References: Multiple sectors of society, e.g. risk transfer, disaster reduction management, planning for disruption of transport, etc. would benefit enormously from potential skill of seasonal forecasts, e.g. November initialised winter (DJF) forecasts. We therefore assess 1. the climatological representation and prediction skill of extra-tropical cyclones and winter windstorms in seasonal prediction models 2. the benefits and limitations of using the North Atlantic Oscillation (NAO) as predictor for European winter windstorms on a seasonal time scale Extra-tropical Cyclones and Windstorms in Seasonal Prediction Models Simon Wild 1 ([email protected]), Daniel J. Befort 1 , Antje Weisheimer 2,3 , Jeff R. Knight 4 , Hazel E. Thornton 4 , Julia F. Lockwood 4 , Leon Hermanson 4 and Gregor C. Leckebusch 1 1. Motivation 3. Results 2. Event Identification AGU Fall Meeting, 2016 A23H – 0333 Cyclone Track Wind Track Windstorm “Daria” (24-26 th Jan 1990); Shadings: number of exceedances of 98 th percentile of wind speed during lifetime of storm • Windstorm identification and tracking algorithm according to Leckebusch et al., (2008) based on ex- ceedances of 98 th percentile of near-surface wind speeds. • Cyclone identification and tracking algorithm according to Murray and Simmonds, (1991) based on Laplacian of MSLP. Our analyses cover the core winter months: December – February; from 1992/93 to 2011/12 Reanalysis: • ECMWF ERA Interim (Dee et al., 2011) Seasonal Prediction Model Suites: • ECMWF System 3, 41 Members (Anderson et al., 2007) • ECMWF System 4, 51 Members (Molteni et al., 2011) • Met Office HadGEM GA3, 24 Members (MacLachlan et al., 2014) • Good agreement of spatial climatological distributions of extra-tropical cyclones and windstorms in comparison with reanalysis • Some biases present depending on the investigated model and region • Positive and significant skill in fore- casting the winter season frequency of extra-tropical cyclones and windstorms • The NAO as predictor for windstorms can be beneficial in some regions while forecast skill of seasonal predictions might be lost elsewhere. [1] Anderson, D. et al., 2007: Development of the ECMWF seasonal forecast System 3. ECMWF Tech. Memo. , 503, 1-58 [2] Dee, D.P. et al., 2011: The ERA-Interim reanalysis: Configuration and performance of the data assimilation system. QJRMS, 137, 553-597 [3] Leckebusch, G.C. et al., 2008: Development and application of an objective storm severity measure for the NE-Atlantic region. MeteorZ., 17(5), 575-587 [4] MacLachlan, C. et al., 2015: Global Seasonal forecast system version 5 (GloSea5): a high-resolution seasonal forecast system. QJRMS , 141 (689), 1072-1084 [5] Molteni, F. et al., 2011: The new ECMWF seasonal forecast system (System 4). ECMWF Tech. Memo. , 656, 1-51 [6] Murray, R. et al., 1991: A numerical scheme for tracking cyclone centres from digital data. Part I. Australian Meteorological Magazine, 39(3), 155-166 4. Summary 1 Extra-tropical Cyclones: Mean Sea Level Pressure (MSLP), 6 hourly Windstorms: Wind Speed at 925hPa, 12 hourly Row 1 (a-b): Direct forecast of windstorms Windstorms in Seasonal Models vs. Windstorms in ERA – Interim Row 3 (g-i): Difference (Row 2 minus Row 1) Blue: direct is better Red: NAO based is better Row 2 (d-f): NAO-regressed forecast of windstorms Regressed Windstorms in Seasonal Models vs. Windstorms in ERA - Interim Regression Slope: Windstorm Trackdensity vs. NAO in ERA-Interim d DATA A B Climatology: All Cyclones # Cyclones per Winter Climatology: Windstorms # Windstorms per Winter Prediction Skill: All Cyclones Prediction Skill: Windstorms B a-f) Kendall Rank Correlation (dots: statistical significance p<0.05) g-i) Difference of Correlation Values A B 2 3 4 A Climatology: Strongest 5% Cyclones # Cyclones per Winter Prediction Skill: Strongest 5% Cyclones Kendall Rank Correlation (dots: statistical significance p<0.05) Befort et al. 2017, QJRMS to be submitted