Virtual Data Set Starter Notebook

Author: Dean Henze, PO.DAAC

Demonstrates basic functionality of virtual data sets (VDS’s). Meant to serve as a modifiable template to explore data sets via this technology.

Sections

1. Obtaining credentials and generating a VDS mapper: Boiler-plate code to get Earthdata Login credentials and apply them to a VDS.
2. Load data, explore, and do computations: Shows three examples of opening a VDS with Xarray, then exploring and doing computations with Xarray built-in functions.
3. Alternate workflows other than Xarray built-in functions: For those who preffer to work in packages other than Xarray, this section shows briefly how to either (a) extract NumPy/Dask arrays from the Xarray Dataset object, or (b) save a subset of the VDS as a local netCDF file.

Keep in mind: For out-of-cloud use, you may run into access errors if trying to do computations with more than ~20 years of data due to limitations on the number of simultaneous data requests that can be made at once. For complex or large computations with VDS’s, please be in the cloud in the same location as the data (for PO.DAAC this is us-west-2).

Environment

Minimal working installation for Python 3.13.3 environment is

fsspec==2024.10.0
earthaccess==0.11.0
virtualizarr==1.2.0
xarray==2025.1.2
zarr==2.18.4
kerchunk==0.2.7
dask==2024.5.2
h5netcdf==1.3.0
netCDF4==1.7.2
matplotlib==3.9.2
jupyterlab==4.4.2

# Filesystem management 
import fsspec
import earthaccess

# Data handling
import numpy as np
import xarray as xr

# Other
import matplotlib.pyplot as plt

%matplotlib inline

1. Obtaining credentials and generating a VDS mapper

Similarly to any access method for NASA Earthdata, access via VDS’s still requires Earthdata Login credentials. Fortunately earthaccess makes this a one-liner. Rerun this line each hour to renew your credentials. If you are doing computations in later sections and suddenly hit access errors, you likely need to rerun this line.

earthaccess.login()

<earthaccess.auth.Auth at 0x104a4d550>

The remaining task of defining some access keywords and mapper to the VDS is generic and can be wrapped into a function (no modification needed). The mapper will contain all the required access credentials and can be passed directly to xarray. In the future this step will likely be built into to earthaccess but for now we must define it in the notebook. The only inputs to the function are:

1. The link to the VDS reference file.
2. Whether or not you are accessing the data in the cloud in the same region as the data. For most beginning users this argument will be set to False.

def get_vds_mapper(vds_link, in_cloud_region=False):
    """
    Produces a virtudal dataset mapper that can be passed to xarray. 
    
    * vds_link: str, link to the mapper
    * in_cloud_region: bool, True if in cloud in the same region as the data, 
        False otherwise.
    """

    if in_cloud_region:
        fs_data = earthaccess.get_s3_filesystem(daac="PODAAC")
        remote_protocol = "s3"
    else:
        fs_data = earthaccess.get_fsspec_https_session()
        remote_protocol = "https"
    
    storage_opts = {"fo": vds_link, "remote_protocol": remote_protocol, "remote_options": fs_data.storage_options}
    fs_ref = fsspec.filesystem('reference', **storage_opts)
    return fs_ref.get_mapper('')

2. Load data, explore, and do computations

The protocol to open the data will be identical for each collection. This section demonstrates the procedure for three data sets, then does some basic computations.

The key action item is locating the VDS file link. In the future there will pathways to discover this information via data set landing pages and CMR metadata searches, but for now, please consult the table of VDS reference file links. If you are out of cloud, make sure the VDS link is for https endpoints and not s3 endpoints.

Loading the data is a one-liner using Xarray. At this point, the ability to utilize the data is limited only by the user’s knowledge of the Xarray package, or subsetting and loading a portion of the data into your package of choice.

2.1 OSTIA-UKMO-L4-GLOB-REP-v2.0

Level 4 sea surface temperature data set, one timestamp per file.

Load data

## Modify vds_link for your desired data set. If in-cloud, set in_cloud_region to True:
vds_link = "https://archive.podaac.earthdata.nasa.gov/podaac-ops-cumulus-public/virtual_collections/OSTIA-UKMO-L4-GLOB-REP-v2.0/OSTIA-UKMO-L4-GLOB-REP-v2.0_virtual_https.json"
vds_mapper = get_vds_mapper(vds_link, in_cloud_region=False)

%%time

## No modification needed!
data = xr.open_dataset(
    vds_mapper, engine="zarr", chunks={},
    backend_kwargs={"consolidated": False}
)
data

CPU times: user 465 ms, sys: 45.4 ms, total: 510 ms
Wall time: 550 ms

<xarray.Dataset> Size: 11TB
Dimensions:           (time: 15340, lat: 3600, lon: 7200)
Coordinates:
  * lat               (lat) float32 14kB -89.97 -89.93 -89.88 ... 89.93 89.97
  * lon               (lon) float32 29kB -180.0 -179.9 -179.9 ... 179.9 180.0
  * time              (time) datetime64[ns] 123kB 1982-01-01T12:00:00 ... 202...
Data variables:
    analysed_sst      (time, lat, lon) float64 3TB dask.array<chunksize=(1, 1200, 2400), meta=np.ndarray>
    analysis_error    (time, lat, lon) float64 3TB dask.array<chunksize=(1, 1200, 2400), meta=np.ndarray>
    mask              (time, lat, lon) float32 2TB dask.array<chunksize=(1, 1800, 3600), meta=np.ndarray>
    sea_ice_fraction  (time, lat, lon) float64 3TB dask.array<chunksize=(1, 1800, 3600), meta=np.ndarray>
Attributes: (12/39)
    Conventions:                CF-1.4, ACDD-1.3
    title:                      Global SST & Sea Ice Analysis, L4 OSTIA, 0.05...
    summary:                    A merged, multi-sensor L4 Foundation SST product
    references:                 Donlon, C.J., Martin, M., Stark, J.D., Robert...
    institution:                UKMO
    history:                    Created from sst.nc; obs_anal.nc; seaice.nc
    ...                         ...
    project:                    Group for High Resolution Sea Surface Tempera...
    publisher_name:             GHRSST Project Office
    publisher_url:              http://www.ghrsst.org
    publisher_email:            ghrsst-po@nceo.ac.uk
    processing_level:           L4
    cdm_data_type:              grid

xarray.Dataset

• Dimensions:
  • time: 15340
  • lat: 3600
  • lon: 7200
• Coordinates: (3)
  • lat
    (lat)
    float32
    -89.97 -89.93 ... 89.93 89.97

    standard_name :

    latitude

    long_name :

    latitude

    units :

    degrees_north

    valid_min :

    -90.0

    valid_max :

    90.0

    axis :

    Y

    comment :

    Latitude geographical coordinates,WGS84 projection

    chunksizes :

    [3600]

    fletcher32 :

    False

    shuffle :

    True

    preferred_chunks :

    {'lat': 3600}

    zlib :

    True

    complevel :

    1

    source :

    <File-like object HTTPFileSystem, https://archive.podaac.earthdata.nasa.gov/podaac-ops-cumulus-protected/OSTIA-UKMO-L4-GLOB-REP-v2.0/1982/001/19820101120000-UKMO-L4_GHRSST-SSTfnd-OSTIA-GLOB_REP-v02.0-fv02.0.nc>

    original_shape :

    [3600]

    dtype :

    float32

    array([-89.975, -89.925, -89.875, ...,  89.875,  89.925,  89.975],
          shape=(3600,), dtype=float32)

  • lon
    (lon)
    float32
    -180.0 -179.9 ... 179.9 180.0

    standard_name :

    longitude

    long_name :

    longitude

    units :

    degrees_east

    valid_min :

    -180.0

    valid_max :

    180.0

    axis :

    X

    comment :

    Longitude geographical coordinates,WGS84 projection

    chunksizes :

    [7200]

    fletcher32 :

    False

    shuffle :

    True

    preferred_chunks :

    {'lon': 7200}

    zlib :

    True

    complevel :

    1

    source :

    <File-like object HTTPFileSystem, https://archive.podaac.earthdata.nasa.gov/podaac-ops-cumulus-protected/OSTIA-UKMO-L4-GLOB-REP-v2.0/1982/001/19820101120000-UKMO-L4_GHRSST-SSTfnd-OSTIA-GLOB_REP-v02.0-fv02.0.nc>

    original_shape :

    [7200]

    dtype :

    float32

    array([-179.975, -179.925, -179.875, ...,  179.875,  179.925,  179.975],
          shape=(7200,), dtype=float32)

  • time
    (time)
    datetime64[ns]
    1982-01-01T12:00:00 ... 2023-12-...

    long_name :

    reference time of sst field

    standard_name :

    time

    axis :

    T

    comment :

    chunksizes :

    [1]

    fletcher32 :

    False

    shuffle :

    True

    preferred_chunks :

    {'time': 1}

    zlib :

    True

    complevel :

    1

    source :

    <File-like object HTTPFileSystem, https://archive.podaac.earthdata.nasa.gov/podaac-ops-cumulus-protected/OSTIA-UKMO-L4-GLOB-REP-v2.0/1982/001/19820101120000-UKMO-L4_GHRSST-SSTfnd-OSTIA-GLOB_REP-v02.0-fv02.0.nc>

    original_shape :

    [1]

    dtype :

    int32

    units :

    seconds since 1981-01-01 00:00:00

    _FillValue :

    1970-01-01T00:00:00.000000000

    array(['1982-01-01T12:00:00.000000000', '1982-01-02T12:00:00.000000000',
           '1982-01-03T12:00:00.000000000', ..., '2023-12-29T12:00:00.000000000',
           '2023-12-30T12:00:00.000000000', '2023-12-31T12:00:00.000000000'],
          shape=(15340,), dtype='datetime64[ns]')

• Data variables: (4)
  • analysed_sst
    (time, lat, lon)
    float64
    dask.array<chunksize=(1, 1200, 2400), meta=np.ndarray>

    long_name :

    analysed sea surface temperature

    standard_name :

    sea_surface_foundation_temperature

    units :

    kelvin

    valid_min :

    -300

    valid_max :

    4500

    reference :

    C.J. Donlon, M. Martin,J.D. Stark, J. Roberts-Jones, E. Fiedler, W. Wimmer. The operational sea surface temperature and sea ice analysis (OSTIA) system. Remote Sensing Environ., 116 (2012), pp. 140-158 http://dx.doi.org/10.1016/j.rse.2010.10.017

    comment :

    OSTIA foundation SST

    Array Chunk Bytes 2.89 TiB 21.97 MiB Shape (15340, 3600, 7200) (1, 1200, 2400) Dask graph 138060 chunks in 2 graph layers Data type float64 numpy.ndarray

  • analysis_error
    (time, lat, lon)
    float64
    dask.array<chunksize=(1, 1200, 2400), meta=np.ndarray>

    long_name :

    estimated error standard deviation of analysed_sst

    standard_name :

    sea_surface_foundation_temperature standard_error

    units :

    kelvin

    valid_min :

    0

    valid_max :

    32767

    comment :

    OSTIA foundation SST analysis standard deviation error

    Array Chunk Bytes 2.89 TiB 21.97 MiB Shape (15340, 3600, 7200) (1, 1200, 2400) Dask graph 138060 chunks in 2 graph layers Data type float64 numpy.ndarray

  • mask
    (time, lat, lon)
    float32
    dask.array<chunksize=(1, 1800, 3600), meta=np.ndarray>

    long_name :

    land sea ice lake bit mask

    valid_min :

    1

    valid_max :

    31

    flag_masks :

    [1, 2, 4, 8, 16]

    flag_meanings :

    water land optional_lake_surface sea_ice optional_river_surface

    source :

    NAVOCEANO_landmask_v1.0 EUMETSAT_OSI-SAF_icemask ARCLake_lakemask

    comment :

    Land/ open ocean/ sea ice /lake mask

    Array Chunk Bytes 1.45 TiB 24.72 MiB Shape (15340, 3600, 7200) (1, 1800, 3600) Dask graph 61360 chunks in 2 graph layers Data type float32 numpy.ndarray

  • sea_ice_fraction
    (time, lat, lon)
    float64
    dask.array<chunksize=(1, 1800, 3600), meta=np.ndarray>

    long_name :

    sea ice area fraction

    standard_name :

    sea_ice_area_fraction

    units :

    1

    valid_min :

    0

    valid_max :

    100

    source :

    EUMETSAT OSI-SAF

    comment :

    Sea ice area fraction

    Array Chunk Bytes 2.89 TiB 49.44 MiB Shape (15340, 3600, 7200) (1, 1800, 3600) Dask graph 61360 chunks in 2 graph layers Data type float64 numpy.ndarray

• Indexes: (3)
  • lat
    PandasIndex

    PandasIndex(Index([ -89.9749984741211, -89.92500305175781,            -89.875,
           -89.82499694824219,  -89.7750015258789,  -89.7249984741211,
           -89.67500305175781,            -89.625, -89.57499694824219,
            -89.5250015258789,
           ...
             89.5250015258789,  89.57499694824219,             89.625,
            89.67500305175781,   89.7249984741211,   89.7750015258789,
            89.82499694824219,             89.875,  89.92500305175781,
             89.9749984741211],
          dtype='float32', name='lat', length=3600))

  • lon
    PandasIndex

    PandasIndex(Index([-179.97500610351562,  -179.9250030517578,            -179.875,
            -179.8249969482422, -179.77499389648438, -179.72500610351562,
            -179.6750030517578,            -179.625,  -179.5749969482422,
           -179.52499389648438,
           ...
            179.52499389648438,   179.5749969482422,             179.625,
             179.6750030517578,  179.72500610351562,  179.77499389648438,
             179.8249969482422,             179.875,   179.9250030517578,
            179.97500610351562],
          dtype='float32', name='lon', length=7200))

  • time
    PandasIndex

    PandasIndex(DatetimeIndex(['1982-01-01 12:00:00', '1982-01-02 12:00:00',
                   '1982-01-03 12:00:00', '1982-01-04 12:00:00',
                   '1982-01-05 12:00:00', '1982-01-06 12:00:00',
                   '1982-01-07 12:00:00', '1982-01-08 12:00:00',
                   '1982-01-09 12:00:00', '1982-01-10 12:00:00',
                   ...
                   '2023-12-22 12:00:00', '2023-12-23 12:00:00',
                   '2023-12-24 12:00:00', '2023-12-25 12:00:00',
                   '2023-12-26 12:00:00', '2023-12-27 12:00:00',
                   '2023-12-28 12:00:00', '2023-12-29 12:00:00',
                   '2023-12-30 12:00:00', '2023-12-31 12:00:00'],
                  dtype='datetime64[ns]', name='time', length=15340, freq=None))

• Attributes: (39)

  Conventions :

  CF-1.4, ACDD-1.3

  title :

  Global SST & Sea Ice Analysis, L4 OSTIA, 0.05 deg daily (METOFFICE-GLO-SST-L4-REP-OBS-SST-V2)

  summary :

  A merged, multi-sensor L4 Foundation SST product

  references :

  Donlon, C.J., Martin, M., Stark, J.D., Roberts-Jones, J., Fiedler, E., Wimmer, W., 2011. The Operational Sea Surface Temperature and Sea Ice Analysis (OSTIA). Remote Sensing of the Environment

  institution :

  UKMO

  history :

  Created from sst.nc; obs_anal.nc; seaice.nc

  comment :

  WARNING Some applications are unable to properly handle signed byte values. If values are encountered > 127, please subtract 256 from this reported value

  license :

  These data are available free of charge under the CMEMS data policy

  id :

  OSTIA-UKMO-L4-GLOB-v2.0

  naming_authority :

  org.ghrsst

  product_version :

  3.2

  uuid :

  a2df4a18-6f19-4772-9532-39307a0e2794

  gds_version_id :

  2.4

  netcdf_version_id :

  4.1

  file_quality_level :

  3

  Metadata_Conventions :

  Unidata Observation Dataset v1.0

  metadata_link :

  http://podaac.jpl.nasa.gov/ws/metadata/dataset?format=gcmd&shortName=UKMO-L4HRfnd-GLOB-OSTIA

  keywords :

  Oceans > Ocean Temperature > Sea Surface Temperature

  keywords_vocabulary :

  NASA Global Change Master Directory (GCMD) Science Keywords

  standard_name_vocabulary :

  NetCDF Climate and Forecast (CF) Metadata Convention

  westernmost_longitude :

  -180.0

  easternmost_longitude :

  180.00001525878906

  southernmost_latitude :

  -90.0

  northernmost_latitude :

  90.0

  spatial_resolution :

  0.05 degree

  geospatial_lat_units :

  degrees_north

  geospatial_lat_resolution :

  0.05000000074505806

  geospatial_lon_units :

  degrees_east

  geospatial_lon_resolution :

  0.05000000074505806

  acknowledgment :

  Please acknowledge the use of these data with the following statement: These data were provided by GHRSST, Met Office and CMEMS

  creator_name :

  Met Office as part of CMEMS

  creator_email :

  servicedesk.cmems@mercator-ocean.eu

  creator_url :

  http://marine.copernicus.eu

  project :

  Group for High Resolution Sea Surface Temperature

  publisher_name :

  GHRSST Project Office

  publisher_url :

  http://www.ghrsst.org

  publisher_email :

  ghrsst-po@nceo.ac.uk

  processing_level :

  L4

  cdm_data_type :

  grid

Plot global sea surface temperature for first timestep

%%time

sst_t0 = data["analysed_sst"].isel(time=0)
plt.figure()
plt.contourf(sst_t0["lon"], sst_t0["lat"], sst_t0.values, levels=np.arange(270, 310, 3))
plt.colorbar()

CPU times: user 2.84 s, sys: 288 ms, total: 3.13 s
Wall time: 5.6 s

Time series of regional mean sea surface temperature

Subset to 10 x 10 degree window off the USA northwest coast, compute mean over this window for each timestep. Currently the time bounds are set for the first decade of data.

time_bounds = ("1982-01-02T00:00:00", "1992-01-02T00:00:00")
lon_bounds = (-135,-125)  # for -180 to 180 domain
lat_bounds = (35, 45)

data_subset = data.sel(time=slice(*time_bounds)).sel(lon=slice(*lon_bounds), lat=slice(*lat_bounds))
data_subset

