[MITgcm-support] Setting ygOrigin in offline mode for Cartesian grids

[Martin Losch]

Hi Takaya,

in general ygOrigin for a cartesian grid should be in meters as you have done correctly. But the model input files do not know anything about ygOrigin, in fact I think ygOrigin is only needed to compute the coriolis parameter (f = y0 + beta * y). So if your resolution is, say, dx=dy=10km, the you still need to provide fields of size 60 x 40 grid points.

I don’t know much about the flt package, but maybe things are simpler if everything is transformed to ygOrigin = 0?

I would need more details (e.g. namelist files and STDOUT) for more detailed advice.

Martin

> On 7. Jul 2019, at 00:12, Uchida Takaya <tu2140 at columbia.edu> wrote:
> 
> Dear MITgcm support,
> 
> I am currently trying to set up an offline run with Cartesian coordinates in which I use horizontal velocity data from a different run over the model domain of Y=[800,1200] km and X=[1400,2000] km as input.
> The example in the documentation only seems to mention cases on a spherical polar grid with ygOrigin and dySpacing being set in latitude (http://mitgcm.org/public/r2_manual/latest/online_documents/node173.html#SECTION004203100000000000000) but when using a Cartesian grid, could you let me know what I should use as ygOrigin?
> 
> I tried setting ygOrigin=8e5 (800km) but STDOUT indicates the model run is stopping there. If I set ygOrigin=0., the model runs but gives no output even though I’ve set the floater package to give me outputs hourly. My assumption for the latter case is that because I’m setting particles to exist over the domain of Y=[800,1200] km and X=[1400,2000] km, the model thinks that there are no particles when ygOrigin=0. but I may be wrong.
> 
> Thank you,
> Takaya
> _______________________________________________
> MITgcm-support mailing list
> MITgcm-support at mitgcm.org
> http://mailman.mitgcm.org/mailman/listinfo/mitgcm-support