Under the Magnifying Glass: multi-scale climate impacts on renewable energy supply and demand in Europe

The deployment of renewable energy technologies is accelerating globally as part of climate mitigation strategies, with capacities projected to triple by 2030 compared to 2023 levels. To be effective, this rapid expansion should be guided by long-term planning that accounts for a changing and varying climate. Such planning efforts are usually supported by energy system models, but these models often lack integration of future climate information. The SPEED2ZERO project addresses this gap by bridging energy and climate research to develop climate-resilient energy transition pathways for Switzerland. To achieve this, a Swiss electricity systems model with high spatial resolution is bounded by outputs of a coarser European model. This multi-scale approach requires coherent climate-informed inputs that align with both spatial resolutions.

To generate these multi-scale inputs, we develop climate-driven projections for hydropower, solar, and wind power, as well as heating and cooling demand for Switzerland and Europe. This comprehensive dataset covers the Representative Concentration Pathways (RCPs) 2.6 and 8.5. We leverage high-resolution regional climate model simulations from the EURO-CORDEX archive that include transient aerosols and bias-correct the relevant variables using CERRA and CERRA-Land reanalysis data. We additionally perform dedicated hydrological simulations for Switzerland using the PREVAH model and route the climate model runoff for Europe using the mRM model. In total, this study considers 3 climate model chains for RCP2.6 and 9 for RCP8.5. These chains consist of global climate model-regional climate model pairs spanning 1991 to 2100. The modeling pipeline employs open-source tools, such as pvlib and windpowerlib, along with technical specifications provided by energy system modelers, to convert the processed climate data into energy quantities. 

Results from the modelling pipeline indicate a consistent increase in PV potential over time, a trend mainly attributable to declining aerosols concentrations over most of Europe. In contrast, wind power and water availability for hydropower exhibit substantial spatial variability and model disagreement, including opposing wetting and drying trends across different climate model chains. As these climate signals are spatially heterogeneous, high resolution climate simulations are essential to explore the nuances that coarser datasets might overlook. This granularity allows, for example, the analysis of climate impacts at individual hydropower plants or reservoirs. First results for Switzerland indicate a decline in hydropower generation in the second half of the century, with nationwide trends not always coinciding with trends at single sites. These findings highlight the need to consider energy system planning both from a big picture and local perspective lens to ensure the system's resilience.