<xarray.Dataset> Size: 4GB
Dimensions:           (time: 3652, lat: 200, lon: 200)
Coordinates:
  * lat               (lat) float32 800B 35.03 35.08 35.12 ... 44.88 44.92 44.97
  * lon               (lon) float32 800B -135.0 -134.9 -134.9 ... -125.1 -125.0
  * time              (time) datetime64[ns] 29kB 1982-01-02T12:00:00 ... 1992...
Data variables:
    analysed_sst      (time, lat, lon) float64 1GB dask.array<chunksize=(1, 200, 200), meta=np.ndarray>
    analysis_error    (time, lat, lon) float64 1GB dask.array<chunksize=(1, 200, 200), meta=np.ndarray>
    mask              (time, lat, lon) float32 584MB dask.array<chunksize=(1, 200, 200), meta=np.ndarray>
    sea_ice_fraction  (time, lat, lon) float64 1GB dask.array<chunksize=(1, 200, 200), meta=np.ndarray>
Attributes: (12/39)
    Conventions:                CF-1.4, ACDD-1.3
    title:                      Global SST & Sea Ice Analysis, L4 OSTIA, 0.05...
    summary:                    A merged, multi-sensor L4 Foundation SST product
    references:                 Donlon, C.J., Martin, M., Stark, J.D., Robert...
    institution:                UKMO
    history:                    Created from sst.nc; obs_anal.nc; seaice.nc
    ...                         ...
    project:                    Group for High Resolution Sea Surface Tempera...
    publisher_name:             GHRSST Project Office
    publisher_url:              http://www.ghrsst.org
    publisher_email:            ghrsst-po@nceo.ac.uk
    processing_level:           L4
    cdm_data_type:              grid

xarray.Dataset

• Dimensions:
  • time: 3652
  • lat: 200
  • lon: 200
• Coordinates: (3)
  • lat
    (lat)
    float32
    35.03 35.08 35.12 ... 44.92 44.97

    standard_name :

    latitude

    long_name :

    latitude

    units :

    degrees_north

    valid_min :

    -90.0

    valid_max :

    90.0

    axis :

    Y

    comment :

    Latitude geographical coordinates,WGS84 projection

    chunksizes :

    [3600]

    fletcher32 :

    False

    shuffle :

    True

    preferred_chunks :

    {'lat': 3600}

    zlib :

    True

    complevel :

    1

    source :

    <File-like object HTTPFileSystem, https://archive.podaac.earthdata.nasa.gov/podaac-ops-cumulus-protected/OSTIA-UKMO-L4-GLOB-REP-v2.0/1982/001/19820101120000-UKMO-L4_GHRSST-SSTfnd-OSTIA-GLOB_REP-v02.0-fv02.0.nc>

    original_shape :

    [3600]

    dtype :

    float32

    array([35.025, 35.075, 35.125, 35.175, 35.225, 35.275, 35.325, 35.375, 35.425,
           35.475, 35.525, 35.575, 35.625, 35.675, 35.725, 35.775, 35.825, 35.875,
           35.925, 35.975, 36.025, 36.075, 36.125, 36.175, 36.225, 36.275, 36.325,
           36.375, 36.425, 36.475, 36.525, 36.575, 36.625, 36.675, 36.725, 36.775,
           36.825, 36.875, 36.925, 36.975, 37.025, 37.075, 37.125, 37.175, 37.225,
           37.275, 37.325, 37.375, 37.425, 37.475, 37.525, 37.575, 37.625, 37.675,
           37.725, 37.775, 37.825, 37.875, 37.925, 37.975, 38.025, 38.075, 38.125,
           38.175, 38.225, 38.275, 38.325, 38.375, 38.425, 38.475, 38.525, 38.575,
           38.625, 38.675, 38.725, 38.775, 38.825, 38.875, 38.925, 38.975, 39.025,
           39.075, 39.125, 39.175, 39.225, 39.275, 39.325, 39.375, 39.425, 39.475,
           39.525, 39.575, 39.625, 39.675, 39.725, 39.775, 39.825, 39.875, 39.925,
           39.975, 40.025, 40.075, 40.125, 40.175, 40.225, 40.275, 40.325, 40.375,
           40.425, 40.475, 40.525, 40.575, 40.625, 40.675, 40.725, 40.775, 40.825,
           40.875, 40.925, 40.975, 41.025, 41.075, 41.125, 41.175, 41.225, 41.275,
           41.325, 41.375, 41.425, 41.475, 41.525, 41.575, 41.625, 41.675, 41.725,
           41.775, 41.825, 41.875, 41.925, 41.975, 42.025, 42.075, 42.125, 42.175,
           42.225, 42.275, 42.325, 42.375, 42.425, 42.475, 42.525, 42.575, 42.625,
           42.675, 42.725, 42.775, 42.825, 42.875, 42.925, 42.975, 43.025, 43.075,
           43.125, 43.175, 43.225, 43.275, 43.325, 43.375, 43.425, 43.475, 43.525,
           43.575, 43.625, 43.675, 43.725, 43.775, 43.825, 43.875, 43.925, 43.975,
           44.025, 44.075, 44.125, 44.175, 44.225, 44.275, 44.325, 44.375, 44.425,
           44.475, 44.525, 44.575, 44.625, 44.675, 44.725, 44.775, 44.825, 44.875,
           44.925, 44.975], dtype=float32)

  • lon
    (lon)
    float32
    -135.0 -134.9 ... -125.1 -125.0

    standard_name :

    longitude

    long_name :

    longitude

    units :

    degrees_east

    valid_min :

    -180.0

    valid_max :

    180.0

    axis :

    X

    comment :

    Longitude geographical coordinates,WGS84 projection

    chunksizes :

    [7200]

    fletcher32 :

    False

    shuffle :

    True

    preferred_chunks :

    {'lon': 7200}

    zlib :

    True

    complevel :

    1

    source :

    <File-like object HTTPFileSystem, https://archive.podaac.earthdata.nasa.gov/podaac-ops-cumulus-protected/OSTIA-UKMO-L4-GLOB-REP-v2.0/1982/001/19820101120000-UKMO-L4_GHRSST-SSTfnd-OSTIA-GLOB_REP-v02.0-fv02.0.nc>

    original_shape :

    [7200]

    dtype :

    float32

    array([-134.975, -134.925, -134.875, -134.825, -134.775, -134.725, -134.675,
           -134.625, -134.575, -134.525, -134.475, -134.425, -134.375, -134.325,
           -134.275, -134.225, -134.175, -134.125, -134.075, -134.025, -133.975,
           -133.925, -133.875, -133.825, -133.775, -133.725, -133.675, -133.625,
           -133.575, -133.525, -133.475, -133.425, -133.375, -133.325, -133.275,
           -133.225, -133.175, -133.125, -133.075, -133.025, -132.975, -132.925,
           -132.875, -132.825, -132.775, -132.725, -132.675, -132.625, -132.575,
           -132.525, -132.475, -132.425, -132.375, -132.325, -132.275, -132.225,
           -132.175, -132.125, -132.075, -132.025, -131.975, -131.925, -131.875,
           -131.825, -131.775, -131.725, -131.675, -131.625, -131.575, -131.525,
           -131.475, -131.425, -131.375, -131.325, -131.275, -131.225, -131.175,
           -131.125, -131.075, -131.025, -130.975, -130.925, -130.875, -130.825,
           -130.775, -130.725, -130.675, -130.625, -130.575, -130.525, -130.475,
           -130.425, -130.375, -130.325, -130.275, -130.225, -130.175, -130.125,
           -130.075, -130.025, -129.975, -129.925, -129.875, -129.825, -129.775,
           -129.725, -129.675, -129.625, -129.575, -129.525, -129.475, -129.425,
           -129.375, -129.325, -129.275, -129.225, -129.175, -129.125, -129.075,
           -129.025, -128.975, -128.925, -128.875, -128.825, -128.775, -128.725,
           -128.675, -128.625, -128.575, -128.525, -128.475, -128.425, -128.375,
           -128.325, -128.275, -128.225, -128.175, -128.125, -128.075, -128.025,
           -127.975, -127.925, -127.875, -127.825, -127.775, -127.725, -127.675,
           -127.625, -127.575, -127.525, -127.475, -127.425, -127.375, -127.325,
           -127.275, -127.225, -127.175, -127.125, -127.075, -127.025, -126.975,
           -126.925, -126.875, -126.825, -126.775, -126.725, -126.675, -126.625,
           -126.575, -126.525, -126.475, -126.425, -126.375, -126.325, -126.275,
           -126.225, -126.175, -126.125, -126.075, -126.025, -125.975, -125.925,
           -125.875, -125.825, -125.775, -125.725, -125.675, -125.625, -125.575,
           -125.525, -125.475, -125.425, -125.375, -125.325, -125.275, -125.225,
           -125.175, -125.125, -125.075, -125.025], dtype=float32)

  • time
    (time)
    datetime64[ns]
    1982-01-02T12:00:00 ... 1992-01-...

    long_name :

    reference time of sst field

    standard_name :

    time

    axis :

    T

    comment :

    chunksizes :

    [1]

    fletcher32 :

    False

    shuffle :

    True

    preferred_chunks :

    {'time': 1}

    zlib :

    True

    complevel :

    1

    source :

    <File-like object HTTPFileSystem, https://archive.podaac.earthdata.nasa.gov/podaac-ops-cumulus-protected/OSTIA-UKMO-L4-GLOB-REP-v2.0/1982/001/19820101120000-UKMO-L4_GHRSST-SSTfnd-OSTIA-GLOB_REP-v02.0-fv02.0.nc>

    original_shape :

    [1]

    dtype :

    int32

    units :

    seconds since 1981-01-01 00:00:00

    _FillValue :

    1970-01-01T00:00:00.000000000

    array(['1982-01-02T12:00:00.000000000', '1982-01-03T12:00:00.000000000',
           '1982-01-04T12:00:00.000000000', ..., '1991-12-30T12:00:00.000000000',
           '1991-12-31T12:00:00.000000000', '1992-01-01T12:00:00.000000000'],
          shape=(3652,), dtype='datetime64[ns]')

• Data variables: (4)
  • analysed_sst
    (time, lat, lon)
    float64
    dask.array<chunksize=(1, 200, 200), meta=np.ndarray>

    long_name :

    analysed sea surface temperature

    standard_name :

    sea_surface_foundation_temperature

    units :

    kelvin

    valid_min :

    -300

    valid_max :

    4500

    reference :

    C.J. Donlon, M. Martin,J.D. Stark, J. Roberts-Jones, E. Fiedler, W. Wimmer. The operational sea surface temperature and sea ice analysis (OSTIA) system. Remote Sensing Environ., 116 (2012), pp. 140-158 http://dx.doi.org/10.1016/j.rse.2010.10.017

    comment :

    OSTIA foundation SST

    Array Chunk Bytes 1.09 GiB 312.50 kiB Shape (3652, 200, 200) (1, 200, 200) Dask graph 3652 chunks in 4 graph layers Data type float64 numpy.ndarray

  • analysis_error
    (time, lat, lon)
    float64
    dask.array<chunksize=(1, 200, 200), meta=np.ndarray>

    long_name :

    estimated error standard deviation of analysed_sst

    standard_name :

    sea_surface_foundation_temperature standard_error

    units :

    kelvin

    valid_min :

    0

    valid_max :

    32767

    comment :

    OSTIA foundation SST analysis standard deviation error

    Array Chunk Bytes 1.09 GiB 312.50 kiB Shape (3652, 200, 200) (1, 200, 200) Dask graph 3652 chunks in 4 graph layers Data type float64 numpy.ndarray

  • mask
    (time, lat, lon)
    float32
    dask.array<chunksize=(1, 200, 200), meta=np.ndarray>

    long_name :

    land sea ice lake bit mask

    valid_min :

    1

    valid_max :

    31

    flag_masks :

    [1, 2, 4, 8, 16]

    flag_meanings :

    water land optional_lake_surface sea_ice optional_river_surface

    source :

    NAVOCEANO_landmask_v1.0 EUMETSAT_OSI-SAF_icemask ARCLake_lakemask

    comment :

    Land/ open ocean/ sea ice /lake mask

    Array Chunk Bytes 557.25 MiB 156.25 kiB Shape (3652, 200, 200) (1, 200, 200) Dask graph 3652 chunks in 4 graph layers Data type float32 numpy.ndarray

  • sea_ice_fraction
    (time, lat, lon)
    float64
    dask.array<chunksize=(1, 200, 200), meta=np.ndarray>

    long_name :

    sea ice area fraction

    standard_name :

    sea_ice_area_fraction

    units :

    1

    valid_min :

    0

    valid_max :

    100

    source :

    EUMETSAT OSI-SAF

    comment :

    Sea ice area fraction

    Array Chunk Bytes 1.09 GiB 312.50 kiB Shape (3652, 200, 200) (1, 200, 200) Dask graph 3652 chunks in 4 graph layers Data type float64 numpy.ndarray

• Indexes: (3)
  • lat
    PandasIndex

    PandasIndex(Index([35.025001525878906,  35.07500076293945,             35.125,
            35.17499923706055, 35.224998474121094, 35.275001525878906,
            35.32500076293945,             35.375,  35.42499923706055,
           35.474998474121094,
           ...
           44.525001525878906,  44.57500076293945,             44.625,
            44.67499923706055, 44.724998474121094, 44.775001525878906,
            44.82500076293945,             44.875,  44.92499923706055,
           44.974998474121094],
          dtype='float32', name='lat', length=200))

  • lon
    PandasIndex

    PandasIndex(Index([-134.97500610351562,  -134.9250030517578,            -134.875,
            -134.8249969482422, -134.77499389648438, -134.72500610351562,
            -134.6750030517578,            -134.625,  -134.5749969482422,
           -134.52499389648438,
           ...
            -125.4749984741211, -125.42500305175781,            -125.375,
           -125.32499694824219,  -125.2750015258789,  -125.2249984741211,
           -125.17500305175781,            -125.125, -125.07499694824219,
            -125.0250015258789],
          dtype='float32', name='lon', length=200))

  • time
    PandasIndex

    PandasIndex(DatetimeIndex(['1982-01-02 12:00:00', '1982-01-03 12:00:00',
                   '1982-01-04 12:00:00', '1982-01-05 12:00:00',
                   '1982-01-06 12:00:00', '1982-01-07 12:00:00',
                   '1982-01-08 12:00:00', '1982-01-09 12:00:00',
                   '1982-01-10 12:00:00', '1982-01-11 12:00:00',
                   ...
                   '1991-12-23 12:00:00', '1991-12-24 12:00:00',
                   '1991-12-25 12:00:00', '1991-12-26 12:00:00',
                   '1991-12-27 12:00:00', '1991-12-28 12:00:00',
                   '1991-12-29 12:00:00', '1991-12-30 12:00:00',
                   '1991-12-31 12:00:00', '1992-01-01 12:00:00'],
                  dtype='datetime64[ns]', name='time', length=3652, freq=None))

• Attributes: (39)

  Conventions :

  CF-1.4, ACDD-1.3

  title :

  Global SST & Sea Ice Analysis, L4 OSTIA, 0.05 deg daily (METOFFICE-GLO-SST-L4-REP-OBS-SST-V2)

  summary :

  A merged, multi-sensor L4 Foundation SST product

  references :

  Donlon, C.J., Martin, M., Stark, J.D., Roberts-Jones, J., Fiedler, E., Wimmer, W., 2011. The Operational Sea Surface Temperature and Sea Ice Analysis (OSTIA). Remote Sensing of the Environment

  institution :

  UKMO

  history :

  Created from sst.nc; obs_anal.nc; seaice.nc

  comment :

  WARNING Some applications are unable to properly handle signed byte values. If values are encountered > 127, please subtract 256 from this reported value

  license :

  These data are available free of charge under the CMEMS data policy

  id :

  OSTIA-UKMO-L4-GLOB-v2.0

  naming_authority :

  org.ghrsst

  product_version :

  3.2

  uuid :

  a2df4a18-6f19-4772-9532-39307a0e2794

  gds_version_id :

  2.4

  netcdf_version_id :

  4.1

  file_quality_level :

  3

  Metadata_Conventions :

  Unidata Observation Dataset v1.0

  metadata_link :

  http://podaac.jpl.nasa.gov/ws/metadata/dataset?format=gcmd&shortName=UKMO-L4HRfnd-GLOB-OSTIA

  keywords :

  Oceans > Ocean Temperature > Sea Surface Temperature

  keywords_vocabulary :

  NASA Global Change Master Directory (GCMD) Science Keywords

  standard_name_vocabulary :

  NetCDF Climate and Forecast (CF) Metadata Convention

  westernmost_longitude :

  -180.0

  easternmost_longitude :

  180.00001525878906

  southernmost_latitude :

  -90.0

  northernmost_latitude :

  90.0

  spatial_resolution :

  0.05 degree

  geospatial_lat_units :

  degrees_north

  geospatial_lat_resolution :

  0.05000000074505806

  geospatial_lon_units :

  degrees_east

  geospatial_lon_resolution :

  0.05000000074505806

  acknowledgment :

  Please acknowledge the use of these data with the following statement: These data were provided by GHRSST, Met Office and CMEMS

  creator_name :

  Met Office as part of CMEMS

  creator_email :

  servicedesk.cmems@mercator-ocean.eu

  creator_url :

  http://marine.copernicus.eu

  project :

  Group for High Resolution Sea Surface Temperature

  publisher_name :

  GHRSST Project Office

  publisher_url :

  http://www.ghrsst.org

  publisher_email :

  ghrsst-po@nceo.ac.uk

  processing_level :

  L4

  cdm_data_type :

  grid

%%time

region_mean_ts = data_subset["analysed_sst"].mean(dim=("lon","lat")).compute()

CPU times: user 1min 21s, sys: 18.4 s, total: 1min 40s
Wall time: 4min 32s

fig = plt.figure(figsize=(8,4), layout="tight")
ax = plt.axes()
plt.plot(region_mean_ts["time"].values, region_mean_ts.values)
plt.xlabel("Year", fontsize=20)
plt.ylabel("SST (K)", fontsize=20)
ax.tick_params(axis='both', which='major', labelsize=14)

2.2 CCMP_WINDS_10M6HR_L4_V3.1

Level 4 dataset with 4 timestamps per file (6-hourly) and longitude range 0-360. The code in this section is mostly identical to the previous section.

