The CGCM for the global seasonal (GloSea) forecasting system is based on the HadCM3 climate model (Gordon et al., 2000). A number of enhancements to HadCM3 were made for GloSea: these include increased vertical ocean resolution, increased meridional ocean resolution in the tropics, and the new Hadley Centre coastal tiling scheme which enables specification of the land-sea mask at the ocean resolution. The atmospheric component (HadAM3) has a horizontal resolution of 2.5° latitude by 3.75° longitude, with 19 levels in the vertical. The ocean component has a zonal grid spacing of 1.25°, while the meridional grid spacing is 0.3° near the equator increasing polewards to 1.25°, and there are 40 vertical levels. Like HadCM3, the CGCM includes an active ice model and detailed land surface processes, with no flux corrections. Initial ocean conditions for the forecasts are obtained by forcing the ocean component with ERA-40 surface fluxes. There is slow timescale relaxation of temperature and salinity to climatology (Levitus and Boyer,1994; Levitus et al., 1995) at all model levels during this run. The FOAM scheme, an optimal interpolation scheme is applied from 1986 onwards to assimilate subsurface data. The performance of FOAM for real-time short-range operational ocean forecasting is described in Bell et al. (2000).Assimilation of thermal data into an ocean model near the equator often results in a dynamically unbalanced state with unrealistic deep overturning circulations. The equatorial bias correction scheme of Bell et al. (2002) is implemented to address this issue (see also Huddleston et al., 2002). Atmosphere and land-surface initial conditions are taken from ERA40.

To generate the ensemble we follow the method developed at ECMWF for use with SYSTEM 2 seasonal forecasts. Perturbations are applied to the forecast system, which aim to represent the uncertainty in observations of wind stress and sea surface temperature. Wind stress perturbations are applied to the ERA40 momentum fluxes that force the ocean model in such a way as to generate several ocean analyses for each hindcast start time. At the start of the hindcasts, sea surface temperature perturbations are added to the top 40m (with ramp to zero at 40m) of the temperature field of the ocean analysis.

Bell, M.J., R.M. Forbes and A. Hines, 2000. Assessment of the FOAM global data assimilation system for real-time operational ocean forecasting. J. Mar. Syst., 24, 249-275.

Bell, M.J., M.J. Martin and N.K. Nichols, 2002. Assimilation of data into an ocean model with systematic errors near the equator. Accepted by Q. J. R. Meteorol. Soc.

Gordon, C., C. Cooper, C.A. Senior, H. Banks, J.M. Gregory, T.C. Johns, J.F.B. Mitchell and R.A. Wood, 2000. The simulation of SST, sea ice extents and ocean heat transports in a version of the Hadley Centre coupled model without flux adjustments. Clim Dyn, 16, 147-168.

Huddleston, M.R., M.J. Bell, M.J. Martin and N.K. Nichols, 2002: Assessment of wind stress errors using bias corrected ocean data assimilation. Submitted to Q. J. R. Meteorol. Soc.

Levitus, S. and T.P. Boyer, 1994. World ocean atlas 1994, vol 4: temperature. NOAA/NESDIS E/OC21, US Department of Commerce, Washington, DC, pp 117.

Levitus, S., R. Burgett and T.P. Boyer, 1995. World ocean atlas 1994, vol 3: salinity. NOAA/NESDIS E/OC21, US Department of Commerce, Washington, DC, p 99.

12.12.2002