result_summaries/TotalCostCountry.csv",
+        global_cost = "results/{scenario}/result_summaries/TotalCostGlobal.csv",
     log:
         log = 'results/{scenario}/logs/node_cost.log'
     script: 
@@ -144,6 +151,7 @@ rule calculate_generation_shares:
     output:
         generation_shares_node = "results/{scenario}/result_summaries/GenerationSharesNode.csv",
         generation_shares_country = "results/{scenario}/result_summaries/GenerationSharesCountry.csv",
+        generation_shares_global = "results/{scenario}/result_summaries/GenerationSharesGlobal.csv",
     log:
         log = 'results/{scenario}/logs/generation_shares.log'
     script: 

workflow/scripts/osemosys_global/summary/carbon_intensity.py

@@ -14,6 +14,7 @@ def format_production(
 ) -> pd.DataFrame:
 
     df = production.copy()
+    df = df.loc[df.index.get_level_values("FUEL").str.startswith('ELC')]
 
     df = df[
         (df.index.get_level_values("TECHNOLOGY").str.startswith("PWR"))
@@ -40,24 +41,35 @@ def format_emissions(annual_emissions: pd.DataFrame) -> pd.DataFrame:
     df["COUNTRY"] = df.EMISSION.str[3:6]
     return df.groupby(["REGION", "EMISSION", "COUNTRY", "YEAR"]).sum()
 
+def format_global_values(production: pd.DataFrame, emissions: pd.DataFrame) -> tuple[pd.DataFrame, pd.DataFrame]:
+    p_global = production.reset_index()[["YEAR", "VALUE"]
+                                       ].groupby(["YEAR"]).sum()
+
+    e_global = emissions.reset_index()[["YEAR", "VALUE"]
+                                       ].groupby(["YEAR"]).sum()
+
+    return p_global, e_global
 
 def calculate_emission_intensity(
-    production: pd.DataFrame, emissions: pd.DataFrame
+    production: pd.DataFrame, 
+    emissions: pd.DataFrame,
+    country: bool
 ) -> pd.DataFrame:
 
     p = production.copy().rename(columns={"VALUE": "PROD_PJ"})
     e = emissions.copy().rename(columns={"VALUE": "EMISSIONS_MT"})
 
     df = p.join(e).fillna(0)
-    df = df.droplevel("COUNTRY")  # emission retains country
+    if country:
+        df = df.droplevel("COUNTRY")  # emission retains country
 
     # 1 PJ (1000 TJ / PJ) (1000 GJ / TJ) (1000 MJ / GJ) (1000 kJ / MJ) (hr / 3600sec)
     df["PROD_kwh"] = df.PROD_PJ.mul(1000).mul(1000).mul(1000).mul(1000).div(3600)
 
     # 1 MT (1,000,000 T / MT) (1000 kg / T) (1000g / kg)
     df["EMISSIONS_g"] = df.EMISSIONS_MT.mul(1000000).mul(1000).mul(1000)
 
-    # intensity in g/kwh
+    # intensity in g/kwh - kg/MWh
     df["VALUE"] = df.EMISSIONS_g.div(df.PROD_kwh)
 
     return df["VALUE"].to_frame()
@@ -68,11 +80,13 @@ def calculate_emission_intensity(
         production_csv = snakemake.input.production_by_technology
         annual_emissions_csv = snakemake.input.annual_emissions
         save = snakemake.output.emission_intensity
+        save_global = snakemake.output.emission_intensity_global
         storage = snakemake.params.storage
     else:
         production_csv = "results/India/results/ProductionByTechnology.csv"
         annual_emissions_csv = "results/India/results/AnnualEmissions.csv"
         save = "results/India/results/AnnualEmissionIntensity.csv"
+        save_global = "results/India/results/AnnualEmissionIntensityGlobal.csv"
         storage = {"SDS": [], "LDS": []}
 
     if storage:
@@ -85,7 +99,14 @@ def calculate_emission_intensity(
 
     production = format_production(production, exclusions)
     emissions = format_emissions(annual_emissions)
+
+    production_global, emissions_global = format_global_values(production, emissions)
 
