Authors James Kuo, [email protected]; Stephanie Petillo, [email protected]; and Christopher Wingard, [email protected]
The Ocean Observatories Initiative (OOI) collects ocean sonar, or bioacoustic sonar data, from most of the sites that comprise the different Coastal and Global arrays. The sensor type used depends on whether the site is cabled or uncabled. The ASL Environmental Sciences Acoustic Zooplankton Fish Profiler (AZFP) is used at the uncabled Coastal and Global (CG) sites, while the Kongsberg Maritime Simrad EK60 Echo Sounder is deployed at two of the cabled sites. The table below provides a full list of the arrays, sites and sensor types where the bioacoustic sonar sensors (a total of 17) are deployed.
Array | Site | Sensor Model | Depth | Description |
---|---|---|---|---|
Coastal Endurance | CE01ISSM | AZFP | 25 m | Oregon Inshore Surface Mooring |
Coastal Endurance | CE02SHBP | EK60 | 80 m | Oregon Shelf Cabled Benthic Experiment Package |
Coastal Endurance | CE04OSPS | EK60 | 200 m | Oregon Offshore Cabled Shallow Profiler Mooring |
Coastal Endurance | CE06ISSM | AZFP | 29 m | Washington Inshore Surface Mooring |
Coastal Endurance | CE07SHSM | AZFP | 87 m | Washington Shelf Surface Mooring |
Coastal Endurance | CE09OSSM | AZFP | 542 m | Washington Offshore Surface Mooring |
Coastal Pioneer | CP03ISSM | AZFP | 95 m | Inshore Surface Mooring |
Coastal Pioneer | CP01CNSM | AZFP | 135 m | Central Surface Mooring |
Coastal Pioneer | CP04OSSM | AZFP | 450 m | Offshore Surface Mooring |
Global Argentine Basin | GA02HYPM | AZFP | 150 m | Apex Profiler Mooring, both upward and downward looking sensors |
Global Irminger Sea | GI02HYPM | AZFP | 150 m | Apex Profiler Mooring, both upward and downward looking sensors |
Global Station Papa | GP02HYPM | AZFP | 150 m | Apex Profiler Mooring, both upward and downward looking sensors |
Global Southern Ocean | GS02HYPM | AZFP | 150 m | Apex Profiler Mooring, both upward and downward looking sensors |
Echopype, the open-source ocean sonar data processing module, is utilized to process the OOI bioacoustic sonar data from the cabled and uncabled sensors to generate weekly echograms that the user community can use to quickly assess the quality and applicability of the data to their research needs. Additionally, preliminary processed data (saved to NetCDF files) is generated to facilitate access to the data in a standard format for others to use for further processing and analysis.
This code was originally developed as part of the ooicgsn-data-tools toolbox, but was extracted out to make it standalone and installable as a package for internal OOI use in processing the realtime cabled data from CE02SHBP and CE04OSPS.
Future releases will make this code available for installation via conda-forge. For now users will need to download (or use git to clone) the code to their local computer, setup a virtual environment for the processing, and then install a development copy of the code into that environment. The installation and setup steps below, assume the user has already installed either anaconda or miniconda to use in executing the python code.
# download the ooi-zpls-echograms code
mkdir -p /home/ooiuser/code
cd /home/ooiuser/code
git clone https://github.com/oceanobservatories/ooi-zpls-echograms.git
cd ooi-zpls-echograms
# configure the OOI python environment
conda env create -f environment.yml
conda activate echopype
# install the package as a local development package
conda develop .
Note, the use of /home/ooiuser/code
above, and /home/ooiuser/data
below, are solely for the purposes of these
examples. It is expected that users will have their own directory structure. Adapt and change as needed.
The code is intended to be used largely from the command line in a shell. Though users can import ooi_zpls_echograms
as a module and run the individual functions directly, the intent is to provide a means of batch processing the
bioacoustic sonar data for internal OOI use. Scripts used to run the processing are available in the utilities
directory. The basic sequence of commands is:
cd /home/ooiuser/code/ooi-zpls-echograms/ooi_zpls_echograms
./zpls_echogram.py -s [site_code] -d [data_directory] -o [output_directory] -dr [dates] -zm [instrument_type]
The inputs to the function are defined below, based on whether they are required or optional.
