CTD profiles - minimal example

📥 Download: CTD profiles - minimal example.ipynb

---

In this notebook

• Load multiple .cnv files containing shipboard profiles from a SeaBird CTD.

• Quick overview and basic plots.

# Import the ctd module from kval.data
from kval.data import ctd 
# Set the plotting backed so we can interact with figures
%matplotlib widget 

/home/docs/checkouts/readthedocs.org/user_builds/kval/conda/latest/lib/python3.12/site-packages/kval/__init__.py:9: UserWarning: NOTE_ Matplotlib >=3.9 may *in some cases* break some ipympl interactive plots. If you experience trouble with plots not displaying correctly, try downgrading to matplotlib 3.8.x.
  warnings.warn(

Load data from .cnv files into a single xarray Dataset

Specify a directory cnv_dir where we have .cnv files collected from a SBE CTD.

---

NOTE: Loading the data in this manner will load all the .cnv files in the directory. Make sure cnv_dir/ only contains the files you want to include.

---

NOTE: In this case, the .cnv files contain pressure-gridded data. It is also possible to load ungridded profiles - profiles will then be pressure gridded within kval.

cnv_dir = ('../../../../tests/test_data/'
           'sbe_files/sbe911plus/atwain_cruise_ctds/')

Load all .cnv files into a single xarray Dataset - the object we will work with and ultimately export to netCDF.

ds = ctd.ctds_from_cnv_dir(cnv_dir)

Found 3 .cnv files in  "../../../../tests/test_data/sbe_files/sbe911plus/atwain_cruise_ctds/".
NOTE: It seems the input data already binned -> using preexisting binning.

/home/docs/checkouts/readthedocs.org/user_builds/kval/conda/latest/lib/python3.12/site-packages/kval/data/ship_ctd_tools/_ctd_tools.py:168: FutureWarning: In a future version of xarray the default value for join will change from join='outer' to join='exact'. This change will result in the following ValueError: cannot be aligned with join='exact' because index/labels/sizes are not equal along these coordinates (dimensions): 'PRES' ('PRES',) The recommendation is to set join explicitly for this case.
  N = xr.concat([N, n], dim = 'TIME', data_vars="all")

The xarray Dataset ds should contain 1-D and 2-D fields gridded on pressure (PRES) and time (TIME).

When available, metadata such as serial numbers and calibration data have been parsed into variable attributes. The PROCESSING variable should include a description of the processing steps and a python script that can be used to reproduce the processing we do here (we add to these fields as we apply methods to the dataset).

Have a look at the data

Take a first look at the dataset

To have a look around, run ds in a cell and browse (if you are running the notebook interactively). Click the ⛃ and 🗎 symbols on the right to display data and metadata.

ds

<xarray.Dataset> Size: 93kB
Dimensions:            (TIME: 3, PRES: 527)
Coordinates:
  * TIME               (TIME) float64 24B 1.743e+04 1.743e+04 1.743e+04
  * PRES               (PRES) float64 4kB 3.0 4.0 5.0 6.0 ... 527.0 528.0 529.0
    STATION            (TIME) <U5 60B 'AT285' 'AT286' 'AT287'
    LATITUDE           (TIME) float64 24B 80.11 80.07 80.04
    LONGITUDE          (TIME) float64 24B 9.875 10.38 10.75
Data variables:
    TEMP               (TIME, PRES) float64 13kB nan 6.362 6.367 ... nan nan nan
    PTEMP              (TIME, PRES) float64 13kB nan 6.362 6.367 ... nan nan nan
    CNDC               (TIME, PRES) float64 13kB nan 34.68 34.69 ... nan nan nan
    PSAL               (TIME, PRES) float64 13kB nan 34.98 34.98 ... nan nan nan
    SIGTH              (TIME, PRES) float64 13kB nan 27.49 27.49 ... nan nan nan
    CHLA_fluorescence  (TIME, PRES) float64 13kB nan 1.39 1.293 ... nan nan nan
    SBE_FLAG           (TIME, PRES) float64 13kB nan 0.0 0.0 0.0 ... nan nan nan
    PROCESSING         object 8B None
Attributes:
    binned:            1 decibars (SBE software)
    source_file:       E.g. AT287.hex, AT287.XMLCON -> AT287.CNV
    history:           2017-09-24 to 2017-09-24: Data collection.\n2018-07-23...
    instrument_model:  Sea-Bird SBE 9
    featureType:       profile

xarray.Dataset

• Dimensions:
  • TIME: 3
  • PRES: 527
• Coordinates: (5)
  • TIME
    (TIME)
    float64
    1.743e+04 1.743e+04 1.743e+04

    units :

