Hybrid EXIOBASE walkthrough#

This notebook is the practical walkthrough for parsing the hybrid EXIOBASE bundle in MARIO.

It focuses on:

  • hybrid HSUT;

  • hybrid HIOT;

  • extension selection.

What this notebook covers#

  • which parser entry points to use for hybrid EXIOBASE;

  • what local bundle MARIO expects;

  • how to parse the hybrid SUT;

  • how to parse the hybrid IOT;

  • how to limit the parsed extensions when needed.

Relevant Zenodo releases#

  • v3.3.18

  • newer consequential release not yet covered by the current MARIO hybrid parser:

Important limitation: the current hybrid parser targets the 3.3.18 bundle. It does not yet include the later consequential developments released separately on Zenodo.

Main entry points#

For normal user workflows, the public entry points are:

  • mario.parse_exiobase(...)

Key arguments and local layout#

The hybrid workflow uses the same mario.parse_exiobase(...) entry point with these main selectors:

  • table: choose "IOT" or "SUT";

  • unit: use "Hybrid";

  • path: extracted 3.3.18 release root;

  • extensions: None, [], "all", or one explicit list of extension groups.

Typical extracted layout:

EXIOBASE/3.3.18/
├── MR_HSUT_*.txt
├── MR_HIOT_*.txt
├── Land_*.txt
├── Emiss_*.txt
├── waste_sup_*.txt
└── waste_use_*.txt

Supported extension groups are resource, Land, Emiss, Emis_unreg_w, waste_sup, waste_use, pack_sup_waste, pack_use_waste, mach_sup_waste, mach_use_waste, stock_addition, and crop_res. Unreg_w is available for HSUT only.

[1]:

import mario

Optional download step#

If you do not already have the bundle locally, MARIO exposes a helper for the hybrid release it supports.

[ ]:

info = mario.download_hybrid_exiobase(
    path="/path/to/3.3.18",
    )
info

Parse the hybrid SUT#

This is the standard starting point for the hybrid supply-use table.

[ ]:

db_sut = mario.parse_exiobase(
    path="/path/to/3.3.18",
    table="SUT",
    unit="Hybrid",
    extensions="all",
)

INFO Parser: reading EXIOBASE hybrid HSUT from /Users/lorenzorinaldi/Library/CloudStorage/OneDrive-SharedLibraries-PolitecnicodiMilano/DENG-SESAM - Documenti/c-Research/a-Datasets/_Input Output Databases/EXIOBASE/3.3.18.
INFO Parser: EXIOBASE hybrid HSUT parsed with 164 activities, 200 commodities, 7 factor rows and 350 extension rows.
INFO Metadata: initialized.

Parse the hybrid IOT#

Use this when you need the hybrid input-output representation instead of the HSUT.

[4]:

db_iot = mario.parse_exiobase(
    path="/path/to/3.3.18",
    table="IOT",
    unit="Hybrid",
    extensions=["resource", "Emiss"],
)

INFO Parser: reading EXIOBASE hybrid HIOT from /Users/lorenzorinaldi/Library/CloudStorage/OneDrive-SharedLibraries-PolitecnicodiMilano/DENG-SESAM - Documenti/c-Research/a-Datasets/_Input Output Databases/EXIOBASE/3.3.18.
INFO Parser: EXIOBASE hybrid HIOT parsed with 164 sectors, 7 factor rows and 119 extension rows.
INFO Metadata: initialized.

Inspect the parsed database#

Once parsed, the result is a standard MARIO database. From here the normal exploration methods apply.

A simple way to inspect a parsed MARIO database is to evaluate the object name in a notebook cell (for example, db_iot or db_sut).
This prints a compact summary of its core properties.
[9]:

db_sut

[9]:

name = None
table = SUT
tech_assumption = industry-based
scenarios = ['baseline']
Activity = 164
Commodity = 200
Factor of production = 7
Satellite account = 350
Consumption category = 6
Region = 48

When dealing with hybrid tables, it’s fundamental to have clear view of the unit of measures. Units are stored in a dictionary, as shown below

[10]:

db_sut.units["Commodity"].head()

[10]:
unit
Paddy rice tonnes
Wheat tonnes
Cereal grains nec tonnes
Vegetables; fruit; nuts tonnes
Oil seeds tonnes

The same exploration, can be done for IOTs

[11]:

db_iot.units["Satellite account"]

[11]:
unit
Employment: Low-skilled male 1000 persons
Employment: Low-skilled female 1000 persons
Employment: Medium-skilled male 1000 persons
Employment: Medium-skilled female 1000 persons
Employment: High-skilled male 1000 persons
... ...
Ni (soil - Emiss) tonnes
C (air - Emiss) tonnes
other emissions (undef - Emiss) tonnes
Carbon dioxide, biogenic (air - Emiss) tonnes
Other emissions nec (air - Emiss) tonnes

119 rows × 1 columns

Notes#

  • hybrid parsing is more dataset-specific than the monetary case, so the direct hybrid entry points are often acceptable in documentation examples;

  • keep extension selection explicit if you want lighter parse results;

  • use the standard MARIO inspection workflow after parsing, exactly as for other databases;

  • if you need the newer consequential developments, the current hybrid parser is not there yet.

Download this notebook