Characterizing solar and wind energy droughts in Spain: Identification and weather patterns

In the near future, wind and solar generation are projected to play a key role in Europe's energy sector. An example is the Spanish roadmap to produce around 80% of its electricity from renewables by 2030, dominated by both sources.

 However, the transition to these envisioned renewable power systems poses a formidable challenge. For instance, with the rising share of variable renewable energy, the power system becomes increasingly vulnerable during periods of very low renewable energy production ("energy droughts") or very high load demand. Specific weather patterns may give rise to so-called compound extreme events, characterized by the simultaneous occurrence of several hazards, such as a scarcity of solar and/or wind resources combined with high load demand. The occurrence of these events and the response of the power system must be carefully evaluated to optimally define mitigation and adaptation strategies. Despite the importance of this issue, there is no consensus on the definition and identification procedures for extreme events.

In this work, the occurrence of solar PV, wind and compound solar PV and wind energy drought events in Spain is analyzed using a novel identification procedure, based on the SPA algorithm. This procedure accounts for the high seasonal variability of the renewable resources in the study region. The study is conducted on a daily basis, using SHIRENDA—an enhanced, open-access database of renewable energy resources in Spain. This database provides wind and solar PV capacity factors (CFs) for the Spanish NUTS3 regions, covering the 1990–2020 period. Special attention is paid to the analysis of the weather patterns associated with the identified events, which are obtained from the ERA5 reanalysis. Finally, this study is part of the WE-Resyst project, which aims to design a renewable energy power system for Spain that is resilient to weather and climate variability and change.

The results showed, firstly, that the proposed event identification method provides more events outside the winter season compared to other reference methods. Secondly, solar PV events showed a mean and maximum duration of 1.2 and 4 days, respectively, compared to 1.4 and 7 days for wind energy. Distinctive weather patterns were identified behind these events. Furthermore, the results showed that: 1) compound solar PV and wind events occur at specific dates, mostly unrelated to the dates of individual solar PV or wind events, and 2) the intensity of compound events, in terms of renewable generation deficit, is lower compared to single-resource events. These results are explained by the existence of a high temporal complementarity between wind and solar resources in the study region during the winter and summer seasons, although with significant interannual variability. As a general conclusion, the characteristics of the solar and wind resources in the study region make the Spanish power system less exposed than others to weather extremes.