[Spoiler alert: I’m not actually going to answer the title question above — not quantitatively at least. But I think there is some merit even just in drawing attention to the fact that there is a question here that can usefully be asked!]

Early in the excellent book Our Renewable Future by Richard Heinberg and David Fridley (full text available online, open access), they present the following “big picture” perspective on the challenge of rapidly decarbonising our energy systems (locally and globally):

[Original image source: Michael Carbajales-Dale, “Fueling the Energy Transition: The Net Energy Perspective,” presentation at the Global Climate and Energy Project Workshop on Net Energy Analysis, Stanford University, Stanford, CA, April 1, 2015.]

Notwithstanding the notional quantitative units on the two axes, the graph is, of course, purely qualitative, but it does still convey an important heuristic insight: building energy infrastructure itself takes energy, and we only get a net energy (or energy service) “surplus” if or when that initial investment is “paid back”.

Unpacking this slightly, given that the technologies of a decarbonised energy system (energy sources, conversions, end uses) are, in significant measure, very different from those of our existing fossil fuel based systems, an awful lot of physical infrastructure will have to be replaced during the decarbonisation transition. And physical infrastructure all involves so-called embodied energy — as opposed to the operational energy that we are more generally familiar with:

• Operational energy is energy used directly to provide some function or service. Thus the heat cooking food in an oven, the electricity lighting a bulb, the liquid fuel burned to produce thrust in an aviation engine are all examples of operational energy.

• By contrast, the energy required to manufacture an oven or a light bulb or an aviation engine (or complete aircraft) is referred to as embodied energy.

Quantification of embodied energy can be contentious as it depends on choices of exact system boundaries: but in general it is intended to include all upstream operational energy required to create the target product or device. The concept of embodied energy applies equally to infrastructure, such as roads, railways, buildings, and energy system components including power stations, electricity grids, oil tankers, pipelines, vehicles, aircraft, refuelling networks etc. In the specific case of renewable energy systems, renewable source technologies such as wind turbines, solar panels, hydroelectric stations, etc. all have associated embodied energy. Similarly, to the extent that use of low-carbon or renewable energy sources may require progressive electrification of heating and transport, then new battery electric vehicles or heat-pumps, or additional grid infrastructure, all with associated embodied energy may be required. And in complementary manner, even interventions to enhance energy efficiency, such as in retrofit and renewal of the built environment, may involve associated embodied energy. (And in relation to energy efficiency, we will not even start on the complicating issues of rebound here…)

Given the current dominance of fossil fuel based infrastructure in energy systems (at global, regional, or national levels) it follows that the decarbonisation transition itself will involve a great deal of embodied energy. While some of that might substitute for energy that would otherwise be embodied in maintenance or refurbishment of the existing energy system, a great deal of it is likely to be additional i.e., it would not be required but for the decarbonisation of the system. At any given point in the decarbonisation of renewable energy transition, it will be the energy system of that time, and the carbon intensity of energy at that time, that determines the emissions associated with that embodied energy. Toward the end of the transition, that intensity will have become low (by definition); but early on, it will still be very high.

In effect then, decarbonisation itself involves a large scale commitment to additional CO₂ emissions, over and above emissions from all “ongoing” human activities (not directly associated with the decarbonisation transition).

As CO₂ is effectively cumulative, any given climate change temperature constraint (such as now embodied in the Paris Agreement) implies a limit on the total further amount of  CO₂ that can be released to the atmosphere, the so-called remaining global carbon budget. But the analysis above indicates that a potentially substantial component of this must be ring-fenced for emissions associated with the embodied energy essential to the construction of the new, decarbonised, energy system.

While precise quantification of this commitment (of embodied energy and associated emissions) is difficult, it strongly suggests a need to absolutely minimise, as far as possible (technically, socially, politically), the amount of operational energy used for any purposes other than for the energy system transition, until that transition is substantially achieved.

As far as I know, this heuristic system-level insight has not, as yet, been incorporated in any explicit way in practical energy transition policy (certainly not in Ireland?)... but comments/pointers are definitely welcome via our project twitter feed @ie_nets or direct email to ienets@dcu.ie.