add new uncertainty switch; addresses USEPA#248

electricitylci/generation.py

@@ -76,11 +76,12 @@
 -   Separate replace egrid function
 -   Fix zero division error in aggregate data
 -   Implement Hawkins-Young uncertainty
+-   Add uncertainty switch
 
 Created:
     2019-06-04
 Last edited:
-    2024-08-09
+    2024-08-12
 """
 __all__ = [
     "add_data_collection_score",
@@ -119,7 +120,11 @@ def _calc_sigma(p_series):
         Assumes a 90% confidence level (see :param:`alpha`).
     """
     alpha = 0.9
-    (is_error, sigma) = hawkins_young_sigma(p_series.values, alpha)
+    if model_specs.calculate_uncertainty:
+        (is_error, sigma) = hawkins_young_sigma(p_series.values, alpha)
+    else:
+        return None
+
     if is_error:
         return None
     else:
@@ -1523,7 +1528,7 @@ def turn_data_to_dict(data, upstream_dict):
     data["flowProperty"] = ""
     data["baseUncertainty"] = ""
     data["provider"] = ""
-    data["FlowType"]="ELEMENTARY_FLOW"
+    data["FlowType"] = "ELEMENTARY_FLOW"
 
     # Effectively rename 'Unit' to 'unit', 'uncertainty Max/Min' to 'Max/Min'
     data["unit"] = data["Unit"]
@@ -1542,7 +1547,7 @@ def turn_data_to_dict(data, upstream_dict):
     # Define products based on compartment label
     product_filter=(
         (data["Compartment"].str.lower().str.contains("technosphere"))
-        |(data["Compartment"].str.lower().str.contains("valuable"))
+        | (data["Compartment"].str.lower().str.contains("valuable"))
     )
     data.loc[product_filter, "FlowType"] = "PRODUCT_FLOW"
 
@@ -1572,7 +1577,7 @@ def turn_data_to_dict(data, upstream_dict):
     data["uncertainty"] = ""
     for index, row in data.iterrows():
         data.at[index, "uncertainty"] = uncertainty_table_creation(
-            data.loc[index:index, :]  # TODO: check this syntax
+            data.loc[index:index, :]
         )
         data.at[index, "flow"] = flow_table_creation(
             data.loc[index:index, :]
@@ -1666,13 +1671,15 @@ def olcaschema_genprocess(database, upstream_dict={}, subregion="BA"):
         "GeomMean",
         "GeomSD",
     ]
+    non_agg_cols = [x for x in non_agg_cols if x in database.columns]
 
     # Create a data frame with one massive column of exchanges
     database_groupby = database.groupby(by=base_cols)
     process_df = pd.DataFrame(
         database_groupby[non_agg_cols].apply(
             turn_data_to_dict,
-            (upstream_dict))
+            (upstream_dict)
+        )
     )
     process_df.columns = ["exchanges"]
     process_df.reset_index(inplace=True)

electricitylci/model_config.py

@@ -126,7 +126,11 @@ class ModelSpecs:
     post_process_generation_emission_factors : bool
         Whether to do post processing.
     gen_mix_from_model_generation_data : bool, optional
+        Influences how generation mix process is read when not replacing
+        eGRID with EIA. Not available in model config.
         Defaults to false.
+    calculate_uncertainty : bool
+        Whether or not to compute the uncertainty of emission flows.
     namestr : str
         Absolute path to JSON-LD zip output file.
         File name includes the model name and current time stamp and is
@@ -185,7 +189,9 @@ def __init__(self, model_specs, model_name):
             os.path.join(data_dir, self.fuel_name_file))
         self.post_process_generation_emission_factors = model_specs[
             "post_process_generation_emission_factors"]
-        self.gen_mix_from_model_generation_data=False
+        self.gen_mix_from_model_generation_data = False
+        self.calculate_uncertainty = model_specs.get(
+            "calculate_uncertainty", True)
         self.namestr = (
             f"{output_dir}/{model_name}_jsonld_"
             f"{datetime.datetime.now().strftime('%Y%m%d_%H%M%S')}.zip"

electricitylci/modelconfig/ELCI_1_config.yml

@@ -21,48 +21,55 @@ regional_aggregation: 'BA'
 # the generation mix.
 egrid_year: 2016
 eia_gen_year: 2016
-replace_egrid: True
+replace_egrid: true
 
 # NETL has developed profiles for renewable generation to capture construction
-# impacts (e.g., solar PV menufacturing).
-include_renewable_generation: True
+# impacts (e.g., solar PV manufacturing).
+include_renewable_generation: true
 
 # NETL has developed profiles for power plant water use.
-include_netl_water: True
+include_netl_water: true
 
 # Upstream data from NETL can be included in the outputs.
-include_upstream_processes: True
+include_upstream_processes: true
 
 inventories_of_interest:
   eGRID: 2016
   TRI: 2016
   NEI: 2016
   RCRAInfo: 2015
 
