OpenIO v2.9

Plastic waste account

Relying on the physical flow account for plastic material, openIO now covers generated plastic waste linked to different plastic purchases as well as the average end-of-life of these wastes (landfilled, incinerated, mismanaged, recycled or leaked into the environment). In addition, the impact on biodiversity of leaked plastic waste is characterized in a new impact category called "Physical effect on biota". We rely on the charcterization factors developed by the MariLCA team. More on their work here: https://doi.org/10.1016/j.jclepro.2023.138197

Disaggregated GHG

In the v2.8, I was not allowed to use the disaggregated GHG accounts from StatCan and so openIO operated with the aggregated version. Now I have the authorization to once again use the disaggergated accounts. As a result, the impact of climate change on human health and ecosystem quality can once again be quantified. Furthermore, the characterization factors from the IPCC AR6 report are now implemented.

Updated requirements

The Python packages used were beginning to be quite old. So I updated the code to work with newer versions of pandas and numpy. Check the new requirements.txt file

OpenIO v2.8

2020 Data

The 2.8 version includes the economic data for the year 2020. Data for emissions were also updated, namely, GHG emissions, energy use, water use, minerals extraction and NPRI emissions.

Aggregated GHG accounts

Per request from the physical flow account team of StatCan, the disaggregated GHG accounts are no longer provided via openIO-Canada as long as these disaggregated accounts are not on the StatCan website. As a result, in v2.8 openIO-Canada operates with aggregated GHG accounts that can be directly found on StatCan's website. What are the implications? Because aggregated accounts are directly available in kgCO2eq

1. we are forced to use the characterization factors used by the physical flow account team, which are 25 for methane and 298 for dinitrogen monoxide. This constitutes a slight underestimation as new values from the latest IPCC report give a 30kgCO2eq CF for methane and a 273kgCO2eq for dinitrogen monoxide.
2. we can no longer cover GTP100, climate change damage indicators (damage of climate change on Human Health and biodiversity) and marine acidification
   We will re-implement disaggregated GHG accounts as soon as they are publicly available on StatCan's website.

Water consumption

In previous versions only water use was covered by openIO-Canada. As of 2.8, openIO-Canada now covers water consumption, i.e., only the water that is actually consumed (a net value between extracted and released water). Now that water consumption is covered, it was also regionalized. Meaning that a cubic meter consumed in Quebec is not as impactant as a cubic meter consumed in Saskatchewan or in the Middle East.

OpenIO v2.7.1

Added value was not properly corrected with endogenization, resulting in a wrong normalization for matrices.

OpenIO v2.7.1 Zenodo sync

Just syncing with Zenodo for DOI.

OpenIO v2.7

Endogenization of capitals

The v2.7 introduces the endogenization of capitals feature. Capital formation (building, machinery, IT equipment, road infrastructure, etc.) are typically contained in the final demand matrix Y. They are thus not directly linked to the commodity/service sector consuming them. Endogenization does this work of linking so that, e.g., the different buildings are linked to sectors buying the buildings. More on endogenization in these articles:
https://doi.org/10.1021/acs.est.8b02791
https://doi.org/10.1111/jiec.12931
We used the updated work of Sodersten of capital endogenization for the exiobase database and endogenized capital formation for the openIO-Canada part.

SF6 / HFC / PFC emissions

While these emissions were already introduced in the v2.6, the data used in the 2.6 was lacking. It was relying on voluntary declarations of companies, which means it did not include all emissions occurring in Canada.
With the v2.7, we used the national inventory of Canada, which is available through the UNFCCC data to cover these GHGs, hence getting a more reliable cover of these emissions.

OpenIO v2.6

New emissions added

Emissions of SF6, as well as 12 HFCs and 7 PFCs were added to openIO-Canada. These emissions come from the Greenhouse Gas Reporting Program (GHGRP) of Canada (https://open.canada.ca/data/en/dataset/a8ba14b7-7f23-462a-bdbb-83b0ef629823). These emissions are therefore coming from declaration of facilities. However, not all facilities within Canada participate to that program. The coverage of these new GHG emissions is thus limited to the reporting facilities and their impact of Canada's carbon footprint is therefore underestimated.

Corrections for meat sector industry

Because the meat sector is aggregated both at the economic level (supply and use tables) and the emissions level (physical flow accounts), the inputs and emissions were allocated following an economic approach. It created inconsistencies in the portrait of the meat sectors in some provinces. For instance, Quebec produces primarily hogs and very little cattle. In the description of the beef products of Quebec, it was actually primarily composed of pig meat, which doe snot make sense. The inputs of the meat sectors were thus adjusted to ensure that all cattle are used for beef meat products, hogs are used for pig meat products, poultry are used for poultry products.

OpenIO v2.5

2019 data available

The data for the year 2019 is now available. This includes economic data, international trade data, exiobase data, GHG emissions data, energy use data. Water use and other pollutants data are not available for the year 2019 yet. The most recent year for these data are used.

Mineral extraction accounts

OpenIO-Canada now includes mineral extraction data. The data comes from the USGS statistics and covers 67 minerals for the Canadian economy. For international imports, we simply use the mineral extension from exiobase. Data is available for years 2014 to 2018. Data for the 2019 is not yet available.

IMPACT World+ update

OpenIO-Canada, by default, operates using the IMPACT World+ impact assessment methodology. The latter has recently been updated to the 2.0 version. This recent version was thus implemented with openIO-Canada (and with exiobase).

OpenIO v2.4

International imports updated

The origin of the imports was determined relying on the global market for the commodity, determined through Exiobase. For example, cars were mainly coming from China, because China is the biggest car producer.

Now, international trade data were added to determine the actual origin of imports. For example, now the biggest exporter of cars to Canada is now the US, as it should be.

Biogenic carbon estimated

The physical flow account used for GHG emissions follow the SEEA guidelines (System of Environmental Economic Accounting) which does not differentiate between fossil and biogenic carbon. To avoid overestimating, we rely on Exiobase biogenic carbon emission levels (which probably stem from the natioral inventory report instead of the physical flow accounts) to estimate the distribution between fossil and biogenic CO2 in each of the 492 sectors of openIO. One notable exception: the final demand, so far we cannot find external data to remove emissions from burning biomass at home or from cars operating with biofuels. Direct emissions from households are therefore most probably overestimated.

Better agriculture GHG emissions

The GHG physical flow accounts only provide an aggregated value for Crop and animal production. OpenIO in previous versions thus distributed GHG emissions to sub-sectors following an economic allocation. Now, it relies on Exiobase crop and animal production distribution between the three GHGs covered by openIO (CO2, CH4 and N2O).

Discontinued

With this release, we officially stop supporting the industry by industry format, as well as IOIC classifications outside of the "Detail level".

OpenIO v2.3

New:

• Provincial energy flow accounts implemented
• Supports 2018 economic data
• Final demand can be disaggregated to its full potential

OpenIO v2.2.2

Correction:

• Calculations without linking to EXIOBASE were not happening correctly due to a matrix from EXIOBASE being called when it should not.
• The rebalancing of international imports was occurring on positive and negative values of final demand sectors. Rebalancing on negative values does not make sense as negative values represent sells (mainly happening in fixed capital formation) and thus do not correspond to an import.