Download notebook

Choose the release, year and native variant together. The E variant needs its own input and variant selector; source CO2eq totals retain their original characterization.

Full-database calculations: install the optional UMFPACK backend. See installation and compute settings. The parser selects solve/direct, with a SciPy fallback when UMFPACK is unavailable.

Import#

import mario

db = mario.parse_emerging(
    path="/path/to/EMERGING_V2_2023_m.mat", table="IOT",
)

Result · A MARIO database. Inspect db, db.get_index("all") and db.units before selecting quantities for your analysis.

Inputs and options#

Selecting a release

Use variant="standard" for numbered releases and variant="E" for EMERGING-E. Point path= to the folder for the chosen version; the database profile above lists the verified years.

Release records: 2.2, 2.0, 1.0 and EMERGING-E.

The local 2.0 collection is incomplete. A companion CO2 file does not add an economic year without its matching MRIO bundle.

For MATLAB v7.3 files, MARIO normalizes the stored axes while the economic matrix is sparse. No manual transpose of Z is needed.

Key arguments

The key public arguments are:

  • path: one EMERGING .mat file or a directory containing multiple yearly bundles;

  • table: currently only "IOT" is supported;

  • year: use it when one directory contains more than one EMERGING year;

  • regions: optional ISO3 subset to keep only one manageable part of the database;

  • load_co2: enable or disable automatic companion CO2 import;

  • co2_path: explicit path to one companion CO2 file when auto-detection is not enough;

  • labels_path: optional Excel workbook with explicit sector labels; for variant="E" it overrides workbook auto-detection next to the bundle;

  • variant: keep "standard" for the original EMERGING releases and use "E" for the EMERGING-E power-disaggregated bundle.

Local file naming conventions

In practice MARIO accepts these local naming conventions:

  • global_mrio_<year>.mat for v1.0;

  • EMERGING_V2_<year>_m.mat for v2.x;

  • EMERGING_V2_<year>.mat for older local v2.x copies when the internal MATLAB structure is still compatible;

  • EMERGING_E_<year>.mat for the EMERGING-E power-disaggregated variant.

For local v2.x files, MARIO does not try to infer the exact sub-version 2.0 versus 2.1 versus 2.2 from the filename alone, because the public naming convention is shared across those releases.

Expected path structure

path can point either to one yearly .mat file or to a directory containing the yearly EMERGING bundle and, optionally, the companion CO2 file.

Typical directory layout:

EMERGING/2.2/
├── EMERGING_V2_2023_m.mat
└── EMERGING_CO2_2023.mat

For the older v1.0 naming convention, the economic bundle is typically named like global_mrio_<year>.mat. For EMERGING-E, the local bundle is typically named like EMERGING_E_<year>.mat and may ship a sibling Figure data.xlsx workbook that MARIO uses to recover ISO3 region codes.

When path is a directory, use year= to select the economic bundle. With load_co2=True, MARIO first looks for the matching CO2 file in the same directory, for example EMERGING_CO2_2023.mat for year=2023. Use co2_path= when the companion file is stored elsewhere or has a non-standard name.

Download workflow

Use mario.download_emerging(...) when you want MARIO to fetch one of the supported official version records, including the newer version="E" EMERGING-E record.

latest currently resolves to v2.2.

import mario
Download one official version explicitly
mario.download_emerging(
    path="/path/to/2.2",
    version="2.2",
    years=[2023],
)
Download the EMERGING-E record

Use version="E" when you want MARIO to fetch the public EMERGING-E Zenodo record. With years=[2018], MARIO keeps the main .mat file and the companion Figure data.xlsx workbook used to recover ISO3 region codes.

mario.download_emerging(
    path="/path/to/emerging-e",
    version="E",
    years=[2018],
)

Also for EMERGING-E, you can use the co2_path argument, passing the path of the CO2 emissions files from standard EMERGING releases.

Parse one local EMERGING bundle
db = mario.parse_emerging(
    path="/path/to/EMERGING_V2_2023_m.mat",
    table="IOT",
)

db
Parse one local EMERGING-E bundle

EMERGING-E is currently a local-file workflow. Use variant="E" and point path to EMERGING_E_<year>.mat.

db_e = mario.parse_emerging(
    path="/path/to/EMERGING_E_2018.mat",
    table="IOT",
    variant="E"
)

db_e
Parse from a directory and select one year
db = mario.parse_emerging(
    path="/path/to/2.2",
    table="IOT",
    year=2023,
)

db
Restrict the region set

The full EMERGING matrix is very large, so regions= is often the right first step.

db = mario.parse_emerging(
    path="/path/to/2.2",
    table="IOT",
    year=2023,
    regions=["DEU", "FRA", "ITA"],
)

db
Control CO2 loading explicitly
db = mario.parse_emerging(
    path="/path/to/2.2",
    table="IOT",
    year=2023,
    co2_path="/path/to/EMERGING_CO2_2023.mat",
)

db
Reading the economic and emissions accounts

V contains the source value-added row, in current million US dollars. Sum its columns for a region or the whole model, and multiply by 1e6 when expressing the result in US dollars. The available bundles do not expose a separate, verified product-tax bridge, so the browser also presents the native va sum as a GDP proxy, without an invented tax adjustment. A note beneath the chart explains that equivalence to GDP at market prices is unverified.

The standard companion files contain seven fossil CO2 accounts, one per fuel. MARIO imports them into E, preserving the native unit and order. They are not a complete GHG inventory and do not supply direct final-demand emissions. The browser presents their sum as a CO₂-only proxy under GHG, with a visible caveat: CH₄, N₂O and F-gases are absent. The native Mt CO₂eq values are multiplied by 1e6 to obtain tonnes; CO₂ has GWP 1 under both AR5 and AR6. EMERGING-E in this collection has economic accounts only.

  • EMERGING currently supports IOT tables.

  • The containing release directory identifies 2.0 versus 2.2; their economic filenames are shared.

  • regions= can restrict the parse when the full model is unnecessary.

  • Use a same-year, same-release CO2 file. load_co2=False reads only the economic accounts.

  • ANT and SER retain historical grouping labels; TMP is the historical code mapped to TLS. Current memberships of the historical groups are not inferred.

db.table_type

db.get_index("Region")[:10]

db.get_index("Satellite account")

db.meta.source

Continue with Inspect and query a database or Calculate matrices.