Using xgcm.transform to interpolate to isopycnal space

Contents

Using `xgcm.transform` to interpolate to isopycnal space#

%load_ext watermark

import numpy as np
import xarray as xr
import xgcm

import pop_tools

%watermark -iv

pop_tools: 0.0.post50+dirty
xarray   : 2024.7.0
xgcm     : 0.8.1
numpy    : 2.0.2

Visualize surface density field#

(xds.rho - 1000).isel(z_t=0).plot(robust=True)

<matplotlib.collections.QuadMesh at 0x7f5e6d456ee0>

../_images/f1d6759d8226a404b72973bd4ebbd0ca4dc93566ee53e2719fc9bf561a99d40e.png

Regrid to density space using `grid.transform`#

See https://xgcm.readthedocs.io/en/latest/transform.html for more

regridded = grid.transform(
    xds.TEMP,
    axis="Z",
    target=np.arange(25, 30, 0.5),
    target_data=xds.rho - 1000,
    method="linear",
)

/home/docs/checkouts/readthedocs.org/user_builds/pop-tools/conda/latest/lib/python3.9/site-packages/xgcm/grid.py:989: FutureWarning: From version 0.8.0 the Axis computation methods will be removed, in favour of using the Grid computation methods instead. i.e. use `Grid.transform` instead of `Axis.transform`
  warnings.warn(

/home/docs/checkouts/readthedocs.org/user_builds/pop-tools/conda/latest/lib/python3.9/site-packages/numba/np/ufunc/gufunc.py:252: RuntimeWarning: invalid value encountered in _interp_1d_linear
  return self.ufunc(*args, **kwargs)

Visualize temperature on a single isopycnal surface#

regridded.isel(rho=5).plot(robust=True)

<matplotlib.collections.QuadMesh at 0x7f5e713573d0>

../_images/e51c7a610e7c41bbff69d5ecb8a60321b7e3341341ec10df85248ca2f7d32932.png

Find depth of isopycnal surface#

regridded.coords["z_t"] = grid.transform(
    xds.z_t.broadcast_like(xds.rho),
    axis="Z",
    target=np.arange(24, 30, 0.5),
    target_data=xds.rho - 1000,
    method="linear",
)
regridded.z_t.isel(rho=5).plot(robust=True)

<matplotlib.collections.QuadMesh at 0x7f5e6cd085e0>

../_images/c0fecb8117bd4fe2685453a618e4dcba2d38816bc03862d9f9cecd4bc42f2f95.png