Improving Subseasonal Wind Speed Forecasts in Europe with a Nonlinear Model

Abstract Subseasonal wind speed forecasts provide valuable guidance for wind power system planning and operations, yet the forecast skills of surface winds decrease sharply after 2 weeks. However, large-scale variables exhibit greater predictability on this time scale. This study explores the potential of leveraging nonlinear relationships between the 500-hPa geopotential height (Z500) and surface wind speed to improve subseasonal wind speed forecast skills in Europe. Our proposed framework uses a multiple linear regression (MLR) or a convolutional neural network (CNN) to regress the surface wind speed from Z500. Evaluations on ERA5 reanalysis indicate that the CNN performs better due to its nonlinearity. Applying these models to subseasonal forecasts from the European Centre for Medium-Range Weather Forecasts, various verification metrics demonstrate the advantages of nonlinearity. Yet, this is partly explained by the fact that these statistical models are underdispersive since they explain only a fraction of the target variable variance. Introducing stochastic perturbations to represent the stochasticity of the unexplained part from the signal helps compensate for this issue. The results show that the perturbed CNN performs better than the perturbed MLR only in the first weeks, while the perturbed MLR’s performance converges toward that of the perturbed CNN after 2 weeks. The study finds that introducing stochastic perturbations can address the issue of insufficient spread in these statistical models, with improvements from the nonlinearity varying with the lead time of the forecasts.