Search

To reach climate mitigation goals, the share of wind power in the electricity production is going to increase substantially in France. In winter, low wind days are challenging for the electricity system if compounded with cold days that are associated with peak electricity demand. The scope of this study is to characterize the evolution of compound low wind and cold events in winter over the 1950–2022 period in France. Compound events are identified at the daily scale using a bottom-up approach based on two indices that are relevant to the French energy sector, derived from temperature and wind observations. The frequency of compound events shows high interannual variability, with some winters having no event and others having up to 13, and a decrease over the 1950–2022 period. Based on a k-means unsupervised classification technique, four weather types are identified, highlighting the diversity of synoptic situations leading to the occurrence of compound events. The weather type associated with the highest frequency of compound events presents pronounced positive sea-level pressure anomalies over Iceland and negative anomalies west of Portugal, limiting the entrance of the westerlies and inducing a north-easterly flow bringing cold air over France and Europe generally. We further show that the atmospheric circulation and its internal variability are likely to play a role in the observed reduction in cold days, suggesting that this negative trend may not be entirely be driven by anthropogenic forcings. Despite this suggested role for cold days, the observed decrease in compound events does not seem to be strongly influenced by the regional atmospheric circulation. [ABSTRACT FROM AUTHOR]

As variable renewable energies are developing, their impacts on the electric system are growing. To anticipate these impacts, prospective studies may use wind power production simulations in the form of 1 h or 30 min time series that are often based on reanalysis wind-speed data. The purpose of this study is to assess how several wind-speed datasets are performing when used to simulate wind-power production at the local scale, when no observation is available to use bias correction methods. The study evaluates two global reanalysis (MERRA-2 from NASA and ERA5 from ECMWF), two high-resolution models (COSMO-REA6 reanalysis from DWD, AROME NWP model from Météo-France) and the New European Wind Atlas mesoscale data. The study is conducted over continental France. In a first part, wind-speed measurements (between 55 and 100 m above ground) at eight locations are directly compared to modelled wind speeds. In a second part, 30 min wind-power productions are simulated for every wind farm in France and compared to two open datasets of observed production published by the distribution and transmission system operators, either at the local scale in terms of annual bias, or aggregated at the regional scale, in terms of bias, correlations and diurnal cycles. ERA5 is very skilled, despite its low resolution compared to the regional models, but it underestimates wind speeds, especially in mountainous areas. AROME and COSMO-REA6 have better skills in complex areas and have generally low biases. MERRA-2 and NEWA have large biases and overestimate wind speeds especially at night. Several problems affecting diurnal cycles are detected in ERA5 and COSMO-REA6. [ABSTRACT FROM AUTHOR]