CROCO output with OpenDrift issues

[carobiomar]

Hi everyone,

I'm Caro, currently working with OpenDrift to simulate particle trajectories using output from the CROCO ocean model. I’ve been testing both the reader_ROMS_native and reader_netCDF_CF_generic, and have already applied several corrections to the CROCO output: renaming scrum_time to time with CF-compliant attributes, adding mask_rho, mask_u, and mask_v, converting them to byte type, and verifying coordinate variables.

Despite these fixes, particles remain stationary during the simulation. The issue seems to be related to vertical interpolation — the horizontal motion appears to be read correctly. Still, no vertical velocity or depth interpolation occurs, and the particles stay in place.

Has anyone successfully used CROCO output with OpenDrift and resolved this issue? Any guidance would be greatly appreciated.

Thanks in advance!

Caro

[knutfrode]

If you can provide a CROCO file, I can have a look.

If the file is smaller than ~ 0.5 MB, we could also include it in the repository for automatic testing for the future.

[carobiomar]

Hi Knut,

Thank you very much for your reply.

We're currently trying to force OpenDrift with CROCO ocean model output, mainly for 3D particle tracking. We've applied several corrections to the CROCO his file, including:

Renaming scrum_time to time with CF-compliant attributes,

Adding and converting mask_rho, mask_u, and mask_v to byte type,

Verifying coordinate consistency.

Despite these adjustments, particles remain stationary throughout the simulation — neither horizontal nor vertical movement occurs. It seems that the fields are not being properly interpolated in the vertical dimension, possibly due to issues reading the sigma-coordinate structure or depth variables from the CROCO file.

To support debugging, I’ve attached a compressed .zip file containing a 2-day CROCO output file. It only has the time dimension reduced (no spatial crop), and includes all original variables.

The zipped file is ~56 KB, well below the 0.5 MB limit.

Thanks again for your time and support!

Best regards,
Caro
croco_his_Y2016M01_cut_date_20-22.zip

[knutfrode]

Sorry for the delayed answer.

I had a look at this file, but the time dimension seems to be cut to zero, thus there is no time at all, and also the current fields (ubar, vbar) are then also empty.
Can you provide instead a CROCO file with e.g. two time steps?