Abstract/Summary

The predictability of extratropical cyclone (ETC) extreme winds is limited by numerical weather prediction (NWP) model physics approximations and initial condition errors. The aim of this research is to improve our understanding of the role played by air-sea fluxes in controlling ETC extreme surface wind speeds, and how accurate modelling of air-sea fluxes can improve ETC forecasts. First, a climatology of observed offshore extreme wind speeds, gusts, and wave heights near the British Isles is created over the period 2012-2020, and extreme events are objectively attributed to ETC conveyor belt jets by an ad-hoc algorithm. The cold conveyor belt is associated with the most hazardous jet, with the largest number of compound wind and wave hazards attributed to it. This jet is also the most underestimated by the latest ECMWF reanalysis, ERA5. Next, the sensitivity of ETCs to air-sea fluxes is explored by running a convective-scale atmosphere-ocean-wave coupled NWP model developed by the Met Office for three ETCs crossing the British Isles. Coupling to waves reduces the ETC extreme surface wind speeds, implying young growing wind waves enhance the air-sea momentum flux by increasing the sea-surface aerodynamic roughness. Finally, the deterministic coupled system is integrated with the Met Office ensemble capabilities into the new Ensemble-RCS framework to assess the respective impacts on ETC extreme winds of coupling and ensemble perturbations. The impact of coupling to waves on ETC extreme wind speeds is at least comparable in size to that of initial condition and stochastic physics perturbations, and is consistent across the ensemble members. Overall, this research demonstrates that coupling to waves is a fundamental aspect of model uncertainty in NWP convective-scale forecasts of ETC extreme wind speeds and that NWPs need to take into account the effect of the dynamic sea state on air-sea fluxes to reduce the model biases.