Load data

## Modify vds_link for your desired data set. If in-cloud, set in_cloud_region to True:
vds_link = "https://archive.podaac.earthdata.nasa.gov/podaac-ops-cumulus-public/virtual_collections/CCMP_WINDS_10M6HR_L4_V3.1/CCMP_WINDS_10M6HR_L4_V3.1_virtual_https.json"
vds_mapper = get_vds_mapper(vds_link, in_cloud_region=False)

%%time

## No modification needed!
data = xr.open_dataset(
    vds_mapper, engine="zarr", chunks={},
    backend_kwargs={"consolidated": False}
)
data

CPU times: user 158 ms, sys: 8.61 ms, total: 167 ms
Wall time: 166 ms

<xarray.Dataset> Size: 781GB
Dimensions:    (latitude: 720, longitude: 1440, time: 47056)
Coordinates:
  * latitude   (latitude) float32 3kB -89.88 -89.62 -89.38 ... 89.38 89.62 89.88
  * longitude  (longitude) float32 6kB 0.125 0.375 0.625 ... 359.4 359.6 359.9
  * time       (time) datetime64[ns] 376kB 1993-01-02 ... 2025-03-31T18:00:00
Data variables:
    nobs       (time, latitude, longitude) float32 195GB dask.array<chunksize=(1, 720, 1440), meta=np.ndarray>
    uwnd       (time, latitude, longitude) float32 195GB dask.array<chunksize=(1, 720, 1440), meta=np.ndarray>
    vwnd       (time, latitude, longitude) float32 195GB dask.array<chunksize=(1, 720, 1440), meta=np.ndarray>
    ws         (time, latitude, longitude) float32 195GB dask.array<chunksize=(1, 720, 1440), meta=np.ndarray>
Attributes: (12/44)
    contact:                       Remote Sensing Systems, support@remss.com
    Conventions:                   CF-1.7 ACDD-1.3
    data_structure:                grid
    title:                         RSS CCMP V3.1 6-hourly surface winds (Leve...
    summary:                       RSS VAM 6-hour analyses using ERA-5 wind r...
    institute_id:                  RSS
    ...                            ...
    processing_level:              L4
    keywords:                      EARTH SCIENCE > ATMOSPHERE > ATMOSPHERIC W...
    keywords_vocabulary:           NASA Global Change Master Directory (GCMD)...
    platform_vocabulary:           GCMD platform keywords
    instrument_vocabulary:         GCMD instrument keywords
    references:                    Mears, C.; Lee, T.; Ricciardulli, L.; Wang...

xarray.Dataset

• Dimensions:
  • latitude: 720
  • longitude: 1440
  • time: 47056
• Coordinates: (3)
  • latitude
    (latitude)
    float32
    -89.88 -89.62 ... 89.62 89.88

    units :

    degrees_north

    long_name :

    latitude

    standard_name :

    latitude

    actual_range :

    [-78.3499984741211, 78.3499984741211]

    valid_min :

    -90.0

    valid_max :

    90.0

    _CoordinateAxisType :

    Lat

    coordinate_defines :

    center

    coverage_content_type :

    coordinate

    chunksizes :

    None

    fletcher32 :

    False

    shuffle :

    False

    source :

    <File-like object HTTPFileSystem, https://archive.podaac.earthdata.nasa.gov/podaac-ops-cumulus-protected/CCMP_WINDS_10M6HR_L4_V3.1/CCMP_Wind_Analysis_19930102_V03.1_L4.nc>

    original_shape :

    [720]

    dtype :

    float32

    array([-89.875, -89.625, -89.375, ...,  89.375,  89.625,  89.875],
          shape=(720,), dtype=float32)

  • longitude
    (longitude)
    float32
    0.125 0.375 0.625 ... 359.6 359.9

    units :

    degrees_east

    long_name :

    longitude

    standard_name :

    longitude

    valid_min :

    -0.0

    valid_max :

    360.0

    _CoordinateAxisType :

    Lon

    coordinate_defines :

    center

    coverage_content_type :

    coordinate

    chunksizes :

    None

    fletcher32 :

    False

    shuffle :

    False

    source :

    <File-like object HTTPFileSystem, https://archive.podaac.earthdata.nasa.gov/podaac-ops-cumulus-protected/CCMP_WINDS_10M6HR_L4_V3.1/CCMP_Wind_Analysis_19930102_V03.1_L4.nc>

    original_shape :

    [1440]

    dtype :

    float32

    array([1.25000e-01, 3.75000e-01, 6.25000e-01, ..., 3.59375e+02, 3.59625e+02,
           3.59875e+02], shape=(1440,), dtype=float32)

  • time
    (time)
    datetime64[ns]
    1993-01-02 ... 2025-03-31T18:00:00

    delta_t :

    0000-00-00 06:00:00

    long_name :

    time of analysis

    standard_name :

    time

    _CoordinateAxisType :

    Time

    coverage_content_type :

    coordinate

    chunksizes :

    [4]

    fletcher32 :

    False

    shuffle :

    False

    preferred_chunks :

    {'time': 4}

    source :

    <File-like object HTTPFileSystem, https://archive.podaac.earthdata.nasa.gov/podaac-ops-cumulus-protected/CCMP_WINDS_10M6HR_L4_V3.1/CCMP_Wind_Analysis_19930102_V03.1_L4.nc>

    original_shape :

    [4]

    dtype :

    float32

    _FillValue :

    1970-01-01T00:00:00.000000000

    units :

    hours since 1987-01-01 00:00:00

    array(['1993-01-02T00:00:00.000000000', '1993-01-02T06:00:00.000000000',
           '1993-01-02T12:00:00.000000000', ..., '2025-03-31T06:00:00.000000000',
           '2025-03-31T12:00:00.000000000', '2025-03-31T18:00:00.000000000'],
          shape=(47056,), dtype='datetime64[ns]')

• Data variables: (4)
  • nobs
    (time, latitude, longitude)
    float32
    dask.array<chunksize=(1, 720, 1440), meta=np.ndarray>

    valid_min :

    0.0

    valid_max :

    100.0

    long_name :

    number of observations used to derive wind vector components

    standard_name :

    number_of_observations

    coverage_content_type :

    auxiliaryInformation

    Array Chunk Bytes 181.75 GiB 3.96 MiB Shape (47056, 720, 1440) (1, 720, 1440) Dask graph 47056 chunks in 2 graph layers Data type float32 numpy.ndarray

  • uwnd
    (time, latitude, longitude)
    float32
    dask.array<chunksize=(1, 720, 1440), meta=np.ndarray>

    units :

    m s-1

    height :

    10 meters above sea-surface

    long_name :

    u-wind vector component at 10 meters

    standard_name :

    eastward_wind

    valid_min :

    -100.0

    valid_max :

    100.0

    ancillary_variables :

    nobs

    coverage_content_type :

    physicalMeasurement

    Array Chunk Bytes 181.75 GiB 3.96 MiB Shape (47056, 720, 1440) (1, 720, 1440) Dask graph 47056 chunks in 2 graph layers Data type float32 numpy.ndarray

  • vwnd
    (time, latitude, longitude)
    float32
    dask.array<chunksize=(1, 720, 1440), meta=np.ndarray>

    units :

    m s-1

    height :

    10 meters above sea-surface

    long_name :

    v-wind vector component at 10 meters

    standard_name :

    northward_wind

    valid_min :

    -100.0

    valid_max :

    100.0

    ancillary_variables :

    nobs

    coverage_content_type :

    physicalMeasurement

    Array Chunk Bytes 181.75 GiB 3.96 MiB Shape (47056, 720, 1440) (1, 720, 1440) Dask graph 47056 chunks in 2 graph layers Data type float32 numpy.ndarray

  • ws
    (time, latitude, longitude)
    float32
    dask.array<chunksize=(1, 720, 1440), meta=np.ndarray>

    units :

    m s-1

    height :

    10 meters above sea-surface

    long_name :

    wind speed at 10 meters

    standard_name :

    wind_speed

    valid_min :

    0.0

    valid_max :

    140.0

    ancillary_variables :

    nobs

    coverage_content_type :

    physicalMeasurement

    Array Chunk Bytes 181.75 GiB 3.96 MiB Shape (47056, 720, 1440) (1, 720, 1440) Dask graph 47056 chunks in 2 graph layers Data type float32 numpy.ndarray

• Indexes: (3)
  • latitude
    PandasIndex

    PandasIndex(Index([-89.875, -89.625, -89.375, -89.125, -88.875, -88.625, -88.375, -88.125,
           -87.875, -87.625,
           ...
            87.625,  87.875,  88.125,  88.375,  88.625,  88.875,  89.125,  89.375,
            89.625,  89.875],
          dtype='float32', name='latitude', length=720))

  • longitude
    PandasIndex

    PandasIndex(Index([  0.125,   0.375,   0.625,   0.875,   1.125,   1.375,   1.625,   1.875,
             2.125,   2.375,
           ...
           357.625, 357.875, 358.125, 358.375, 358.625, 358.875, 359.125, 359.375,
           359.625, 359.875],
          dtype='float32', name='longitude', length=1440))

  • time
    PandasIndex

    PandasIndex(DatetimeIndex(['1993-01-02 00:00:00', '1993-01-02 06:00:00',
                   '1993-01-02 12:00:00', '1993-01-02 18:00:00',
                   '1993-01-03 00:00:00', '1993-01-03 06:00:00',
                   '1993-01-03 12:00:00', '1993-01-03 18:00:00',
                   '1993-01-05 00:00:00', '1993-01-05 06:00:00',
                   ...
                   '2025-03-29 12:00:00', '2025-03-29 18:00:00',
                   '2025-03-30 00:00:00', '2025-03-30 06:00:00',
                   '2025-03-30 12:00:00', '2025-03-30 18:00:00',
                   '2025-03-31 00:00:00', '2025-03-31 06:00:00',
                   '2025-03-31 12:00:00', '2025-03-31 18:00:00'],
                  dtype='datetime64[ns]', name='time', length=47056, freq=None))

• Attributes: (44)

  contact :

  Remote Sensing Systems, support@remss.com

  Conventions :

  CF-1.7 ACDD-1.3

  data_structure :

  grid

  title :

  RSS CCMP V3.1 6-hourly surface winds (Level 4)

  summary :

  RSS VAM 6-hour analyses using ERA-5 wind reanalyses as background. Satellite winds (vector and scalar) are assimilated using a variational analysis identical to that used in CCMP V1.1. The ERA5 winds are adjusted to match winds from scatteromters before use in the analysis.

  institute_id :

  RSS

  institution :

  Remote Sensing Systems (RSS)

  source :

  blend of satellite and ERA-5 winds

  comment :

  none

  license :

  This work is licensed under CC BY 4.0. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/

  product_version :

  3.1

  standard_name_vocabulary :

  NetCDF Climate and Forecast (CF) Metadata Convention

  naming_authority :

  gov.nasa

  geospatial_lat_units :

  degrees_north

  geospatial_lat_resolution :

  0.25 degrees

  geospatial_lat_min :

  -78.375

  geospatial_lat_max :

  78.375

  geospatial_lon_units :

  degrees_east

  geospatial_lon_resolution :

  0.25 degrees

  geospatial_lon_min :

  0.125

  geospatial_lon_max :

  359.875

  geospatial_vertical_min :

  10.0

  geospatial_vertical_max :

  10.0

  geospatial_vertical_positive :

  up

  geospatial_vertical_units :

  meters

  creator_name :

  Remote Sensing Systems

  creator_type :

  institution

  creator_email :

  support@remss.com

  creator_url :

  http://www.remss.com/

  program :

  NASA/Ocean Vector Winds Science Team

  project :

  RSS Cross-Calibrated Multi-Platform Ocean Surface Wind Project

  publisher_name :

  Physical Oceanography Distributed Active Archive Center, Jet Propulsion Laboratory, NASA

  publisher_type :

  institution

  publisher_institution :

  NASA/JPL/PO.DAAC

  publisher_email :

  podaac@podaac.jpl.nasa.gov

  publisher_url :

  https://podaac.jpl.nasa.gov

  contributor_name :

  Carl Mears, Tong Lee, Frank Wentz

  contributor_role :

  Principal-Investigator, Co-Investigator, Co-Investigator

  processing_level :

  L4

  keywords :

  EARTH SCIENCE > ATMOSPHERE > ATMOSPHERIC WINDS > SURFACE WINDS, EARTH SCIENCE > OCEANS > OCEAN WINDS, EARTH SCIENCE > ATMOSPHERIC WINDS > SURFACE WINDS > WIND SPEED, EARTH SCIENCE > ATMOSPHERE WINDS > SURFACE WINDS > WIND DIRECTION, 10km - < 50km or approximately 0.09 degree - < 0.5 degree

  keywords_vocabulary :

  NASA Global Change Master Directory (GCMD) Science Keywords

  platform_vocabulary :

  GCMD platform keywords

  instrument_vocabulary :

  GCMD instrument keywords

  references :

  Mears, C.; Lee, T.; Ricciardulli, L.; Wang, X.; Wentz, F. Improving the Accuracy of the Cross-Calibrated Multi-Platform (CCMP) Ocean Vector Winds. Remote Sens. 2022, 14, 4230. https://doi.org/10.3390/rs14174230, Atlas, R., R. N. Hoffman, J. Ardizzone, S. M. Leidner, J. C. Jusem, D. K. Smith, D. Gombos, 2011: A cross-calibrated, multiplatform ocean surface wind velocity product for meteorological and oceanographic applications. Bull. Amer. Meteor. Soc., 92, 157-174. https://doi.org/10.1175/2010BAMS2946.1

Plot global ocean surface wind for first timestep

%%time

ws_t0 = data["ws"].isel(time=0)
plt.figure()
plt.contourf(ws_t0["longitude"], ws_t0["latitude"], ws_t0.values, levels=np.arange(0, 20, 2))
plt.colorbar()

CPU times: user 196 ms, sys: 36.8 ms, total: 232 ms
Wall time: 1.64 s

Time series of regional mean wind speed

Subset to 10 x 10 degree window off the USA northwest coast, compute mean over this window for each day. For the CCMP data set which is 6-hourly, this includes resampling to once daily data. Currently the time bounds are set for the first decade of data.

time_bounds = ("1993-01-02T00:00:00", "2003-01-02T00:00:00")
lon_bounds = (225,235)  # for 0 to 360 domain
lat_bounds = (35, 45)

# First subset by time then by space.
# For temporal subset, also resample the 6-hourly frequency to once daily:
data_subset = data.sel(time=slice(*time_bounds)).resample(time="1d").nearest()
data_subset = data_subset.sel(longitude=slice(*lon_bounds), latitude=slice(*lat_bounds))
data_subset

<xarray.Dataset> Size: 94MB
Dimensions:    (latitude: 40, longitude: 40, time: 3653)
Coordinates:
  * latitude   (latitude) float32 160B 35.12 35.38 35.62 ... 44.38 44.62 44.88
  * longitude  (longitude) float32 160B 225.1 225.4 225.6 ... 234.4 234.6 234.9
  * time       (time) datetime64[ns] 29kB 1993-01-02 1993-01-03 ... 2003-01-02
Data variables:
    nobs       (time, latitude, longitude) float32 23MB dask.array<chunksize=(1, 40, 40), meta=np.ndarray>
    uwnd       (time, latitude, longitude) float32 23MB dask.array<chunksize=(1, 40, 40), meta=np.ndarray>
    vwnd       (time, latitude, longitude) float32 23MB dask.array<chunksize=(1, 40, 40), meta=np.ndarray>
    ws         (time, latitude, longitude) float32 23MB dask.array<chunksize=(1, 40, 40), meta=np.ndarray>
Attributes: (12/44)
    contact:                       Remote Sensing Systems, support@remss.com
    Conventions:                   CF-1.7 ACDD-1.3
    data_structure:                grid
    title:                         RSS CCMP V3.1 6-hourly surface winds (Leve...
    summary:                       RSS VAM 6-hour analyses using ERA-5 wind r...
    institute_id:                  RSS
    ...                            ...
    processing_level:              L4
    keywords:                      EARTH SCIENCE > ATMOSPHERE > ATMOSPHERIC W...
    keywords_vocabulary:           NASA Global Change Master Directory (GCMD)...
    platform_vocabulary:           GCMD platform keywords
    instrument_vocabulary:         GCMD instrument keywords
    references:                    Mears, C.; Lee, T.; Ricciardulli, L.; Wang...

xarray.Dataset

• Dimensions:
  • latitude: 40
  • longitude: 40
  • time: 3653
• Coordinates: (3)
  • latitude
    (latitude)
    float32
    35.12 35.38 35.62 ... 44.62 44.88

    units :

    degrees_north

    long_name :

    latitude

    standard_name :

    latitude

    actual_range :

    [-78.3499984741211, 78.3499984741211]

    valid_min :

    -90.0

    valid_max :

    90.0

    _CoordinateAxisType :

    Lat

    coordinate_defines :

    center

    coverage_content_type :

    coordinate

    chunksizes :

    None

    fletcher32 :

    False

    shuffle :

    False

    source :

    <File-like object HTTPFileSystem, https://archive.podaac.earthdata.nasa.gov/podaac-ops-cumulus-protected/CCMP_WINDS_10M6HR_L4_V3.1/CCMP_Wind_Analysis_19930102_V03.1_L4.nc>

    original_shape :

    [720]

    dtype :

    float32

    array([35.125, 35.375, 35.625, 35.875, 36.125, 36.375, 36.625, 36.875, 37.125,
           37.375, 37.625, 37.875, 38.125, 38.375, 38.625, 38.875, 39.125, 39.375,
           39.625, 39.875, 40.125, 40.375, 40.625, 40.875, 41.125, 41.375, 41.625,
           41.875, 42.125, 42.375, 42.625, 42.875, 43.125, 43.375, 43.625, 43.875,
           44.125, 44.375, 44.625, 44.875], dtype=float32)

  • longitude
    (longitude)
    float32
    225.1 225.4 225.6 ... 234.6 234.9

    units :

    degrees_east

    long_name :

    longitude

    standard_name :

    longitude

    valid_min :

    -0.0

    valid_max :

    360.0

    _CoordinateAxisType :

    Lon

    coordinate_defines :

    center

    coverage_content_type :

    coordinate

    chunksizes :

    None

    fletcher32 :

    False

    shuffle :

    False

    source :

    <File-like object HTTPFileSystem, https://archive.podaac.earthdata.nasa.gov/podaac-ops-cumulus-protected/CCMP_WINDS_10M6HR_L4_V3.1/CCMP_Wind_Analysis_19930102_V03.1_L4.nc>

    original_shape :

    [1440]

    dtype :

    float32

    array([225.125, 225.375, 225.625, 225.875, 226.125, 226.375, 226.625, 226.875,
           227.125, 227.375, 227.625, 227.875, 228.125, 228.375, 228.625, 228.875,
           229.125, 229.375, 229.625, 229.875, 230.125, 230.375, 230.625, 230.875,
           231.125, 231.375, 231.625, 231.875, 232.125, 232.375, 232.625, 232.875,
           233.125, 233.375, 233.625, 233.875, 234.125, 234.375, 234.625, 234.875],
          dtype=float32)

  • time
    (time)
    datetime64[ns]
    1993-01-02 ... 2003-01-02

    delta_t :

    0000-00-00 06:00:00

    long_name :

    time of analysis

    standard_name :

    time

    _CoordinateAxisType :

    Time

    coverage_content_type :

    coordinate

    chunksizes :

    [4]

    fletcher32 :

    False

    shuffle :

    False

    preferred_chunks :

    {'time': 4}

    source :

    <File-like object HTTPFileSystem, https://archive.podaac.earthdata.nasa.gov/podaac-ops-cumulus-protected/CCMP_WINDS_10M6HR_L4_V3.1/CCMP_Wind_Analysis_19930102_V03.1_L4.nc>

    original_shape :

    [4]

    dtype :

    float32

    _FillValue :

    1970-01-01T00:00:00.000000000

    units :

    hours since 1987-01-01 00:00:00

    array(['1993-01-02T00:00:00.000000000', '1993-01-03T00:00:00.000000000',
           '1993-01-04T00:00:00.000000000', ..., '2002-12-31T00:00:00.000000000',
           '2003-01-01T00:00:00.000000000', '2003-01-02T00:00:00.000000000'],
          shape=(3653,), dtype='datetime64[ns]')

