Difference between revisions of "One way nesting"
| Line 1: | Line 1: | ||
The need for grid nesting is greatly reduced with unstructured grid models. | The need for grid nesting is greatly reduced with unstructured grid models. | ||
| − | Occasionally, 1 way nesting may be useful; e.g., you don't usually have velocity b.c. at an open boundary. You can effectively use | + | Occasionally, 1 way nesting may be useful; e.g., you don't usually have velocity b.c. at an open boundary. You can effectively use SCHISM's 1-way nesting capability to get around this issue. |
| − | You can first run the efficient | + | You can first run the efficient SCHISM 2D in a larger or same grid, and then use the 1-way nesting scripts in the source code bundle (v3.2 and up) to extract results from this run and prepare elev3D.th, uv3D.th (if you first run a coarse-grid SCHISM 3D baroclinic simulation, you can also prepare salt3D.th and temp3D.th). Then simple use *3D.th to drive the small-domain 3D runs. Note that this approach only works if the b.c. does not significantly affect the internal dynamics. |
Revision as of 12:37, 18 December 2014
The need for grid nesting is greatly reduced with unstructured grid models.
Occasionally, 1 way nesting may be useful; e.g., you don't usually have velocity b.c. at an open boundary. You can effectively use SCHISM's 1-way nesting capability to get around this issue.
You can first run the efficient SCHISM 2D in a larger or same grid, and then use the 1-way nesting scripts in the source code bundle (v3.2 and up) to extract results from this run and prepare elev3D.th, uv3D.th (if you first run a coarse-grid SCHISM 3D baroclinic simulation, you can also prepare salt3D.th and temp3D.th). Then simple use *3D.th to drive the small-domain 3D runs. Note that this approach only works if the b.c. does not significantly affect the internal dynamics.
