MONC
Data Types | Modules | Functions/Subroutines
stencil.F90 File Reference

Go to the source code of this file.

Data Types

type  stencil_mod::grid_stencil_type
 Configuration for a specific stencil interpolation to perform. More...
 

Modules

module  stencil_mod
 Performs the interpolation between the primal and dual grids via a stencil approach. For performance reasons, for each field we store the entirety of the y dimension and the number of x slices required by the stencil. Therefore a new interpolation is simply the calculation of one point and reuse of existing computed points (unless this is the first x or y.) The applicable interpolation stenciled data is copied out, with :,1,1 being the central point, minus and plus in each y and x dimension as determined by the stencil size. This is done so that the stencil size can easily be different for each flow field and this is the case with u (where we need u-2 in the X.)
 

Functions/Subroutines

type(grid_stencil_type) function, public stencil_mod::create_stencil (local_grid, fields, nfields, interpolations_to_perform, sizes, xdim, ydim)
 Creates a stencil configuration which will then be used for interpolation. More...
 
subroutine, public stencil_mod::free_stencil (stencil)
 Frees up the memory allocated to a stencil. More...
 
subroutine, public stencil_mod::interpolate_to_dual (local_grid, field, stencil, x, y, interpolated_fields, interpolation_id)
 Interpolates the (vector) flow fields from the primal to dual grid based upon a specific field interpolation. More...
 
subroutine stencil_mod::calculate_interpolated_cell_value (local_grid, j, i, interpolate_in_dimension, field, column_stencil_values)
 Calculates the interpolated values for each point in a column. The exact method of interpolation depends on whether we are interpolating in the w, y or x dimension. More...
 
subroutine stencil_mod::copy_interpolated_to_prognostic (interpolated_fields, stencil, local_grid, x, y, interpolation_id)
 Copies stencil number of interpolated values from the store used by this functionality to preallocated prognostics. THe central point is always :,1,1 in the prognostic, with negative and in the x and y. More...