    Days since 1970-01-01 00:00

    standard_name :

    time

    long_name :

    Start time of profile

    comment :

    Source: "start_time" header line

    array([17433.745012, 17433.797002, 17433.844711])

  • PRES
    (PRES)
    float64
    3.0 4.0 5.0 ... 527.0 528.0 529.0

    units :

    dbar

    sensor_serial_number :

    0972

    sensor_calibration_date :

    20-Feb-14

    coverage_content_type :

    coordinate

    array([  3.,   4.,   5., ..., 527., 528., 529.], shape=(527,))

  • STATION
    (TIME)
    <U5
    'AT285' 'AT286' 'AT287'

    long_name :

    CTD station ID

    cf_role :

    profile_id

    array(['AT285', 'AT286', 'AT287'], dtype='<U5')

  • LATITUDE
    (TIME)
    float64
    80.11 80.07 80.04

    standard_name :

    latitude

    units :

    degree_north

    long_name :

    latitude

    array([80.112 , 80.0681, 80.0429])

  • LONGITUDE
    (TIME)
    float64
    9.875 10.38 10.75

    standard_name :

    longitude

    units :

    degree_east

    long_name :

    longitude

    array([ 9.8747, 10.3808, 10.7464])

• Data variables: (8)
  • TEMP
    (TIME, PRES)
    float64
    nan 6.362 6.367 ... nan nan nan

    units :

    degree_Celsius

    sensor_serial_number :

    2400

    sensor_calibration_date :

    11-Apr-17

    coverage_content_type :

    physicalMeasurement

    sensor_mount :

    mounted_on_shipborne_profiler

    array([[   nan, 6.3619, 6.3672, ..., 3.7716, 3.7715, 3.767 ],
           [6.3235, 6.3271, 6.3348, ...,    nan,    nan,    nan],
           [6.0014, 6.0089, 6.0013, ...,    nan,    nan,    nan]],
          shape=(3, 527))

  • PTEMP
    (TIME, PRES)
    float64
    nan 6.362 6.367 ... nan nan nan

    units :

    degree_Celsius

    sensor_serial_number :

    2400

    sensor_calibration_date :

    11-Apr-17

    coverage_content_type :

    physicalMeasurement

    sensor_mount :

    mounted_on_shipborne_profiler

    array([[   nan, 6.3615, 6.3668, ..., 3.7341, 3.7339, 3.7293],
           [6.3233, 6.3268, 6.3344, ...,    nan,    nan,    nan],
           [6.0012, 6.0086, 6.0009, ...,    nan,    nan,    nan]],
          shape=(3, 527))

  • CNDC
    (TIME, PRES)
    float64
    nan 34.68 34.69 ... nan nan nan

    units :

    mS cm-1

    sensor_serial_number :

    3234

    sensor_calibration_date :

    12Apr-17

    coverage_content_type :

    physicalMeasurement

    sensor_mount :

    mounted_on_shipborne_profiler

    array([[      nan, 34.678238, 34.685049, ..., 32.645796, 32.646174,
            32.642579],
           [34.588011, 34.593203, 34.602327, ...,       nan,       nan,
                  nan],
           [34.170361, 34.178967, 34.171042, ...,       nan,       nan,
                  nan]], shape=(3, 527))

  • PSAL
    (TIME, PRES)
    float64
    nan 34.98 34.98 ... nan nan nan

    units :

    1

    sensor_serial_number :

    2400, 3234

    sensor_calibration_date :

    11-Apr-17, 12Apr-17

    coverage_content_type :

    physicalMeasurement

    sensor_mount :

    mounted_on_shipborne_profiler

    array([[    nan, 34.9766, 34.9782, ..., 35.0679, 35.0679, 35.068 ],
           [34.9153, 34.917 , 34.9188, ...,     nan,     nan,     nan],
           [34.776 , 34.7775, 34.7758, ...,     nan,     nan,     nan]],
          shape=(3, 527))

