bonsai-ipcc 0.3.2

Implementation of the IPCC Guidelines for National GHG Inventories.

bonsai_ipcc

The bonsai_ipcc python package enables users to calculate national greenhouse gas (GHG) inventories based on the guidelines provided by the International Panel on Climate Change.

By using volumes and chapters, the python package follows the structure of the IPCC guidelines and allows users to access individual equations and sequences of equations as bonsai_ipcc.<volume>.<chapter>.elementary.<equation> or bonsai_ipcc.<volume>.<chapter>.sequence.<tier_method>. The package allows users to access the data as Pandas dataframes in bonsai_ipcc.<volume>.<chapter>.dimension.<table> or bonsai_ipcc.<volume>.<chapter>.parameter.<table>: dimensions list valid coordinates to access parameters; parameters are values to be used in equations. When using the bonsai_ipcc python package, it may helpful to also use the pdf documents for additional information.

The package also allows uncertainty information to be taken into account. Within a sequence, the user can choose between analytical error propagation and Monte Carlo simulation. Thereby, the values of an equation are transformed into ufloat or numpy.array, respectively.

A comrehensive documentation is available here.

Installation for users

You can install the package from PyPi sing pip:

pip install bonsai_ipcc

You can also download the package from gitlab.com Replace the keyword tag by the specific version, e.g., v0.3.0.

pip install git+ssh://[e-mail]/bonsamurais/bonsai/util/ipcc.git@tag

Change pip to pip3 in Linux. Note that the path may change in future versions.

Installation for developers

Create a python environment

The bonsai_ipcc package requires python>=3.9. If you use conda as your python package manager, you could do something like:

conda create --name py311 python=3.11
conda activate py311

Install the package in editable mode

git clone [e-mail]:bonsamurais/bonsai/util/ipcc.git
cd bonsai_ipcc
pip install -e .

Basic use

Inside a Python console or notebook, create an instance of the IPCC class like this:

import bonsai_ipcc
my_ipcc = bonsai_ipcc.IPCC()

With my_ipcc.<volume>.<chapter> the bonsai_ipcc package follows the structure of the IPCC guidelines. To show the elementary equations and sequences of a certain chapter in a specific volume:

dir(my_ipcc.waste.swd.elementary)
# dir(my_ipcc.waste.swd.sequence)

To find information about a elementary equation:

help(my_ipcc.waste.swd.elementary.ddoc_from_wd_data)

The following will print the docstring with information on the parameters and the reqiured units:

Help on function ddoc_from_wd_data in module bonsai_ipcc.waste.swd.elementary:

ddoc_from_wd_data_tier1(waste, doc, doc_f, mcf)
    Equation 3.2 (tier 1)

    Calculates the decomposable doc (ddocm) from waste disposal data.

    Argument
    ---------
    waste (tonnes) : float
        Amount of waste
        (either wet or dry-matter, but attention to doc!)
    doc (kg/kg) : float
        Fraction of degradable organic carbon in waste.
    doc_F (kg/kg) : float
        Fraction of doc that can decompose.
    mcf (kg/kg) : float
        CH4 correction factor for aerobic decomposition in the year of decompostion.

    Returns
    -------
    VALUE: float
        Decomposable doc (tonnes/year)

To show the dimensions of a certain parameter:

my_ipcc.waste.swd.parameter.mcf.index.names

FrozenList(['swds_type', 'property'])

To find the possible values of a dimension:

my_ipcc.waste.swd.dimension.swds_type.index

Index(['managed', 'managed_well_s-a', 'managed_poorly_s-a', 'managed_well_a-a',
       'managed_poorly_a-a', 'unmanaged_deep', 'unmanaged_shallow',
       'uncharacterised'],
      dtype='object', name='code')

To retrieve the value and the unit of a certain parameter.

my_ipcc.waste.swd.parameter.mcf.loc[("managed","def")]

value      1.0
unit     kg/kg
Name: (managed, def), dtype: object

Run a tier sequence

Despite the fact that various default data for parameter tables is provided within the bonsai_ipcc package, in most cases, the user still needs to collect data to calculate the greenhouse gas inventories. For the tier1_co2 sequence in the incineration chapter of volume waste, data for urban population is required. The data can be added as a pandas DataFrame.

import bonsai_ipcc
import pandas as pd

# urban population
d = {
    "year": [2010,2010,2010,2010,2010],
    "region": ["DE","DE","DE","DE","DE"],
    "property": [
        "def","min","max","abs_min","abs_max"
    ],
    "value": [
        62940432,61996325.52,63884538.48,0.0,"inf",
    ],
    "unit": [
    "cap/yr","cap/yr","cap/yr","cap/yr","cap/yr",
    ],
}
urb_pop = pd.DataFrame(d).set_index(["year", "region", "property"])

my_ipcc=bonsai_ipcc.IPCC()
my_ipcc.waste.incineration.parameter.urb_population=urb_pop

NOTE: When adding own data, the user is encouraged to also specify uncertainty information. Property "def" is always required and specifies the mean value. For uncertainty analysis "min", "max", "abs_min" and "abs_max" are required ("min": 2.5 percentile, "max": 97.5 percentile, "abs_min": absolute minimum, "abs_max": absolute maximum).

To get a list of all parameters involved in the sequence, you can do:

my_ipcc.inspect(my_ipcc.waste.incineration.sequence.tier1_co2)

To calculate the GHG inventory based on a tier method, specifiy the keywords of the sequence:

my_tier = my_ipcc.waste.incineration.sequence.tier1_co2(
          year=2010, region="DE",wastetype="msw_plastics", incintype="inc_unspecified", uncertainty="def")

# show the list of steps of the sequence
my_tier.__dict__

For uncertainty calculation based on Monte Carlo use uncertainty="monte_carlo", for analytical error propagation use uncertainty="analytical".

To retrieve the result's value of a sequence's step, type:

my_tier.co2_emissions.value

NOTE: The type of value depends on the uncertainty assessment. For uncertainty = "def": type = float, for uncertainty = "analytical": type = ufloat and for uncertainty = "monte-carlo": type = numpy.array. Furthermore, some tier sequences provide time series instead of one single value. If so, value is of type numpy.array, including the values for different years. The type of each years' value also depend on the uncertainty assessment.