Question on SST nudging

run / config issues
1

Hi everyone,

I am currently working on nudging my CROCO model with GLORYS-based SST data but encountered an issue where GET_TCLIMA is unable to read the temp variable from the climatology forcing file (see attached figure). I created the climatology forcing using the MATLAB make_clim program, and it shares the same dimensions as my restart file, except for the 4th dimension, which represents the monthly climatology.

Screen Shot 2024-09-18 at 10.56.22 AM
Screen Shot 2024-09-18 at 10.56.22 AM1350×1332 115 KB

For tclm_time, I have tried both “360.0 days per year” and real-time values, but the issue persists regardless.

In my cppdefs.h file, these are the options I have enabled. I also enabled FRC_BRY. (I came across a suggestion online that enabling both CLIMATOLOGY and FRC_BRY might cause a conflict, so I tried disabling FRC_BRY, but the issue remained.)

#define CLIMATOLOGY
#ifdef CLIMATOLOGY
#undef ZCLIMATOLOGY
#undef M2CLIMATOLOGY
#undef M3CLIMATOLOGY
#define TCLIMATOLOGY

#undef ZNUDGING
#undef M2NUDGING
#undef M3NUDGING
#define TNUDGING
#define ROBUST_DIAG
#endif

P.S.: When I undefine `CLIMATOLOGY’ flag the model run smoothly.

Has anyone encountered a similar issue? I would greatly appreciate any advice or suggestions!

Thanks,
Xing Zhou

2

you tried SST. I think its temp.
I will just do extrapolation with some reference drivers. 4DVAR.
start with Kriging & IDW.
I know ROBUST_DIAG
Best

3

Hi Smaishal,

Thanks for the quick response. The original data is the GLORYS SST data, and I used make clim to create a NetCDF file that can be read by CROCO. I don’t think it’s a naming issue, as if it a name issue the error would be diffirent, something like couldn’t find the variable “temp.” I’m also unsure if extrapolation is relevant here; it seems more like an I/O problem.

4

@xzhou473 I/O what is it? I don’t know it?
diode and transistor, finding. :slight_smile:
Best,