The goal of SACADA is to provide a powerful and efficient method for chemical data assimilation, which then shall be used for the interpretation of ENVISAT data at the DFD of the DLR. For the operative use of meteorological programmes with complex numeric algorithms and large amount of data, the executing site needs high-capacity hardware. Its efficient use makes great demands on the used data structure, the algorithms and the programming concept. This in the context of the task of SCAI. The project shall contribute to an improvement of the overall efficiency of data assimilation algorithms through a multitude of single problems. The contributions listed here are directly connected to the joint project SACADA. In particular the following aims can be formulated:

Use of a discretisations mesh which avoids unnecessarily high resolutions and at the same time allows for local refinements

  • Reduction of the numerical effort for solving the minimisation problem
  • Elaboration of a parallelisation strategy for the complex assimilation system
  • Evaluation of further solving methods, particularly multigrid approaches