Preparing boundary conditions for CCS1

Generation of Lateral Open Boundary Conditions for CCS1

Open boundary conditions for CCS1 are generated using the brushcutter Python package. Follow the link for installation instructions.

The preprocessing script is here

Script breakdown

For this example, OBC data are generated using output from a global MOM6 simulation. The following outlines the full script using sea surface height as an example

1. Download input data

2. Start a Python session and import required modules

   from brushcutter import lib_obc_variable as lov

   from brushcutter import lib_obc_segments as los

   from brushcutter import lib_ioncdf as ncdf

   import netCDF4 as nc

   import numpy as np

   import subprocess as sp

3. Define the source and target grids

   momgrd = 'ocean_hgrid.nc'

   srcgrd = 'input_data/ocean_hgrid.nc'

4. Define the source sea surface height data

   ssh_data = 'input_data/20140101.ocean_month.nc'

5. Define the boundary using the los.obc_segment method

   south = los.obc_segment('segment_001',momgrd,istart=0,iend=360,jstart=0,jend=0)

   The label attached to this segment is segment_001, it can be anything you choose. momgrd is the path to the regional model FMS supergrid file. In this example, the model supergrid size equals 360x960 (a reminder that the supergrid has double the refinement of the tracer grid) and the boundary is defined along the south edge of the domain.

6. Define the sea surface height (zeta) field to be interpolated to the boundary

   zeta_south = lov.obc_variable(south,'zeta',geometry='line',\

   obctype='flather',use_locstream=True)

   zeta will be the name given to the interpolated field in the file

7. Load the source grid using the source supergrid file

   xsrc=nc.Dataset(srcgrd).variables['x'][1::2,1::2]

   ysrc=nc.Dataset(srcgrd).variables['y'][1::2,1::2]

8. Now interpolate sea surface heights from the source model and save each time level in a separate file

   for k in np.arange(12):

   zeta_south.interpolate_from(ssh_data,'ssh',frame=k,from_global=True,\

   x_coords=xsrc,y_coords=ysrc, drown='ncl')

   time = zeta_south.timesrc

   time.calendar = nc.Dataset(ssh_data).variables['time'].calendar

   ncdf.write_obc_file([south],[zeta_south],[],time,output='obc_CCS1_'+str(k).zfill(2)+'.nc')

9. Combine individual netCDF files

   cmd = 'ncrcat obc_CCS1_??.nc -O -o obc_CCS1.nc' sp.call(cmd,shell=True)

10. Add the modulo attribute to the time axis

    cmd = "ncatted -h -a modulo,time,c,c,' ' obc_CCS1.nc" sp.call(cmd,shell=True)

Here is a visualization of the resulting surface heights at the CCS1 domain boundaries.