Section interpolation and masking.

[eocene]

Running run_rapidmoc.py on output from various models on their native grids, for some models the program fails and returns:

`

thetao: using mask information from {DATA_LOC}/thetao_RAW.nc.

so: using mask information from {DATA_LOC}/so_RAW.nc.

tauuo: using mask information from {DATA_LOC}/tauuo_RAW.nc.

vo: using mask information from {DATA_LOC}/vo_RAW.nc.

Traceback (most recent call last):

File "{CODE_LOC}/RapidMoc/run_rapidmoc.py", line 16, in <module>

main()

File "{CODE_LOC}/RapidMoc/rapidmoc/rapidmoc.py", line 112, in main

t_on_v = sections.interpolate(t, v)

File "{CODE_LOC}RapidMoc/rapidmoc/sections.py", line 446, in interpolate

sinterp.data = np.ma.MaskedArray(s1.interp_along_section(dist, x0, y0), mask=mask)    

File "{PYTHON_LOC}/lib/python3.9/site-packages/numpy/ma/core.py", line 2909, in __new__

raise MaskError(msg % (nd, nm))

numpy.ma.core.MaskError: Mask and data not compatible: data size is 3061620, mask size is 1270200.

`

Has anyone had experience with this error? It works on some models and not others, but it's not clear why.

Thank you for your help in advance!

[cdr30]

Hi, it looks like the mask and data definitions are incompatible for some reason. By default masks are inferred from the missing data in the files. Could you attach/paste your config file so I can see if there are any obvious issues? Also, could you paste the output of "ncdump -h" applied to each of the input data files.

In the meantime you can try specifying missing data masks manually using the following options (see description of config file in README):

[temperature/salinity/tau/meridional_velocity]
maskf = path to netcdf file containing explicit data mask [string, optional]
maskvar = netcdf variable name for mask in maskf [string, optional]
maskmdi = missing data indicator for maskvar [float, optional]

[eocene]

Hi - thanks so much for the quick response!

Attached are those 5 files:
ncdump_h_tauuo.txt
ncdump_h_vo.txt
ncdump_h_so.txt
ncdump_h_thetao.txt
config.ini.ACCESS-ESM1-5.txt

I'll try the manual masking in the meantime like you suggest!

Thank you very much for your time!

[cdr30]

Hi - it looks like the time dimension is different in the salinity data file (time=370) compared to other files (time=900). This will definitely cause problems as data must be the same shape.

Depending on the grid staggering, you may also want to specify j2 = j1 + 1 for t/s/tau files so that these data are averaged onto the velocity grid (though it probably won't make much difference). See README for more details: "Note that it is possible to specify a range of indices in the y-direction. If j1 != j2, then data is extracted over a range of j-indices and then averaged to create a single zonal section. This allows temperature and salinity to be averaged onto the velocity section in a way that respects the staggering of the original model grid."

There are also example netcdf and config files in the test_data folder if you want to compare with your data to make sure you have the data dimensions/config set consistently.

Chris