Compound wind and rainfall extremes: Drivers and future changes over the UK and Ireland

Abstract

The co-occurrence of wind and rainfall extremes can yield larger impacts than when either hazard occurs in isolation. This study assesses compound extremes produced by Extra-tropical cyclones (ETCs) during winter from two perspectives. Firstly, we assess ETCs with extreme footprints of wind and rainfall; footprint severity is measured using the wind severity index (WSI) and rain severity index (RSI) which account for the intensity, duration, and area of either hazard. Secondly, we assess local co-occurrences of 6-hourly wind and rainfall extremes within ETCs. We quantify the likelihood of compound extremes in these two perspectives and characterise a number of their drivers (jet stream, cyclone tracks, and fronts) in control (1981-2000) and future (2060-2081, RCP8.5) climate simulations from a 12-member ensemble of local convection-permitting 2.2 km climate projections over the UK and Ireland. Simulations indicate an increased probability of ETCs producing extremely severe WSI and RSI in the same storm in the future, occurring 3.6 times more frequently (every 5 years compared to every 18 years in the control). This frequency increase is mainly driven by increased rainfall intensities, pointing to a predominantly thermodynamic driver. However, future winds also increase alongside a strengthened jet stream, while a southward displaced jet and cyclone track in these events leads to a dynamically-enhanced increase in temperature. This intensifies rainfall in line with Clausius-Clapeyron, and potentially wind speeds due to additional latent heat energy. Future simulations also indicate an increase in the land area experiencing locally co-occurring wind and rainfall extremes; largely explained by increased rainfall within warm and cold fronts, although the relative increase is highest near cold fronts suggesting increased convective activity. These locally co-occurring extremes are more likely in storms with severe WSI and RSI, but not exclusively so as local co-occurrence requires the coincidence of separate drivers within ETCs. Overall, our results reveal many contributing factors to compound wind and rainfall extremes and their future changes. Further work is needed to understand the uncertainty in the future response by sampling additional climate models.