• Data variables: (4)
  • nobs
    (time, latitude, longitude)
    float32
    dask.array<chunksize=(1, 40, 40), meta=np.ndarray>

    valid_min :

    0.0

    valid_max :

    100.0

    long_name :

    number of observations used to derive wind vector components

    standard_name :

    number_of_observations

    coverage_content_type :

    auxiliaryInformation

    Array Chunk Bytes 22.30 MiB 6.25 kiB Shape (3653, 40, 40) (1, 40, 40) Dask graph 3653 chunks in 5 graph layers Data type float32 numpy.ndarray

  • uwnd
    (time, latitude, longitude)
    float32
    dask.array<chunksize=(1, 40, 40), meta=np.ndarray>

    units :

    m s-1

    height :

    10 meters above sea-surface

    long_name :

    u-wind vector component at 10 meters

    standard_name :

    eastward_wind

    valid_min :

    -100.0

    valid_max :

    100.0

    ancillary_variables :

    nobs

    coverage_content_type :

    physicalMeasurement

    Array Chunk Bytes 22.30 MiB 6.25 kiB Shape (3653, 40, 40) (1, 40, 40) Dask graph 3653 chunks in 5 graph layers Data type float32 numpy.ndarray

  • vwnd
    (time, latitude, longitude)
    float32
    dask.array<chunksize=(1, 40, 40), meta=np.ndarray>

    units :

    m s-1

    height :

    10 meters above sea-surface

    long_name :

    v-wind vector component at 10 meters

    standard_name :

    northward_wind

    valid_min :

    -100.0

    valid_max :

    100.0

    ancillary_variables :

    nobs

    coverage_content_type :

    physicalMeasurement

    Array Chunk Bytes 22.30 MiB 6.25 kiB Shape (3653, 40, 40) (1, 40, 40) Dask graph 3653 chunks in 5 graph layers Data type float32 numpy.ndarray

  • ws
    (time, latitude, longitude)
    float32
    dask.array<chunksize=(1, 40, 40), meta=np.ndarray>

    units :

    m s-1

    height :

    10 meters above sea-surface

    long_name :

    wind speed at 10 meters

    standard_name :

    wind_speed

    valid_min :

    0.0

    valid_max :

    140.0

    ancillary_variables :

    nobs

    coverage_content_type :

    physicalMeasurement

    Array Chunk Bytes 22.30 MiB 6.25 kiB Shape (3653, 40, 40) (1, 40, 40) Dask graph 3653 chunks in 5 graph layers Data type float32 numpy.ndarray

• Indexes: (3)
  • latitude
    PandasIndex

    PandasIndex(Index([35.125, 35.375, 35.625, 35.875, 36.125, 36.375, 36.625, 36.875, 37.125,
           37.375, 37.625, 37.875, 38.125, 38.375, 38.625, 38.875, 39.125, 39.375,
           39.625, 39.875, 40.125, 40.375, 40.625, 40.875, 41.125, 41.375, 41.625,
           41.875, 42.125, 42.375, 42.625, 42.875, 43.125, 43.375, 43.625, 43.875,
           44.125, 44.375, 44.625, 44.875],
          dtype='float32', name='latitude'))

  • longitude
    PandasIndex

    PandasIndex(Index([225.125, 225.375, 225.625, 225.875, 226.125, 226.375, 226.625, 226.875,
           227.125, 227.375, 227.625, 227.875, 228.125, 228.375, 228.625, 228.875,
           229.125, 229.375, 229.625, 229.875, 230.125, 230.375, 230.625, 230.875,
           231.125, 231.375, 231.625, 231.875, 232.125, 232.375, 232.625, 232.875,
           233.125, 233.375, 233.625, 233.875, 234.125, 234.375, 234.625, 234.875],
          dtype='float32', name='longitude'))

  • time
    PandasIndex

    PandasIndex(DatetimeIndex(['1993-01-02', '1993-01-03', '1993-01-04', '1993-01-05',
                   '1993-01-06', '1993-01-07', '1993-01-08', '1993-01-09',
                   '1993-01-10', '1993-01-11',
                   ...
                   '2002-12-24', '2002-12-25', '2002-12-26', '2002-12-27',
                   '2002-12-28', '2002-12-29', '2002-12-30', '2002-12-31',
                   '2003-01-01', '2003-01-02'],
                  dtype='datetime64[ns]', name='time', length=3653, freq='D'))

• Attributes: (44)

  contact :

  Remote Sensing Systems, support@remss.com

  Conventions :

  CF-1.7 ACDD-1.3

  data_structure :

  grid

  title :

  RSS CCMP V3.1 6-hourly surface winds (Level 4)

  summary :

  RSS VAM 6-hour analyses using ERA-5 wind reanalyses as background. Satellite winds (vector and scalar) are assimilated using a variational analysis identical to that used in CCMP V1.1. The ERA5 winds are adjusted to match winds from scatteromters before use in the analysis.

  institute_id :

  RSS

  institution :

  Remote Sensing Systems (RSS)

  source :

  blend of satellite and ERA-5 winds

  comment :

  none

  license :

  This work is licensed under CC BY 4.0. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/

  product_version :

  3.1

  standard_name_vocabulary :

  NetCDF Climate and Forecast (CF) Metadata Convention

  naming_authority :

  gov.nasa

  geospatial_lat_units :

  degrees_north

  geospatial_lat_resolution :

  0.25 degrees

  geospatial_lat_min :

  -78.375

  geospatial_lat_max :

  78.375

  geospatial_lon_units :

  degrees_east

  geospatial_lon_resolution :

  0.25 degrees

  geospatial_lon_min :

  0.125

  geospatial_lon_max :

  359.875

  geospatial_vertical_min :

  10.0

  geospatial_vertical_max :

  10.0

  geospatial_vertical_positive :

  up

  geospatial_vertical_units :

  meters

  creator_name :

  Remote Sensing Systems

  creator_type :

  institution

  creator_email :

  support@remss.com

  creator_url :

  http://www.remss.com/

  program :

  NASA/Ocean Vector Winds Science Team

  project :

  RSS Cross-Calibrated Multi-Platform Ocean Surface Wind Project

  publisher_name :

  Physical Oceanography Distributed Active Archive Center, Jet Propulsion Laboratory, NASA

  publisher_type :

  institution

  publisher_institution :

  NASA/JPL/PO.DAAC

  publisher_email :

  podaac@podaac.jpl.nasa.gov

  publisher_url :

  https://podaac.jpl.nasa.gov

  contributor_name :

  Carl Mears, Tong Lee, Frank Wentz

  contributor_role :

  Principal-Investigator, Co-Investigator, Co-Investigator

  processing_level :

  L4

  keywords :

  EARTH SCIENCE > ATMOSPHERE > ATMOSPHERIC WINDS > SURFACE WINDS, EARTH SCIENCE > OCEANS > OCEAN WINDS, EARTH SCIENCE > ATMOSPHERIC WINDS > SURFACE WINDS > WIND SPEED, EARTH SCIENCE > ATMOSPHERE WINDS > SURFACE WINDS > WIND DIRECTION, 10km - < 50km or approximately 0.09 degree - < 0.5 degree

  keywords_vocabulary :

  NASA Global Change Master Directory (GCMD) Science Keywords

  platform_vocabulary :

  GCMD platform keywords

  instrument_vocabulary :

  GCMD instrument keywords

  references :

  Mears, C.; Lee, T.; Ricciardulli, L.; Wang, X.; Wentz, F. Improving the Accuracy of the Cross-Calibrated Multi-Platform (CCMP) Ocean Vector Winds. Remote Sens. 2022, 14, 4230. https://doi.org/10.3390/rs14174230, Atlas, R., R. N. Hoffman, J. Ardizzone, S. M. Leidner, J. C. Jusem, D. K. Smith, D. Gombos, 2011: A cross-calibrated, multiplatform ocean surface wind velocity product for meteorological and oceanographic applications. Bull. Amer. Meteor. Soc., 92, 157-174. https://doi.org/10.1175/2010BAMS2946.1

%%time

region_mean_ts = data_subset["ws"].mean(dim=("longitude","latitude")).compute()

CPU times: user 1min 32s, sys: 22.5 s, total: 1min 54s
Wall time: 5min 14s

# The wind speed seasonal cycle is noisier than e.g. SST, so also compute a 
# monthly rolling mean to plot:
rolling_mean_ts = region_mean_ts.rolling(time=30).mean()

fig = plt.figure(figsize=(8,4), layout="tight")
ax = plt.axes()
plt.plot(region_mean_ts["time"].values, region_mean_ts.values)
plt.plot(rolling_mean_ts["time"].values, rolling_mean_ts.values)
plt.xlabel("Year", fontsize=20)
plt.ylabel("Wind Speed (m s-1)", fontsize=20)
ax.tick_params(axis='both', which='major', labelsize=14)

2.3 ECCO_L4_TEMP_SALINITY_05DEG_DAILY_V4R4

Level 4 model output for ocean temperature and salinity with one timestamp per file and dimensions (time, latitude, longitude, depth). The code in this section is mostly identical to the previous section.

Load data

## Modify vds_link for your desired data set. If in-cloud, set in_cloud_region to True:
vds_link = "https://archive.podaac.earthdata.nasa.gov/podaac-ops-cumulus-public/virtual_collections/ECCO_L4_TEMP_SALINITY_05DEG_DAILY_V4R4/ECCO_L4_TEMP_SALINITY_05DEG_DAILY_V4R4_virtual_https.json"
vds_mapper = get_vds_mapper(vds_link, in_cloud_region=False)

%%time

## No modification needed!
data = xr.open_dataset(
    vds_mapper, engine="zarr", chunks={},
    backend_kwargs={"consolidated": False}
)
data

CPU times: user 162 ms, sys: 13.3 ms, total: 175 ms
Wall time: 3.69 s

<xarray.Dataset> Size: 985GB
Dimensions:         (time: 9497, Z: 50, latitude: 360, longitude: 720, nv: 2)
Coordinates:
  * Z               (Z) float32 200B -5.0 -15.0 -25.0 ... -5.461e+03 -5.906e+03
    Z_bnds          (Z, nv) float32 400B dask.array<chunksize=(50, 2), meta=np.ndarray>
  * latitude        (latitude) float32 1kB -89.75 -89.25 -88.75 ... 89.25 89.75
    latitude_bnds   (latitude, nv) float32 3kB dask.array<chunksize=(360, 2), meta=np.ndarray>
  * longitude       (longitude) float32 3kB -179.8 -179.2 -178.8 ... 179.2 179.8
    longitude_bnds  (longitude, nv) float32 6kB dask.array<chunksize=(720, 2), meta=np.ndarray>
  * time            (time) datetime64[ns] 76kB 1992-01-01T18:00:00 ... 2017-1...
    time_bnds       (time, nv) datetime64[ns] 152kB dask.array<chunksize=(1, 2), meta=np.ndarray>
Dimensions without coordinates: nv
Data variables:
    SALT            (time, Z, latitude, longitude) float32 492GB dask.array<chunksize=(1, 25, 180, 360), meta=np.ndarray>
    THETA           (time, Z, latitude, longitude) float32 492GB dask.array<chunksize=(1, 25, 180, 360), meta=np.ndarray>
Attributes: (12/53)
    acknowledgement:                 This research was carried out by the Jet...
    author:                          Ian Fenty and Ou Wang
    cdm_data_type:                   Grid
    comment:                         Fields provided on a regular lat-lon gri...
    Conventions:                     CF-1.8, ACDD-1.3
    coordinates_comment:             Note: the global 'coordinates' attribute...
    ...                              ...
    source:                          The ECCO V4r4 state estimate was produce...
    standard_name_vocabulary:        NetCDF Climate and Forecast (CF) Metadat...
    summary:                         This dataset provides daily-averaged oce...
    time_coverage_duration:          P1D
    time_coverage_resolution:        P1D
    title:                           ECCO Ocean Temperature and Salinity - Da...

xarray.Dataset

• Dimensions:
  • time: 9497
  • Z: 50
  • latitude: 360
  • longitude: 720
  • nv: 2
• Coordinates: (8)
  • Z
    (Z)
    float32
    -5.0 -15.0 ... -5.906e+03

    axis :

    Z

    bounds :

    Z_bnds

    comment :

    Non-uniform vertical spacing.

    coverage_content_type :

    coordinate

    long_name :

    depth of grid cell center

    positive :

    up

    standard_name :

    depth

    units :

    m

    array([-5.000000e+00, -1.500000e+01, -2.500000e+01, -3.500000e+01,
           -4.500000e+01, -5.500000e+01, -6.500000e+01, -7.500500e+01,
           -8.502500e+01, -9.509500e+01, -1.053100e+02, -1.158700e+02,
           -1.271500e+02, -1.397400e+02, -1.544700e+02, -1.724000e+02,
           -1.947350e+02, -2.227100e+02, -2.574700e+02, -2.999300e+02,
           -3.506800e+02, -4.099300e+02, -4.774700e+02, -5.527100e+02,
           -6.347350e+02, -7.224000e+02, -8.144700e+02, -9.097400e+02,
           -1.007155e+03, -1.105905e+03, -1.205535e+03, -1.306205e+03,
           -1.409150e+03, -1.517095e+03, -1.634175e+03, -1.765135e+03,
           -1.914150e+03, -2.084035e+03, -2.276225e+03, -2.491250e+03,
           -2.729250e+03, -2.990250e+03, -3.274250e+03, -3.581250e+03,
           -3.911250e+03, -4.264250e+03, -4.640250e+03, -5.039250e+03,
           -5.461250e+03, -5.906250e+03], dtype=float32)

  • Z_bnds
    (Z, nv)
    float32
    dask.array<chunksize=(50, 2), meta=np.ndarray>

    comment :

    One pair of bounds for each vertical level.

    coverage_content_type :

    coordinate

    long_name :

    depths of grid cell upper and lower interfaces

    Array Chunk Bytes 400 B 400 B Shape (50, 2) (50, 2) Dask graph 1 chunks in 2 graph layers Data type float32 numpy.ndarray

  • latitude
    (latitude)
    float32
    -89.75 -89.25 ... 89.25 89.75

    axis :

    Y

    bounds :

    latitude_bnds

    comment :

    uniform grid spacing from -89.75 to 89.75 by 0.5

    coverage_content_type :

    coordinate

    long_name :

    latitude at grid cell center

    standard_name :

    latitude

    units :

    degrees_north

    array([-89.75, -89.25, -88.75, ...,  88.75,  89.25,  89.75],
          shape=(360,), dtype=float32)

  • latitude_bnds
    (latitude, nv)
    float32
    dask.array<chunksize=(360, 2), meta=np.ndarray>

    coverage_content_type :

    coordinate

    long_name :

    latitude bounds grid cells

    Array Chunk Bytes 2.81 kiB 2.81 kiB Shape (360, 2) (360, 2) Dask graph 1 chunks in 2 graph layers Data type float32 numpy.ndarray

  • longitude
    (longitude)
    float32
    -179.8 -179.2 ... 179.2 179.8

    axis :

    X

    bounds :

    longitude_bnds

    comment :

    uniform grid spacing from -179.75 to 179.75 by 0.5

    coverage_content_type :

    coordinate

    long_name :

    longitude at grid cell center

    standard_name :

    longitude

    units :

    degrees_east

    array([-179.75, -179.25, -178.75, ...,  178.75,  179.25,  179.75],
          shape=(720,), dtype=float32)

  • longitude_bnds
    (longitude, nv)
    float32
    dask.array<chunksize=(720, 2), meta=np.ndarray>

    coverage_content_type :

    coordinate

    long_name :

    longitude bounds grid cells

    Array Chunk Bytes 5.62 kiB 5.62 kiB Shape (720, 2) (720, 2) Dask graph 1 chunks in 2 graph layers Data type float32 numpy.ndarray

  • time
    (time)
    datetime64[ns]
    1992-01-01T18:00:00 ... 2017-12-...

    axis :

    T

    bounds :

    time_bnds

    coverage_content_type :

    coordinate

    long_name :

    center time of averaging period

    standard_name :

    time

    chunksizes :

    None

    fletcher32 :

    False

    shuffle :

    False

    source :

    <File-like object HTTPFileSystem, https://archive.podaac.earthdata.nasa.gov/podaac-ops-cumulus-protected/ECCO_L4_TEMP_SALINITY_05DEG_DAILY_V4R4/OCEAN_TEMPERATURE_SALINITY_day_mean_1992-01-01_ECCO_V4r4_latlon_0p50deg.nc>

    original_shape :

    [1]

    dtype :

    int32

    array(['1992-01-01T18:00:00.000000000', '1992-01-02T12:00:00.000000000',
           '1992-01-03T12:00:00.000000000', ..., '2017-12-29T12:00:00.000000000',
           '2017-12-30T18:00:00.000000000', '2017-12-31T06:00:00.000000000'],
          shape=(9497,), dtype='datetime64[ns]')

  • time_bnds
    (time, nv)
    datetime64[ns]
    dask.array<chunksize=(1, 2), meta=np.ndarray>

    comment :

    Start and end times of averaging period.

    coverage_content_type :

    coordinate

    long_name :

    time bounds of averaging period

    Array Chunk Bytes 148.39 kiB 16 B Shape (9497, 2) (1, 2) Dask graph 9497 chunks in 2 graph layers Data type datetime64[ns] numpy.ndarray

• Data variables: (2)
  • SALT
    (time, Z, latitude, longitude)
    float32
    dask.array<chunksize=(1, 25, 180, 360), meta=np.ndarray>

    coverage_content_type :

    modelResult

    long_name :

    Salinity

    standard_name :

    sea_water_salinity

    units :

    1e-3

    valid_min :

    16.73577880859375

    valid_max :

    41.321231842041016

    comment :

    Defined using CF convention 'Sea water salinity is the salt content of sea water, often on the Practical Salinity Scale of 1978. However, the unqualified term 'salinity' is generic and does not necessarily imply any particular method of calculation. The units of salinity are dimensionless and the units attribute should normally be given as 1e-3 or 0.001 i.e. parts per thousand.' see https://cfconventions.org/Data/cf-standard-names/73/build/cf-standard-name-table.html

    Array Chunk Bytes 458.51 GiB 6.18 MiB Shape (9497, 50, 360, 720) (1, 25, 180, 360) Dask graph 75976 chunks in 2 graph layers Data type float32 numpy.ndarray

  • THETA
    (time, Z, latitude, longitude)
    float32
    dask.array<chunksize=(1, 25, 180, 360), meta=np.ndarray>

    coverage_content_type :

    modelResult

    long_name :

    Potential temperature

    standard_name :

    sea_water_potential_temperature

    units :

    degree_C

    comment :

    Sea water potential temperature is the temperature a parcel of sea water would have if moved adiabatically to sea level pressure. Note: the equation of state is a modified UNESCO formula by Jackett and McDougall (1995), which uses the model variable potential temperature as input assuming a horizontally and temporally constant pressure of $p_0=-g ho_{0} z$.

