OM4: GFDL’s ¼ degree ice-ocean model component for CMIP6

Alistair Adcroft1, John P Dunne2, Stephen Griffies3, Robert Hallberg4, Matt Harrison5, Malte Jansen6, Sonya Legg7, Brandon G Reichl4 and Rong Zhang8, (1)Princeton University, Program in Atmospheric and Oceanic Sciences, Princeton, NJ, United States, (2)NOAA Geophys Fluid Dynamic, Princeton, United States, (3)Geophysical Fluid Dynamics Laboratory, Princeton, United States, (4)NOAA/Geophysical Fluid Dynamics Laboratory, Princeton, United States, (5)Geophysical Fluid Dynamics Laboratory, Princeton, NJ, United States, (6)University of Chicago, Department of the Geophysical Sciences, Chicago, IL, United States, (7)Princeton University, Princeton, NJ, United States, (8)Princeton University, Program in Atmospheric and Oceanic Sciences, Princeton, United States
Abstract:
We describe OM4, a new ice-ocean model developed at the Geophysical Fluid Dynamics Laboratory (GFDL). OM4.0 serves as the ocean/sea-ice component for the GFDL CMIP6 generation climate and Earth system models and uses version 6 of the Modular Ocean Model (MOM6) and version 2 of the Sea Ice Simulator (SIS2). This contribution follows the Coordinated Ocean-sea ice Reference Experiments protocol to assess OM4.0 simulation quality across a broad suite of climate-relevant features. MOM6 makes use of a vertical Lagrangian-remap algorithm that enables general vertical coordinates. We show that use of a hybrid depth-isopycnal coordinate reduces the mid-depth ocean warming drift commonly found in pure z-coordinate models. Analysis of overflows reveals that excessive parameterized near-bottom vertical mixing is likely the dominant cause of a shallow overturning. The ¼ degree resolution allows the model to be eddy-rich in much of the ocean, albeit far from eddy-resolving. We show additional results using parameterizations aimed at improving the representation of both the resolved and unresolved mesoscale eddies. In coupled configurations, OM4.0 exhibits dramatically reduced ocean biases compared with previous generations of GFDL coupled models.