Modelling exercises such as those presented in this document have a number of limitations that derive from the simplifications required and the uncertainty inherent in forward-looking initiatives. Dealing with these limitations requires careful assumptions; beyond some of the main assumptions presented in section 1.2 above, a few key points are set out below.
In the 2018 edition of this Outlook, we covered only energy-related emissions, in effect excluding other important sources such as industrial processes and agriculture (outside of fuel consumption). While we treated this limitation with care in the previous edition, the inclusion of—and indeed, our main focus on—NZ scenarios forces us to broaden the coverage, especially given that these sources present significant challenges and that fewer technological options are available at the moment. As a result, even though our main focus remains the transformation of the Canadian energy system, emission coverage goes beyond energy. It should also be noted that we now include fugitive emissions from the energy production sector, which is another addition to the model compared with its 2018 version.
Some GHG emissions are still missing from the model: emissions from land-use, land-use changes and forestry (LULUCF) are not covered, although they are partly touched on through the use of “negative emissions”, a concept used to describe the use of biomass coupled with carbon capture. Emissions from international aviation and marine bunkers are also excluded from the model. Moreover, ancillary costs for electricity grids with a very high level of variable generation (such as from wind and solar installations) are simplified through the use of a cost premium on these as well as on storage technologies.
The broadening of our emission coverage does not eliminate uncertainty about the likelihood of the advent of disruptive technologies that could be game changers in some sectors, affecting the pace of some of the results. This uncertainty is typical in this kind of modelling and must certainly be kept in mind when interpreting the results. Since technology pathways with a high degree of uncertainty, such as hydrogen applications or certain niche technologies, are difficult to model by nature, results are necessarily uncertain.
Focusing on energy issues, CEO2021 has also left aside the important issue of adaptation to climate change that will affect energy consumption/production and the choice of investments in infrastructure. Certainly, the energy transition is as much about technological and economic development as it is about reducing the risks and costs associated with accelerated climate change caused by rising GHG levels in the atmosphere.
Finally, it must be noted that, to a certain extent, our discussion downplays the issue of displaced emissions. Not all technologies required for the extensive transformation of energy services following the different scenarios will be produced in Canada, and, following the Paris Agreement, we do not evaluate the impact that this shift will have on global GHG emissions. Another example is emissions from oil refineries, which the optimization of the model may reduce by importing refined products instead, even though these emissions will still reach the atmosphere. Although we do not take this issue lightly, this shortcoming is inevitable given our Canadian focus, as well as beyond the scope of our analysis—like many others with a national focus.
Despite these caveats, modelling allows for the identification of general trends, which we believe to be fundamental in setting the bases for a discussion of net-zero pathways for the Canadian energy system. We return to these issues in Chapter 15 in light of our results.