Evaluation of wind power forecast uncertainty at German transmission grid buses: spatial patterns and ensemble spread reliability

The strong increase of renewable energy share and their weather dependence introduce forecast uncertainty as a key challenge for grid operators. Knowing the forecast reliability at the relevant aggregation level for operational grid planning is essential. As a key methodological contribution, this paper investigates forecast uncertainty between reanalysis and day-ahead forecasts. Within this analysis, we investigate spatial variations in the reliability of wind power forecast uncertainty at the German transmission grid bus aggregation level and compare two weather ensemble forecast models. The results indicate distinct spatial patterns: Overdispersion of the forecasts in Germany’s lowland region and good dispersion agreement - with a slight tendency to underestimate the upper bound of extreme values - in southern more hilly and mountainous regions for the ensemble model of the European Centre for Medium-Range Weather Forecast (IFS-EPS) and an overall overdispersion for the European-nested ensemble model of the German Weather Service (ICON-EU EPS). The findings highlight the importance of region-specific forecast calibrations, particularly in areas with high wind capacity that significantly influence grid calculations.