Simulated Data


This is the HiGEM codebase

HiGEM V6.1 Control (xbpjt)



RelatedModel, HiGEM: This is the first HiGEM climate run


The atmospheric component of HiGEM has 38 vertical levels with a horizontal resolution of 1.25 degrees of latitude by 0.83 degrees of longitude, which produces a global grid of 288 x 217 grid cells. This is equivalent to a surface resolution of about 139 km x 92 km at the Equator, reducing to 98 km x 92 km at 45 degrees of latitude (comparable to a spectral resolution of Nblah). The atmospheric timestep period is 20 minutes (72 timesteps per 1 days).

Radiation Scheme

A general 2-stream radiation code including cloud microphysics. The radiation scheme uses 6 spectral bands in the solar (shortwave) wavelenths and 9 bands in the terrestrial thermal (longwave) wavelengths. The radiative effects of CO2 and ozone are explicitly represented as well as oxygen, methane, N2O, CFC-11 and CFC-12. The LW and SW radiative effects of climatological distributions of sulphate, seasalt, soot and biomass aerosols are included. A cloud area parameterisation produces an Area Cloud Fraction which replaces the bulk value used in the radiation code. Mixed phase clouds containing both ice and water are segregated into separate sub-clouds for radiation calculations.


  • JM Edwards, Slingo A, 1996: Studies with a flexible new radiation code. 1. Choosing a configuration for a large-scale model QUARTERLY JOURNAL OF THE ROYAL METEOROLOGICAL SOCIETY, 122(531) 689-719

Land Surface Scheme

The surface albedo is a function of snow depth and the temperature of the snow over sea ice. The surface hydrology uses the MOSES-II (Met Office Surface Exchange Scheme). The vegetation distribution is fixed. Using coastal tiling allows both land and sea to co-exist in the same gridbox. There are 9 land surface tiles per grid cell.



Boundary Layer Scheme

The boundary layer scheme explicitly parameterises the top-of-mixed-layer entrainment. It uses a formulation of the surface exchange coefficients based directly on Monin-Obukhov stability functions. It uses a subgrid diagnosis of cloud-base height in order to improve the accuracy of the buoyancy flux integral which is used to diagnose the depth of mixing in stratocumulus clouds. The boundary layer scheme splits the radiative heating increments into separate LW and SW contributions. It uses a Richardson number based mixing scheme and orographic roughness fields. The scheme accounts for the radiative coupling and the thermal capacity of the vegetation canopy.


  • Lock, A. P. 2001: The numerical representation of entrainment in parametrizations of boundary layer turbulent mixing. MWR, 129, 1148-1163
  • Lock, A. P., A. R. Brown, M. R. Bush, G. M. Martin, R. N. B. Smith et al. 2000: A new boundary layer mixing scheme. Part I: scheme description and SCM tests. MWR, 128, 3187-3199

Convection Scheme

Convection is able to transport momentum in the vertical. The inital convective plume mass flux is determined by a CAPE based clousure scheme. The radiative representation of anvils modifies the convective cloud amount (CCA) to vary with height during deep convection. Excluding precipitation from the water path means that the radiation scheme does not 'see' the convective rain and snow. The accurate treatment of precipitation phase change ensures that precipitation does not change phase if the associated latent cooling would take the temperature below the freezing point again.


  • None
  • None

Gravity Wave Scheme

The orographic gravitity wave scheme also includes flow blocking. The gravity wave constant is 1.00e+05 and defines the magnitude of the parametrized response. The critical Froude number is 4.00 and determines the proportion of that drag attributed to flow blocking and gravity wave drag respectively. The spectral gravity wave scheme is not used.


  • Webster S., A.R. Brown, D.R. Cameron and C.P. Jones, 2003: Improvements to the Representation of Orography in the Met Office Unified Model. Quarterly Journal of the Royal Meteorological Society, 129 (591): 1989-2010 Part B.

Precipitation and Cloud Scheme

The large scale precipitation scheme contains a full microphysical calculation of the cloud phase and generation of precipitation with water vapour, cloud liquid water and ice particle content as prognostic variables. Microphysical processes are treated as transfer terms between water vapour, liquid, ice, and rain. The fraction of cloud ice content that is pristine ice crystals and snow aggregate particles are treated seperately in the microphysical transfer terms. Condensation can occur before grid scale supersaturation and the vapour is condensed to cloud water. The conversion from vapour to liquid or frozen cloud water is reversible. A RHcrit parametrization causes the cloud scheme to use 3D diagnosed critical relative humidity. A cloud area parametrization produces an Area Cloud Fraction which replaces the Bulk value in much of the radiation code.


