Modelling the EU’s Long-Term Strategy towards a carbon-neutral energy system

The COP21 UNFCCC conference in Paris in December 2015 flagged a new era for energy and climate policy. Climate change mitigation turned from being the wish of a few, to the reality of almost all nations around the globe. The signed agreement has invited all parties to submit, by 2020, mid-century low-emission strategies compatible with the goal of limiting the rise in average global temperatures to well below 2^oC above pre-industrial levels and pursuing efforts to limit it to 1.5^oC.

The aim of a low carbon economy has been on the EU policy agenda since the release of its “Low carbon economy roadmap” in 2011 that introduced an 80% GHG emission reduction target for the year 2050 relative to 1990 levels. However, pursuing the 1.5^oC temperature increase limit requires boosting even more the ambitious climate target and aiming at a carbon-neutral economy by mid-century. Given the new climate policy regime, the discussions around low, zero, or even negative carbon policy options have been intensified in the years following the Paris milestone; new quantitative analysis, in the form of detailed policy scenarios, strategies and quantitative pathways, is necessary to assist EU policy makers in assessing the available options from both a technological and economic perspective, while also considering societal and governance issues. Many research discussions, debates and synergies in the modelling communities, have been triggered after the COP21 conference, as new innovative concepts need to be introduced and enrich existing modelling frameworks, in order for the latter to be able to perform the appropriate deep decarbonisation scenarios and provide sound input to impact assessment studies.

Certain key policy elements and technologies are considered pillars of the low-carbon transition and are treated as “no-regret” options in all recent EU climate and energy policy discussions. Such policy options are the strong electrification of final energy demand sectors, accelerated energy efficiency mainly via the renovation of the existing buildings’ stock, advanced appliances, heat recovery in the industrial sector and intelligent transportation, and the strong push of variable renewable energy sources (RES) in the power system. From a modelling point of view, their assessment is well established in the literature, and does not pose any considerable, unpresented difficulties. However, the modelling of disruptive technologies and policy instruments that could be proven essential for the transition towards a carbon-neutral EU economy (a case that requires GHG emission reductions beyond the 80% target) is far from being considered as mature in the existing literature or academic research. Modelling a power sector with renewables above 80% is a challenge as it requires representing variability in some detail along with the various balancing resources including cross-border trade. Modelling strong energy efficiency in buildings implies representing the decision of individuals about renovating the buildings deeply; this is also a challenging modelling task give the large variety of building cases and idiosyncrasies of the individuals in decision making. Electrifying heat and mobility in market segments where this is cost-efficient is among the no-regrets option. Modelling the pace of transition and the role of policy drivers from internal combustion engines to electric cars and from boilers to heat pumps are also challenging tasks due to the large heterogeneity of circumstances that the modelling will have to capture. So, the amplitude of the three main no-regrets options is challenging the modelling despite the significant experience accumulated so far. 

The carbon-neutrality target poses considerable additional challenges for the modelling. The no-regrets options are not sufficient to deliver carbon-neutrality by mid-century. Mobility, heating, high-temperature industrial uses and the industrial processes would emit significant amounts of CO2 by 2050 if conventional wisdom policies and measures only apply. To abate the remaining emissions in these sectors, disruptive changes are necessary, regarding the origin of primary energy and the way energy is used and distributed. The disruptive options are surrounded by high uncertainty due to the low technology readiness levels of the technologies involved. The disruptive options are antagonistic to each other because they require large funding resources to achieve industrial maturity and economies of scale of the yet immature technologies. Such concentration of resources requires long-term visibility for the investors and infrastructure investment, which both require clear strategic choices among the disruptive options.

We consider the “disruptive” options grouped in stylised categories, as shown below. The modelling has to include assumptions regarding the future evolution of costs and performance of a plethora of technologies and options for alternative sets of disruptive changes had to represent. Each stylized category of disruptive changes has its own challenges in terms of modelling considerations, as presented below:

1. Extreme Efficiency and Circularity: The aim of the options included in this category is to push energy efficiency savings close to its maximum potential, introducing circularity aspects and further improving material efficiency in the EU economy, increasing the intelligence of the transport system, sharing of vehicles, achieving near-zero energy building stock, etc. Even though these concepts are present in the literature, introducing them in a large-scale applied modelling framework poses significant difficulties as, for instance, estimating the maximum industrial output reductions that can be realised at a sectoral level. It also required estimating the upper boundaries of the impact of behavioural and restructuring changes in the transport sector in reducing transport activity. Similarly, modelling a near-zero energy building stock involves great difficulties regarding the driving policies and the idiosyncratic behaviours of individuals. Almost all options included in this category are associated with non-linearly increasing costs, beyond a certain level of deployment that needs to be captured in the quantitative assessment. Some of the options might also create so-called “disutility” to the consumers, as they alter their behaviour to an eventually less “comfortable” patterns. Not capturing such effects, would underestimate the difficulty in the introduction of such policy options.

