Change energy-carrier structure for heat generation fuel mix

[danielhuppmann]

While working on #33, I noticed that the variable hierarchy for heat generation in the ENGAGE and NAVIGATE templates does not account for any "green" fuels like biofuels, biomethane or hydrogen.

The current structure (omitting the CCS-subcategories) is:

• Secondary Energy|Heat|Biomass
• Secondary Energy|Heat|Electricity
• Secondary Energy|Heat|Fossil
• Secondary Energy|Heat|Coal
• Secondary Energy|Heat|Gas
• Secondary Energy|Heat|Oil
• Secondary Energy|Heat|Geothermal
• Secondary Energy|Heat|Nuclear
• Secondary Energy|Heat|Solar
• Secondary Energy|Heat|Other

where for "Gas" is not clear whether this only refers to fossil methane. (We usually use "Gases" for the aggregate of fossil and renewable forms of methane).

I suggest to apply the "Final Energy"-convention to the heat generation fuel mix and use "Liquids", "Solids" and "Gases" instead of "Oil", "Coal" and "Gas" (with possible sub-categories to distinguish between fossil and renewable types of the fuel).

Any comments @IAMconsortium/common-definitions-coordination?

[Renato-Rodrigues]

I second the suggestion of using the final energy convention and I would like to also suggest adding a further split of all of those sub-categories, in final and secondary energy, into Fossil, Biomass and Electricity (e-fuels), as it is currently being done at the ECEMF template:

e.g.

Secondary Energy|Heat|Liquids|Fossil, Biomass or Electricity
Secondary Energy|Heat|Gases|Fossil, Biomass or Electricity
Secondary Energy|Heat|Solids|Fossil or Biomass
Secondary Energy|Heat|Electricity
Secondary Energy|Heat|Geothermal
Secondary Energy|Heat|Nuclear
Secondary Energy|Heat|Solar
Secondary Energy|Heat|Other

[danielhuppmann]

In this variable template so far, I have not included that type of information. Copying (and slightly adapting) a blurb from my email to the PRISMA consortium on August 31, 2023:

1. Most final-energy variables (in ENGAGE, NAVIGATE and other projects) have a split by sector, secondary energy carrier, and primary energy from which that secondary energy carrier was produced, e.g., "Final Energy|Industry|Gases|Biomass“. However, in practice, the source of the secondary-energy carrier cannot be known exactly after the fuel is blended into the transmission system, e.g., methane molecules in a pipeline. Including all these combinations leads to an enormous number of variables that only offer spurious precision. So I have not included this level of detail - instead, users should look at the Secondary-Energy-variables for information on the source-mix of a secondary-energy-level fuel.

[Renato-Rodrigues]

I understand the logic, but I am not sure I agree with it.

Let's assume that you have an industrial park in some specific country in which biomass gases are directed directly to the chemicals industry.
If we do not have the variable Final Energy|Industry|Gases|Biomass available in the report, the most anyone could do is to assume that industry and all other sectors receive the same proportion of bio-gases from the total Secondary Energy|Gases|Biomass supply, which would be wrong.

This hypothetical case is not so hypothetical if we consider that most secondary energy infrastructure in the World is geographically segregated, with congestion points and/or insufficient interconnections which create independent systems even within the same country. This is even worst if we assume that bulk energy imports are usually directed to specific sectors and activities, and not simply injected to a economy wide secondary energy grid. Just as an example, bio-liquids in Brazil are primarily used in the transport sector due to legislation reasons and market maturity reasons.

I understand that this repository aims to set primary common variables, but I think that defining a common structure to represent this level of detail is not detrimental to this goal, and leaving this undefined can create more divergences between project results and variables mappings in the future.

Models that can represent sectoral differentiation of biomass use for example would be able to provide this additional information if these variables are included, meanwhile models that can't could use the secondary energy approximation if necessary for any analysis.

[danielhuppmann]

Ok, this discussion goes beyond the scope of this PR (which only changes one aspect of the heat supply. Let's continue this conversation in #39 and discuss whether to add fuel-source-sub-categories for secondary- and final-energy.

[strefler]

"Gas" is always fossil and always refers to the primary energy level. On the secondary energy level, gases are called "gases".
I don't really understand why you think that biofuels or biomethane are not covered, as there is Secondary Energy|Heat|Biomass? The thought behind this structure is to always distinguish Biomass, Fossil, and Hydrogen / Synthetic fuels summarized as "Electricity" on the secondary energy level. The name "Electricity" should probably be changed, but the level of disaggregation seems useful to me. However, Heat can be generated from PE and from SE, and therefore reflects both disaggregations.

[danielhuppmann]

Heat can be generated from PE and from SE, and therefore reflects both disaggregations.

It doesn't reflect both disaggregations, NAVIGATE/ENGAGE only covers primary energy carriers. Electricity was added when developing this variable template.

I guess the perspective hinges on the relevant research question:

1. what was the original source of the energy (which fossil fuel or nuclear or anything green)
2. what kind of heat technology was used (burning a liquid or a gas or something else)

Both perspectives have their merits in a "Secondary Energy" view. Perspective 1 is used in other secondary-energy variables (like production of liquid fuels), Perspective 2 is used in the final-energy structure.

But only perspective 2 makes sense if you look at capacity variables (which was the starting point of this issue). And "Capacity|Heat|..." did not exist in ENGAGE/NAVIGATE...

[danielhuppmann]

Just talked with @volker-krey, he is also in favor of keeping perspective 1 for Secondary-Energy variables for legacy-consistency. So #41 adds a clarification per the comment by @strefler above about biogas being included in biomass.