Changes between Version 143 and Version 144 of doc/app/initialization_parameters
- Timestamp:
- Jan 3, 2011 3:58:01 PM (14 years ago)
Legend:
- Unmodified
- Added
- Removed
- Modified
-
doc/app/initialization_parameters
v143 v144 760 760 The 5 th order upwind scheme of Wicker and Skamarock (2002, Mon. Wea. Rev, 130, 2088-2097) is used. The dispersion error is much smaller than the dispersion error of 'pw-scheme'. The 5th order schemes implies a small numerical dissipation that stabilized the solution. To assure a stable numerical solution the time integration has to be carry out with [#timestep_scheme timestep_scheme] = 'runge-kutta-3'. This scheme is based on a formultion of the advection term in flux form, which requires a vanishing divergence of the flow field, else a stable numerical solution is not given. So [#call_psolver_at_all_substeps call_psolver_at_all_substeps] = 'True' has to be used. When [#psolver psolver] = 'multigrid' the [#residual_limit residual_limit] should be smaller than 1.0E-6 for a sufficient reduction of the divergences. For the moment the scheme can only be used in conjunction with [#topography topography] = 'flat'. So far [#bc_lr bc_lr] /= 'cyclic' or [#bc_ns bc_ns] /= 'cyclic' are not allowed with [#loop_optimization loop_optimization] = 'vector' in combination with 'ws-scheme'. 761 761 762 Note: Due to the larger stencil of this scheme vertical grid stretching should be handled with care.\\\\ 762 '''Note''': Due to the larger stencil of this scheme vertical grid stretching should be handled with care. 763 The computation of turbulent fluxes takes place inside the advection routines to get a statistical evaluation consistent to the numerical solution.\\\\ 763 764 '' 'pw-scheme' ''\\\\ 764 765 The scheme of Piascek and Williams (1970, J. Comp. Phys., 6, 392-405) with central differences in the form C3 is used. … … 953 954 The 5th order upwind scheme of Wicker and Skamarock (2002, Mon. Wea. Rev, 130, 2088-2097) is used. The dispersion error is much smaller than the dispersion error of 'pw-scheme'. The 5th order schemes implies a small numerical dissipation that stabilized the solution. To assure a stable numerical solution the time integration has to be carry out with [#timestep_scheme timestep_scheme] = 'runge-kutta-3' . This scheme is based on a formultion of the advection term in flux form, which requires a vanishing divergence of the flow field, else a stable numerical solution is not given. So [#call_psolver_at_all_substeps call_psolver_at_all_substeps] = 'True' has to be used. When [#psolver psolver] = 'multigrid' the [#residual_limit residual_limit] should be smaller than 1.0E-6 for a sufficient reduction of the divergences. For the moment the scheme can only be used in conjunction with [#topography topography] = 'flat'. So far [#bc_lr bc_lr] /= 'cyclic' or [#bc_ns bc_ns] /= 'cyclic' are not allowed with [#loop_optimization loop_optimization] = 'vector' in combination with 'ws-scheme'. 954 955 955 Note: Due to the larger stencil of this scheme vertical grid stretching should be handled with care.\\\\ 956 '''Note''': Due to the larger stencil of this scheme vertical grid stretching should be handled with care. 957 The computation of turbulent fluxes takes place inside the advection routines to get a statistical evaluation consistent to the numerical solution.\\\\ 956 958 '' 'pw-scheme' ''\\\\ 957 959 The scheme of Piascek and Williams (1970, J. Comp. Phys., 6, 392-405) with central differences in the form C3 is used.