  • Wood et al. ,2002: Atmos. Res., 65, 109-128

Advection and Diffusion

A semi-lagrangian advection scheme is used. The advection of potential temperature, moisture, density and winds are treated separately. Moisture is conserved using a non-hydrostatic scheme. A conservative horizontal diffusion scheme is used. Vertical diffusion is switched off.




The aerosol parameterisation includes a sulphur cycle, soot scheme and biomass aerosol scheme. The sulphur cycle includes SO2 emissions from the surface, chimneys and volcanoes. The sulphur cycle also uses an interactive dimethyl sulphide scheme. The biomass scheme includes emissions from the surface and from high levels.


All rivers flow with an effective velocity of 0.4 m/s and a meander ratio of 1.4. River outflow to the ocean is included.


The oceanic component of HiGEM has 40 vertical levels with a horizontal resolution of 0.333 degrees of latitude by 0.333 degrees of longitude, which produces a global grid of 1082 x 540 grid cells. This is equivalent to a surface resolution of about 37 km x 37 km at the Equator, reducing to 26 km x 37 km at 45 degrees of latitude (comparable to a spectral resolution of Nblah). The atmospheric timestep period is 20 minutes (72 timesteps per 1 days). The ocean GCM includes a polar island as standard. The ocean GCM uses the McDougall equation of state.



Tracer advection and diffusion

The advection of active tracers, temperature and salinity, uses a fourth order differencing scheme (Pacanowski and Griffies, 1998) which uses a fourth order estimate of the tracer gradients together with the second order advective fluxes. The option to use upwind advection in the bottom gridcell at each point avoids instabilities found in high resolution runs. The Griffies diffusion scheme orientates the mixing tensor to lie along isopycnal rather than horizontal sufarces (Griffies et al., 1998). Isopycnal diffusivity is 5.00e+02 (m*m/s) and is constant with depth. The Gent and McWilliams (GM) Scheme parametrises the effect of mesoscale eddies on tracer transports. The Visbeck scheme allows the diffusivity for the GM scheme to be spatially and temporally variable, so that it can take large values in eddy-generation regions and small values elsewhere. The HADCM4 version of the Visbeck scheme uses large-scale density gradients to pick out eddy-generation regions. The isopycnal diffusivity is tapered as the slope of the isopycnals increases using a hyperbolic tangent function. A scale-selective version of the Gent and McWilliams scheme (Roberts and Marshall 1998) removes small-scale noise from the tracer fields without affecting their large-scale distribution and without causing any mixing across isopycnal surfaces.


  • Pacanowski and Griffies, MOM 3.0 Manual, 1998
  • Griffies et al 1998
  • Roberts and Marshall, 1998


Fourier filtering is used to decrease the effective resolution of the model at high latitudes, allowing a longer timestep to be used. See UMDP 40. Different filtered regions can be chosen for tracers and velocity and for the northern and southern hemispheres. In the northern hemisphere, filtering starts at 'First tracer/velocity row in northern hemisphere to be filtered' and goes right up to the north pole. The filtering removes scales less than the grid scale on the row defined by 'Tracer/velocity row used to define basic zonal dimension'. The equator-most row to be filtered in each hemisphere determines the minimum effective gridlength retained by the filtering. The first tracer/velocity row in the northern hemisphere: 510/509 Tracer/velocity row used to define basic zonal dimensions: 510/509 The last tracer/velocity row in the southern hemisphere:34/34



Mixed Layer and vertical diffusion

A Kraus-Turner (1967) type mixed layer model is used to parameterise the effects of surface generated turbulence. Vertical diffusion is dependent on the Ricardson Number (Peters et al, ?) The quadratic Large scheme calculates the vertical diffusion coefficient in the mixed layer (Large et al 1994) The quadratic Large scheme is applied where the Richardson number is less than 0.3 upto a maximum depth of 80 m.


  • Kraus Turner, 1967
  • Peters et al, ?
  • W.G.Large et al 1994, Oceanic Vertical Mixing : A review and a model with a nonlocal boundary layer parametrisation, Rev Geophys, 32, 363-403.

