Assessing CMIP6 uncertainties at global warming levels

Abstract IPCC reports and climate change impact studies generally exploit ensembles of climate projections based on different socio-economic pathways and climate models, which provide the temporal evolution of plausible future climates. However, The Paris Agreement and many national and international commitments consider adaptation and mitigation plans targeting future global warming levels. Model uncertainty and scenario uncertainty typically affect both the crossing-time of future warming levels and the climate features at a given global warming level. In this study, we assess the uncertainties in a multi-model multi-member CMIP6 ensemble (MME) of seasonal and regional temperature and precipitation projections. In particular, we show that the uncertainties of regional temperature projections are considerably reduced if considered at a specific global warming level, with a limited effect of the emission scenarios and a reduced influence of GCM sensitivity. We also describe in detail the large uncertainties related to the different behavior of the GCMs in some regions.