-    emission_intensity = calculate_emission_intensity(production, emissions).round(2)
+    emission_intensity = calculate_emission_intensity(
+        production, emissions, country = True).round(2)
+
+    emission_intensity_global = calculate_emission_intensity(
+        production_global, emissions_global, country = False).round(2)
 
     emission_intensity.to_csv(save, index=True)
+    emission_intensity_global.to_csv(save_global, index=True)

workflow/scripts/osemosys_global/summary/constants.py

@@ -1,7 +1,7 @@
 """Constants for summary statistics"""
 
 RENEWABLES = ["BIO", "GEO", "HYD", "SPV", "CSP", "WAS", "WAV", "WON", "WOF"]
-CLEAN = ["BIO", "GEO", "HYD", "SPV", "CSP", "WAS", "WAV", "WON", "WOF", "URN"]
+CLEAN = ["BIO", "CCS", "GEO", "HYD", "SPV", "CSP", "WAS", "WAV", "WON", "WOF", "URN"]
 FOSSIL = ["CCG", "COA", "COG", "OCG", "OIL", "OTH", "PET"]
 
 MONTH_NAMES = {

workflow/scripts/osemosys_global/summary/costs.py

@@ -47,6 +47,42 @@ def get_storage_cost(
         .sum()
     )
 
+def get_transmission_cost(discounted_cost_tech: pd.DataFrame, country: bool) -> pd.DataFrame:
+
+    df = discounted_cost_tech.copy()
+
+    df1 = df[
+        (df.index.get_level_values("TECHNOLOGY").str.startswith("TRN"))
+    ].copy()
+
+    if country:
+        r = "COUNTRY"
+        df1[r] = df1.index.get_level_values("TECHNOLOGY").str[3:6]
+    else:
+        r = "NODE"
+        df1[r] = df1.index.get_level_values("TECHNOLOGY").str[3:8]
+
+    df2 = df.copy()[
+        (df.index.get_level_values("TECHNOLOGY").str.startswith("TRN"))
+    ]
+
+    if country:
+        r = "COUNTRY"
+        df2[r] = df2.index.get_level_values("TECHNOLOGY").str[8:11]
+    else:
+        r = "NODE"
+        df2[r] = df2.index.get_level_values("TECHNOLOGY").str[8:13]
+
+    for df in [df1, df2]:
+        df['VALUE'] = df['VALUE'] / 2
+
+    df = pd.concat([df1, df2])
+
+    return (
+        df.reset_index()[["REGION", r, "YEAR", "VALUE"]]
+        .groupby(["REGION", r, "YEAR"])
+        .sum()
+    )
 
 def get_demand(demand: pd.DataFrame, country: bool) -> pd.DataFrame:
 
@@ -84,17 +120,22 @@ def get_pwr_cost(demand: pd.DataFrame, cost: pd.DataFrame) -> pd.DataFrame:
         demand_csv = snakemake.input.demand
         power_cost_node_csv = snakemake.output.node_pwr_cost
         power_cost_country_csv = snakemake.output.country_pwr_cost
+        power_cost_global_csv = snakemake.output.global_pwr_cost        
         total_cost_node_csv = snakemake.output.node_cost
         total_cost_country_csv = snakemake.output.country_cost
+        total_cost_global_csv = snakemake.output.global_cost
+
     else:
         discounted_cost_by_technology_csv = (
             "results/India/results/DiscountedCostByTechnology.csv"
         )
         demand_csv = "results/India/results/Demand.csv"
-        power_cost_node_csv = "results/India/result_summaries/NodePowerCost.csv"
-        power_cost_country_csv = "results/India/result_summaries/CountryPowerCost.csv"
-        total_cost_node_csv = "results/India/result_summaries/NodeCost.csv"
-        total_cost_country_csv = "results/India/result_summaries/CountryCost.csv"
+        power_cost_node_csv = "results/India/result_summaries/PowerCostNode.csv"
+        power_cost_country_csv = "results/India/result_summaries/PowerCostCountry.csv"
+        power_cost_global_csv = "results/India/result_summaries/PowerCostGlobal.csv"
+        total_cost_node_csv = "results/India/result_summaries/TotalCostNode.csv"
+        total_cost_country_csv = "results/India/result_summaries/TotalCostCountry.csv"
+        total_cost_global_csv = "results/India/result_summaries/TotalCostGlobal.csv"
 
