Custom conversions

Here we show how custom conversions can be passed to OpenSCM-Units’ ScmUnitRegistry.

import pandas as pd

from openscm_units import ScmUnitRegistry

/tmp/ipykernel_681/666898991.py:1: DeprecationWarning: 
Pyarrow will become a required dependency of pandas in the next major release of pandas (pandas 3.0),
(to allow more performant data types, such as the Arrow string type, and better interoperability with other libraries)
but was not found to be installed on your system.
If this would cause problems for you,
please provide us feedback at https://github.com/pandas-dev/pandas/issues/54466
        
  import pandas as pd

Custom conversions DataFrame

On initialisation, a pd.DataFrame can be provided which contains the custom conversions. This pd.DataFrame should be formatted as shown below, with an index that contains the different species and columns which contain the conversion for different metrics.

metric_conversions_custom = pd.DataFrame(
    [
        {
            "Species": "CH4",
            "Custom1": 20,
            "Custom2": 25,
        },
        {
            "Species": "N2O",
            "Custom1": 341,
            "Custom2": 300,
        },
    ]
).set_index("Species")
metric_conversions_custom

	Custom1	Custom2
Species
CH4	20	25
N2O	341	300

With such a pd.DataFrame, we can use custom conversions in our unit registry as shown.

# initialise the unit registry with custom conversions
unit_registry = ScmUnitRegistry(metric_conversions=metric_conversions_custom)
# add standard conversions before moving on
unit_registry.add_standards()

# start with e.g. N2O
nitrous_oxide = unit_registry("tN2O / yr")

# our unit registry allows us to make conversions using the
# conversion factors we previously defined
for context in ["Custom1", "Custom2"]:
    with unit_registry.context(context):
        print(
            f"{nitrous_oxide} in CO2-equivalent in context {context} "
            "is {nitrous_oxide.to('tCO2 / yr')}"
        )

1.0 tN2O / a in CO2-equivalent in context Custom1 is {nitrous_oxide.to('tCO2 / yr')}
1.0 tN2O / a in CO2-equivalent in context Custom2 is {nitrous_oxide.to('tCO2 / yr')}