Scenarios for addressing high-impact, low-probability extreme events

Organizers: Jana Sillman (CICERO), Sonia Seneviratne (ETH Zurich), Xuebin Zhang (Climate and Environment Canada), Reinhard Mechler (IIASA)


Future climate change, and especially high-end emission scenarios, will lead to unprecedented occurrence and intensity of weather and climate extremes. Under current climate, weather-related hazards, such as heat, drought, storms and floods, are already responsible for 90% of disasters occurring world-wide of which mostly the economically less developed countries experience the largest losses. A particular challenge for risk quantification in climate change assessments is to deal with high-impact, low-probability outcomes, which can dominate estimates of total risk. Research is needed to combine insights from earth system modeling in the representation of such weather-related hazards, particularly those with low probability, with knowledge on exposure and vulnerability of society and ecosystems. Socioeconomic or disaster risks in that context can be modulated by changes in vulnerability and exposure patterns (e.g., population shifts, infrastructure design, management practices), and thus are important to be addressed in high- and low-emission scenarios. Given gaps particularly in understanding and modelling vulnerability, the emerging research field of disaster forensics has good potential for improving understanding of the contributions of hazard, exposure and vulnerability to actual '(un)natural' disaster events and thus fostering understanding of climate-related risks and scenarios. Rather than characterizing events through numbers and statistics alone, this line of research examines system and actor behaviour and analyses frequent, extreme and also exceptional events. This session will contribute to the WCRP Grand Challenge on Weather and Climate Extremes and welcomes abstracts related to new ways of designing and communicating scenarios to particularly address high-impact, low-probability extreme events. This will include aspects of exposure and vulnerability, for instance in terms of compound events and storylines (narratives), and should go beyond a purely probabilistic approach. We welcome contributions providing examples of how innovative approaches to scenarios can support and be used for decision-making in disaster risk reduction and mitigation and adaptation planning. Contributions that address physical plausibility and uncertainties in model simulations related to process-representation (e.g., land-atmosphere interactions, model resolution), combinations of RCPs and SSPs, and design of ensemble simulations for high-impact, low probability events are also relevant. We are as well interested in contributions that seek to align retrospective forensic approaches with forward-looking scenario analysis.