You are here: Home Research Transformation Pathways Research Data-centric modeling of cross-sectoral impacts

Data-centric modeling of cross-sectoral impacts

At today's 1°C of global warming the local signals of climate change emerge from natural variability and manifest in more frequent heat waves and droughts, heavy rainfall events, long-term local trends in temperature and precipitation as well as sea-level rise and associated shifts in extreme sea levels. A global synthesis and understanding of societies’ sensitivities to these changes is still severely limited due to  fragmentary observational records and the concurrent direct human influence through, for example, land-use change, pollution, or changes in exposure and vulnerability due to population and protection measures. In principle, process-based climate impact simulations are ideal tools to address the issues of fragmentary observations and concurrent human factors. Many impact models do however not yet stand up to this challenge due to missing detail in the representation of socio-economic conditions and dynamics but also due to missing representation of physical processes. For example, the approaches to estimate global impacts of sea-level rise do not resolve the flooding events from tropical cyclones, though these events cause a large part of losses and produce rich - yet largely untapped - observational data.

This working group aims at generating central access to critical impact-driving and impact-evaluation data along the emissions-to-impacts simulation chain. In the emerging era of data richness we especially aim at sourcing non-traditional datasets like Open Street Maps or social media to increase the regional relevance of global climate impact simulation, addressing also the growing demands for financial climate risk assessment. We aim to better understand and separate the contribution of weather fluctuation (extreme events or long term trends) versus direct human influence as drivers of bio-physical or social impacts. We will pursue this work in close collaboration with impact modelers from the ISIMIP community as an efficient way to generate community-wide progress in impact modelling across sectors. While part of the work will be dedicated to the analysis of impact simulations generated within ISIMIP, we will develop a modelling approach for spatially and temporally explicit simulations of coastal system change, overcoming the limitations of present methods.

The work will finally put us in the position to push forward the attribution of climate-change impacts, going beyond physical climate change attribution and paving the path to climate liability. To better quantify the link between greenhouse gas emissions and past, present and future impacts we will streamline the development of the global emission database and simple climate modeling at PIK. ISIMIP-based impact functions linking impacts to global mean temperature change and the ability to trace incremental temperature increases back to emission sources will help to ultimately attribute past, present and future damages to major global greenhouse gas emitters.

Team

Matthias Mengel (leader)

Matthias Büchner

Robert Gieseke

Stefan Lange

Benjamin Schmidt

Simon Treu

Iliusi Vega de Valle

Jan Volkholz

Document Actions