Operational Extreme Event Monitoring and Attribution Service: a multi-method comparison

The objective of this project of the ECMWF is to create an “Operational Extreme Event Monitoring and Attribution Service”, building on established methodologies and previous collaborations, including C3S, EUCLEIA, EUPHEME, and XAIDA. The service is enabled by a flexible, globally applicable framework, based on scientific, operational, and communication expertise. In this study, we apply the different methodologies available in the operational framework and beyond to different climate extremes, to compare attribution across different types of extreme weather events.

The main methodologies in the framework are probabilistic attribution and analogue-based dynamical attribution. We also include storyline-based methods in our comparison, to provide a more comprehensive picture of all key methodologies used by the research community. Each methodology has their unique strengths and may therefore be useful in specific user cases. Furthermore, the advantages and drawbacks of the methods are dependent on the type of extreme weather event considered. For example, extreme rainfall events are relatively short-lived, whereas droughts generally occur for several months. It is therefore crucial to have a comparison of the framework across different types of climate extremes - both univariate and compound (e.g. fire weather). We therefore aim to include a wide range of extreme events, including a heatwave, drought, fire weather, and extreme rainfall event. Similarly, we apply a range of methods including probabilistic attribution, dynamical attribution using analogues, and storyline attribution using nudged climate model simulations from DestinE, and potentially more. By doing so, we provide a coherent case study comparison using the Operational Extreme Event Monitoring and Attribution Service, as part of the C3S project.