Project

Discipline

climate extremes; copulas; multivariate extremes; carbon cycle; climate risk; climate model; compound events

Lead
Klimaextreme wie Dürre und Hitze richten regelmässig beträchtliche Schaden an. Die Schäden sind oft noch um einiges schlimmer, wenn mehrere Extremereignisse zusammenfallen. Für Risikoabschätzungen von solchen Ereignissen werden oft Klimamodelle verwendet. Allerdings ist im Moment unbekannt, wie gut moderne Klimamodelle solche gleichzeitig auftretenden Extremereignisse simulieren können. Wenn Risikoabschätzungen für solche Ereignisse ungenau oder verzerrt sind, kann das zu falschen Anpassungsstrategien führen.
Lay summary
Inhalt und Ziel des Forschungsprojekts Dieses Projekt wird neue statistische Masse entwickeln, um zu evaluieren, wie gut Klimamodelle kombinierte Extremereignisse und andere Wetterrisiken simulieren können. Hierbei werden wir den Fokus auf folgende Ereignisse legen: a) gleichzeitiges Auftreten von Dürre und Hitze, gleichzeitiges Auftreten von Starkwind und Extremniederschlag, c) Hitzestress und d) Feuerrisiko. Hitzestress und Feuerrisiko lassen sich aus der Kombination von Temperatur und relativer Feuchte abschätzen. Die neuen Masse werden dann verwendet, um Unsicherheiten in Zukunftsprognosen von Wetterrisiken zu reduzieren. Wissenschaftlicher und gesellschaftlicher Kontext des Forschungsprojekts Unsere Arbeit wird einen wichtigen Beitrag zu einer verbesserten Abschätzung von Klimarisiken durch extreme Wetterereignisse leisten. Das ist insbesondere relevant für Zukunftsplaner in vielen Bereichen der Politik, um mögliche Anpassungsstrategien zu entwickeln. Es ist aber auch relevant für einzelne Industriezweige wie zum Beispiel Wasserversorger oder Versicherungsunternehmen.

Direct link to Lay Summary

Last update: 20.09.2018

Climate-related hazards such as heat waves, droughts, and floods can lead to devastating impacts on human societies and ecosystems. The impacts are often particularly severe when several hazards occur at the same time. Many hazards result from a combination of physical processes that interact on multiple spatial and temporal scales. Climate change will alter many of these processes and their interactions, making projections of future hazards based on single driver analyses difficult. Impact studies that consider only one climatic driver usually fail to assess the full extent of the impacts associated with multiple dependent drivers. Furthermore, it is not clear whether current climate models can capture major changes in risk associated with climate-related hazards. Existing modelling approaches used to assess risk may therefore lead to serious mal-adaptation. This project will (i) develop new metrics to evaluate climate models with respect to multivariate relationships, extremes, and compound hazards, (ii) constrain model ensembles with observations to improve projections of hazards and compound hazards, and (iii) relate hazard probabilities with impacts to quantify the importance of climate in comparison to vulnerability and exposure for high-impact climate events. The project will focus on the compound hazards a) drought and heat, b) wind and precipitation extremes, as well as on the hazards c) human heat stress and d) fire risk, which both can be expressed in terms of temperature and relative humidity. As a data basis for the development of new metrics, the project will largely rely on pre-industrial control simulations of climate models. These long simulations do not contain trends and other non-stationarities associated with human-induced climate change and thus provide an excellent environment to develop new robust statistical approaches. Subsequently, the new metrics will be used to constrain hazards in present-day and future simulations from climate model ensembles with observation-based gridded climate and reanalysis datasets. Furthermore, estimated hazard probabilities will be confronted with modeled and observed impacts such as extremely anomalous carbon fluxes, extremely low crop yields, human mortality, damaged infrastructure, and large wildfires. We will further conduct experiments with a dynamical vegetation model to study the influence of different drought-heat signatures on carbon dynamics such as interannual variability of carbon fluxes and cumulative carbon uptake. The newly developed metrics will serve a wide community of Earth system modelers for better evaluating process-based models against multivariate dependencies that are critical for large impacts. These metrics will complement and extend established model evaluation systems. Improved projections of hazards, in particular compound hazards, are highly relevant for assessing future risks associated with climate extremes. Comparing hazard probabilities with actual impacts will put the contribution of climate drivers into perspective and reveal in which regions and for which hazards other non-climatic factors related to vulnerability and exposure play an important role. These insights are critical for a broad community including risk managers, decision-makers and private businesses such as the insurance sector.