    valid_min :

    -2.9179372787475586

    valid_max :

    36.425140380859375

    Array Chunk Bytes 458.51 GiB 6.18 MiB Shape (9497, 50, 360, 720) (1, 25, 180, 360) Dask graph 75976 chunks in 2 graph layers Data type float32 numpy.ndarray

• Indexes: (4)
  • Z
    PandasIndex

    PandasIndex(Index([               -5.0,               -15.0,               -25.0,
                         -35.0,               -45.0,               -55.0,
                         -65.0,  -75.00499725341797,   -85.0250015258789,
            -95.09500122070312, -105.30999755859375, -115.87000274658203,
            -127.1500015258789, -139.74000549316406, -154.47000122070312,
           -172.39999389648438, -194.73500061035156,  -222.7100067138672,
            -257.4700012207031, -299.92999267578125, -350.67999267578125,
           -409.92999267578125,  -477.4700012207031,  -552.7100219726562,
            -634.7349853515625,  -722.4000244140625,   -814.469970703125,
             -909.739990234375,  -1007.155029296875,  -1105.905029296875,
           -1205.5350341796875, -1306.2049560546875, -1409.1500244140625,
            -1517.094970703125,  -1634.175048828125,  -1765.135009765625,
           -1914.1500244140625,  -2084.034912109375,   -2276.22509765625,
                      -2491.25,            -2729.25,            -2990.25,
                      -3274.25,            -3581.25,            -3911.25,
                      -4264.25,            -4640.25,            -5039.25,
                      -5461.25,            -5906.25],
          dtype='float32', name='Z'))

  • latitude
    PandasIndex

    PandasIndex(Index([-89.75, -89.25, -88.75, -88.25, -87.75, -87.25, -86.75, -86.25, -85.75,
           -85.25,
           ...
            85.25,  85.75,  86.25,  86.75,  87.25,  87.75,  88.25,  88.75,  89.25,
            89.75],
          dtype='float32', name='latitude', length=360))

  • longitude
    PandasIndex

    PandasIndex(Index([-179.75, -179.25, -178.75, -178.25, -177.75, -177.25, -176.75, -176.25,
           -175.75, -175.25,
           ...
            175.25,  175.75,  176.25,  176.75,  177.25,  177.75,  178.25,  178.75,
            179.25,  179.75],
          dtype='float32', name='longitude', length=720))

  • time
    PandasIndex

    PandasIndex(DatetimeIndex(['1992-01-01 18:00:00', '1992-01-02 12:00:00',
                   '1992-01-03 12:00:00', '1992-01-04 12:00:00',
                   '1992-01-05 12:00:00', '1992-01-06 12:00:00',
                   '1992-01-07 12:00:00', '1992-01-08 12:00:00',
                   '1992-01-09 12:00:00', '1992-01-10 12:00:00',
                   ...
                   '2017-12-22 12:00:00', '2017-12-23 12:00:00',
                   '2017-12-24 12:00:00', '2017-12-25 12:00:00',
                   '2017-12-26 12:00:00', '2017-12-27 12:00:00',
                   '2017-12-28 12:00:00', '2017-12-29 12:00:00',
                   '2017-12-30 18:00:00', '2017-12-31 06:00:00'],
                  dtype='datetime64[ns]', name='time', length=9497, freq=None))

• Attributes: (53)

  acknowledgement :

  This research was carried out by the Jet Propulsion Laboratory, managed by the California Institute of Technology under a contract with the National Aeronautics and Space Administration.

  author :

  Ian Fenty and Ou Wang

  cdm_data_type :

  Grid

  comment :

  Fields provided on a regular lat-lon grid. They have been mapped to the regular lat-lon grid from the original ECCO lat-lon-cap 90 (llc90) native model grid.

  Conventions :

  CF-1.8, ACDD-1.3

  coordinates_comment :

  Note: the global 'coordinates' attribute describes auxillary coordinates.

  creator_email :

  ecco-group@mit.edu

  creator_institution :

  NASA Jet Propulsion Laboratory (JPL)

  creator_name :

  ECCO Consortium

  creator_type :

  group

  creator_url :

  https://ecco-group.org

  geospatial_bounds_crs :

  EPSG:4326

  geospatial_lat_max :

  90.0

  geospatial_lat_min :

  -90.0

  geospatial_lat_resolution :

  0.5

  geospatial_lat_units :

  degrees_north

  geospatial_lon_max :

  180.0

  geospatial_lon_min :

  -180.0

  geospatial_lon_resolution :

  0.5

  geospatial_lon_units :

  degrees_east

  geospatial_vertical_max :

  0.0

  geospatial_vertical_min :

  -6134.5

  geospatial_vertical_positive :

  up

  geospatial_vertical_resolution :

  variable

  geospatial_vertical_units :

  meter

  history :

  Inaugural release of an ECCO Central Estimate solution to PO.DAAC

  id :

  10.5067/ECG5D-OTS44

  institution :

  NASA Jet Propulsion Laboratory (JPL)

  instrument_vocabulary :

  GCMD instrument keywords

  keywords_vocabulary :

  NASA Global Change Master Directory (GCMD) Science Keywords

  license :

  Public Domain

  metadata_link :

  https://cmr.earthdata.nasa.gov/search/collections.umm_json?ShortName=ECCO_L4_TEMP_SALINITY_05DEG_DAILY_V4R4

  naming_authority :

  gov.nasa.jpl

  platform :

  ERS-1/2, TOPEX/Poseidon, Geosat Follow-On (GFO), ENVISAT, Jason-1, Jason-2, CryoSat-2, SARAL/AltiKa, Jason-3, AVHRR, Aquarius, SSM/I, SSMIS, GRACE, DTU17MDT, Argo, WOCE, GO-SHIP, MEOP, Ice Tethered Profilers (ITP)

  platform_vocabulary :

  GCMD platform keywords

  processing_level :

  L4

  product_time_coverage_end :

  2018-01-01T00:00:00

  product_time_coverage_start :

  1992-01-01T12:00:00

  product_version :

  Version 4, Release 4

  program :

  NASA Physical Oceanography, Cryosphere, Modeling, Analysis, and Prediction (MAP)

  project :

  Estimating the Circulation and Climate of the Ocean (ECCO)

  publisher_email :

  podaac@podaac.jpl.nasa.gov

  publisher_institution :

  PO.DAAC

  publisher_name :

  Physical Oceanography Distributed Active Archive Center (PO.DAAC)

  publisher_type :

  institution

  publisher_url :

  https://podaac.jpl.nasa.gov

  references :

  ECCO Consortium, Fukumori, I., Wang, O., Fenty, I., Forget, G., Heimbach, P., & Ponte, R. M. 2020. Synopsis of the ECCO Central Production Global Ocean and Sea-Ice State Estimate (Version 4 Release 4). doi:10.5281/zenodo.3765928

  source :

  The ECCO V4r4 state estimate was produced by fitting a free-running solution of the MITgcm (checkpoint 66g) to satellite and in situ observational data in a least squares sense using the adjoint method

  standard_name_vocabulary :

  NetCDF Climate and Forecast (CF) Metadata Convention

  summary :

  This dataset provides daily-averaged ocean potential temperature and salinity interpolated to a regular 0.5-degree grid from the ECCO Version 4 Release 4 (V4r4) ocean and sea-ice state estimate. Estimating the Circulation and Climate of the Ocean (ECCO) state estimates are dynamically and kinematically-consistent reconstructions of the three-dimensional, time-evolving ocean, sea-ice, and surface atmospheric states. ECCO V4r4 is a free-running solution of a global, nominally 1-degree configuration of the MIT general circulation model (MITgcm) that has been fit to observations in a least-squares sense. Observational data constraints used in V4r4 include sea surface height (SSH) from satellite altimeters [ERS-1/2, TOPEX/Poseidon, GFO, ENVISAT, Jason-1,2,3, CryoSat-2, and SARAL/AltiKa]; sea surface temperature (SST) from satellite radiometers [AVHRR], sea surface salinity (SSS) from the Aquarius satellite radiometer/scatterometer, ocean bottom pressure (OBP) from the GRACE satellite gravimeter; sea-ice concentration from satellite radiometers [SSM/I and SSMIS], and in-situ ocean temperature and salinity measured with conductivity-temperature-depth (CTD) sensors and expendable bathythermographs (XBTs) from several programs [e.g., WOCE, GO-SHIP, Argo, and others] and platforms [e.g., research vessels, gliders, moorings, ice-tethered profilers, and instrumented pinnipeds]. V4r4 covers the period 1992-01-01T12:00:00 to 2018-01-01T00:00:00.

  time_coverage_duration :

  P1D

  time_coverage_resolution :

  P1D

  title :

  ECCO Ocean Temperature and Salinity - Daily Mean 0.5 Degree (Version 4 Release 4)

Plot global ocean salinity at 5m depth for first timestep

%%time

salt_t0 = data["SALT"].isel(time=0, Z=0)
plt.figure()
plt.contourf(salt_t0["longitude"], salt_t0["latitude"], salt_t0.values, levels=np.arange(24, 40, 0.5))
plt.colorbar()

CPU times: user 299 ms, sys: 68.7 ms, total: 368 ms
Wall time: 1.45 s

Time series of regional mean ocean salinity at 5m depth

Subset to 10 x 10 degree window off the USA northwest coast, compute mean over this window for each day. Currently the time bounds are set for the first decade of data.

time_bounds = ("1992-01-02T00:00:00", "2002-01-02T00:00:00")
lon_bounds = (-135,-125)  # for -180 to 180 domain
lat_bounds = (35, 45)

# First subset by time then by space (horizontal and vertical):
data_subset = data.sel(time=slice(*time_bounds))
data_subset = data_subset.sel(longitude=slice(*lon_bounds), latitude=slice(*lat_bounds))
data_subset = data_subset.isel(Z=0)
data_subset

<xarray.Dataset> Size: 12MB
Dimensions:         (time: 3653, latitude: 20, longitude: 20, nv: 2)
Coordinates:
    Z               float32 4B -5.0
    Z_bnds          (nv) float32 8B dask.array<chunksize=(2,), meta=np.ndarray>
  * latitude        (latitude) float32 80B 35.25 35.75 36.25 ... 44.25 44.75
    latitude_bnds   (latitude, nv) float32 160B dask.array<chunksize=(20, 2), meta=np.ndarray>
  * longitude       (longitude) float32 80B -134.8 -134.2 ... -125.8 -125.2
    longitude_bnds  (longitude, nv) float32 160B dask.array<chunksize=(20, 2), meta=np.ndarray>
  * time            (time) datetime64[ns] 29kB 1992-01-02T12:00:00 ... 2002-0...
    time_bnds       (time, nv) datetime64[ns] 58kB dask.array<chunksize=(1, 2), meta=np.ndarray>
Dimensions without coordinates: nv
Data variables:
    SALT            (time, latitude, longitude) float32 6MB dask.array<chunksize=(1, 20, 20), meta=np.ndarray>
    THETA           (time, latitude, longitude) float32 6MB dask.array<chunksize=(1, 20, 20), meta=np.ndarray>
Attributes: (12/53)
    acknowledgement:                 This research was carried out by the Jet...
    author:                          Ian Fenty and Ou Wang
    cdm_data_type:                   Grid
    comment:                         Fields provided on a regular lat-lon gri...
    Conventions:                     CF-1.8, ACDD-1.3
    coordinates_comment:             Note: the global 'coordinates' attribute...
    ...                              ...
    source:                          The ECCO V4r4 state estimate was produce...
    standard_name_vocabulary:        NetCDF Climate and Forecast (CF) Metadat...
    summary:                         This dataset provides daily-averaged oce...
    time_coverage_duration:          P1D
    time_coverage_resolution:        P1D
    title:                           ECCO Ocean Temperature and Salinity - Da...

xarray.Dataset

• Dimensions:
  • time: 3653
  • latitude: 20
  • longitude: 20
  • nv: 2
• Coordinates: (8)
  • Z
    ()
    float32
    -5.0

    axis :

    Z

    bounds :

    Z_bnds

    comment :

    Non-uniform vertical spacing.

    coverage_content_type :

    coordinate

    long_name :

    depth of grid cell center

    positive :

    up

    standard_name :

    depth

    units :

    m

    array(-5., dtype=float32)

  • Z_bnds
    (nv)
    float32
    dask.array<chunksize=(2,), meta=np.ndarray>

    comment :

    One pair of bounds for each vertical level.

    coverage_content_type :

    coordinate

    long_name :

    depths of grid cell upper and lower interfaces

    Array Chunk Bytes 8 B 8 B Shape (2,) (2,) Dask graph 1 chunks in 3 graph layers Data type float32 numpy.ndarray

  • latitude
    (latitude)
    float32
    35.25 35.75 36.25 ... 44.25 44.75

    axis :

    Y

    bounds :

    latitude_bnds

    comment :

    uniform grid spacing from -89.75 to 89.75 by 0.5

    coverage_content_type :

    coordinate

    long_name :

    latitude at grid cell center

    standard_name :

    latitude

    units :

    degrees_north

    array([35.25, 35.75, 36.25, 36.75, 37.25, 37.75, 38.25, 38.75, 39.25, 39.75,
           40.25, 40.75, 41.25, 41.75, 42.25, 42.75, 43.25, 43.75, 44.25, 44.75],
          dtype=float32)

  • latitude_bnds
    (latitude, nv)
    float32
    dask.array<chunksize=(20, 2), meta=np.ndarray>

    coverage_content_type :

    coordinate

    long_name :

    latitude bounds grid cells

    Array Chunk Bytes 160 B 160 B Shape (20, 2) (20, 2) Dask graph 1 chunks in 3 graph layers Data type float32 numpy.ndarray

  • longitude
    (longitude)
    float32
    -134.8 -134.2 ... -125.8 -125.2

    axis :

    X

    bounds :

    longitude_bnds

    comment :

    uniform grid spacing from -179.75 to 179.75 by 0.5

    coverage_content_type :

    coordinate

    long_name :

    longitude at grid cell center

    standard_name :

    longitude

    units :

    degrees_east

    array([-134.75, -134.25, -133.75, -133.25, -132.75, -132.25, -131.75, -131.25,
           -130.75, -130.25, -129.75, -129.25, -128.75, -128.25, -127.75, -127.25,
           -126.75, -126.25, -125.75, -125.25], dtype=float32)

  • longitude_bnds
    (longitude, nv)
    float32
    dask.array<chunksize=(20, 2), meta=np.ndarray>

    coverage_content_type :

    coordinate

    long_name :

    longitude bounds grid cells

    Array Chunk Bytes 160 B 160 B Shape (20, 2) (20, 2) Dask graph 1 chunks in 3 graph layers Data type float32 numpy.ndarray

  • time
    (time)
    datetime64[ns]
    1992-01-02T12:00:00 ... 2002-01-...

    axis :

    T

    bounds :

    time_bnds

    coverage_content_type :

    coordinate

    long_name :

    center time of averaging period

    standard_name :

    time

    chunksizes :

    None

    fletcher32 :

    False

    shuffle :

    False

    source :

    <File-like object HTTPFileSystem, https://archive.podaac.earthdata.nasa.gov/podaac-ops-cumulus-protected/ECCO_L4_TEMP_SALINITY_05DEG_DAILY_V4R4/OCEAN_TEMPERATURE_SALINITY_day_mean_1992-01-01_ECCO_V4r4_latlon_0p50deg.nc>

    original_shape :

    [1]

    dtype :

    int32

    array(['1992-01-02T12:00:00.000000000', '1992-01-03T12:00:00.000000000',
           '1992-01-04T12:00:00.000000000', ..., '2001-12-30T12:00:00.000000000',
           '2001-12-31T12:00:00.000000000', '2002-01-01T12:00:00.000000000'],
          shape=(3653,), dtype='datetime64[ns]')

  • time_bnds
    (time, nv)
    datetime64[ns]
    dask.array<chunksize=(1, 2), meta=np.ndarray>

    comment :

    Start and end times of averaging period.

    coverage_content_type :

    coordinate

    long_name :

    time bounds of averaging period

    Array Chunk Bytes 57.08 kiB 16 B Shape (3653, 2) (1, 2) Dask graph 3653 chunks in 3 graph layers Data type datetime64[ns] numpy.ndarray

• Data variables: (2)
  • SALT
    (time, latitude, longitude)
    float32
    dask.array<chunksize=(1, 20, 20), meta=np.ndarray>

    coverage_content_type :

    modelResult

    long_name :

    Salinity

    standard_name :

    sea_water_salinity

    units :

    1e-3

    valid_min :

    16.73577880859375

    valid_max :

    41.321231842041016

    comment :

    Defined using CF convention 'Sea water salinity is the salt content of sea water, often on the Practical Salinity Scale of 1978. However, the unqualified term 'salinity' is generic and does not necessarily imply any particular method of calculation. The units of salinity are dimensionless and the units attribute should normally be given as 1e-3 or 0.001 i.e. parts per thousand.' see https://cfconventions.org/Data/cf-standard-names/73/build/cf-standard-name-table.html

    Array Chunk Bytes 5.57 MiB 1.56 kiB Shape (3653, 20, 20) (1, 20, 20) Dask graph 3653 chunks in 5 graph layers Data type float32 numpy.ndarray

  • THETA
    (time, latitude, longitude)
    float32
    dask.array<chunksize=(1, 20, 20), meta=np.ndarray>

    coverage_content_type :

    modelResult

    long_name :

    Potential temperature

    standard_name :

    sea_water_potential_temperature

    units :

    degree_C

    comment :

    Sea water potential temperature is the temperature a parcel of sea water would have if moved adiabatically to sea level pressure. Note: the equation of state is a modified UNESCO formula by Jackett and McDougall (1995), which uses the model variable potential temperature as input assuming a horizontally and temporally constant pressure of $p_0=-g ho_{0} z$.

    valid_min :

    -2.9179372787475586

    valid_max :

    36.425140380859375

    Array Chunk Bytes 5.57 MiB 1.56 kiB Shape (3653, 20, 20) (1, 20, 20) Dask graph 3653 chunks in 5 graph layers Data type float32 numpy.ndarray

• Indexes: (3)
  • latitude
    PandasIndex

    PandasIndex(Index([35.25, 35.75, 36.25, 36.75, 37.25, 37.75, 38.25, 38.75, 39.25, 39.75,
           40.25, 40.75, 41.25, 41.75, 42.25, 42.75, 43.25, 43.75, 44.25, 44.75],
          dtype='float32', name='latitude'))

  • longitude
    PandasIndex

    PandasIndex(Index([-134.75, -134.25, -133.75, -133.25, -132.75, -132.25, -131.75, -131.25,
           -130.75, -130.25, -129.75, -129.25, -128.75, -128.25, -127.75, -127.25,
           -126.75, -126.25, -125.75, -125.25],
          dtype='float32', name='longitude'))

  • time
    PandasIndex

    PandasIndex(DatetimeIndex(['1992-01-02 12:00:00', '1992-01-03 12:00:00',
                   '1992-01-04 12:00:00', '1992-01-05 12:00:00',
                   '1992-01-06 12:00:00', '1992-01-07 12:00:00',
                   '1992-01-08 12:00:00', '1992-01-09 12:00:00',
                   '1992-01-10 12:00:00', '1992-01-11 12:00:00',
                   ...
                   '2001-12-23 12:00:00', '2001-12-24 12:00:00',
                   '2001-12-25 12:00:00', '2001-12-26 12:00:00',
                   '2001-12-27 12:00:00', '2001-12-28 12:00:00',
                   '2001-12-29 12:00:00', '2001-12-30 12:00:00',
                   '2001-12-31 12:00:00', '2002-01-01 12:00:00'],
                  dtype='datetime64[ns]', name='time', length=3653, freq=None))

• Attributes: (53)

  acknowledgement :

  This research was carried out by the Jet Propulsion Laboratory, managed by the California Institute of Technology under a contract with the National Aeronautics and Space Administration.

  author :

  Ian Fenty and Ou Wang

  cdm_data_type :

  Grid

  comment :

  Fields provided on a regular lat-lon grid. They have been mapped to the regular lat-lon grid from the original ECCO lat-lon-cap 90 (llc90) native model grid.