     discounted_cost_by_technology = pd.read_csv(
         discounted_cost_by_technology_csv, index_col=[0, 1, 2]
@@ -119,7 +160,11 @@ def get_pwr_cost(demand: pd.DataFrame, cost: pd.DataFrame) -> pd.DataFrame:
 
     tech_cost = get_tech_cost(discounted_cost_by_technology, country=False)
     storage_cost = get_storage_cost(discounted_cost_by_storage, country=False)
-    cost = tech_cost.add(storage_cost, fill_value=0)
+    transmission_cost = get_transmission_cost(discounted_cost_by_technology, country=False)
+
+    cost = tech_cost.add(storage_cost, fill_value = 0
+                         ).add(transmission_cost, fill_value = 0)
+
     demand = get_demand(demand_raw, country=False)
     pwr_cost = get_pwr_cost(demand, cost)
 
@@ -128,11 +173,31 @@ def get_pwr_cost(demand: pd.DataFrame, cost: pd.DataFrame) -> pd.DataFrame:
 
     # country level metrics
 
-    tech_cost = get_tech_cost(discounted_cost_by_technology, country=True)
-    storage_cost = get_storage_cost(discounted_cost_by_storage, country=True)
-    cost = tech_cost.add(storage_cost, fill_value=0)
-    demand = get_demand(demand_raw, country=True)
-    pwr_cost = get_pwr_cost(demand, cost)
-
-    pwr_cost.to_csv(power_cost_country_csv, index=True)
-    cost.to_csv(total_cost_country_csv, index=True)
+    tech_cost_country = get_tech_cost(discounted_cost_by_technology, country=True)
+    storage_cost_country = get_storage_cost(discounted_cost_by_storage, country=True)
+    transmission_cost_country = get_transmission_cost(discounted_cost_by_technology, country=True)    
+
+    cost_country = tech_cost_country.add(storage_cost_country, fill_value = 0
+                                         ).add(transmission_cost_country, fill_value = 0)
+
+    demand_country = get_demand(demand_raw, country=True)
+    pwr_cost_country = get_pwr_cost(demand_country, cost_country)
+
+    pwr_cost_country.to_csv(power_cost_country_csv, index=True)
+    cost_country.to_csv(total_cost_country_csv, index=True)
+
+    # global level metrics
+
+    cost_global = tech_cost_country.add(storage_cost_country, 
+                                        fill_value = 0
+                                        ).add(transmission_cost_country, fill_value = 0
+                                              ).groupby(["YEAR"]).sum()
+
+    demand_global = get_demand(demand_raw, country=True)
+    demand_global = demand_global.groupby([
+        demand_global.index.get_level_values("YEAR")]).sum()
+
+    pwr_cost_global = get_pwr_cost(demand_global, cost_global)
+
+    pwr_cost_global.to_csv(power_cost_global_csv, index=True)
+    cost_global.to_csv(total_cost_global_csv, index=True)

workflow/scripts/osemosys_global/summary/gen_shares.py

@@ -12,6 +12,7 @@ def _get_gen_by_node(
 ) -> pd.DataFrame:
 
     df = production.copy()
+    df = df.loc[df.index.get_level_values("FUEL").str.startswith('ELC')]
 
     assert "TECHNOLOGY" in df.index.names
 
@@ -38,6 +39,7 @@ def _get_gen_by_country(
 ) -> pd.DataFrame:
 
     df = production.copy()
+    df = df.loc[df.index.get_level_values("FUEL").str.startswith('ELC')]
 
     assert "TECHNOLOGY" in df.index.names
 
@@ -56,6 +58,31 @@ def _get_gen_by_country(
         .sum()
     )
 
