Diversity of options to eliminate fossil fuels and reach carbon neutrality across the entire European energy system

What is redundant and what is not? Computational trade-offs in modelling to generate alternatives for energy infrastructure deployment

“…The model is based on the well-established open-source energy system modelling framework Calliope [26] and takes the name of Euro-Calliope. Originally conceived by Tröndle et al [27], the version we use here is based on a previous work of ours [16] in which we updated the initial (brown eld) electricity grid topology to mirror the most relevant real-world transmission constraints as identi ed by the e-Highway 2050 project. Besides various hydroelectric technologies whose capacity is assumed to remain constant, the model features seven main technologies for capacity expansion: roof-mounted and open-eld solar photovoltaic, on-shore and off-shore wind, bioenergy power plants, and battery and hydrogen-to-power storage.…”

“…These add up to the 50 alternatives generated within the main batch, leading to an overall 210 SPORES. For all SPORES, we adopt a slack cost of 10%, in line with previous work [20,16,19]. We refer the readers to the Supplementary Methods for further details about the model and how it has been customised for this work.…”

“…It also allows readers to compare our ndings with those of other recent MGA studies applied to energy system models of Europe [19,22,21], which are all grounded on a similar model set up. What is more, the analysis of the available decision space for the full decarbonisation of Europe including all energy-consuming sectors has already been undertaken using Euro-Calliope [16] and would therefore be redundant to repeat here.…”

“…This difference can have very concrete implications. In line with previous work [20,16], let us assume that stakeholders may be interested in ltering out the decision space based on multiple criteria at once. For instance, minimising the deployment of bioenergy power plants while simultaneously reducing on-shore wind capacity concentration (with the same threshold as de ned in subsection 3.2).…”

“…Yet, MGA displays limitations when applied to suf ciently large models, such as state-of-the-art energy system optimisation models covering the whole of Europe at high temporal and spatial resolution and with many technological options [16,17]. The number of meaningful alternatives that can be generated for such models is de facto in nite, and the conventional form of applying MGA often fails to represent diversity well, for two reasons.…”