  Conventions :

  CF-1.8, ACDD-1.3

  coordinates_comment :

  Note: the global 'coordinates' attribute describes auxillary coordinates.

  creator_email :

  ecco-group@mit.edu

  creator_institution :

  NASA Jet Propulsion Laboratory (JPL)

  creator_name :

  ECCO Consortium

  creator_type :

  group

  creator_url :

  https://ecco-group.org

  geospatial_bounds_crs :

  EPSG:4326

  geospatial_lat_max :

  90.0

  geospatial_lat_min :

  -90.0

  geospatial_lat_resolution :

  0.5

  geospatial_lat_units :

  degrees_north

  geospatial_lon_max :

  180.0

  geospatial_lon_min :

  -180.0

  geospatial_lon_resolution :

  0.5

  geospatial_lon_units :

  degrees_east

  geospatial_vertical_max :

  0.0

  geospatial_vertical_min :

  -6134.5

  geospatial_vertical_positive :

  up

  geospatial_vertical_resolution :

  variable

  geospatial_vertical_units :

  meter

  history :

  Inaugural release of an ECCO Central Estimate solution to PO.DAAC

  id :

  10.5067/ECG5D-OTS44

  institution :

  NASA Jet Propulsion Laboratory (JPL)

  instrument_vocabulary :

  GCMD instrument keywords

  keywords_vocabulary :

  NASA Global Change Master Directory (GCMD) Science Keywords

  license :

  Public Domain

  metadata_link :

  https://cmr.earthdata.nasa.gov/search/collections.umm_json?ShortName=ECCO_L4_TEMP_SALINITY_05DEG_DAILY_V4R4

  naming_authority :

  gov.nasa.jpl

  platform :

  ERS-1/2, TOPEX/Poseidon, Geosat Follow-On (GFO), ENVISAT, Jason-1, Jason-2, CryoSat-2, SARAL/AltiKa, Jason-3, AVHRR, Aquarius, SSM/I, SSMIS, GRACE, DTU17MDT, Argo, WOCE, GO-SHIP, MEOP, Ice Tethered Profilers (ITP)

  platform_vocabulary :

  GCMD platform keywords

  processing_level :

  L4

  product_time_coverage_end :

  2018-01-01T00:00:00

  product_time_coverage_start :

  1992-01-01T12:00:00

  product_version :

  Version 4, Release 4

  program :

  NASA Physical Oceanography, Cryosphere, Modeling, Analysis, and Prediction (MAP)

  project :

  Estimating the Circulation and Climate of the Ocean (ECCO)

  publisher_email :

  podaac@podaac.jpl.nasa.gov

  publisher_institution :

  PO.DAAC

  publisher_name :

  Physical Oceanography Distributed Active Archive Center (PO.DAAC)

  publisher_type :

  institution

  publisher_url :

  https://podaac.jpl.nasa.gov

  references :

  ECCO Consortium, Fukumori, I., Wang, O., Fenty, I., Forget, G., Heimbach, P., & Ponte, R. M. 2020. Synopsis of the ECCO Central Production Global Ocean and Sea-Ice State Estimate (Version 4 Release 4). doi:10.5281/zenodo.3765928

  source :

  The ECCO V4r4 state estimate was produced by fitting a free-running solution of the MITgcm (checkpoint 66g) to satellite and in situ observational data in a least squares sense using the adjoint method

  standard_name_vocabulary :

  NetCDF Climate and Forecast (CF) Metadata Convention

  summary :

  This dataset provides daily-averaged ocean potential temperature and salinity interpolated to a regular 0.5-degree grid from the ECCO Version 4 Release 4 (V4r4) ocean and sea-ice state estimate. Estimating the Circulation and Climate of the Ocean (ECCO) state estimates are dynamically and kinematically-consistent reconstructions of the three-dimensional, time-evolving ocean, sea-ice, and surface atmospheric states. ECCO V4r4 is a free-running solution of a global, nominally 1-degree configuration of the MIT general circulation model (MITgcm) that has been fit to observations in a least-squares sense. Observational data constraints used in V4r4 include sea surface height (SSH) from satellite altimeters [ERS-1/2, TOPEX/Poseidon, GFO, ENVISAT, Jason-1,2,3, CryoSat-2, and SARAL/AltiKa]; sea surface temperature (SST) from satellite radiometers [AVHRR], sea surface salinity (SSS) from the Aquarius satellite radiometer/scatterometer, ocean bottom pressure (OBP) from the GRACE satellite gravimeter; sea-ice concentration from satellite radiometers [SSM/I and SSMIS], and in-situ ocean temperature and salinity measured with conductivity-temperature-depth (CTD) sensors and expendable bathythermographs (XBTs) from several programs [e.g., WOCE, GO-SHIP, Argo, and others] and platforms [e.g., research vessels, gliders, moorings, ice-tethered profilers, and instrumented pinnipeds]. V4r4 covers the period 1992-01-01T12:00:00 to 2018-01-01T00:00:00.

  time_coverage_duration :

  P1D

  time_coverage_resolution :

  P1D

  title :

  ECCO Ocean Temperature and Salinity - Daily Mean 0.5 Degree (Version 4 Release 4)

%%time

region_mean_ts = data_subset["SALT"].mean(dim=("longitude","latitude")).compute()

CPU times: user 1min 40s, sys: 16 s, total: 1min 56s
Wall time: 4min 33s

fig = plt.figure(figsize=(8,4), layout="tight")
ax = plt.axes()
plt.plot(region_mean_ts["time"].values, region_mean_ts.values)
plt.xlabel("Year", fontsize=20)
plt.ylabel("SALT (PSU)", fontsize=20)
ax.tick_params(axis='both', which='major', labelsize=14)

3. Alternate workflows other than Xarray built-in functions

Only Section 1 needs to be run prior to this section. Section 2 can be skipped.

Xarray built-in functions are very convenient for subsetting and many standard computations (grouping, averaging, statistics, etc.). However, if you prefer to use Xarray only to open the data but then work with the data in other ways, this section provides a few methods, using the ECCO model’s ocean salinity variable (4D: time, latitude, longitude, depth).

Open the data using VDS as before

## Modify vds_link for your desired data set. If in-cloud, set in_cloud_region to True:
vds_link = "https://archive.podaac.earthdata.nasa.gov/podaac-ops-cumulus-public/virtual_collections/ECCO_L4_TEMP_SALINITY_05DEG_DAILY_V4R4/ECCO_L4_TEMP_SALINITY_05DEG_DAILY_V4R4_virtual_https.json"
vds_mapper = get_vds_mapper(vds_link, in_cloud_region=False)

%%time

data = xr.open_dataset(
    vds_mapper, engine="zarr", chunks={},
    backend_kwargs={"consolidated": False}
)
data

CPU times: user 81.4 ms, sys: 14.9 ms, total: 96.3 ms
Wall time: 4.24 s

<xarray.Dataset> Size: 985GB
Dimensions:         (time: 9497, Z: 50, latitude: 360, longitude: 720, nv: 2)
Coordinates:
  * Z               (Z) float32 200B -5.0 -15.0 -25.0 ... -5.461e+03 -5.906e+03
    Z_bnds          (Z, nv) float32 400B dask.array<chunksize=(50, 2), meta=np.ndarray>
  * latitude        (latitude) float32 1kB -89.75 -89.25 -88.75 ... 89.25 89.75
    latitude_bnds   (latitude, nv) float32 3kB dask.array<chunksize=(360, 2), meta=np.ndarray>
  * longitude       (longitude) float32 3kB -179.8 -179.2 -178.8 ... 179.2 179.8
    longitude_bnds  (longitude, nv) float32 6kB dask.array<chunksize=(720, 2), meta=np.ndarray>
  * time            (time) datetime64[ns] 76kB 1992-01-01T18:00:00 ... 2017-1...
    time_bnds       (time, nv) datetime64[ns] 152kB dask.array<chunksize=(1, 2), meta=np.ndarray>
Dimensions without coordinates: nv
Data variables:
    SALT            (time, Z, latitude, longitude) float32 492GB dask.array<chunksize=(1, 25, 180, 360), meta=np.ndarray>
    THETA           (time, Z, latitude, longitude) float32 492GB dask.array<chunksize=(1, 25, 180, 360), meta=np.ndarray>
Attributes: (12/53)
    acknowledgement:                 This research was carried out by the Jet...
    author:                          Ian Fenty and Ou Wang
    cdm_data_type:                   Grid
    comment:                         Fields provided on a regular lat-lon gri...
    Conventions:                     CF-1.8, ACDD-1.3
    coordinates_comment:             Note: the global 'coordinates' attribute...
    ...                              ...
    source:                          The ECCO V4r4 state estimate was produce...
    standard_name_vocabulary:        NetCDF Climate and Forecast (CF) Metadat...
    summary:                         This dataset provides daily-averaged oce...
    time_coverage_duration:          P1D
    time_coverage_resolution:        P1D
    title:                           ECCO Ocean Temperature and Salinity - Da...

xarray.Dataset

• Dimensions:
  • time: 9497
  • Z: 50
  • latitude: 360
  • longitude: 720
  • nv: 2
• Coordinates: (8)
  • Z
    (Z)
    float32
    -5.0 -15.0 ... -5.906e+03

    axis :

    Z

    bounds :

    Z_bnds

    comment :

    Non-uniform vertical spacing.

    coverage_content_type :

    coordinate

    long_name :

    depth of grid cell center

    positive :

    up

    standard_name :

    depth

    units :

    m

    array([-5.000000e+00, -1.500000e+01, -2.500000e+01, -3.500000e+01,
           -4.500000e+01, -5.500000e+01, -6.500000e+01, -7.500500e+01,
           -8.502500e+01, -9.509500e+01, -1.053100e+02, -1.158700e+02,
           -1.271500e+02, -1.397400e+02, -1.544700e+02, -1.724000e+02,
           -1.947350e+02, -2.227100e+02, -2.574700e+02, -2.999300e+02,
           -3.506800e+02, -4.099300e+02, -4.774700e+02, -5.527100e+02,
           -6.347350e+02, -7.224000e+02, -8.144700e+02, -9.097400e+02,
           -1.007155e+03, -1.105905e+03, -1.205535e+03, -1.306205e+03,
           -1.409150e+03, -1.517095e+03, -1.634175e+03, -1.765135e+03,
           -1.914150e+03, -2.084035e+03, -2.276225e+03, -2.491250e+03,
           -2.729250e+03, -2.990250e+03, -3.274250e+03, -3.581250e+03,
           -3.911250e+03, -4.264250e+03, -4.640250e+03, -5.039250e+03,
           -5.461250e+03, -5.906250e+03], dtype=float32)

  • Z_bnds
    (Z, nv)
    float32
    dask.array<chunksize=(50, 2), meta=np.ndarray>

    comment :

    One pair of bounds for each vertical level.

    coverage_content_type :

    coordinate

    long_name :

    depths of grid cell upper and lower interfaces

    Array Chunk Bytes 400 B 400 B Shape (50, 2) (50, 2) Dask graph 1 chunks in 2 graph layers Data type float32 numpy.ndarray

  • latitude
    (latitude)
    float32
    -89.75 -89.25 ... 89.25 89.75

    axis :

    Y

    bounds :

    latitude_bnds

    comment :

    uniform grid spacing from -89.75 to 89.75 by 0.5

    coverage_content_type :

    coordinate

    long_name :

    latitude at grid cell center

    standard_name :

    latitude

    units :

    degrees_north

    array([-89.75, -89.25, -88.75, ...,  88.75,  89.25,  89.75],
          shape=(360,), dtype=float32)

  • latitude_bnds
    (latitude, nv)
    float32
    dask.array<chunksize=(360, 2), meta=np.ndarray>

    coverage_content_type :

    coordinate

    long_name :

    latitude bounds grid cells

    Array Chunk Bytes 2.81 kiB 2.81 kiB Shape (360, 2) (360, 2) Dask graph 1 chunks in 2 graph layers Data type float32 numpy.ndarray

  • longitude
    (longitude)
    float32
    -179.8 -179.2 ... 179.2 179.8

    axis :

    X

    bounds :

    longitude_bnds

    comment :

    uniform grid spacing from -179.75 to 179.75 by 0.5

    coverage_content_type :

    coordinate

    long_name :

    longitude at grid cell center

    standard_name :

    longitude

    units :

    degrees_east

    array([-179.75, -179.25, -178.75, ...,  178.75,  179.25,  179.75],
          shape=(720,), dtype=float32)

  • longitude_bnds
    (longitude, nv)
    float32
    dask.array<chunksize=(720, 2), meta=np.ndarray>

    coverage_content_type :

    coordinate

    long_name :

    longitude bounds grid cells

    Array Chunk Bytes 5.62 kiB 5.62 kiB Shape (720, 2) (720, 2) Dask graph 1 chunks in 2 graph layers Data type float32 numpy.ndarray

  • time
    (time)
    datetime64[ns]
    1992-01-01T18:00:00 ... 2017-12-...

    axis :

    T

    bounds :

    time_bnds

    coverage_content_type :

    coordinate

    long_name :

    center time of averaging period

    standard_name :

    time

    chunksizes :

    None

    fletcher32 :

    False

    shuffle :

    False

    source :

    <File-like object HTTPFileSystem, https://archive.podaac.earthdata.nasa.gov/podaac-ops-cumulus-protected/ECCO_L4_TEMP_SALINITY_05DEG_DAILY_V4R4/OCEAN_TEMPERATURE_SALINITY_day_mean_1992-01-01_ECCO_V4r4_latlon_0p50deg.nc>

    original_shape :

    [1]

    dtype :

    int32

    array(['1992-01-01T18:00:00.000000000', '1992-01-02T12:00:00.000000000',
           '1992-01-03T12:00:00.000000000', ..., '2017-12-29T12:00:00.000000000',
           '2017-12-30T18:00:00.000000000', '2017-12-31T06:00:00.000000000'],
          shape=(9497,), dtype='datetime64[ns]')

  • time_bnds
    (time, nv)
    datetime64[ns]
    dask.array<chunksize=(1, 2), meta=np.ndarray>

    comment :

    Start and end times of averaging period.

    coverage_content_type :

    coordinate

    long_name :

    time bounds of averaging period

    Array Chunk Bytes 148.39 kiB 16 B Shape (9497, 2) (1, 2) Dask graph 9497 chunks in 2 graph layers Data type datetime64[ns] numpy.ndarray

• Data variables: (2)
  • SALT
    (time, Z, latitude, longitude)
    float32
    dask.array<chunksize=(1, 25, 180, 360), meta=np.ndarray>

    coverage_content_type :

    modelResult

    long_name :

    Salinity

    standard_name :

    sea_water_salinity

    units :

    1e-3

    valid_min :

    16.73577880859375

    valid_max :

    41.321231842041016

    comment :

    Defined using CF convention 'Sea water salinity is the salt content of sea water, often on the Practical Salinity Scale of 1978. However, the unqualified term 'salinity' is generic and does not necessarily imply any particular method of calculation. The units of salinity are dimensionless and the units attribute should normally be given as 1e-3 or 0.001 i.e. parts per thousand.' see https://cfconventions.org/Data/cf-standard-names/73/build/cf-standard-name-table.html

    Array Chunk Bytes 458.51 GiB 6.18 MiB Shape (9497, 50, 360, 720) (1, 25, 180, 360) Dask graph 75976 chunks in 2 graph layers Data type float32 numpy.ndarray

  • THETA
    (time, Z, latitude, longitude)
    float32
    dask.array<chunksize=(1, 25, 180, 360), meta=np.ndarray>

    coverage_content_type :

    modelResult

    long_name :

    Potential temperature

    standard_name :

    sea_water_potential_temperature

    units :

    degree_C

    comment :

    Sea water potential temperature is the temperature a parcel of sea water would have if moved adiabatically to sea level pressure. Note: the equation of state is a modified UNESCO formula by Jackett and McDougall (1995), which uses the model variable potential temperature as input assuming a horizontally and temporally constant pressure of $p_0=-g ho_{0} z$.

    valid_min :

    -2.9179372787475586

    valid_max :

    36.425140380859375

    Array Chunk Bytes 458.51 GiB 6.18 MiB Shape (9497, 50, 360, 720) (1, 25, 180, 360) Dask graph 75976 chunks in 2 graph layers Data type float32 numpy.ndarray

• Indexes: (4)
  • Z
    PandasIndex

    PandasIndex(Index([               -5.0,               -15.0,               -25.0,
                         -35.0,               -45.0,               -55.0,
                         -65.0,  -75.00499725341797,   -85.0250015258789,
            -95.09500122070312, -105.30999755859375, -115.87000274658203,
            -127.1500015258789, -139.74000549316406, -154.47000122070312,
           -172.39999389648438, -194.73500061035156,  -222.7100067138672,
            -257.4700012207031, -299.92999267578125, -350.67999267578125,
           -409.92999267578125,  -477.4700012207031,  -552.7100219726562,
            -634.7349853515625,  -722.4000244140625,   -814.469970703125,
             -909.739990234375,  -1007.155029296875,  -1105.905029296875,
           -1205.5350341796875, -1306.2049560546875, -1409.1500244140625,
            -1517.094970703125,  -1634.175048828125,  -1765.135009765625,
           -1914.1500244140625,  -2084.034912109375,   -2276.22509765625,
                      -2491.25,            -2729.25,            -2990.25,
                      -3274.25,            -3581.25,            -3911.25,
                      -4264.25,            -4640.25,            -5039.25,
                      -5461.25,            -5906.25],
          dtype='float32', name='Z'))

  • latitude
    PandasIndex

    PandasIndex(Index([-89.75, -89.25, -88.75, -88.25, -87.75, -87.25, -86.75, -86.25, -85.75,
           -85.25,
           ...
            85.25,  85.75,  86.25,  86.75,  87.25,  87.75,  88.25,  88.75,  89.25,
            89.75],
          dtype='float32', name='latitude', length=360))

  • longitude
    PandasIndex

    PandasIndex(Index([-179.75, -179.25, -178.75, -178.25, -177.75, -177.25, -176.75, -176.25,
           -175.75, -175.25,
           ...
            175.25,  175.75,  176.25,  176.75,  177.25,  177.75,  178.25,  178.75,
            179.25,  179.75],
          dtype='float32', name='longitude', length=720))

  • time
    PandasIndex

    PandasIndex(DatetimeIndex(['1992-01-01 18:00:00', '1992-01-02 12:00:00',
                   '1992-01-03 12:00:00', '1992-01-04 12:00:00',
                   '1992-01-05 12:00:00', '1992-01-06 12:00:00',
                   '1992-01-07 12:00:00', '1992-01-08 12:00:00',
                   '1992-01-09 12:00:00', '1992-01-10 12:00:00',
                   ...
                   '2017-12-22 12:00:00', '2017-12-23 12:00:00',
                   '2017-12-24 12:00:00', '2017-12-25 12:00:00',
                   '2017-12-26 12:00:00', '2017-12-27 12:00:00',
                   '2017-12-28 12:00:00', '2017-12-29 12:00:00',
                   '2017-12-30 18:00:00', '2017-12-31 06:00:00'],
                  dtype='datetime64[ns]', name='time', length=9497, freq=None))

• Attributes: (53)

  acknowledgement :

  This research was carried out by the Jet Propulsion Laboratory, managed by the California Institute of Technology under a contract with the National Aeronautics and Space Administration.

  author :

  Ian Fenty and Ou Wang

  cdm_data_type :

  Grid

  comment :

  Fields provided on a regular lat-lon grid. They have been mapped to the regular lat-lon grid from the original ECCO lat-lon-cap 90 (llc90) native model grid.