Barotropic Solution, Momentum Flux and Diffusion

A free-surface barotropic solution is used with Delphus-Delcross smoothing for the surface height field. A modifed Cox scheme is used for calculating velocity fluxes. Horizontal momentum diffusion uses viscosity coeffiecients that are constant in latitude: 0.00. Biharmonic momentum diffusion allows scale-selective damping to be applied to the velocities without affecting the large-scale velocity field. It is useful in helping the removal of grid-scale noise in the velocity field.


A Rahmstorf's full convection scheme is used which is guaranteed to produce a profile having complete static stability.

Salinity Control

There is no reference salinity, instead salinity limits are applied. Upper salinity limit: 4.50000e-02 (psu/1000). Lower salinity limit: 5.00000e-03 (psu/1000).

Ocean Straits

A generalised strait exchange scheme is used that advects water from a marginal sea into the main ocean, with a corresponding return flow. There is 1 strait in this set up with end coordinates (i,j) at (62, 378) and (65, 378).

Sea Ice

The prognostic sea ice model contains ice thermodynamics based on Semtner's "zero-layer" and calculates prognostic ice depth, ice concentration and snow depth. The multiple ice categories model allows the sub-grid scale ice thickness distribution to be represented. The EVP (elastic-viscous-plastic) dynamics based on Hibler's sea-ice rheology calculates velocities that are used to advect sea-ice. A north polar island is included and sea ice can be advected over it.


  • Semtner, A. J., 1976: A model for the thermodynamic growth of sea ice in numerical investigations of climate. J. Phys. Oceanogr., 6, 379-389.
  • Hibler, W. D., 1979: Dynamic Thermodynamic Sea Ice Model. Journal of Physical Oceanography, 9(4), 815-846.

Sea Ice Thermodynamics

Ocean to ice heat flux parameterisation uses the 'McPhee scheme' (McPhee, 1992), which uses both the ocean-ice temperature difference and the friction velocity in the flux parameterisation. The 'McPhee scheme' produces a flux proportional to the ice concentration above a marginal sea ice concentration of 0.05. For lower concentrations, the heat flux is constant. Number of sea ice categories is 5. Minimum local ice depth is 0.1 m. Min local snow thickness: 1.0E-5 m Min local thickness of new ice growing in leads: 0.05m


  • McPhee, M. G., 1992: Turbulent heat-flux in the upper ocean under sea ice. Journal of Geophysical Research-Oceans, 97(C4), 5365-5379.
  • Maykut G. A., M.G. McPhee, 1995: Solar heating of the Arctic mixed layer Journal of Geophysical Research-Oceans, 100(C12), 24691-24703.

Sea Ice Dynamics

The sea ice velocity arises from a balance of windstress, ocean drag, coriolis and internal ice stresses. It is based on the viscous-plastic sea-ice rheology of Hibler (1979), and recommended for use in climate modelling by the Sea Ice Model Intercomparison Project [Kreyscher et al, 2000]. Convergence of ice is impeded or prevented when the ice is thick. The ice ridging scheme converts thinner ice to thicker ice, and if the ice is converging, the scheme ensures that enough ice ridges to keep the ice concentration equal or below 1 (Hunke and Lipscomb). Maximum compressive strength of ice per unit thickness is 2.00e+04 (N/m**2) Ice strength is smoothed to avoid instabilities at high northern latitudes polewards of 87.5 lat.. Ice velocities are filtered at high northern latiitudes to prevent excessive ridging and buildup of ice. The Quadratic ice-ocean drag coefficient is 1.50e-02


  • Hibler, W. D., 1979: Dynamic Thermodynamic Sea Ice Model. Journal of Physical Oceanography, 9(4), 815-846.
  • Kreyscher M et al., 2000: Results of the Sea Ice Model Intercomparison Project: Evaluation of sea ice rheology schemes for use in climate simulations Journal of Geophysical Research-Oceans, 105 (C5): 11299-11320.
  • Thorndike, A. S., D. A. Rothrock, G. A. Maykut et al., 1975: Thickness Distribution of Sea Ice. Journal fo Geophysical Research-Oceans and Atmosphereres, 80(33), 4501-4513.
  • Flato, G. M. and W. D. Hibler, 1995: Ridging and strength in modeling the thickness distribution of arctic sea-ice. Journal of Geophysical Research-Oceans, 100 (C9), 18611-18626.
  • Lipscomb, W.H. and E. C. Hunke, 2004: Modeling sea ice transport using incremental remapping. Monthly Weather Review, 132 (6), 1341-1354.

Atmos-Ocean Coupler