Implementation and evaluation of the LMCSL scheme in Enviro-HIRLAM using data from the ETEX-project – Niels Bohr Institute - University of Copenhagen

Forward this page to a friend Resize Print kalender-ikon Bookmark and Share

Niels Bohr Institute > Calendar > 2009 > Implementation and eva...

Implementation and evaluation of the LMCSL scheme in Enviro-HIRLAM using data from the ETEX-project

Master Thesis by Allan Christensen

The talk will be in Danish.

The semi-Lagrangian approach is widely used in numerical weather prediction because of its stability and accuracy. Though, a traditional semi-Lagrangian approach does not conserve the mass.

The aim of this thesis is to evaluate and implement a locally mass conserving semi-Lagrangian scheme into an existing three dimensional forecast model. The LMCSL scheme, which has been developed by Eigil Kaas, is equivalent to any other cell integrated semi-Lagrangian transport schemes. The scheme uses a bi-cubic interpolation method to find the departure point and keep track of how much each Eulerian cell gives away mass. The scheme will be tested with a High resolution limited area model (HIRLAM) developed at Danish meteorological Institute and to test the scheme the European tracer experiment (ETEX-1) will be used where a chemical inert tracer was emitted to the atmosphere. The given version of HIRLAM has implemented another cell integrated semi-Lagrangian scheme (CISL) and a traditional semi-implicit semi-Lagrangian scheme, which will be compared with the LMCSL scheme using the ETEX-project. The given HIRLAM-version also has a further improvement of the given CISL scheme by using an average pressure to compute an implicit vertical velocity used for the energy conversion term in the discretization of the thermodynamic equation. This will also be implemented and tested for the LMCSL scheme. The conservation of mass has been achived, if one ignores small changes generated by vertical interpolation. By comparison of precipiation and cross sections of specific humidity the CISL scheme seems more local.

Supervisor: Eigil Kaas, NBI