You can run this notebook in a live session Binder or view it on Github.

ROMS Ocean Model Example

The Regional Ocean Modeling System (ROMS) is an open source hydrodynamic model that is used for simulating currents and water properties in coastal and estuarine regions. ROMS is one of a few standard ocean models, and it has an active user community.

ROMS uses a regular C-Grid in the horizontal, similar to other structured grid ocean and atmospheric models, and a stretched vertical coordinate (see the ROMS documentation for more details). Both of these require special treatment when using xarray to analyze ROMS ocean model output. This example notebook shows how to create a lazily evaluated vertical coordinate, and make some basic plots. The xgcm package is required to do analysis that is aware of the horizontal C-Grid.

[1]:
import numpy as np
import cartopy.crs as ccrs
import cartopy.feature as cfeature
import matplotlib.pyplot as plt

%matplotlib inline

import xarray as xr

Load a sample ROMS file. This is a subset of a full model available at

http://barataria.tamu.edu/thredds/catalog.html?dataset=txla_hindcast_agg

The subsetting was done using the following command on one of the output files:

#open dataset
ds = xr.open_dataset('/d2/shared/TXLA_ROMS/output_20yr_obc/2001/ocean_his_0015.nc')

# Turn on chunking to activate dask and parallelize read/write.
ds = ds.chunk({'ocean_time': 1})

# Pick out some of the variables that will be included as coordinates
ds = ds.set_coords(['Cs_r', 'Cs_w', 'hc', 'h', 'Vtransform'])

# Select a a subset of variables. Salt will be visualized, zeta is used to
# calculate the vertical coordinate
variables = ['salt', 'zeta']
ds[variables].isel(ocean_time=slice(47, None, 7*24),
                   xi_rho=slice(300, None)).to_netcdf('ROMS_example.nc', mode='w')

So, the ROMS_example.nc file contains a subset of the grid, one 3D variable, and two time steps.

Load in ROMS dataset as an xarray object

[2]:
# load in the file
ds = xr.tutorial.open_dataset("ROMS_example.nc", chunks={"ocean_time": 1})

# This is a way to turn on chunking and lazy evaluation. Opening with mfdataset, or
# setting the chunking in the open_dataset would also achieve this.
ds
---------------------------------------------------------------------------
PermissionError                           Traceback (most recent call last)
File /usr/lib/python3/dist-packages/pooch/utils.py:262, in make_local_storage(path, env)
    258 if action == "create":
    259     # When running in parallel, it's possible that multiple jobs will
    260     # try to create the path at the same time. Use exist_ok to avoid
    261     # raising an error.
--> 262     os.makedirs(path, exist_ok=True)
    263 else:

File /usr/lib/python3.12/os.py:215, in makedirs(name, mode, exist_ok)
    214 try:
--> 215     makedirs(head, exist_ok=exist_ok)
    216 except FileExistsError:
    217     # Defeats race condition when another thread created the path

File /usr/lib/python3.12/os.py:215, in makedirs(name, mode, exist_ok)
    214 try:
--> 215     makedirs(head, exist_ok=exist_ok)
    216 except FileExistsError:
    217     # Defeats race condition when another thread created the path

File /usr/lib/python3.12/os.py:225, in makedirs(name, mode, exist_ok)
    224 try:
--> 225     mkdir(name, mode)
    226 except OSError:
    227     # Cannot rely on checking for EEXIST, since the operating system
    228     # could give priority to other errors like EACCES or EROFS

PermissionError: [Errno 13] Permission denied: '/sbuild-nonexistent'

The above exception was the direct cause of the following exception:

PermissionError                           Traceback (most recent call last)
Cell In[2], line 2
      1 # load in the file
----> 2 ds = xr.tutorial.open_dataset("ROMS_example.nc", chunks={"ocean_time": 1})
      4 # This is a way to turn on chunking and lazy evaluation. Opening with mfdataset, or
      5 # setting the chunking in the open_dataset would also achieve this.
      6 ds

File /usr/lib/python3/dist-packages/xarray/tutorial.py:161, in open_dataset(name, cache, cache_dir, engine, **kws)
    158     url = f"{base_url}/raw/{version}/{path.name}"
    160 # retrieve the file
--> 161 filepath = pooch.retrieve(url=url, known_hash=None, path=cache_dir)
    162 ds = _open_dataset(filepath, engine=engine, **kws)
    163 if not cache:

File /usr/lib/python3/dist-packages/pooch/core.py:227, in retrieve(url, known_hash, fname, path, processor, downloader, progressbar)
    222 action, verb = download_action(full_path, known_hash)
    224 if action in ("download", "update"):
    225     # We need to write data, so create the local data directory if it
    226     # doesn't already exist.
--> 227     make_local_storage(path)
    229     get_logger().info(
    230         "%s data from '%s' to file '%s'.",
    231         verb,
    232         url,
    233         str(full_path),
    234     )
    236     if downloader is None:

File /usr/lib/python3/dist-packages/pooch/utils.py:276, in make_local_storage(path, env)
    272 if env is not None:
    273     message.append(
    274         f"Use environment variable '{env}' to specify a different location."
    275     )
--> 276 raise PermissionError(" ".join(message)) from error

PermissionError: [Errno 13] Permission denied: '/sbuild-nonexistent' | Pooch could not create data cache folder '/sbuild-nonexistent/.cache/xarray_tutorial_data'. Will not be able to download data files.

