Its global ocean configuration used in both versions of the coupled climate model is known as ORCA2. It has a tripolar, quasi-isotropic grid: a combination of an isotropic Mercator grid south of 20 °N, and a non-geographic quasi-isotropic grid north of it, in which the North Pole singularity is replaced by a line between points in Canada and Siberia. A nominal resolution of 2° at the equator is chosen to which a latitudinal grid refinement of 1/2° is added in the tropics. ORCA2 uses realistic bottom topography and coastlines, derived from Smith and Sandwell (1997) up to 60° of latitude and ETOPO5 elsewhere. The maximum depth of 5000 m is spanned by 31 z-levels ranging from 10 m in thickness in the

upper 120 m to a maximum of 500 m at the bottom. Vertical mixing is computed http://www.selleckchem.com/products/AZD2281(Olaparib).html from

a turbulence closure scheme based on a prognostic vertical turbulent kinetic equation (TKE scheme), which performs well in the tropics ( Blanke and Delecluse, 1993). Lateral diffusivity is parameterized by an iso-neutral Laplacian operator with an eddy diffusivity coefficient of 2,000 m2 s−1. In addition a bolus velocity is applied on temperature and salinity ( Gent and McWilliams, 1990) with the NEMO default of a spatially and temporally varying coefficient (calculated from the local growth rate of baroclinic instability and, between 20°N and 20°S, forced to decrease to vanish at the Equator), as described in Treguier et al. (1997). Lateral viscosity is parameterized by a horizontal laplacian operator and an eddy viscosity coefficient of 4.104 m2 s−1 histone deacetylase activity except in the tropics where it reduces to 2.103 m2 s−1 (except along western boundaries) (). The ocean model is coupled Adenosine triphosphate to the LIM-2 sea-ice model ( Timmermann et al., 2005), which is unchanged in all simulations considered in

this study. In spite of these common aspects, IPSL-CM4 and IPSL-CM5A ocean component has evolved from OPA8 (Madec et al., 1999) to NEMOv3.2 (Madec, 2008) respectively, which implies the implementation of several additional parameterizations related to bottom topography and vertical mixing, as described in the following section, as well as the use of a state-of-the-art biological model, PISCES. The PISCES model is derived from the Hamburg Model of Carbon Cycle version 5 (HAMOCC5) (Aumont et al., 2003). A detailed description of the model parameterizations can be found in Séférian et al. (2012). The coupled simulations combine the OPA oceanic component to the LMDZ4 (Hourdin et al., 2006) for IPSL-CM4 or LMDZ5A atmospheric model (Hourdin et al., 2012) for IPSL-CM5A. Evolutions between these two models are described in detail in Hourdin et al., 2012). In terms of resolution, given the increasing recognition of the role of the stratosphere in controlling some aspects of the tropospheric climate (e.g. Nikulin and Lott, 2010), priority has been given to vertical resolution increase (from 19 to 39 levels) rather than horizontal resolution.