  Conventions :

  CF-1.8, ACDD-1.3

  coordinates_comment :

  Note: the global 'coordinates' attribute describes auxillary coordinates.

  creator_email :

  ecco-group@mit.edu

  creator_institution :

  NASA Jet Propulsion Laboratory (JPL)

  creator_name :

  ECCO Consortium

  creator_type :

  group

  creator_url :

  https://ecco-group.org

  geospatial_bounds_crs :

  EPSG:4326

  geospatial_lat_max :

  90.0

  geospatial_lat_min :

  -90.0

  geospatial_lat_resolution :

  0.5

  geospatial_lat_units :

  degrees_north

  geospatial_lon_max :

  180.0

  geospatial_lon_min :

  -180.0

  geospatial_lon_resolution :

  0.5

  geospatial_lon_units :

  degrees_east

  geospatial_vertical_max :

  0.0

  geospatial_vertical_min :

  -6134.5

  geospatial_vertical_positive :

  up

  geospatial_vertical_resolution :

  variable

  geospatial_vertical_units :

  meter

  history :

  Inaugural release of an ECCO Central Estimate solution to PO.DAAC

  id :

  10.5067/ECG5D-OTS44

  institution :

  NASA Jet Propulsion Laboratory (JPL)

  instrument_vocabulary :

  GCMD instrument keywords

  keywords_vocabulary :

  NASA Global Change Master Directory (GCMD) Science Keywords

  license :

  Public Domain

  metadata_link :

  https://cmr.earthdata.nasa.gov/search/collections.umm_json?ShortName=ECCO_L4_TEMP_SALINITY_05DEG_DAILY_V4R4

  naming_authority :

  gov.nasa.jpl

  platform :

  ERS-1/2, TOPEX/Poseidon, Geosat Follow-On (GFO), ENVISAT, Jason-1, Jason-2, CryoSat-2, SARAL/AltiKa, Jason-3, AVHRR, Aquarius, SSM/I, SSMIS, GRACE, DTU17MDT, Argo, WOCE, GO-SHIP, MEOP, Ice Tethered Profilers (ITP)

  platform_vocabulary :

  GCMD platform keywords

  processing_level :

  L4

  product_time_coverage_end :

  2018-01-01T00:00:00

  product_time_coverage_start :

  1992-01-01T12:00:00

  product_version :

  Version 4, Release 4

  program :

  NASA Physical Oceanography, Cryosphere, Modeling, Analysis, and Prediction (MAP)

  project :

  Estimating the Circulation and Climate of the Ocean (ECCO)

  publisher_email :

  podaac@podaac.jpl.nasa.gov

  publisher_institution :

  PO.DAAC

  publisher_name :

  Physical Oceanography Distributed Active Archive Center (PO.DAAC)

  publisher_type :

  institution

  publisher_url :

  https://podaac.jpl.nasa.gov

  references :

  ECCO Consortium, Fukumori, I., Wang, O., Fenty, I., Forget, G., Heimbach, P., & Ponte, R. M. 2020. Synopsis of the ECCO Central Production Global Ocean and Sea-Ice State Estimate (Version 4 Release 4). doi:10.5281/zenodo.3765928

  source :

  The ECCO V4r4 state estimate was produced by fitting a free-running solution of the MITgcm (checkpoint 66g) to satellite and in situ observational data in a least squares sense using the adjoint method

  standard_name_vocabulary :

  NetCDF Climate and Forecast (CF) Metadata Convention

  summary :

  This dataset provides daily-averaged ocean potential temperature and salinity interpolated to a regular 0.5-degree grid from the ECCO Version 4 Release 4 (V4r4) ocean and sea-ice state estimate. Estimating the Circulation and Climate of the Ocean (ECCO) state estimates are dynamically and kinematically-consistent reconstructions of the three-dimensional, time-evolving ocean, sea-ice, and surface atmospheric states. ECCO V4r4 is a free-running solution of a global, nominally 1-degree configuration of the MIT general circulation model (MITgcm) that has been fit to observations in a least-squares sense. Observational data constraints used in V4r4 include sea surface height (SSH) from satellite altimeters [ERS-1/2, TOPEX/Poseidon, GFO, ENVISAT, Jason-1,2,3, CryoSat-2, and SARAL/AltiKa]; sea surface temperature (SST) from satellite radiometers [AVHRR], sea surface salinity (SSS) from the Aquarius satellite radiometer/scatterometer, ocean bottom pressure (OBP) from the GRACE satellite gravimeter; sea-ice concentration from satellite radiometers [SSM/I and SSMIS], and in-situ ocean temperature and salinity measured with conductivity-temperature-depth (CTD) sensors and expendable bathythermographs (XBTs) from several programs [e.g., WOCE, GO-SHIP, Argo, and others] and platforms [e.g., research vessels, gliders, moorings, ice-tethered profilers, and instrumented pinnipeds]. V4r4 covers the period 1992-01-01T12:00:00 to 2018-01-01T00:00:00.

  time_coverage_duration :

  P1D

  time_coverage_resolution :

  P1D

  title :

  ECCO Ocean Temperature and Salinity - Daily Mean 0.5 Degree (Version 4 Release 4)

3.1 Extract NumPy or Dask arrays from the Xarray Dataset

NumPy arrays are more or less a fundamental data structure for scientific computing in Python. Dask arrays are the analog to NumPy arrays, but for arrays that are too large to load into memory at once at need to be analyzed in chunks. Documentation for both these packages are great and easily findable, e.g. consult the Resources section of the Using Virtual Datasets chapter of the PO.DAAC Cookbook.

Extracting a NumPy array will take longer than extracting a Dask Array because in the former, the data are fully loaded, while in the latter they are only lazy loaded.

## Get the first year of salinity data off the USA west coast to work with:
time_bounds = ("1992-01-02T00:00:00", "1993-01-02T00:00:00")
lon_bounds = (-135,-115)
lat_bounds = (25, 45)

## Data is "lazy-subsetted"
data_subset = data.sel(time=slice(*time_bounds)).sel(longitude=slice(*lon_bounds), latitude=slice(*lat_bounds))

%%time
## Load as a numpy array (takes longer because data is fully loaded):
salt_np = data_subset["SALT"].values

print(type(salt_np))
print(np.nanmin(salt_np))
print(np.nanmax(salt_np))

<class 'numpy.ndarray'>
31.433891
35.17998

%%time
## Load as a Dask array
salt_da = data_subset["SALT"].data
print(type(salt_da))
salt_da

<class 'dask.array.core.Array'>
CPU times: user 1.05 ms, sys: 1.5 ms, total: 2.55 ms
Wall time: 2.71 ms

Array Chunk Bytes 111.69 MiB 156.25 kiB Shape (366, 50, 40, 40) (1, 25, 40, 40) Dask graph 732 chunks in 4 graph layers Data type float32 numpy.ndarray

%%time
np.nanmin(salt_da).compute()

np.float32(31.433891)

3.2 Downloading a subset of the data as a netCDF locally

Note that currently the Python package used to create our VDS’s (virtualizarr) adds some extra variable-level attributes to the coordinate variables that don’t play well with netCDF standards and causes netCDF to complain when we try to save a subset of the VDS to netCDF. This should be fixed in future versions of the virtualizarr package, but for now we include some additional boiler-plate code to deal with these problem attributes. No modification of that code should be necessary from the user.

## Same subset as Section 3.1:
time_bounds = ("1992-01-02T00:00:00", "1993-01-02T00:00:00")
lon_bounds = (-135,-115)
lat_bounds = (25, 45)

data_for_downloading = data.sel(time=slice(*time_bounds)).sel(longitude=slice(*lon_bounds), latitude=slice(*lat_bounds))
data_for_downloading = data_for_downloading["SALT"]

## Function to deal with problem attributes in the VDS before saving to netCDF. No modification should be needed:
def fix_coord_attrs(ds):

    for coord in ds.coords:

        attrs = ds[coord].attrs.copy()
    
        # Convert any booleans to strings:
        for atr, value in attrs.items():
            if type(value) == bool:
                attrs[atr] = str(value)
        
        # Fix the 'preferred_chunks' field (should not be dict):
        if 'preferred_chunks' in attrs:
            pref_chunks = attrs['preferred_chunks']
            if type(pref_chunks) == dict:
                pref_chunks_list = [e[1] for e in pref_chunks.items()]
                attrs['preferred_chunks'] = pref_chunks_list

        # 'chunksizes' attr cannot be None:
        if 'chunksizes' in attrs:
            if attrs['chunksizes'] is None:
                attrs['chunksizes'] = 'none'
    
        if coord == "time":
            # Remove problematic attributes for time encoding,
            # (Likely due to time already being decoded in VDS?)
            if 'units' in attrs:
                del attrs['units']
            if '_FillValue' in attrs:
                del attrs['_FillValue']

        ds[coord].attrs = attrs

    return ds

data_for_downloading = fix_coord_attrs(data_for_downloading)

%%time

data_for_downloading.to_netcdf("data_local_subset.nc")

CPU times: user 43.4 s, sys: 13.6 s, total: 57 s
Wall time: 2min 39s

Then reload the data locally are plot a couple things

data_local = xr.open_dataset("data_local_subset.nc")
data_local

<xarray.Dataset> Size: 234MB
Dimensions:         (time: 366, Z: 50, latitude: 40, longitude: 40, nv: 2)
Coordinates:
  * Z               (Z) float32 200B -5.0 -15.0 -25.0 ... -5.461e+03 -5.906e+03
    Z_bnds          (Z, nv) float32 400B ...
  * latitude        (latitude) float32 160B 25.25 25.75 26.25 ... 44.25 44.75
    latitude_bnds   (latitude, nv) float32 320B ...
  * longitude       (longitude) float32 160B -134.8 -134.2 ... -115.8 -115.2
    longitude_bnds  (longitude, nv) float32 320B ...
  * time            (time) datetime64[ns] 3kB 1992-01-02T12:00:00 ... 1993-01...
    time_bnds       (time, nv) datetime64[ns] 6kB ...
Dimensions without coordinates: nv
Data variables:
    SALT            (time, Z, latitude, longitude) float32 117MB ...
    THETA           (time, Z, latitude, longitude) float32 117MB ...
Attributes: (12/53)
    acknowledgement:                 This research was carried out by the Jet...
    author:                          Ian Fenty and Ou Wang
    cdm_data_type:                   Grid
    comment:                         Fields provided on a regular lat-lon gri...
    Conventions:                     CF-1.8, ACDD-1.3
    coordinates_comment:             Note: the global 'coordinates' attribute...
    ...                              ...
    source:                          The ECCO V4r4 state estimate was produce...
    standard_name_vocabulary:        NetCDF Climate and Forecast (CF) Metadat...
    summary:                         This dataset provides daily-averaged oce...
    time_coverage_duration:          P1D
    time_coverage_resolution:        P1D
    title:                           ECCO Ocean Temperature and Salinity - Da...

xarray.Dataset

• Dimensions:
  • time: 366
  • Z: 50
  • latitude: 40
  • longitude: 40
  • nv: 2
• Coordinates: (8)
  • Z
    (Z)
    float32
    -5.0 -15.0 ... -5.906e+03

    axis :

    Z

    bounds :

    Z_bnds

    comment :

    Non-uniform vertical spacing.

    coverage_content_type :

    coordinate

    long_name :

    depth of grid cell center

    positive :

    up

    standard_name :

    depth

    units :

    m

    array([-5.000000e+00, -1.500000e+01, -2.500000e+01, -3.500000e+01,
           -4.500000e+01, -5.500000e+01, -6.500000e+01, -7.500500e+01,
           -8.502500e+01, -9.509500e+01, -1.053100e+02, -1.158700e+02,
           -1.271500e+02, -1.397400e+02, -1.544700e+02, -1.724000e+02,
           -1.947350e+02, -2.227100e+02, -2.574700e+02, -2.999300e+02,
           -3.506800e+02, -4.099300e+02, -4.774700e+02, -5.527100e+02,
           -6.347350e+02, -7.224000e+02, -8.144700e+02, -9.097400e+02,
           -1.007155e+03, -1.105905e+03, -1.205535e+03, -1.306205e+03,
           -1.409150e+03, -1.517095e+03, -1.634175e+03, -1.765135e+03,
           -1.914150e+03, -2.084035e+03, -2.276225e+03, -2.491250e+03,
           -2.729250e+03, -2.990250e+03, -3.274250e+03, -3.581250e+03,
           -3.911250e+03, -4.264250e+03, -4.640250e+03, -5.039250e+03,
           -5.461250e+03, -5.906250e+03], dtype=float32)

  • Z_bnds
    (Z, nv)
    float32
    ...

    comment :

    One pair of bounds for each vertical level.

    coverage_content_type :

    coordinate

    long_name :

    depths of grid cell upper and lower interfaces

    [100 values with dtype=float32]

  • latitude
    (latitude)
    float32
    25.25 25.75 26.25 ... 44.25 44.75

    axis :

    Y

    bounds :

    latitude_bnds

    comment :

    uniform grid spacing from -89.75 to 89.75 by 0.5

    coverage_content_type :

    coordinate

    long_name :

    latitude at grid cell center

    standard_name :

    latitude

    units :

    degrees_north

    array([25.25, 25.75, 26.25, 26.75, 27.25, 27.75, 28.25, 28.75, 29.25, 29.75,
           30.25, 30.75, 31.25, 31.75, 32.25, 32.75, 33.25, 33.75, 34.25, 34.75,
           35.25, 35.75, 36.25, 36.75, 37.25, 37.75, 38.25, 38.75, 39.25, 39.75,
           40.25, 40.75, 41.25, 41.75, 42.25, 42.75, 43.25, 43.75, 44.25, 44.75],
          dtype=float32)

  • latitude_bnds
    (latitude, nv)
    float32
    ...

    coverage_content_type :

    coordinate

    long_name :

    latitude bounds grid cells

    [80 values with dtype=float32]

  • longitude
    (longitude)
    float32
    -134.8 -134.2 ... -115.8 -115.2

    axis :

    X

    bounds :

    longitude_bnds

    comment :

    uniform grid spacing from -179.75 to 179.75 by 0.5

    coverage_content_type :

    coordinate

    long_name :

    longitude at grid cell center

    standard_name :

    longitude

    units :

    degrees_east

    array([-134.75, -134.25, -133.75, -133.25, -132.75, -132.25, -131.75, -131.25,
           -130.75, -130.25, -129.75, -129.25, -128.75, -128.25, -127.75, -127.25,
           -126.75, -126.25, -125.75, -125.25, -124.75, -124.25, -123.75, -123.25,
           -122.75, -122.25, -121.75, -121.25, -120.75, -120.25, -119.75, -119.25,
           -118.75, -118.25, -117.75, -117.25, -116.75, -116.25, -115.75, -115.25],
          dtype=float32)

  • longitude_bnds
    (longitude, nv)
    float32
    ...

    coverage_content_type :

    coordinate

    long_name :

    longitude bounds grid cells

    [80 values with dtype=float32]

  • time
    (time)
    datetime64[ns]
    1992-01-02T12:00:00 ... 1993-01-...

    axis :

    T

    bounds :

    time_bnds

    coverage_content_type :

    coordinate

    long_name :

    center time of averaging period

    standard_name :

    time

    chunksizes :

    none

    fletcher32 :

    False

    shuffle :

    False

    source :

    <File-like object HTTPFileSystem, https://archive.podaac.earthdata.nasa.gov/podaac-ops-cumulus-protected/ECCO_L4_TEMP_SALINITY_05DEG_DAILY_V4R4/OCEAN_TEMPERATURE_SALINITY_day_mean_1992-01-01_ECCO_V4r4_latlon_0p50deg.nc>

    original_shape :

    1

    dtype :

    int32

    array(['1992-01-02T12:00:00.000000000', '1992-01-03T12:00:00.000000000',
           '1992-01-04T12:00:00.000000000', ..., '1992-12-30T12:00:00.000000000',
           '1992-12-31T12:00:00.000000000', '1993-01-01T12:00:00.000000000'],
          shape=(366,), dtype='datetime64[ns]')

  • time_bnds
    (time, nv)
    datetime64[ns]
    ...

    comment :

    Start and end times of averaging period.

    coverage_content_type :

    coordinate

    long_name :

    time bounds of averaging period

    [732 values with dtype=datetime64[ns]]

• Data variables: (2)
  • SALT
    (time, Z, latitude, longitude)
    float32
    ...

    coverage_content_type :

    modelResult

    long_name :

    Salinity

    standard_name :

    sea_water_salinity

    units :

    1e-3

    valid_min :

    16.73577880859375

    valid_max :

    41.321231842041016

    comment :

    Defined using CF convention 'Sea water salinity is the salt content of sea water, often on the Practical Salinity Scale of 1978. However, the unqualified term 'salinity' is generic and does not necessarily imply any particular method of calculation. The units of salinity are dimensionless and the units attribute should normally be given as 1e-3 or 0.001 i.e. parts per thousand.' see https://cfconventions.org/Data/cf-standard-names/73/build/cf-standard-name-table.html

    [29280000 values with dtype=float32]

  • THETA
    (time, Z, latitude, longitude)
    float32
    ...

    coverage_content_type :

    modelResult

    long_name :