1. Extreme Electrification: Strong electrification is a “no-regret” option, but extreme electrification is a relative newly established concept. This category adopts electricity as the single energy vector in all sectors in the long-term, with bioenergy complementing electricity only in sectors where full electrification is not technically feasible with currently known technologies, such as aviation and maritime. From a modelling perspective, deep electrification requires to studying the technoeconomic of innovative technological options such as full-electric long-distance road freight vehicles, electric aircrafts for short-haul flights and high temperature heat pumps. Using electricity as the only vector for the heating of buildings will broaden the seasonal demand gap between the summer and winter seasons for many regions, requiring either significant investments in power storage solutions or in power capacity that will have low utilisation. Capturing this in the modelling requires establishing electricity load patterns that differ by scenario at an appropriate time resolution.

1. Hydrogen as a carrier: This category assumes that hydrogen production and distribution infrastructure will develop allowing hydrogen to become a universal energy commodity, covering all end-uses including transport and high-temperature industrial uses. Hydrogen can also provide a versatile electricity storage service with daily up to seasonal storage cycles. Hydrogen is assumed to replace distributed gas after an extensive overhaul of the pipeline system and gas storage facilities. From a modelling perspective it requires identifying the industrial processes that can be decarbonised using hydrogen-based solutions (and the relative boundary conditions), assessing the techno-economics of a variety of carbon-free hydrogen production facilities (for both blue and green hydrogen) and the costs associated with the upgrade of the gas distribution and storage system. All the above, should be established in a modelling framework that is available to co-optimise the operation of the power system and the hydrogen production facilities, enhancing in this way energy storage and coupling various sectors of the energy system and the economy.

1. GHG-neutral hydrocarbons (liquid and gaseous): In this case, the paradigm of using and distributing energy commodities, along with the respective infrastructure, is maintained. The nature and origin of the hydrocarbon molecules is modified in order to ensure carbon neutrality from a lifecycle emissions perspective, using synthetic molecules rather than fossil ones. The production of synthetic methane and liquid fuels requires carbon-free hydrogen production and an appropriate carbon dioxide (CO₂) feedstock source. The last element hints at the emergence of CO₂ as a commodity, therefore it is important that energy-economy models are modified in order to treat CO₂ not only as a by-product of fossil fuel combustion, as it was done till recently, but as a product that can be used for different purposes (e.g. apart from producing GHG-neutral fuels, creating carbon sinks and inducing net negative emissions via embodying it in materials or storing it underground). The origin of CO₂ can either “from air” via using Direct Air Capture (DAC) technologies or by biogenic sources; both of them have uncertainties related to its learning potential and maximum availability potential, respectively. For the production of hydrogen, similar modelling considerations as in the previous case apply. A model able to represent effectively this strategy should have a rich technology database that explicitly includes the major pathways for the production of synthetic fuels; in turn populating such a database is a difficult and time-consuming exercise given the uncertainty regarding the learning potential of the various technologies. Ideally, a model should be in a position to optimise of the location of clean-fuel production facilities in Europe or elsewhere, as it is more likely that such commodities will be traded extensively.

The analytical assessment, which has provided input to the “Clean Planet for All” strategy by the European Commission in November 2018, has confirmed that a carbon-neutral EU economy by mid-century (2050) is viable both from a technological but also an economic perspective, should a number of key technologies evolve as anticipated. The analysis should be perceived as the first step of a complex assessment procedure to bolster the decision-making process regarding the definition of the EU’s long term energy and climate strategy.

Next steps should focus on the assessment of several uncertainties associated with the various pathways studied. For instance, emphasis should be given to identifying the appropriate policy instruments that could be used to materialise the emergence of technologies and energy carriers as long-term visibility of future markets is crucial for their deployment. The characteristics and costs of several disruptive technologies are also worthy of further research with emphasis on the potential of learning and economies of scale. The modelling and data improvements to be realised in INNOPATHS will enhance the model representation of disruptive options and sectoral transformation, and will enable further improvement of the design of deep decarbonisation pathways.

For more information on the analytical work behind the “Clean Planet for All” communications, please click here.