Sensitivity of forecast errors to initial and lateral boundary conditions

Article English OPEN
Gustafsson, Nils ; Källén, Erland ; Thorsteinsson, Sigurdur (2011)

The adjoint of a limited area model has been used to study the sensitivity of 12 h forecast errorsto initial and lateral boundary conditions. Upper troposphere potential vorticity and mean sealevel pressure verification scores for 1 month of operational forecasts from the SwedishMeteorological and Hydrological Institute were used to select 2 cases with particularly poorforecast performance. The sensitivity experiments show that errors in initial data is the mostlikely explanation for one of the forecast failures, while errors in initial as well as lateralboundary data can explain the 2nd forecast failure. Results from the sensitivity experimentswith respect to the lateral boundary conditions indicate that poor quality lateral boundaryconditions may be improved by utilizing subsequent downstream observations within the modelintegration area. This result is of great significance with regard to the possibilities for applying4-dimensional variational data assimilation (4DVAR) for limited area forecast models. Resultsfrom the sensitivity experiments also reveal, however, that the lateral boundary treatment inoperational limited area models needs to be improved with regard to the mathematical formulation.It is furthermore shown that modifications to be applied to the lateral boundary conditionsneed to be determined with appropriate time resolution and that some filtering of these lateralboundary modifications has to be introduced to avoid enhanced high-frequency gravity wavenoise in the vicinity of the lateral boundaries.DOI: 10.1034/j.1600-0870.1998.t01-1-00002.x
  • References (16)
    16 references, page 1 of 2

    Davies, H. C. 1983. Limitations of some common lateral boundary schemes used in NWP models. Mon. Wea. Rev. 111, 1002-1012.

    Eliassen, A. and Thorsteinsson, S. 1984. Numerical studies of stratified air flow over a mountain ridge on the rotating earth. T ellus 36A, 172-186.

    Errico, R. M., Vukic´evic´, T. and Raeder, K. 1993. Comparison of initial and lateral boundary condition sensitivity for a limited-area model. T ellus 45A, 539-557.

    Gustafsson, N. 1990. Sensitivity of limited area model data assimilation to lateral boundary condition fields. T ellus 42A, 109-115.

    Gustafsson, N. 1991. The HIRLAM model. Proceedings of the ECMW F Seminar on Numerical methods in atmospheric models, 9-13 September, 1991; available from the European Centre for Medium Range Weather Forecasting, Shinfield Park, Reading, England.

    Gustafsson, N. and Huang, X.-Y. 1996. Sensitivity experiments with the spectral HIRLAM and its adjoint. T ellus 48A, 501-517.

    Gustafsson, N. and McDonald, A. 1996. A comparison of the HIRLAM grid point and spectral semiLagrangian models. Mon. Wea. Rev. 124, 2008-2022.

    Gustafsson, N., L o¨nnberg, P. and Pailleux, J. 1997. Data assimilation for high resolution limited area models. J. Met. Soc. Japan 75B, 367-382.

    Huang, X.-Y., Gustafsson, N. and Ka¨lle´n, E. 1997. Using an adjoint model to improve an optimum interpolation based data assimilation system. T ellus 49A, 161-176.

    Ka¨lle´n, E. and Huang, X.-Y. 1988. The influence of isolated observations on short-range numerical weather forecasts. T ellus 40A, 324-336.

  • Metrics
    No metrics available
Share - Bookmark