+def _get_gen_global(
+    production: pd.DataFrame,
+    include: Optional[list[str]] = None,
+    exclude: Optional[list[str]] = None,
+) -> pd.DataFrame:
+
+    df = production.copy()
+    df = df.loc[df.index.get_level_values("FUEL").str.startswith('ELC')]
+
+    assert "TECHNOLOGY" in df.index.names
+
+    df["TECH"] = df.index.get_level_values("TECHNOLOGY").str[3:6]
+    #df["COUNTRY"] = df.index.get_level_values("TECHNOLOGY").str[6:9]
+
+    if exclude:
+        df = df[~df.TECH.isin(exclude)].copy()
+
+    if include:
+        df = df[df.TECH.isin(include)].copy()
+
+    return (
+        df.reset_index()[["REGION", "YEAR", "VALUE"]]
+        .groupby(["REGION", "YEAR"])
+        .sum()
+    )
 
 def calc_generation_shares_node(
     production_by_technology: pd.DataFrame, exclusions: Optional[list[str]] = None
@@ -72,10 +99,13 @@ def calc_generation_shares_node(
     ]
 
     total = _get_gen_by_node(df, exclude=exclusions).rename(columns={"VALUE": "TOTAL"})
+
     clean = _get_gen_by_node(df, include=CLEAN).rename(columns={"VALUE": "CLEAN"})
+
     renewable = _get_gen_by_node(df, include=RENEWABLES).rename(
         columns={"VALUE": "RENEWABLE"}
     )
+
     fossil = _get_gen_by_node(df, include=FOSSIL).rename(columns={"VALUE": "FOSSIL"})
 
     shares = (
@@ -128,13 +158,50 @@ def calc_generation_shares_country(
 
     return shares[["CLEAN", "RENEWABLE", "FOSSIL"]].round(1)
 
+def calc_generation_shares_global(
+    production_by_technology: pd.DataFrame, exclusions: Optional[list[str]] = None
+) -> pd.DataFrame:
+
+    if not exclusions:
+        exclusions = []
+
+    df = production_by_technology.copy()
+
+    df = df[
+        (df.index.get_level_values("TECHNOLOGY").str.startswith("PWR"))
+        & ~(df.index.get_level_values("TECHNOLOGY").str.contains("TRN"))
+    ]
+
+    total = _get_gen_global(df, exclude=exclusions).rename(
+        columns={"VALUE": "TOTAL"}
+    )
+    clean = _get_gen_global(df, include=CLEAN).rename(columns={"VALUE": "CLEAN"})
+    renewable = _get_gen_global(df, include=RENEWABLES).rename(
+        columns={"VALUE": "RENEWABLE"}
+    )
+    fossil = _get_gen_global(df, include=FOSSIL).rename(columns={"VALUE": "FOSSIL"})
+
+    shares = (
+        total.join(clean, how="outer")
+        .join(renewable, how="outer")
+        .join(fossil, how="outer")
+        .fillna(0)
+    )
+
+    shares["CLEAN"] = shares.CLEAN.div(shares.TOTAL).mul(100)
+    shares["RENEWABLE"] = shares.RENEWABLE.div(shares.TOTAL).mul(100)
+    shares["FOSSIL"] = shares.FOSSIL.div(shares.TOTAL).mul(100)
+
+    return shares[["CLEAN", "RENEWABLE", "FOSSIL"]].round(1)
+
 
 if __name__ == "__main__":
     if "snakemake" in globals():
         production_by_technology_annual_csv = snakemake.input.production_by_technology
         storage = snakemake.params.storage
         gen_shares_node = snakemake.output.generation_shares_node
         gen_shares_country = snakemake.output.generation_shares_country
+        gen_shares_global = snakemake.output.generation_shares_global
     else:
         production_by_technology_annual_csv = (
             "results/India/results/ProductionByTechnologyAnnual.csv"
@@ -144,6 +211,10 @@ def calc_generation_shares_country(
         gen_shares_country = (
             "results/India/result_summaries/GenerationSharesCountry.csv"
         )
+        gen_shares_global = (
+            "results/India/result_summaries/GenerationSharesGlobal.csv"
+        )
+
 
     production_by_technology_annual = pd.read_csv(
         production_by_technology_annual_csv, index_col=[0, 1, 2, 3]
@@ -155,9 +226,9 @@ def calc_generation_shares_country(
         exclusions = []
 
     nodes = calc_generation_shares_node(production_by_technology_annual, exclusions)
-    country = calc_generation_shares_country(
-        production_by_technology_annual, exclusions
-    )
+    country = calc_generation_shares_country(production_by_technology_annual, exclusions)
+    glob = calc_generation_shares_global(production_by_technology_annual, exclusions)
 
     nodes.to_csv(gen_shares_node, index=True)
     country.to_csv(gen_shares_country, index=True)
+    glob.to_csv(gen_shares_global, index=True)