Required Inputs | Description |
---|---|
-s, -site_code | The OOI 8-letter site name for where the ZPLSC/G is located. |
-d, --data_directory | The path to the root data directory, below which the .01A or .raw files may be found. |
-o, --output_directory | The path to the root data directory where .nc file and .png plot will be saved. |
-dr, --date_range | The date range to be plotted, e.g.:-dr "20200118" will plot the day (01/18/2020),-dr "202001" will plot the whole month (01/01/2020 to 01/31/2020),-dr "202001 202002" will plot (01/01/2020 to 02/29/2020),-dr "20200115 20200216" will plot (01/15/2020 to 02/16/2020) |
-zm, --zpls_model | Sensor model, either an EK60 for cabled data, or an AZFP for uncabled |
-xf, --xml_file | The path to .xml file to use if the sensor model is an AZFP. Required if the sensor model is AZFP |
Optional flags | Description |
---|---|
-tc, --tilt_correction | Apply tilt correction in degree(s) |
-dd, --deployed_depth | The depth where the ZPLSC/G is located at |
-cr, --colorbar_range | Set colorbar range. Usage: "min max" |
-vr, --vertical_range | Set the range for the y-axis. Usage: "min max" |
-
The root data directory depends on the data source. If looking at cabled data and the EK60, the path to the root data directory ends with the year (e.g.
/home/ooiuser/data/CE04OSPS/PC01B/05-ZPLSCB102/2017
). For uncabled data, the path to the root data directory ends withDATA
(e.g./home/ooiuser/data/CP01CNSM/R00012/instruments/dcl37/ZPLSC_sn55080/DATA
). -
The path to the
.xml
file is required if the ZPLS model is an AZFP. The.xml
file contains instrument specific settings and calibration coefficients needed to process the data. -
In the case of the global Apex Profiler (subsurface) moorings, the OOI 8-letter site name is appended with either an "_Upper" or "_Lower" to indicate which of the sensors is being used (there are two on the 150 m sphere, one looking up towards the surface, and the other looking down).
-
The preset colorbar ranges are the same for every array. Feel free to change it via the optional colorbar_range input.
The raw ZPLS data can be found on the OOI Raw Data server, with the directory structure dependent on the ZPLS instrument type (either an EK60 or AZFP).
For the AZFP, the data are organized by the instrument in a folder called DATA
, below which the files are organized
in folders named according the year and month (e.g. 201610
) the data was collected. The files are named according to
the year, month, day and hour the data were recorded (e.g. 16101401.01A
). The higher level directory structure above
DATA
is created by OOI technicians after the mooring is recovered. If the ZPLS data from the
Global Argentine Basin, upward looking sensor for Deployment 3
were downloaded to a local folder in the users /home/ooiuser/data
directory named GA
, the folders would look like:
/home/ooiuser/data/GA/ZPLSG_sn55067/DATA/201610
GA/ZPLSG_sn55067/DATA/201610/16101401.01A
GA/ZPLSG_sn55067/DATA/201610/16101402.01A
GA/ZPLSG_sn55067/DATA/201610/16101403.01A
GA/ZPLSG_sn55067/DATA/201610/*.01A
GA/ZPLSG_sn55067/DATA/201610/16101417.XML
/home/ooiuser/data/GA/ZPLSG_sn55067/DATA/201611
GA/ZPLSG_sn55067/DATA/201611/16111401.01A
GA/ZPLSG_sn55067/DATA/201611/16111402.01A
GA/ZPLSG_sn55067/DATA/201611/16111403.01A
GA/ZPLSG_sn55067/DATA/201611/*.01A
For the EK60, the data are organized by OOI technicians in folders based on the year, month and day (e.g. 2017/01/01
)
the data was collected. Files are named by the instrument according to the date and time the file was created (e.g.
OOI-D20170831-T113333.raw
). If the ZPLS data from the Coastal Endurance, Oregon Offshore Cabled Shallow Profiler
Mooring, 200 m Platform from 2017 were
downloaded to local folder in the user's /home/ooiuser/data
directory named CE04
, the folders would look like:
/home/ooiuser/data/CE04/PC01B/05-ZPLSCB102/2017/01/01/
CE04/PC01B/05-ZPLSCB102/2017/01/01/OOI-D20170101-T000000.bot
CE04/PC01B/05-ZPLSCB102/2017/01/01/OOI-D20170101-T000000.idx
CE04/PC01B/05-ZPLSCB102/2017/01/01/OOI-D20170101-T000000.raw
/home/ooiuser/data/CE04/PC01B/05-ZPLSCB102/2017/01/02
CE04/PC01B/05-ZPLSCB102/2017/01/02/OOI-D20170102-T000000.bot
CE04/PC01B/05-ZPLSCB102/2017/01/02/OOI-D20170102-T000000.idx
CE04/PC01B/05-ZPLSCB102/2017/01/02/OOI-D20170102-T000000.raw
For the AZFP example above, the root data directory would be /home/ooiuser/data/GA/ZPLSG_sn55067/DATA
. For the EK60
example, the root data directory would be /home/ooiuser/data/CE04/PC01B/05-ZPLSCB102/2017
.
