The Brunt-Vaisala frequency N can now be read from a field in theforcing dataset. To do so N_from_forcing in the linear theory options must be set to true and the name of the field that contains the squared Brunt Vaisala frequency must be set in the ICAR options with the option nvar.
Note: the field in the forcing dataset must store N^2.
The following plot show the resulting difference in the Brunt-Väisälä frequency field due to different calculation methods:
The uppermost panel plots log(N^2) used by ICAR for the case where ICAR calculates it from the state of the atmosphere in the ICAR domain - this is the default behavior so far.
The middle panel plots log(N^2) used by ICAR for the case where it is read from a field in the forcing dataset. This is the result if the options added with these pull-request are activated.
The lower panel shows log(N^2) as stored in a field of the forcing dataset. It is extracted from the grid points closest to the ICAR cross sections shown in the other two panels above.
The center and bottom panel show, as expected, a very close agreement. The remaining differences are explicable due to the horizontal grid points in the forcing and the ICAR simulation not overlapping exactly.
The Brunt-Vaisala frequency N can now be read from a field in theforcing dataset. To do so N_from_forcing in the linear theory options must be set to true and the name of the field that contains the squared Brunt Vaisala frequency must be set in the ICAR options with the option nvar.
Note: the field in the forcing dataset must store N^2.
The following plot show the resulting difference in the Brunt-Väisälä frequency field due to different calculation methods:
The center and bottom panel show, as expected, a very close agreement. The remaining differences are explicable due to the horizontal grid points in the forcing and the ICAR simulation not overlapping exactly.