xiaoshir / premise

Coupling REMIND output to ecoinvent LCA databases.

Coupling Brightway2 & Wurst Future Ecoinvent Toolset to the REMIND IAM.

Introduction

rmnd-lca allows to align the life cycle inventories contained in the ecoinvent 3.5 and 3.6 cutoff databases with the output results of the REMIND IAM, in order to produce life cycle inventories under future policy scenarios for any year between 2005 and 2150.

In the current version, this includes:

• electricity generation: alignment of regional electricity production mixes as well as efficiencies for a number of electricity production technologies, including Carbon Capture and Storage technologies.

In upcoming versions, important sectors such as cement and steel will also be updated.

Documentation

https://rmnd-lca.readthedocs.io/en/latest/

Objective

The objective is to produce life cycle inventories under future energy policies, by modifying the inventory database ecoinvent 3 to reflect projected energy policy trajectories.

Requirements

• Python language interpreter 3.x
• License for ecoinvent 3
• Brightway2
• REMIND IAM output files come with the library ("xxx.mif" and "GAINS emission factors.csv") and are located by default in the subdirectory "/data/Remind output files". A file path can be specified to fetch the REMIND IAM output files elsewhere on your computer.

How to install this package?

In a terminal, from Github:

pip install git+https://github.com/romainsacchi/rmnd-lca.git

will install the package and the required dependencies.

Alternatively, from Conda:

conda install -c romainsacchi/label/nightly rmnd-lca-dev

How to use it?

Extract (using brightway2)

A preliminary requirement to the use this library is to have a brightway2 project created and opened, with the ecoinvent 3.5 cutoff or ecoinvent 3.6 cutoff database registered, so that:

    import brightway2 as bw
    bw.projects.set_current('remind')
    bw.databases

returns

    Databases dictionary with 2 object(s):
	biosphere3
	ecoinvent 3.5 cutoff

Then, for a chosen policy and year between 2005 and 2150, the following function will:

• extract the ecoinvent database, clean it, add additional inventories for carbon capture and storage, biofuels, etc.

For example, here with the year 2028 and the policy "Business-as-usual":

    from rmnd_lca import *
    ndb = NewDatabase(scenario = 'BAU',
              year = 2028,
              source_db = 'ecoinvent 3.6 cutoff',
              source_version = 3.6,
             )

Note that, by default, the library will look for REMIND output files ("xxx.mif" files and "GAINS emission factors.csv") in the "data/Remind output files" subdirectory. If those are not located there, you need to specify the path to the correct directory, as such::

    from rmnd_lca import *
    ndb = NewDatabase(scenario = 'BAU',
              year = 2028,
              source_db = 'ecoinvent 3.6 cutoff',
              source_version = 3.6,
              r"C:\Users\username\Documents\Remind output files"
             )

Extract (without brightway2)

If you are not using brightway2, you may load the ecoinvent database from its ecospold2 files (available from the ecoinvent website), like shown in the example below, by specifying source_type = 'ecospold' and the file path to the ecospold files in source_file_path.

    from rmnd_lca import *
    ndb = NewDatabase(scenario = 'BAU',
                  year = 2028,
                  source_db = 'ecoinvent 3.5 cutoff',
                  source_version = 3.5,
                  source_type = 'ecospold',
                  source_file_path = r"C:\Users\path\ecoinvent 3.5_cutoff_ecoSpold02\datasets"
                 )

Transform

A series of transformations can be performed on the extracted database. Currently, only the transformation regarding electricity generation and distribution is implemented.

Electricity

The following function will:

• remove existing electricity markets
• replace them by regional markets (high, medium and low voltage) with a geographical scope and production mix defined by the REMIND model for that year,
• relink electricity-consuming activities to the newly created electricity markets,
• update the efficiency of electricity-producing technologies (fuel-input-to-energy-output ratio), according to the projections given by REMIND,
• and rescale fuel-related emissions of electricity-producing technologies according to their newly defined efficiency.

    ndb.update_electricity_to_remind_data()

returns

    Remove old electricity datasets
    Create high voltage markets.
    Create medium voltage markets.
    Create low voltage markets.
    Link activities to new electricity markets.
    Log of deleted electricity markets saved in C:\Users\username\Documents\GitHub\rmnd-lca\rmnd_lca\data\logs
    Log of created electricity markets saved in C:\Users\username\Documents\GitHub\rmnd-lca\rmnd_lca\data\logs
    Rescale inventories and emissions for Coal IGCC
    Rescale inventories and emissions for Coal IGCC CCS
    Rescale inventories and emissions for Coal PC
    Rescale inventories and emissions for Coal PC CCS
    Rescale inventories and emissions for Coal CHP
    Rescale inventories and emissions for Gas OC
    Rescale inventories and emissions for Gas CC
    Rescale inventories and emissions for Gas CHP
    Rescale inventories and emissions for Gas CCS
    Rescale inventories and emissions for Oil
    Rescale inventories and emissions for Biomass CHP
    Rescale inventories and emissions for Biomass IGCC CCS
    Rescale inventories and emissions for Biomass IGCC

Note that logs of deleted and created electricity markets are created in the data/logs/ directory as MS Excel files, within rmnd_lca working directory.

Load (export back to brightway2)

Once the process is completed, the resulting database is registered back into the current Brightway2 project:

    ndb.write_db_to_brightway()

returns

    Write new database to Brightway2.
    15223 datasets
    540424 exchanges
    0 unlinked exchanges

    Writing activities to SQLite3 database:
    Created database: ecoinvent_BAU_2028

Load (export to matrices)

If you do not use brightway2, it is possible to export the transformed database into matrices.

    ndb.write_db_to_matrices()

returns

    Write new database to matrix.
    Matrices saved in C:\Users\username\Documents\GitHub\rmnd-lca\rmnd_lca\data\matrices.

Two matrices are created:

• matrix A: contains product exchanges
• matrix B: contains exchanges between activities and the biosphere

Two other files are exported:

• A_matrix_index: maps row/column index of A_matrix to activity label
• B_matrix_index: maps row index of B_matrix to biosphere flow label

The column indices of B_matrix are similar to the row/column indices of A_matrix.