+# Provide uncertainty estimates for emissions.
+calculate_uncertainty: true
+
 
 # GENERATOR FILTERS
 # These parameters determine if any power plants are filtered out.
-include_only_egrid_facilities_with_positive_generation: True
-filter_on_efficiency: True
+include_only_egrid_facilities_with_positive_generation: true
+filter_on_efficiency: true
 egrid_facility_efficiency_filters:
   lower_efficiency: 10
   upper_efficiency: 100
 
 # ELCI creates life cycle processes for each fuel type. If you only want to
 # include power plants with a minimum amount of generation from a single fuel,
 # make sure this parameter is set to true.
-filter_on_min_plant_percent_generation_from_primary_fuel: True
+filter_on_min_plant_percent_generation_from_primary_fuel: true
 min_plant_percent_generation_from_primary_fuel_category: 90
 
 # If the parameter above is set to true, one may still want to keep the plants
-# that do not meet the threshhold. This setting will mark those power plants as
+# that do not meet the threshold. This setting will mark those power plants as
 # MIXED and generate life cycle processes for them. Note that the definition
 # of mixed is purposefully ambiguous and will mean different things depending
 # on the region of aggregation.
-keep_mixed_plant_category: True
+keep_mixed_plant_category: true
 
-# NEEDS A GOOD DESCRIPTION OF THIS PARAMETER
-filter_non_egrid_emission_on_NAICS: True
+# To ensure consistency with eGRID, the model applies a filter to remove
+# industrial cogeneration facilities from the EIA Form 923 data based on
+# their NAICS code (i.e., code 22: Electricity Utilities and Independent
+# Power Producers) and by EIA sector numbers 1 and 2. This filter is
+# designed to reproduce the same power plants that are listed in eGRID.
+filter_non_egrid_emission_on_NAICS: true
 
 
 # TRANSMISSION AND ELECTRICITY TRADING
@@ -78,16 +85,16 @@ efficiency_of_distribution_grid: 0.95
 # without a surplus trading pool) that is based on eGRID regions or a method
 # from NETL that uses an input-output approach using exchange data between
 # balancing authorities from EIA.
-EPA_eGRID_trading: False
+EPA_eGRID_trading: false
 
 # Select between net trading (true) or gross trading (false) between NERC and
 # eGRID regions for consumption mix calculations.
-net_trading: False
+net_trading: false
 
 NETL_IO_trading_year: 2016
 
 # OTHER PARAMETERS
 
 fedelemflowlist_version: 1.0.2
-post_process_generation_emission_factors: False
+post_process_generation_emission_factors: false
 fuel_name_file: fuelname_1.csv

electricitylci/modelconfig/ELCI_2020_config.yml

@@ -21,18 +21,18 @@ regional_aggregation: 'BA'
 # the generation mix.
 egrid_year: 2020
 eia_gen_year: 2020
-replace_egrid: True
+replace_egrid: true
 
 # NETL has developed profiles for renewable generation to capture construction
 # impacts (e.g., solar PV manufacturing). Set to true to include these impacts.
-include_renewable_generation: True
+include_renewable_generation: true
 
 # NETL has developed profiles for power plant water use.
 # Set to true to include these impacts.
-include_netl_water: True
+include_netl_water: true
 
 # Upstream data from NETL can be included in the outputs.
-include_upstream_processes: True
+include_upstream_processes: true
 
 # Inventory vintages taken from StEWI.
 # https://github.com/USEPA/standardizedinventories
@@ -42,34 +42,37 @@ inventories_of_interest:
   NEI: 2020
   RCRAInfo: 2019
 
+# Provide uncertainty estimates for emissions.
+calculate_uncertainty: true
+
 
 # GENERATOR FILTERS
 # These parameters determine if any power plants are filtered out
-include_only_egrid_facilities_with_positive_generation: True
-filter_on_efficiency: True
+include_only_egrid_facilities_with_positive_generation: true
+filter_on_efficiency: true
 egrid_facility_efficiency_filters:
   lower_efficiency: 10
   upper_efficiency: 100
 
 # ELCI creates life cycle processes for each fuel type. If you only want to
 # include power plants with a minimum amount of generation from a single fuel,
 # make sure this parameter is set to true.
-filter_on_min_plant_percent_generation_from_primary_fuel: True
+filter_on_min_plant_percent_generation_from_primary_fuel: true
 min_plant_percent_generation_from_primary_fuel_category: 90
 
 # If the parameter above is set to true, one may still want to keep the plants
 # that do not meet the threshold. This setting will mark those power plants as
 # MIXED and generate life cycle processes for them. Note that the definition
 # of mixed is purposefully ambiguous and will mean different things depending
 # on the region of aggregation.
-keep_mixed_plant_category: True
+keep_mixed_plant_category: true
 
