EDGAR Inventory Processing¶

This tutorial demonstrates how to process EDGAR inventory data.

We will perform the following steps:

Load the data
Create a grid for the output
Remap the inventory to the grid
Process the data
Save the output

Visualizing the Inventory¶

Let’s examine the contents of our inventory.

emiproc provides several functions to assist us in this process.

[5]:

# We can look at the total emissions of the inventory (units are in kg/year)
inv.total_emissions.T

[5]:

	CH4	CO2	CO2bio
Agricultural soils	3.678327e+10	1.430920e+11	NaN
Agricultural waste burning	2.359133e+09	NaN	1.323736e+12
Chemical processes	3.591839e+08	7.536434e+11	NaN
Power Industry	5.637066e+08	1.469337e+13	8.824078e+11
Enteric fermentation	1.127170e+11	NaN	NaN
Combustion for manufacturing	6.869048e+08	6.223605e+12	1.980250e+12
Iron and steel production	1.648024e+08	2.898203e+11	2.250683e+08
Manure management	1.237949e+10	NaN	NaN
Fuel exploitation	9.676207e+10	3.818153e+11	2.763252e+11
Energy for buildings	9.904553e+09	3.385634e+12	2.975195e+12
Oil refineries and Transformation industry	6.970340e+09	2.220603e+12	4.190084e+11
Solid waste incineration	1.634158e+09	1.786001e+10	7.102761e+09
Solid waste landfills	3.855265e+10	NaN	NaN
Aviation climbing_and_descent	1.967106e+06	2.808891e+11	1.820282e+06
Aviation cruise	2.653998e+06	3.789719e+11	2.455900e+06
Aviation landing_and_takeoff	6.291362e+05	8.983632e+10	5.821781e+05
Railways, pipelines, off-road transport	9.078780e+06	2.732929e+11	2.795448e+09
Shipping	8.084582e+07	8.810432e+11	1.611196e+09
Road transportation	9.797972e+08	6.039474e+12	2.933062e+11
Waste water handling	2.412476e+10	NaN	NaN
__total__	3.450370e+11	3.824663e+13	8.161967e+12
Non energy use of fuels	NaN	2.767136e+10	NaN
Non-ferrous metals production	NaN	1.228438e+11	NaN
Non-metallic minerals production	NaN	1.980856e+12	NaN
Solvents and products use	NaN	6.231152e+10	NaN

Plot the inventory using built-in functions.

[6]:

import matplotlib.pyplot as plt
from emiproc.plots import plot_inventory

plt.style.use("default")

plot_inventory(inv, total_only=True)

../_images/tutos_edgar_processing_11_0.png

../_images/tutos_edgar_processing_11_1.png

../_images/tutos_edgar_processing_11_2.png

../_images/tutos_edgar_processing_11_3.png

From these results, we can observe how emissions are distributed globally.

In the totals, the largest categories of methane emissions are Enteric fermentation and Fuel exploitation. For CO2, the Power Industry is the largest category, followed by Road Transportation, Combustion for manufacturing, and Energy for buildings.

Most of these categories can be mitigated through the electrification of the economy and the adoption of renewable power sources. However, Enteric fermentation, which refers to the emissions from cows and other ruminants digesting food, presents a unique challenge that can be relatively easier and faster to address by simply reducing or stopping meat consumption.

It is also important to note that many “natural” CO2 emissions are associated with meat production due to deforestation (not accounted for in EDGAR), making dietary changes a doubly beneficial approach.

Processing the Inventory¶

In this step, we will demonstrate some standard operations that can be performed on the inventory.

Remapping the Inventory¶

In this section, we will remap the inventory to a grid.

Suppose we want to perform a simulation over Australia. First, we need to create a custom grid.

[7]:

from emiproc.grids import RegularGrid

# The grid can be defined by various parameters
# See the documentation for more details
# https://emiproc.readthedocs.io/en/master/api/grids.html#emiproc.grids.RegularGrid
australian_grid = RegularGrid(xmin=110, xmax=160, ymin=-45, ymax=-10, nx=200, ny=160)
australian_grid

[7]:

RegularGrid(unnamed)_nx(200)_ny(160)_dx(0.25)_dy(0.21875)_x(110,160.0)_y(-45,-10.0)_

In some cases, you may already have a grid for your simulation, which you could import into emiproc with appropriate code. A tutorial for developper is available.

[8]:

# Remap the inventory on the grid
from emiproc.regrid import remap_inventory

remapped = remap_inventory(inv, australian_grid)
remapped

C:\Users\coli\Documents\emiproc\emiproc\regrid.py:251: UserWarning: Geometry is in a geographic CRS. Results from 'area' are likely incorrect. Use 'GeoSeries.to_crs()' to re-project geometries to a projected CRS before this operation.

  gdf_weights.geometry_inter.area / gdf_weights.geometry.area

[8]:

Inventory(Inventory)

Let’s examine our remapped inventory.

[9]:

plot_inventory(remapped, total_only=True)

../_images/tutos_edgar_processing_19_0.png

../_images/tutos_edgar_processing_19_1.png

../_images/tutos_edgar_processing_19_2.png

../_images/tutos_edgar_processing_19_3.png

For this domain, the panorama is quite different. For example, waste is well-managed in Australia, so its fraction of the total emissions is smaller. Maybe Australia exports its waste to other regions such as Asia or Africa ? Who knows ?

Grouping Categories¶

Many categories are similar and have very small emissions.

In such cases, we can group some categories together to reduce the number of emission sectors.

[10]:

from emiproc.inventories.utils import group_categories

grouped = group_categories(
    remapped,
    categories_group={
        "agriculture": [
            "Agricultural soils",
            "Agricultural waste burning",
            "Manure management",
        ],
        "industry": [
            "Chemical processes",
            "Power Industry",
            "Oil refineries and Transformation industry",
            "Fuel exploitation",
            "Energy for buildings",
            "Combustion for manufacturing",
            "Iron and steel production",
            "Non energy use of fuels",
            "Solvents and products use",
            "Non-ferrous metals production",
            "Non-metallic minerals production",
        ],
        "livestock": ["Enteric fermentation"],
        "waste": [
            "Waste water handling",
            "Solid waste incineration",
            "Solid waste landfills",
        ],
        "transportation": [
            "Aviation climbing_and_descent",
            "Aviation cruise",
            "Aviation landing_and_takeoff",
            "Railways, pipelines, off-road transport",
            "Shipping",
            "Road transportation",
        ],
    },
)
grouped.total_emissions.T

[10]:

	CO2bio	CH4	CO2
agriculture	1.436499e+10	2.274178e+08	3.350383e+09
industry	4.741786e+10	1.683938e+09	2.830813e+11
transportation	5.580335e+08	8.589536e+06	1.137983e+11
__total__	6.234089e+10	4.749143e+09	4.002463e+11
livestock	NaN	2.403860e+09	NaN
waste	NaN	4.253374e+08	1.629228e+07

Merging CO2 Emissions¶

Ultimately, we aim to obtain the total CO2 emissions.

To achieve this, we will sum the biogenic CO2 with the fossil CO2 emissions.

[11]:

from emiproc.speciation import merge_substances


merged = merge_substances(grouped, {"CO2": ["CO2", "CO2bio"]})
merged.total_emissions.T

[11]:

	CH4	CO2
agriculture	2.274178e+08	1.771538e+10
industry	1.683938e+09	3.304992e+11
livestock	2.403860e+09	NaN
waste	4.253374e+08	1.629228e+07
transportation	8.589536e+06	1.143563e+11
__total__	4.749143e+09	4.625872e+11