    Potential temperature

    standard_name :

    sea_water_potential_temperature

    units :

    degree_C

    comment :

    Sea water potential temperature is the temperature a parcel of sea water would have if moved adiabatically to sea level pressure. Note: the equation of state is a modified UNESCO formula by Jackett and McDougall (1995), which uses the model variable potential temperature as input assuming a horizontally and temporally constant pressure of $p_0=-g ho_{0} z$.

    valid_min :

    -2.9179372787475586

    valid_max :

    36.425140380859375

    [29280000 values with dtype=float32]

• Indexes: (4)
  • Z
    PandasIndex

    PandasIndex(Index([               -5.0,               -15.0,               -25.0,
                         -35.0,               -45.0,               -55.0,
                         -65.0,  -75.00499725341797,   -85.0250015258789,
            -95.09500122070312, -105.30999755859375, -115.87000274658203,
            -127.1500015258789, -139.74000549316406, -154.47000122070312,
           -172.39999389648438, -194.73500061035156,  -222.7100067138672,
            -257.4700012207031, -299.92999267578125, -350.67999267578125,
           -409.92999267578125,  -477.4700012207031,  -552.7100219726562,
            -634.7349853515625,  -722.4000244140625,   -814.469970703125,
             -909.739990234375,  -1007.155029296875,  -1105.905029296875,
           -1205.5350341796875, -1306.2049560546875, -1409.1500244140625,
            -1517.094970703125,  -1634.175048828125,  -1765.135009765625,
           -1914.1500244140625,  -2084.034912109375,   -2276.22509765625,
                      -2491.25,            -2729.25,            -2990.25,
                      -3274.25,            -3581.25,            -3911.25,
                      -4264.25,            -4640.25,            -5039.25,
                      -5461.25,            -5906.25],
          dtype='float32', name='Z'))

  • latitude
    PandasIndex

    PandasIndex(Index([25.25, 25.75, 26.25, 26.75, 27.25, 27.75, 28.25, 28.75, 29.25, 29.75,
           30.25, 30.75, 31.25, 31.75, 32.25, 32.75, 33.25, 33.75, 34.25, 34.75,
           35.25, 35.75, 36.25, 36.75, 37.25, 37.75, 38.25, 38.75, 39.25, 39.75,
           40.25, 40.75, 41.25, 41.75, 42.25, 42.75, 43.25, 43.75, 44.25, 44.75],
          dtype='float32', name='latitude'))

  • longitude
    PandasIndex

    PandasIndex(Index([-134.75, -134.25, -133.75, -133.25, -132.75, -132.25, -131.75, -131.25,
           -130.75, -130.25, -129.75, -129.25, -128.75, -128.25, -127.75, -127.25,
           -126.75, -126.25, -125.75, -125.25, -124.75, -124.25, -123.75, -123.25,
           -122.75, -122.25, -121.75, -121.25, -120.75, -120.25, -119.75, -119.25,
           -118.75, -118.25, -117.75, -117.25, -116.75, -116.25, -115.75, -115.25],
          dtype='float32', name='longitude'))

  • time
    PandasIndex

    PandasIndex(DatetimeIndex(['1992-01-02 12:00:00', '1992-01-03 12:00:00',
                   '1992-01-04 12:00:00', '1992-01-05 12:00:00',
                   '1992-01-06 12:00:00', '1992-01-07 12:00:00',
                   '1992-01-08 12:00:00', '1992-01-09 12:00:00',
                   '1992-01-10 12:00:00', '1992-01-11 12:00:00',
                   ...
                   '1992-12-23 12:00:00', '1992-12-24 12:00:00',
                   '1992-12-25 12:00:00', '1992-12-26 12:00:00',
                   '1992-12-27 12:00:00', '1992-12-28 12:00:00',
                   '1992-12-29 12:00:00', '1992-12-30 12:00:00',
                   '1992-12-31 12:00:00', '1993-01-01 12:00:00'],
                  dtype='datetime64[ns]', name='time', length=366, freq=None))

• Attributes: (53)

  acknowledgement :

  This research was carried out by the Jet Propulsion Laboratory, managed by the California Institute of Technology under a contract with the National Aeronautics and Space Administration.

  author :

  Ian Fenty and Ou Wang

  cdm_data_type :

  Grid

  comment :

  Fields provided on a regular lat-lon grid. They have been mapped to the regular lat-lon grid from the original ECCO lat-lon-cap 90 (llc90) native model grid.

  Conventions :

  CF-1.8, ACDD-1.3

  coordinates_comment :

  Note: the global 'coordinates' attribute describes auxillary coordinates.

  creator_email :

  ecco-group@mit.edu

  creator_institution :

  NASA Jet Propulsion Laboratory (JPL)

  creator_name :

  ECCO Consortium

  creator_type :

  group

  creator_url :

  https://ecco-group.org

  geospatial_bounds_crs :

  EPSG:4326

  geospatial_lat_max :

  90.0

  geospatial_lat_min :

  -90.0

  geospatial_lat_resolution :

  0.5

  geospatial_lat_units :

  degrees_north

  geospatial_lon_max :

  180.0

  geospatial_lon_min :

  -180.0

  geospatial_lon_resolution :

  0.5

  geospatial_lon_units :

  degrees_east

  geospatial_vertical_max :

  0.0

  geospatial_vertical_min :

  -6134.5

  geospatial_vertical_positive :

  up

  geospatial_vertical_resolution :

  variable

  geospatial_vertical_units :

  meter

  history :

  Inaugural release of an ECCO Central Estimate solution to PO.DAAC

  id :

  10.5067/ECG5D-OTS44

  institution :

  NASA Jet Propulsion Laboratory (JPL)

  instrument_vocabulary :

  GCMD instrument keywords

  keywords_vocabulary :

  NASA Global Change Master Directory (GCMD) Science Keywords

  license :

  Public Domain

  metadata_link :

  https://cmr.earthdata.nasa.gov/search/collections.umm_json?ShortName=ECCO_L4_TEMP_SALINITY_05DEG_DAILY_V4R4

  naming_authority :

  gov.nasa.jpl

  platform :

  ERS-1/2, TOPEX/Poseidon, Geosat Follow-On (GFO), ENVISAT, Jason-1, Jason-2, CryoSat-2, SARAL/AltiKa, Jason-3, AVHRR, Aquarius, SSM/I, SSMIS, GRACE, DTU17MDT, Argo, WOCE, GO-SHIP, MEOP, Ice Tethered Profilers (ITP)

  platform_vocabulary :

  GCMD platform keywords

  processing_level :

  L4

  product_time_coverage_end :

  2018-01-01T00:00:00

  product_time_coverage_start :

  1992-01-01T12:00:00

  product_version :

  Version 4, Release 4

  program :

  NASA Physical Oceanography, Cryosphere, Modeling, Analysis, and Prediction (MAP)

  project :

  Estimating the Circulation and Climate of the Ocean (ECCO)

  publisher_email :

  podaac@podaac.jpl.nasa.gov

  publisher_institution :

  PO.DAAC

  publisher_name :

  Physical Oceanography Distributed Active Archive Center (PO.DAAC)

  publisher_type :

  institution

  publisher_url :

  https://podaac.jpl.nasa.gov

  references :

  ECCO Consortium, Fukumori, I., Wang, O., Fenty, I., Forget, G., Heimbach, P., & Ponte, R. M. 2020. Synopsis of the ECCO Central Production Global Ocean and Sea-Ice State Estimate (Version 4 Release 4). doi:10.5281/zenodo.3765928

  source :

  The ECCO V4r4 state estimate was produced by fitting a free-running solution of the MITgcm (checkpoint 66g) to satellite and in situ observational data in a least squares sense using the adjoint method

  standard_name_vocabulary :

  NetCDF Climate and Forecast (CF) Metadata Convention

  summary :

  This dataset provides daily-averaged ocean potential temperature and salinity interpolated to a regular 0.5-degree grid from the ECCO Version 4 Release 4 (V4r4) ocean and sea-ice state estimate. Estimating the Circulation and Climate of the Ocean (ECCO) state estimates are dynamically and kinematically-consistent reconstructions of the three-dimensional, time-evolving ocean, sea-ice, and surface atmospheric states. ECCO V4r4 is a free-running solution of a global, nominally 1-degree configuration of the MIT general circulation model (MITgcm) that has been fit to observations in a least-squares sense. Observational data constraints used in V4r4 include sea surface height (SSH) from satellite altimeters [ERS-1/2, TOPEX/Poseidon, GFO, ENVISAT, Jason-1,2,3, CryoSat-2, and SARAL/AltiKa]; sea surface temperature (SST) from satellite radiometers [AVHRR], sea surface salinity (SSS) from the Aquarius satellite radiometer/scatterometer, ocean bottom pressure (OBP) from the GRACE satellite gravimeter; sea-ice concentration from satellite radiometers [SSM/I and SSMIS], and in-situ ocean temperature and salinity measured with conductivity-temperature-depth (CTD) sensors and expendable bathythermographs (XBTs) from several programs [e.g., WOCE, GO-SHIP, Argo, and others] and platforms [e.g., research vessels, gliders, moorings, ice-tethered profilers, and instrumented pinnipeds]. V4r4 covers the period 1992-01-01T12:00:00 to 2018-01-01T00:00:00.

  time_coverage_duration :

  P1D

  time_coverage_resolution :

  P1D

  title :

  ECCO Ocean Temperature and Salinity - Daily Mean 0.5 Degree (Version 4 Release 4)

map of salinity at 5 m depth

salt_t0z0 = data_local["SALT"].isel(time=0, Z=0)
plt.figure()
plt.contourf(salt_t0z0["longitude"], salt_t0z0["latitude"], salt_t0z0.values, levels=np.arange(32, 36, 0.25))
plt.colorbar()

salt_upper100m = data_local["SALT"].sel(Z=slice(0,-100))
salt_upper100m

<xarray.DataArray 'SALT' (time: 366, Z: 10, latitude: 40, longitude: 40)> Size: 23MB
array([[[[35.023365, ..., 33.73043 ],
         ...,
         [32.692387, ...,       nan]],

        ...,

        [[34.989086, ..., 33.54993 ],
         ...,
         [32.905792, ...,       nan]]],


       ...,


       [[[35.101208, ..., 33.75153 ],
         ...,
         [32.76833 , ...,       nan]],

        ...,

        [[35.093216, ..., 33.53298 ],
         ...,
         [32.952007, ...,       nan]]]], shape=(366, 10, 40, 40), dtype=float32)
Coordinates:
  * Z          (Z) float32 40B -5.0 -15.0 -25.0 -35.0 ... -75.0 -85.03 -95.1
  * latitude   (latitude) float32 160B 25.25 25.75 26.25 ... 43.75 44.25 44.75
  * longitude  (longitude) float32 160B -134.8 -134.2 -133.8 ... -115.8 -115.2
  * time       (time) datetime64[ns] 3kB 1992-01-02T12:00:00 ... 1993-01-01T1...
Attributes:
    coverage_content_type:  modelResult
    long_name:              Salinity
    standard_name:          sea_water_salinity
    units:                  1e-3
    valid_min:              16.73577880859375
    valid_max:              41.321231842041016
    comment:                Defined using CF convention 'Sea water salinity i...

xarray.DataArray

'SALT'

• time: 366
• Z: 10
• latitude: 40
• longitude: 40

• 35.02 35.02 34.97 34.97 34.9 34.9 34.83 ... nan nan nan nan nan nan

  array([[[[35.023365, ..., 33.73043 ],
           ...,
           [32.692387, ...,       nan]],
  
          ...,
  
          [[34.989086, ..., 33.54993 ],
           ...,
           [32.905792, ...,       nan]]],
  
  
         ...,
  
  
         [[[35.101208, ..., 33.75153 ],
           ...,
           [32.76833 , ...,       nan]],
  
          ...,
  
          [[35.093216, ..., 33.53298 ],
           ...,
           [32.952007, ...,       nan]]]], shape=(366, 10, 40, 40), dtype=float32)

• Coordinates: (4)
  • Z
    (Z)
    float32
    -5.0 -15.0 -25.0 ... -85.03 -95.1

    axis :

    Z

    bounds :

    Z_bnds

    comment :

    Non-uniform vertical spacing.

    coverage_content_type :

    coordinate

    long_name :

    depth of grid cell center

    positive :

    up

    standard_name :

    depth

    units :

    m

    array([ -5.   , -15.   , -25.   , -35.   , -45.   , -55.   , -65.   , -75.005,
           -85.025, -95.095], dtype=float32)

  • latitude
    (latitude)
    float32
    25.25 25.75 26.25 ... 44.25 44.75

    axis :

    Y

    bounds :

    latitude_bnds

    comment :

    uniform grid spacing from -89.75 to 89.75 by 0.5

    coverage_content_type :

    coordinate

    long_name :

    latitude at grid cell center

    standard_name :

    latitude

    units :

    degrees_north

    array([25.25, 25.75, 26.25, 26.75, 27.25, 27.75, 28.25, 28.75, 29.25, 29.75,
           30.25, 30.75, 31.25, 31.75, 32.25, 32.75, 33.25, 33.75, 34.25, 34.75,
           35.25, 35.75, 36.25, 36.75, 37.25, 37.75, 38.25, 38.75, 39.25, 39.75,
           40.25, 40.75, 41.25, 41.75, 42.25, 42.75, 43.25, 43.75, 44.25, 44.75],
          dtype=float32)

  • longitude
    (longitude)
    float32
    -134.8 -134.2 ... -115.8 -115.2

    axis :

    X

    bounds :

    longitude_bnds

    comment :

    uniform grid spacing from -179.75 to 179.75 by 0.5

    coverage_content_type :

    coordinate

    long_name :

    longitude at grid cell center

    standard_name :

    longitude

    units :

    degrees_east

    array([-134.75, -134.25, -133.75, -133.25, -132.75, -132.25, -131.75, -131.25,
           -130.75, -130.25, -129.75, -129.25, -128.75, -128.25, -127.75, -127.25,
           -126.75, -126.25, -125.75, -125.25, -124.75, -124.25, -123.75, -123.25,
           -122.75, -122.25, -121.75, -121.25, -120.75, -120.25, -119.75, -119.25,
           -118.75, -118.25, -117.75, -117.25, -116.75, -116.25, -115.75, -115.25],
          dtype=float32)

  • time
    (time)
    datetime64[ns]
    1992-01-02T12:00:00 ... 1993-01-...

    axis :

    T

    bounds :

    time_bnds

    coverage_content_type :

    coordinate

    long_name :

    center time of averaging period

    standard_name :

    time

    chunksizes :

    none

    fletcher32 :

    False

    shuffle :

    False

    source :

    <File-like object HTTPFileSystem, https://archive.podaac.earthdata.nasa.gov/podaac-ops-cumulus-protected/ECCO_L4_TEMP_SALINITY_05DEG_DAILY_V4R4/OCEAN_TEMPERATURE_SALINITY_day_mean_1992-01-01_ECCO_V4r4_latlon_0p50deg.nc>

    original_shape :

    1

    dtype :

    int32

    array(['1992-01-02T12:00:00.000000000', '1992-01-03T12:00:00.000000000',
           '1992-01-04T12:00:00.000000000', ..., '1992-12-30T12:00:00.000000000',
           '1992-12-31T12:00:00.000000000', '1993-01-01T12:00:00.000000000'],
          shape=(366,), dtype='datetime64[ns]')

• Indexes: (4)
  • Z
    PandasIndex

    PandasIndex(Index([              -5.0,              -15.0,              -25.0,
                        -35.0,              -45.0,              -55.0,
                        -65.0, -75.00499725341797,  -85.0250015258789,
           -95.09500122070312],
          dtype='float32', name='Z'))

  • latitude
    PandasIndex

    PandasIndex(Index([25.25, 25.75, 26.25, 26.75, 27.25, 27.75, 28.25, 28.75, 29.25, 29.75,
           30.25, 30.75, 31.25, 31.75, 32.25, 32.75, 33.25, 33.75, 34.25, 34.75,
           35.25, 35.75, 36.25, 36.75, 37.25, 37.75, 38.25, 38.75, 39.25, 39.75,
           40.25, 40.75, 41.25, 41.75, 42.25, 42.75, 43.25, 43.75, 44.25, 44.75],
          dtype='float32', name='latitude'))

  • longitude
    PandasIndex

    PandasIndex(Index([-134.75, -134.25, -133.75, -133.25, -132.75, -132.25, -131.75, -131.25,
           -130.75, -130.25, -129.75, -129.25, -128.75, -128.25, -127.75, -127.25,
           -126.75, -126.25, -125.75, -125.25, -124.75, -124.25, -123.75, -123.25,
           -122.75, -122.25, -121.75, -121.25, -120.75, -120.25, -119.75, -119.25,
           -118.75, -118.25, -117.75, -117.25, -116.75, -116.25, -115.75, -115.25],
          dtype='float32', name='longitude'))

  • time
    PandasIndex

    PandasIndex(DatetimeIndex(['1992-01-02 12:00:00', '1992-01-03 12:00:00',
                   '1992-01-04 12:00:00', '1992-01-05 12:00:00',
                   '1992-01-06 12:00:00', '1992-01-07 12:00:00',
                   '1992-01-08 12:00:00', '1992-01-09 12:00:00',
                   '1992-01-10 12:00:00', '1992-01-11 12:00:00',
                   ...
                   '1992-12-23 12:00:00', '1992-12-24 12:00:00',
                   '1992-12-25 12:00:00', '1992-12-26 12:00:00',
                   '1992-12-27 12:00:00', '1992-12-28 12:00:00',
                   '1992-12-29 12:00:00', '1992-12-30 12:00:00',
                   '1992-12-31 12:00:00', '1993-01-01 12:00:00'],
                  dtype='datetime64[ns]', name='time', length=366, freq=None))

• Attributes: (7)

  coverage_content_type :

  modelResult

  long_name :

  Salinity

  standard_name :

  sea_water_salinity

  units :

  1e-3

  valid_min :

  16.73577880859375

  valid_max :

  41.321231842041016

  comment :

  Defined using CF convention 'Sea water salinity is the salt content of sea water, often on the Practical Salinity Scale of 1978. However, the unqualified term 'salinity' is generic and does not necessarily imply any particular method of calculation. The units of salinity are dimensionless and the units attribute should normally be given as 1e-3 or 0.001 i.e. parts per thousand.' see https://cfconventions.org/Data/cf-standard-names/73/build/cf-standard-name-table.html

Compute and plot time series of mean salinity profile at top 100 meters from ocean surface

salt_upper100m = data_local["SALT"].sel(Z=slice(0,-100))
salt_mean = salt_upper100m.mean(dim=["longitude", "latitude"])

fig = plt.figure(figsize=(12,4))
plt.contourf(salt_mean["time"], salt_mean["Z"], salt_mean.values.T)
plt.xlabel("time")
plt.ylabel("depth (m)")
plt.title("Mean salinity profile off USA west coast (PSU)")
plt.colorbar()