 # To ensure consistency with eGRID, the model applies a filter to remove
 # industrial cogeneration facilities from the EIA Form 923 data based on
 # their NAICS code (i.e., code 22: Electricity Utilities and Independent
 # Power Producers) and by EIA sector numbers 1 and 2. This filter is
 # designed to reproduce the same power plants that are listed in eGRID.
-filter_non_egrid_emission_on_NAICS: True
+filter_non_egrid_emission_on_NAICS: true
 
 
 # TRANSMISSION AND ELECTRICITY TRADING
@@ -85,18 +88,18 @@ efficiency_of_distribution_grid: 0.95
 # without a surplus trading pool) that is based on eGRID regions or a method
 # from NETL that uses an input-output approach using exchange data between
 # balancing authorities from EIA. If false, net trading is used.
-EPA_eGRID_trading: False
+EPA_eGRID_trading: false
 
 # For net trading, select between two options for for consumption mix
 # calculations: net trading (true) or gross trading (false) between NERC
 # and eGRID regions.
-net_trading: False
+net_trading: false
 
 # Trading year for EIA data used for net trading.
 NETL_IO_trading_year: 2020
 
 # OTHER PARAMETERS
 
 fedelemflowlist_version: 1.0.3
-post_process_generation_emission_factors: False
+post_process_generation_emission_factors: false
 fuel_name_file: fuelname_1.csv

electricitylci/modelconfig/ELCI_2021_config.yml

@@ -21,18 +21,18 @@ regional_aggregation: 'BA'
 # the generation mix.
 egrid_year: 2021
 eia_gen_year: 2021
-replace_egrid: True
+replace_egrid: true
 
 # NETL has developed profiles for renewable generation to capture construction
 # impacts (e.g., solar PV manufacturing). Set to true to include these impacts.
-include_renewable_generation: True
+include_renewable_generation: true
 
 # NETL has developed profiles for power plant water use.
 # Set to true to include these impacts.
-include_netl_water: True
+include_netl_water: true
 
 # Upstream data from NETL can be included in the outputs.
-include_upstream_processes: True
+include_upstream_processes: true
 
 # Inventory vintages taken from StEWI.
 # https://github.com/USEPA/standardizedinventories
@@ -42,34 +42,37 @@ inventories_of_interest:
   NEI: 2020
   RCRAInfo: 2019
 
+# Provide uncertainty estimates for emissions.
+calculate_uncertainty: true
+
 
 # GENERATOR FILTERS
 # These parameters determine if any power plants are filtered out
-include_only_egrid_facilities_with_positive_generation: True
-filter_on_efficiency: True
+include_only_egrid_facilities_with_positive_generation: true
+filter_on_efficiency: true
 egrid_facility_efficiency_filters:
   lower_efficiency: 10
   upper_efficiency: 100
 
 # ELCI creates life cycle processes for each fuel type. If you only want to
 # include power plants with a minimum amount of generation from a single fuel,
 # make sure this parameter is set to true.
-filter_on_min_plant_percent_generation_from_primary_fuel: True
+filter_on_min_plant_percent_generation_from_primary_fuel: true
 min_plant_percent_generation_from_primary_fuel_category: 90
 
 # If the parameter above is set to true, one may still want to keep the plants
 # that do not meet the threshold. This setting will mark those power plants as
 # MIXED and generate life cycle processes for them. Note that the definition
 # of mixed is purposefully ambiguous and will mean different things depending
 # on the region of aggregation.
-keep_mixed_plant_category: True
+keep_mixed_plant_category: true
 
 # To ensure consistency with eGRID, the model applies a filter to remove
 # industrial cogeneration facilities from the EIA Form 923 data based on
 # their NAICS code (i.e., code 22: Electricity Utilities and Independent
 # Power Producers) and by EIA sector numbers 1 and 2. This filter is
 # designed to reproduce the same power plants that are listed in eGRID.
-filter_non_egrid_emission_on_NAICS: True
+filter_non_egrid_emission_on_NAICS: true
 
 
 # TRANSMISSION AND ELECTRICITY TRADING
@@ -85,18 +88,18 @@ efficiency_of_distribution_grid: 0.95
 # without a surplus trading pool) that is based on eGRID regions or a method
 # from NETL that uses an input-output approach using exchange data between
 # balancing authorities from EIA. If false, net trading is used.
-EPA_eGRID_trading: False
+EPA_eGRID_trading: false
 
 # For net trading, select between two options for for consumption mix
 # calculations: net trading (true) or gross trading (false) between NERC
 # and eGRID regions.
-net_trading: False
+net_trading: false
 
 # Trading year for EIA data used for net trading.
 NETL_IO_trading_year: 2021
 
 # OTHER PARAMETERS
 
 fedelemflowlist_version: 1.0.3
-post_process_generation_emission_factors: False
+post_process_generation_emission_factors: false
 fuel_name_file: fuelname_1.csv