  • SIGTH
    (TIME, PRES)
    float64
    nan 27.49 27.49 ... nan nan nan

    units :

    kg m-3

    sensor_serial_number :

    2400, 3234

    sensor_calibration_date :

    11-Apr-17, 12Apr-17

    coverage_content_type :

    physicalMeasurement

    sensor_mount :

    mounted_on_shipborne_profiler

    array([[    nan, 27.4872, 27.4878, ..., 27.8678, 27.8679, 27.8684],
           [27.4438, 27.4447, 27.4451, ...,     nan,     nan,     nan],
           [27.3755, 27.3758, 27.3754, ...,     nan,     nan,     nan]],
          shape=(3, 527))

  • CHLA_fluorescence
    (TIME, PRES)
    float64
    nan 1.39 1.293 ... nan nan nan

    units :

    mg m-3

    sensor_serial_number :

    FLRTD-1547

    sensor_calibration_date :

    4-Jan-16

    coverage_content_type :

    physicalMeasurement

    sensor_mount :

    mounted_on_shipborne_profiler

    array([[    nan,  1.3904,  1.2928, ..., -0.101 , -0.1005, -0.1034],
           [ 1.3349,  1.3979,  1.4288, ...,     nan,     nan,     nan],
           [ 1.2438,  1.3095,  1.4271, ...,     nan,     nan,     nan]],
          shape=(3, 527))

  • SBE_FLAG
    (TIME, PRES)
    float64
    nan 0.0 0.0 0.0 ... nan nan nan nan

    units :

    array([[nan,  0.,  0., ...,  0.,  0.,  0.],
           [ 0.,  0.,  0., ..., nan, nan, nan],
           [ 0.,  0.,  0., ..., nan, nan, nan]], shape=(3, 527))

  • PROCESSING
    ()
    object
    None

    long_name :

    Empty variable whose attributes describe processing history of the dataset.

    comment :

    # source_file #: List of files produced by SBE processing. # post_processing #: Description of post-processing starting with `source_file`. (Note: Indexing in the PRES dimension starts at 0 - for MATLAB add 1 to the index). # python_script #: Python script for reproducing post-processing from `source_file`.

    SBE_processing :

    SBE SOFTWARE PROCESSING STEPS (extracted from .cnv file header): 1. Raw data read from e.g. AT287.hex, AT287.XMLCON. 2. Bin averaged (1 decibars). > Bad scans not excluded, skipped over 0 initial scans. > (No surface bin).

    source_file :

    AT285.cnv, AT286.cnv, AT287.cnv

    post_processing :

    python_script :

    from kval import data # Path to directory containing *source_file* (MUST BE SET BY THE USER!) cnv_dir = "./" # Load all .cnv files and join together into a single xarray Dataset: ds = data.ctd.ctds_from_cnv_dir( cnv_dir, station_from_filename=False, start_time_NMEA=False, processing_variable=True )

    array(None, dtype=object)

• Attributes: (5)

  binned :

  1 decibars (SBE software)

  source_file :

  E.g. AT287.hex, AT287.XMLCON -> AT287.CNV

  history :

  2017-09-24 to 2017-09-24: Data collection. 2018-07-23: Processed to .cnv using SBE software. 2025-11-26: Post-processing.

  instrument_model :

  Sea-Bird SBE 9

  featureType :

  profile

Quick map

Show a quick map of the profile locations. May or may not be useful depending on your dataset.

ctd.map(ds)

/home/docs/checkouts/readthedocs.org/user_builds/kval/conda/latest/lib/python3.12/site-packages/cartopy/io/__init__.py:242: DownloadWarning: Downloading: https://naturalearth.s3.amazonaws.com/50m_physical/ne_50m_land.zip
  warnings.warn(f'Downloading: {url}', DownloadWarning)

Contour plots

The ctd.contour() function displays two contour plots. You can interactively change which variables are displayed and adjust the axes.

ctd.contour(ds)

Profile plots

Have a quick look at individual profiles. Interactively select profiles and variables.

ctd.inspect_profiles(ds)