Report documenting the developments in conditional attribution and results from initial application

Abstract

One of the main goals of the COMPASS project is to extend attribution methodologies from single-driver extremes to complex extremes, aiding development of a harmonized methodological framework for climate impact attribution of compound events. Deliverable 2.5 ‘Report documenting the developments in conditional attribution and results from initial application’, documents some of the key developments in conditional attribution as applied to more complex extremes. It draws upon developments in work package 2, WP2: “Attribution frameworks in a physical context”, around large-scale drivers and complex extremes, as well as developments from the COMPASS phase 1 Use Cases in WP4: “ Developing actionable climate information for events”. This report finds that the application of attribution methods to more complex extremes and specifically impact attribution requires extra choices around conditioning, since that can have a significant impact on the results. Therefore, it is important for impact attribution that stakeholder engagement and a clear focus on which science questions the study aims to answer is addressed at an early stage, to provide the most useful output. The report also highlights that the choice of impact attribution methods and hence conditionality is still very limited for certain complex extremes such as Tropical Cyclones (TC), where very strong conditioning is required to combat the extra uncertainty introduced in the attribution results from the impacts side. Furthermore, this report documents a framework developed in the COMPASS project in WP2 linking large-scale drivers to complex extremes and its application to one of the Use Cases. This framework has strong potential to guide the choice of conditionality for impact attribution studies looking at complex extremes and adding valuable input to attribution studies.