Add a lazilly calculated vertical coordinates

Write equations to calculate the vertical coordinate. These will be only evaluated when data is requested. Information about the ROMS vertical coordinate can be found (here)[https://www.myroms.org/wiki/Vertical_S-coordinate]

In short, for Vtransform==2 as used in this example,

\(Z_0 = (h_c \, S + h \,C) / (h_c + h)\)

\(z = Z_0 (\zeta + h) + \zeta\)

where the variables are defined as in the link above.

[3]:
if ds.Vtransform == 1:
    Zo_rho = ds.hc * (ds.s_rho - ds.Cs_r) + ds.Cs_r * ds.h
    z_rho = Zo_rho + ds.zeta * (1 + Zo_rho / ds.h)
elif ds.Vtransform == 2:
    Zo_rho = (ds.hc * ds.s_rho + ds.Cs_r * ds.h) / (ds.hc + ds.h)
    z_rho = ds.zeta + (ds.zeta + ds.h) * Zo_rho

ds.coords["z_rho"] = z_rho.transpose()  # needing transpose seems to be an xarray bug
ds.salt
---------------------------------------------------------------------------
NameError                                 Traceback (most recent call last)
Cell In[3], line 1
----> 1 if ds.Vtransform == 1:
      2     Zo_rho = ds.hc * (ds.s_rho - ds.Cs_r) + ds.Cs_r * ds.h
      3     z_rho = Zo_rho + ds.zeta * (1 + Zo_rho / ds.h)

NameError: name 'ds' is not defined

A naive vertical slice

Creating a slice using the s-coordinate as the vertical dimension is typically not very informative.

[4]:
ds.salt.isel(xi_rho=50, ocean_time=0).plot()
---------------------------------------------------------------------------
NameError                                 Traceback (most recent call last)
Cell In[4], line 1
----> 1 ds.salt.isel(xi_rho=50, ocean_time=0).plot()

NameError: name 'ds' is not defined

We can feed coordinate information to the plot method to give a more informative cross-section that uses the depths. Note that we did not need to slice the depth or longitude information separately, this was done automatically as the variable was sliced.

[5]:
section = ds.salt.isel(xi_rho=50, eta_rho=slice(0, 167), ocean_time=0)
section.plot(x="lon_rho", y="z_rho", figsize=(15, 6), clim=(25, 35))
plt.ylim([-100, 1]);
---------------------------------------------------------------------------
NameError                                 Traceback (most recent call last)
Cell In[5], line 1
----> 1 section = ds.salt.isel(xi_rho=50, eta_rho=slice(0, 167), ocean_time=0)
      2 section.plot(x="lon_rho", y="z_rho", figsize=(15, 6), clim=(25, 35))
      3 plt.ylim([-100, 1]);

NameError: name 'ds' is not defined

A plan view

Now make a naive plan view, without any projection information, just using lon/lat as x/y. This looks OK, but will appear compressed because lon and lat do not have an aspect constrained by the projection.

[6]:
ds.salt.isel(s_rho=-1, ocean_time=0).plot(x="lon_rho", y="lat_rho")
---------------------------------------------------------------------------
NameError                                 Traceback (most recent call last)
Cell In[6], line 1
----> 1 ds.salt.isel(s_rho=-1, ocean_time=0).plot(x="lon_rho", y="lat_rho")

NameError: name 'ds' is not defined

And let’s use a projection to make it nicer, and add a coast.

[7]:
proj = ccrs.LambertConformal(central_longitude=-92, central_latitude=29)
fig = plt.figure(figsize=(15, 5))
ax = plt.axes(projection=proj)
ds.salt.isel(s_rho=-1, ocean_time=0).plot(
    x="lon_rho", y="lat_rho", transform=ccrs.PlateCarree()
)

coast_10m = cfeature.NaturalEarthFeature(
    "physical", "land", "10m", edgecolor="k", facecolor="0.8"
)
ax.add_feature(coast_10m)
---------------------------------------------------------------------------
NameError                                 Traceback (most recent call last)
Cell In[7], line 4
      2 fig = plt.figure(figsize=(15, 5))
      3 ax = plt.axes(projection=proj)
----> 4 ds.salt.isel(s_rho=-1, ocean_time=0).plot(
      5     x="lon_rho", y="lat_rho", transform=ccrs.PlateCarree()
      6 )
      8 coast_10m = cfeature.NaturalEarthFeature(
      9     "physical", "land", "10m", edgecolor="k", facecolor="0.8"
     10 )
     11 ax.add_feature(coast_10m)

NameError: name 'ds' is not defined
../_images/examples_ROMS_ocean_model_15_1.png
[ ]: