Benthic Pod Data

The Ocean Tracking Network (OTN) deploys sensors on a series of benthic pods (BP) on the Scotian Shelf. These pods were designed in collaboration with Satlantic to monitor and report key oceanographic parameters to provide environmental context to animal movements and validation data to sophisticated biogeochemical and physical ocean models. Data stored on the pods can be recovered via acoustic modem from either surface vessels or autonomous vehicles.

Mounted on the BPs are:

  • Satlantic STOR-X data/power manager (http://satlantic.com/stor-x)
  • Digiquartz Paroscientific pressure sensor (http://www.paroscientific.com/depthsensors.htm)
  • VEMCO VR4 acoustic receiver (http://vemco.com/products/vr4-uwm/)
  • Sea-Bird 37SIP Microcat (http://www.seabird.com/products/spec_sheets/37sipdata.htm)
  • Aanderaa dissolved oxygen sensor (3830 & 4330; http://www.aanderaa.com/productsdetail.php?Oxygen-Optodes-2)
  • Satlantic 15V, 102Ah alkaline battery pack (http://satlantic.com/sites/default/files/documents/AlkalineBatteryPack-RevE-Manual.pdf)

The Paroscientific Digiquartz pressure sensor is housed inside the STOR-X, the Aanderaa oxygen optode is connected to the top of the STOR-X, and the Sea-Bird SBE 37-SIP Microcat is connected by cable to the STOR-X. As configured the BPs can be deployed to a depth of 330m.

The STOR-X controls the sampling of the oceanographic instrumentation. The sensors sample for 30s every hour on the hour. The STOR-X stores the requested sampling data from the three connected sensors (C:\Data\*.RAW) and also processes the resulting 30 seconds of sampling into single averaged hourly record in a file called PROCESS.DAT (also found at C:\DATA).

The OTN Benthic Pods are deployed with a stable bottom mount so that the data from the sensitive pressure sensor is not aliased by movements of the frame.

Download unprocessed data

StationPod IDLast Record / Offload DateDownload Link
HFX0080812011-06-10 14:00PROCESS-POD0081A.dat
HFX008b0952011-06-10 14:00PROCESS-POD0095a.dat
HFX0080902012-11-16 20:00PROCESS-POD0090a.dat
HFX0280992012-11-12 18:00PROCESS-POD0099a.dat
HFX0480892012-11-20 17:00PROCESS-POD0089a.dat
HFX0690922012-11-20 20:00PROCESS-POD0092a.dat
HFX0970942012-12-14 14:00PROCESS-POD0094a.dat
HFX1261162013-05-29 12:00PROCESS-POD0116A.dat
HFX1531182013-05-19 17:00PROCESS-POD0118A.dat
HFX1800952013-05-28 13:00BENTHIC_POD_95_20130528_1-POD0095a.dat
HFX2120962013-05-19 20:00PROCESS-POD0096a.dat
HFX2450912013-05-19 22:00PROCESS-POD0091a.dat

Once the data are collected they are sent to the Bedford Institute of Oceanography (BIO) for processing. Below is information provided by BIO about the data processing.


Input Data

The Dalhousie OTN program sends files to BIO as a zip archive. Contained in the archive are Excel files with a deployment log sheet for each instrument, files with instrument metadata, and a deployment and recovery mission report. There is also one data file (extension “.dat”) for each instrument deployed. The filename format is generally “PROCESS-PODXXXXa.dat”, where XXXX is the serial number of the pod. The input data file has a 5 line header. Included is the name of the Satlantic calibration file used to convert the data to physical units, and the list of parameters in the file, along with the units for each parameter.

The data are in physical units. Each processed record is an average of a number of raw data records, sampled as a burst just before the reported time. The number of raw data records in the average is reported in field 3 of the data file, referred to as “SIZE”.

Processing Steps

The Excel files containing the deployment information are opened in Excel and converted into “pdf” documents for submission to the Ocean Data and Information Services (ODIS) data shop.

The relevant information in the log sheets is identified for use in the processing script used to convert the information into DFO’s ODF data format. This includes the mooring position, mooring depth, station name, deployment and recovery time, sensor model numbers and serial numbers, mission name and chief scientist name.

A Matlab script, named convert_NNNN_to_ODF.m, is used to read the data and output a single “ODF” file. The user is prompted for the input file name. Before running the script, the instrument and deployment information identified above is entered into the relevant section of code. In addition to creating the ODF file, a number of other functions are performed, as listed below

