Identification of extreme sea-level events in climate model data for the Swedish West Coast (Kattegatt)

Abstract

<p>Coastal flooding due to extreme sea level events have hazardous socio-economic impact on coastal communities, and water, energy and transport infrastructures. Predicting changes in frequency and magnitude of occurrence of such extreme events on timescales ranging from the next season to the coming decade(s) is of high importance for timely preparation and decision-making. This requires having climate information at time scales beyond the weekly weather forecasts at long range as well as century-long timescales. Such data, however, are normally provided on the global scale, and thus have a relatively coarse resolution compared to the local events such as extreme sea-level. Here, a new empirical event-identification method is developed to detect extreme sea-level events for the western Swedish coast (Kattegatt) using data from global climate models and reanalysis. It, therefore, serves as an add-on to climate model forecast/projection data. The event-identification method considers both dynamical and statistical relations between high sea-level events and multivariate atmospheric variables by using peak-over-threshold wind speeds in combination with a catalogue of surge-relevant sea-level-pressure and wind-field patterns at both local as well as large scales. The identification performs well when tested for ERA5 data and captures significant number of observed high sea-level events in the Swedish West Coast. Using our method, we investigate the changes in the frequency of occurrence and magnitude of extreme events in the coming decade and until the end of the century. For this purpose, we use data from EC-Earth model experiments from the Decadal Climate Prediction Project (DCPP) and the SMHI Large Ensemble (LENS) for future climate scenarios (1970-2100; 50 members). Our new approach is also used to explore compound events when <span>extreme sea-level and heavy precipitation occurring together </span>under current and climate change conditions.</p>