Research on SI is at an early stage, and to the authors’ knowledge no previous studies have systematically explored what resolution is required to resolve it in ocean models. As computational power increases, models are able to simultaneously Rucaparib resolve a richer set of dynamics by running at higher spatial resolution and incorporating more complex physical and biogeochemical parameterizations. However, higher spatial resolution introduces a new set of challenges as well, the first among

these being the issue of double-counting (Delworth et al., 2012). It is commonly thought that as models enter an “eddy-permitting” regime, where some (but not all) of the mesoscale eddies are explicitly resolved, parameterizations

should either be turned off or minimized in order to prevent both resolving and parameterizing the same eddies. One reason for this is that parameterizations can out-compete the resolved eddies for the energy sources required to grow, leaving the resolved eddies weak and ineffectual (Henning and Vallis, 2004). Therefore, one of the first steps to developing a skillful parameterization is to know when its use is appropriate, and when it should be turned off to avoid double-counting. The Venetoclax supplier issue of double-counting is not confined to just mesoscale eddies, however. Submesoscales develop at scales less than 10 km,

and these in turn will become partially resolved as GCM resolution becomes even finer in upcoming model generations. SI is one such submesoscale process, and ocean models will increasingly pass into a regime that could be described as “SI-permitting”. As is the case with mesoscale eddies, explicitly resolving only some of the SI modes can be expected to present a challenge in preventing double-counting by a parameterization. As of the writing of this paper no parameterization exists for SI in the oceanic mixed layer, and any forthcoming attempt at one will require knowledge of how SI OSBPL9 behaves when it is partially resolved. Symmetric instability in a stably stratified flow occurs when the Ertel PV takes on the opposite sign of f ( Hoskins, 1974). Fronts in the surface mixed layer of the ocean feature strong lateral density gradients, which in conjunction with wind forcing and/or buoyancy fluxes create conditions favorable to the development of SI ( Thomas and Taylor, 2010). SI is capable of restratifying the mixed layer on timescales shorter than that of baroclinic instability ( Haine and Marshall, 1998, Boccaletti et al., 2007 and Li et al., 2012), and both types of instability are central to setting the stratification of the surface ocean at strong fronts.