  • Conductivity in mmho/cm is converted to conductivity ratio, by dividing by 4.2914
  • Pressure is corrected for atmospheric pressure, by subtracting 1 ATM (10.1325 dbar).
  • Salinity is recomputed, compensated for temperature and pressure. This is done by function “sw_salt”, in the Matlab Seawater toolbox.
  • Oxygen in micromolar/L is compensated for salinity, temperature and pressure by function “o2comp.m”. This function was written by Roger Pettipas (DFO) and was derived from Aanderaa Data Instruments, “Compensation of Oxygen Measurements for Salinity and Depth’1, contained in a spreadsheet named “Oxygen Optode Calculations-2.xls”.
  • Oxygen in micromolar/L is converted to ml/l by multiplying by 0.022391.
  • Oxygen saturation value is recalculated with the corrected salinity using function “sw_satO2” in the Matlab Sea Water toolbox.
  • Oxygen percent saturation is recomputed with the corrected oxygen value and the oxygen saturation value.
  • The “status” flag value2 for each data record is applied. This is a decimal number between 0 and 127. A flag of zero indicates that all sensors recorded data, a non- zero flag indicates that one or more sensors did not record. Applying the flag sets missing values to null data, instead of zero (as used in Satlantic input file).
  • Time is converted from the format YYYYDDD (YYYY=year, DDD=day number) plus decimal hour in the PROCESS.dat file into ODF “julian” time format.
  • The Satlantic 5 line header is written to a text file, with the same name as the input file with “_QCRP” appended to the end. This will be used later to export the quality controlled data back to a text file for OTN.

Download QCed data

StationPod IDOffload DateDownload Link
HFX245POD0091A2013-05-23PROCESS_POD0091A_HFX245_2013-05-23_QCRP.dat
HFX180POD0095A2013-10-10PROCESS_POD0095A_HFX180_2013-10-10_QCRP.dat
HFX212POD0096A2013-05-23PROCESS_POD0096A_HFX212_2013-05-23_QCRP.dat
SBXPODLBPOD0114A2013-08-07PROCESS_POD0114A_SBXPODLB_2013-08-07_QCRP.dat
SBXPODNLPOD0115A2013-08-07PROCESS_POD0115A_SBXPODNL_2013-08-07_QCRP.dat
HFX126POD0116A2013-05-30PROCESS_POD0116A_HFX126_2013-05-30_QCRP.dat
HFX153POD0118A2013-05-23PROCESS_POD0118A_HFX153_2013-05-23_QCRP.dat

The ODF Output fields are:

  • SYTM_01 (ODF time format)
  • TEMP_01 (Temperature)
  • CRAT_01 (Conductivity Ratio)
  • PSAL_01 (Practical Salinity, UNESCO 1983)
  • SIGT_01 (Sigma-T, kg/m3)
  • PRES_01 (Sea Pressure, Dbar)
  • DOXY_01 (Dissolved oxygen, in ml/l)
  • OSAT_01 (Oxygen saturation, in percent)
  • VOLT_01 (Sensor voltage, as in PROCESS.dat file)
  • QQQQ_01 (Satlantic data status flag, as in PROCESS.dat file)
  • CNTR_01 (Record counter, as in PROCESS.dat file)
  • CNTR_02 (Instrument serial number, as in PROCESS.dat file)
  • CNTR_03 (Number of samples in data average, as in PROCESS.dat file)

Quality Control

Once the ODF file is created, the pre and post deployment records are removed using a text editor. The start and end data times should agree reasonably well with the times recorded in the deployment and recovery mission reports. Otherwise, it may be necessary to contact OTN for clarification or to report an error. The ODF file is regenerated by using a Matlab “read_odf” command, followed by a “write_odf” command, to refresh the start and end times, number of records, etc.

A time series plot of the data is generated. If any spikes or suspicious data are found, they are removed, and their values set to null.

Whenever possible, values are compared with deployments of nearby CTDs or other Microcat instruments, to ensure they are reasonable. The DFO AZMP program spring and fall surveys usually have nearby CTD casts.

Data Output and Archiving

Once the data have been edited and the accuracy checked, an ascii file is generated from the ODF file, for submission to OTN. This is done by Matlab script “OTN_flatfile.m”. The user is prompted for the ODF file, and also for the “_QCRP.dat” file created by the ODF creation script. This ascii file is similar in format to the original “PROCESS.dat” file, with oxygen in micromolar/L, and conductivity in mmho/cm. Missing data values are set to the string “NaN’.

A final ODF file is submitted to the ODIS data shop for archiving, along with the mission reports, instrument logs and instrument metadata. Before sending the ODF file, parameter fields containing the instrument serial number (CNTR_02), record counter (CNTR_01), number of samples (CNTR_03), and status flag (QQQQ_01) may be removed. They were retained for re-generation of the flat file for OTN, and may not be needed for archiving.

References

  1. Solubility and salinity compensation calculation based on Garcia and Gordon. 1992. Oxygen solubility in seawater: Better fitting equations Limnology and Oceanography: 37(6) :1307-1312.
  2. Pressure compensation based on Hiroshi Uchida, Takeshi Kawano, Ikuo Kaneko and Masao Fukasawa. In-Situ calibration of optode-based oxygen sensors. Journal of Atmospheric and Oceanic Technology December 2008.
  3. For other calcualtions refer to AADI Operating Manual TD218 and TD269
  4. Satlantic, Benthic Pod Interface Control Document SAT-DN-04437, Revision L, 2011-11-20, Section 4.4.2.2, Pages 24-25



Data available: