Intraseasonal prediction of monthly storminess in the North Sea with the ACE2 atmospheric emulator and Random Forests

Abstract. This research explores the predictability of seasonal storminess in the North Sea using machine learning methods and the weather model emulator ACE2, focusing on how the stratosphere and upper troposphere influence winter storms. Understanding the drivers of winter storminess is essential for improving sub-seasonal prediction skill in regions strongly affected by extratropical cyclones. Using ERA5 reanalysis data (1940–2024), we built a storminess index based on storm event frequency, examined its relationship with large-scale atmospheric fields, and explored its predictability at seasonal timescales.

We aim to predict North Sea storminess using two approaches: one based on the ACE2 climate emulator and another on the Random Forest machine learning algorithm. For the ACE2 model, we used modified air temperature and zonal and meridional wind patterns at 70 hPa as predictors, which were imposed as initial conditions on the 1^st of each winter month, and tested changes in storminess over the ensuing weeks and months of the winter season. For the Random Forest regression model, we used monthly means of air temperature, zonal wind at 70 hPa, and geopotential height at 200 hPa as predictors to predict storminess in the following months. The ACE2 simulations show that by modifying the stratospheric initial conditions on 1^st December, we can increase the emulated mean January surface wind speeds by about 0.5–3 ms⁻¹ across much of the North Sea. Similar sensitivity emulations initialised at the start of other months, e.g. November 1^st, failed to produce a meaningful response in the ensuing month. This suggests a dynamic link between early-winter stratospheric conditions and increased mid-winter surface storminess.

The Random Forest Regression model was applied after dimensionality reduction using Principal Components Analysis (PCA). The best results were obtained by predicting January storminess from the mean December fields, yielding a correlation between prediction and target of approximately 0.55–0.60. For other month pairs, the correlation ranges from 0.20 to 0.36 for November–to-December and January–to-February predictions, but it becomes negative (–0.44 to –0.05) for October–to-November and February–to-March predictions.

This seasonal predictability pattern, derived from both the ACE2 emulations and the Random Forest model, follows the seasonal cycle of the polar vortex's average intensity. The circumpolar westerly jet strengthens from autumn and peaks in winter, when predictability is highest. This higher skill is likely linked to stronger stratosphere–troposphere coupling between November and January, as polar vortex anomalies develop and begin to descend toward the surface. Both indicate that the seasonal predictability of storminess peaks in mid-winter, with a predictability lead time of about 4 to 6 weeks. It fades for the earlier and late winter periods. 

Overall, this research shows that stratospheric conditions play an essential role in shaping North Sea winter storminess and that machine learning methods can improve sub-seasonal predictions in this region.