Using lciafmt

Complete Examples for LCIA Methods

LCIA Methods as generated for the Federal LCA Commons can be generated using the scripts in the examples folder.

Loading method data

A data frame with the data of a method can be loaded with the get_method(<method ID>) function

import lciafmt

traci = lciafmt.get_method(lciafmt.Method.TRACI)

This will download and cache the raw method data in a temporary folder (~/temp/lciafmt). Before downloading method data, lciafmt will first check if the method data are available in this cache folder. Alternatively, a file path or web URL can be passed as arguments to the get_method function to load the data from other locations:

import lciafmt

traci = lciafmt.get_method(lciafmt.Method.TRACI,
                           file="path/to/traci_2.1.xlsx")
traci = lciafmt.get_method(lciafmt.Method.TRACI,
                           url="http://.../path/to/traci_2.1.xlsx")

Also, it is possible to clear the cache to ensure that the newest version is downloaded from the internet:

import lciafmt

lciafmt.clear_cache()
traci = lciafmt.get_method(lciafmt.Method.TRACI)

The function supported_methods returns a list of meta data objects that each contain the information of LCIA methods that are currently supported:

import lciafmt

lciafmt.supported_methods()

Apply flow mappings

The flow mappings defined in the fedelemflowlist can be directly applied on a data frame to generate a new data frame with mapped flows. A specific source system can be selected via the system parameter. The system parameter must match a specific mapping file available in fedelemflowlist :

import lciafmt

traci = lciafmt.get_method(lciafmt.Method.TRACI)
traci_mapped = lciafmt.map_flows(traci, system="TRACI2.1")

The available systems can be retrieved via the supported_mapping_systems() function:

import lciafmt

lciafmt.supported_mapping_systems()

Additionally, the map_flows function accepts the following optional parameters:

• mapping: a data frame in the fedelemflowlist mapping format that contains the mapping that should be applied. Used instead of specifying system
• preserve_unmapped: a Boolean value that indicates whether the unmapped flows should be preserved in the resulting data frame.
• case_insensitive: a Boolean value that indicates if mapping should be case insensitive (True) or by default case sensitive (False)

Generating custom endpoint methods

Midpoint or endpoint indicators can be converted to custom endpoint or valuation methods based on user supplied data. For example, an indicator that tracks ecosystem impacts in species-years can be converted to a valuation. The input data for this conversion must be provided within the data subfolder as a csv file and in the format specified in the endpoint format specs.

import lciafmt

valuation_method = lciafmt.generate_endpoints(file = 'file_name', name = 'My Valuation Method')

The optional parameter matching_fields can be used to indicate whether the new method should be converted and aggregated for each 'Indicator' (default) or each listed 'Indicator unit'. An example using 'Indicator unit' is shown for the Weidema Valuation method (Weidema 2009).

Data export

The method data are stored in a standard pandas data frame so that the standard export functions of pandas can be directly used:

import lciafmt

traci = lciafmt.get_method(lciafmt.Method.TRACI)
traci.to_csv("output/traci_2.1.csv", index=False)

Additionally, a method can be stored as JSON-LD package that can be imported into an openLCA database:

import lciafmt

traci = lciafmt.get_method(lciafmt.Method.TRACI)
lciafmt.to_jsonld(traci, "output/traci_2.1_jsonld.zip")

When also elementary flows should be written to the JSON-LD package the write_flows flag can be passed to the export call:

import lciafmt

traci = lciafmt.get_method(lciafmt.Method.TRACI)
lciafmt.to_jsonld(traci, "output/traci_2.1_jsonld.zip", write_flows=True)

Note that unit groups and flow properties are currently not added to the JSON-LD package so that the package can only be imported into a database where at least the standard openLCA unit groups and flow properties are available.

Logging details

The lciafmt module writes messages to the default logger of the logging package. In order to see more details, you can set the log level to a finer level:

import logging as log
log.basicConfig(level=log.INFO)