Processed data are created in a folder of the users choice. We recommend placing the data at the same level as the root
data directory, but in a separate folder called processed
. The data in the processed
folder are organized into
sub-folders named according to the date range entered for processing. For example, using the AZFP data from above, if
the date range was 2017-01-01 through 2017-01-08, then the processed data would end up in a folder 20170101-20170108
.
The full path would be /home/ooiuser/data/GA/ZPLSG_sn55067/processed/20170101-20170108/
There are three types of processed data created by the zpls_echogram script:
- Converted: These are files converted from the raw format (either
.01A
or.raw
) to NetCDF format (.nc
) by the echopype convert function. They are 1:1 with the source raw data file. These.nc
files use the SONAR-NetCDF4 formatting convention. - Processed: The echopype process function, taking the converted files as inputs, is used to calculate the volume
acoustic backscatter strength (Sv), and the range (distance from the sensor face) in meters. This subset of data is
saved in daily
.nc
files that use the site and date range to construct the file name. Continuing to use the AZFP example from above, data for 2017-01-01 would be saved inGA02HYPM_Upper_Bioacoustic_Echogram_20170101-20170108_Calibrated_Sv_Full_20170107.nc
. - Averaged: Takes the processed data one step further and applies temporal averaging (15 minute median averages for
the coastal ZPLSC and 60 minute median averages for the global ZPLSG) for use in creating the echograms. Missing data
is filled in via linear interpolation for gaps less than 3x the averaging time window. The averaged data is saved
in a single
.nc
file covering the date range specified using a similar file naming convention as the processed data above except the termFull_20170101
in the file name above is replaced withAveraged
(e.g.GA02HYPM_Upper_Bioacoustic_Echogram_20170101-20170108_Calibrated_Sv_Averaged.nc
).
An echogram covering the time period specified by the date range is created from the averaged data and saved along with
the three data types listed above. The files share the same naming convention as the averaged data (e.g.,
GA02HYPM_Upper_Bioacoustic_Echogram_20170101-20170108_Calibrated_Sv_Averaged.png
).
The examples below show how to use the python script with some of the different options from above. The examples are all built around using the AZFP data from the Global Argentine Basin upward looking sensor.
Example 1:
# set default data directories and the XML file
GA_DATA="/home/ooiuser/data/GA/ZPLSG_sn55067/DATA"
GA_PROC="/home/ooiuser/data/GA/ZPLSG_sn55067/processed"
GA_XML="/home/ooiuser/data/GA/ZPLSG_sn55067/DATA/201610/16101417.XML"
# set the site, the script will look at the default mooring configuration and
# apply the tilt angle correction (15 degrees for GA02HYPM_Upper) and the
# deployment depth (150 m for GA02HYPM_Upper).
SITE="GA02HYPM_Upper"
# convert and process the data
conda activate echopype
cd /home/ooiuser/code/ooi-zpls-echograms/ooi_zpls_echograms
./zpls_echogram.py -s $SITE -d $GA_DATA -o $GA_PROC -dr "20161115 20161116" -zm AZFP -xf $GA_XML
Process data from 2016-11-15 through 2016-11-16. The .nc
files and .png
plot will be stored under
/home/ooiuser/data/GA/ZPLSG_sn55067/processed/20161115-20161116/
.
Example 2:
./zpls_echogram.py -s $SITE -d $GA_DATA -o $GA_PROC -dr "201611" -zm AZFP -xf $GA_XML
Process the data for the entire month of November, from 2016-11-01 through 2016-11-30. The .nc
files and .png
plot
will be stored under /home/ooiuser/data/GA/ZPLSG_sn55067/processed/20161101-20161130/
.
Example 3:
./zpls_echogram.py -s $SITE -d $GA_DATA -o $GA_PROC -dr "201611" -zm AZFP -xf $GA_XML -tc 10
Change the default tilt correction angle to 10 degrees via the optional -tc
flag.
Example 4:
./zpls_echogram.py -s $SITE -d $GA_DATA -o $GA_PROC -dr "201611" -zm AZFP -xf $GA_XML -dd 160
If the instrument is deployed at 160 m, instead of the design depth of 150 m. The optional flag -dd 160
will
override the mooring configuration.
Example 5:
./zpls_echogram.py -s $SITE -d $GA_DATA -o $GA_PROC -dr "201611" -zm AZFP -xf $GA_XML -cr -120 -20 -dd 160
Change both the deployed depth to 160 m and the default colorbar range to -120 to -20 dB via the optional -cr
flag.
Example 6:
./zpls_echogram.py -s $SITE -d $GA_DATA -o $GA_PROC -xf $GA_XML -dr "201611" -zm AZFP -vr 0 160 -dd 160
Change the default vertical range to use 0 to 160 m for the new deployment depth of 160 m via the optional -vr
flag.