Regional ocean zooms, enhancements and hybrid methods for climate extremes and impacts at global and local scales

Regional ocean zooms, enhancements and hybrid methods for climate extremes and impacts at global and local scales (Razorege) aims to develop and implement Earth System Models (ESMs) that accurately capture regional aspects of the climate, while remaining computationally affordable for global, multi-century simulations.Projecting regional climate variability, change, and extreme events remains a challenge for current ESMs. These difficulties stem from inaccuracies in representing both local processes and distant teleconnections, as well as naturally higher variability of regional climates - which typically require larger ensembles than are currently feasible.For Europe, key uncertainties in climate prediction include the representation of the Gulf Stream, the Atlantic Meridional Overturning Circulation, and the Mediterranean Sea, particularly air–sea interactions. Similarly, other major ocean boundary currents—such as the Kuroshio in the Pacific and the Antarctic Circumpolar Current in the Southern Ocean—are critical for forecasting climate impacts in East Asia and South Africa, respectively.Our approach involves targeted improvements to ocean and sea-ice models, guided by the latest scientific understanding of how regional climate indices relate to model biases. We will apply techniques such as region-specific adjustments to horizontal and vertical resolution, as well as new parameterisations. A central goal is to keep the models efficient enough to support simulations with Earth System complexity, while also improving physical and biogeochemical processes.Once validated, these model enhancements will be integrated into Earth System Models to contribute to the next Coupled Model Intercomparison Project (CMIP7) and IPCC assessments. Because these upgrades are focused and cost-effective, they can also be applied to other modelling systems—such as seasonal-to-decadal forecasts and Digital Twins—supporting broader collaboration across climate modelling communities.