1.2 Overview of CAM 3.0

The CAM 3.0 is the fifth generation of the NCAR atmospheric GCM. The name of the model series has been changed from Community Climate Model to Community Atmosphere Model to reflect the role of CAM 3.0 in the fully coupled climate system. In contrast to previous generations of the atmospheric model, CAM 3.0 has been designed through a collaborative process with users and developers in the Atmospheric Model Working Group (AMWG). The AMWG includes scientists from NCAR, the university community, and government laboratories. For CAM 3.0, the AMWG proposed testing a variety of dynamical cores and convective parameterizations. The data from these experiments has been freely shared among the AMWG, particularly with member organizations (e.g. PCMDI) with methods for comparing modeled climates against observations. The proposed model configurations have also been extensively evaluated using a new diagnostics package developed by M. Stevens and J. Hack (CMS). The consensus of the AMWG is to retain the spectral Eulerian dynamical core for the first official release of CAM 3.0, although the code includes the option to run with semi-Lagrange dynamics (section 3.2) or with finite-volume dynamics (FV; section 3.3). The addition of FV is a major extension to the model provided through a collaboration between NCAR and NASA Goddard's Data Assimilation Office (DAO). The AMWG also has decided to retain the Zhang and McFarlane [199] parameterization for deep convection (section 4.1) in CAM 3.0.

The major changes in the physics include:

• Treatment of cloud condensed water using a prognostic treatment (section 4.5): The original formulation is introduced in Rasch and Kristjánsson [144]. Revisions to the parameterization to deal more realistically with the treatment of the condensation and evaporation under forcing by large scale processes and changing cloud fraction are described in Zhang et al. [200].The parameterization has two components: 1) a macroscale component that describes the exchange of water substance between the condensate and the vapor phase and the associated temperature change arising from that phase change [200]; and 2) a bulk microphysical component that controls the conversion from condensate to precipitate [144].
• A new thermodynamic package for sea ice (chapter 6): The philosophy behind the design of the sea ice formulation of CAM 3.0 is to use the same physics, where possible, as in the sea ice model within CCSM, which is known as CSIM for Community Sea Ice Model. In the absence of an ocean model, uncoupled simulations with CAM 3.0 require sea ice thickness and concentration to be specified. Hence the primary function of the sea ice formulation in CAM 3.0 is to compute surface fluxes. The new sea ice formulation in CAM 3.0 uses parameterizations from CSIM for predicting snow depth, brine pockets, internal shortwave radiative transfer, surface albedo, ice-atmosphere drag, and surface exchange fluxes.
• Explicit representation of fractional land and sea-ice coverage (section 7.2): Earlier versions of the global atmospheric model (the CCM series) included a simple land-ocean-sea ice mask to define the underlying surface of the model. It is well known that fluxes of fresh water, heat, and momentum between the atmosphere and underlying surface are strongly affected by surface type. The CAM 3.0 provides a much more accurate representation of flux exchanges from coastal boundaries, island regions, and ice edges by including a fractional specification for land, ice, and ocean. That is, the area occupied by these surface types is described as a fractional portion of the atmospheric grid box. This fractional specification provides a mechanism to account for flux differences due to sub-grid inhomogeneity of surface types.
• A new, general, and flexible treatment of geometrical cloud overlap in the radiation calculations (section 4.8.5): The new parameterizations compute the shortwave and longwave fluxes and heating rates for random overlap, maximum overlap, or an arbitrary combination of maximum and random overlap. The specification of the type of overlap is identical for the two bands, and it is completely separated from the radiative parameterizations. In CAM 3.0, adjacent cloud layers are maximally overlapped and groups of clouds separated by cloud-free layers are randomly overlapped. The introduction of the generalized overlap assumptions permits more realistic treatments of cloud-radiative interactions. The parameterizations are based upon representations of the radiative transfer equations which are more accurate than previous approximations in the literature. The methodology has been designed and validated against calculations based upon the independent column approximation (ICA).
• A new parameterization for the longwave absorptivity and emissivity of water vapor (section 4.9.2): This updated treatment preserves the formulation of the radiative transfer equations using the absorptivity/emissivity method. However, the components of the absorptivity and emissivity related to water vapor have been replaced with new terms calculated with the General Line-by-line Atmospheric Transmittance and Radiance Model (GENLN3). Mean absolute differences between the cooling rates from the original method and GENLN3 are typically 0.2 K/day. These differences are reduced by at least a factor of 3 using the updated parameterization. The mean absolute errors in the surface and top-of-atmosphere clear-sky longwave fluxes for standard atmospheres are reduced to less than 1 W/m${}^2$. The updated parameterization increases the longwave cooling at 300 mb by 0.3 to 0.6 K/day, and it decreases the cooling near 800 mb by 0.1 to 0.5 K/day. The increased cooling is caused by line absorption and the foreign continuum in the rotation band, and the decreased cooling is caused by the self continuum in the rotation band.
• The near-infrared absorption by water vapor has been updated (section 4.8.2). In the original shortwave parameterization for CAM [27], the absorption by water vapor is derived from the LBL calculations by Ramaswamy and Freidenreich [140]. In turn, these LBL calculations are based upon the 1983 AFGL line data [152]. The original parameterization did not include the effects of the water-vapor continuum in the visible and near-infrared. In the new version of CAM, the parameterization is based upon the HITRAN2k line database [153], and it incorporates the CKD 2.4 prescription for the continuum. The magnitude of errors in flux divergences and heating rates relative to modern LBL calculations have been reduced by approximately seven times compared to the old CAM parameterization.
• The uniform background aerosol has been replaced with a present-day climatology of sulfate, sea-salt, carbonaceous, and soil-dust aerosols (section 4.8.3). The climatology is obtained from a chemical transport model forced with meteorological analysis and constrained by assimilation of satellite aerosol retrievals. These aerosols affect the shortwave energy budget of the atmosphere. CAM 3.0 also includes a mechanism for treating the shortwave and longwave effects of volcanic aerosols. A time history for the mass of stratospheric sulfuric acid for volcanic eruptions in the recent past is included with the standard model.
• Evaporation of convective precipitation (section 4.1) following Sundqvist [169]: The enhancement of atmospheric moisture through this mechanism offsets the drying introduced by changes in the longwave absorptivity and emissivity.
• A careful formulation of vertical diffusion of dry static energy (section 4.11).

Other major enhancements include:

• A new, extensible sea-surface temperature boundary data set (section 7.2): This dataset prescribes analyzed monthly mid-point mean values of SST and ice concentration for the period 1950 through 2001. The dataset is a blended product, using the global HadISST OI dataset prior to 1981 and the Smith/Reynolds EOF dataset post-1981. In addition to the analyzed time series, a composite of the annual cycle for the period 1981-2001 is also available in the form of a mean ``climatological'' dataset.
• Clean separation between the physics and dynamics (chapter 2): The dynamical core can be coupled to the parameterization suite in a purely time split manner or in a purely process split one. The distinction is that in the process split approximation the physics and dynamics are both calculated from the same past state, while in the time split approximations the dynamics and physics are calculated sequentially, each based on the state produced by the other.