Data

Your access to Sea Surface temperature Sensors for Australian Vessels data discovery and exploration is through the Australian Ocean Data Network (AODN) Portal.

As part of IMOS, the Bureau of Meteorology (Bureau) has instrumented ten commercial volunteer observing ships with hull temperature sensors (Sea Bird SBE 48), supplying high-quality bulk SST observations every hour.  In addition, the Bureau has provided near real-time access to one minute averaged SST and salinity data streams from seven research vessels (RV Southern Surveyor, RSV Aurora Australis, RV L’Astrolabe, RV Solander, RV Cape Ferguson, RV Tangaroa and RV Linnaeus), two tourist ferries (PV SeaFlyte and PV Fantasea One) and one commercial vessel (MV Pacific Celebes).  In total, 20 vessels have contributed near real-time data to IMOS and the Global Telecommunications System (GTS).  Due to sales of vessels over the years, currently only 12 of these vessels provide data to IMOS and the GTS. 

All SST data are quality assured (Beggs et al., 2012) and placed in real-time on the Global Telecommunications System (GTS) as either SHIP or TRACKOB reports.  The quality controlled (QC’d) SST data are also available in netCDF format with QC flags and metadata via the AODN Portal.

Map of cruise tracks of volunteer observing ships providing SST data to IMOS as part of the SOOP SST Sensors sub-facility up to 3 August 2015 (source: IMOS Ocean Portal). Note: other research vessels provide SST to the IMOS Ocean Portal and their tracks are not shown here.

Further information relating to the call signs of the 20 vessels that have previously and currently collected IMOS SST data and the information about the sensors and dates can be found here.

Quality Control and Validation

The sea surface temperature (SST) quality control (QC) procedure is a fully automated process, and is based on the system developed by the Center for Ocean-Atmospheric Prediction Studies (COAPS), Florida State University, for the Shipboard Automated Meteorological and Oceanographic System Initiative (SAMOS), with small differences due to varying IMOS/Bureau requirements.   The QC system flags data that fail to pass the following QC tests, in order of application:

  1. Verify the existence of time, latitude and longitude data for every record;
  2. Flag data that are not within physically possible bounds;
  3. Flag non-sequential and/or duplicate times;
  4. Flag positions where the vessel is over land;
  5. Flag vessel speeds that are unrealistic;
  6. Flag data that exceeds 3°C above/below the Bureau’s most recent operational daily SST analysis (blended from satellite and in situ SST data);
  7. For vessels instrumented with a hull-contact sensor, flag data where ship speed is below 2.5ms-1 (since the hull-contact sensors exhibit anomalous SST values when in port).

Once any datum’s flag is changed, it will not be altered further by any subsequent test.

Comparisons were performed between IMOS ship SST observations and those from buoys and the Advanced Along-Track Scanning Radiometer (AATSR) on the EnviSat satellite, over the region 60°E – 190°E, 70°S – 20°N, during 1 December 2008 to 30 May 2011. This study indicated that 12 of the then 13 IMOS ship SST data streams  (including all those from hull-contact temperature sensors)  had comparable uncertainties to drifting and moored buoys, with standard deviations within ±0.06°C and biases generally within ±0.1°C to those obtained using buoys (Beggs et al., 2012). Exceptions were the fast tourist ferries PV SeaFlyte*  and PV Fantasea One, installed with calibrated thermistors in engine water intakes, possibly due to a combination of the predominantly daytime cruises and warming by the engine of the water being sampled.  The SST data streams from SBE 48 hull-contact sensors installed on six commercial vessels exhibited overall lower standard deviations when matched with AATSR SST (0.37°C) compared with buoy SST match-ups with AATSR SST in the same region (0.44°C) but slightly higher warm bias (0.14°C cf 0.07°C). The SBE 48 SST data from the six AVOF vessels had a quarter of the standard deviation of non-IMOS ship SSTs over the same region reported to the GTS (0.37°C cf 1.51°C). Night-time comparisons between Advanced Very High Resolution Radiometer (AVHRR), drifting buoy and IMOS ship SST observations also indicated that the standard deviation of the match-ups with hull-contact sensor SSTs were within 0.15°C of those from drifting buoys and the biases within ±0.2°C (Beggs et al., 2012). Hull-contact temperature sensors have therefore been demonstrated to be capable of producing validation-quality SST data, provided that one obtains a good thermal contact with the hull, locate the sensor away from on-ship heat sources, and thermally insulate the sensor and surrounding hull to a distance of at least 0.5m from the sensor.

 

* Note that during the second half of 2012 the SBE 38 SST sensor was re-installed in a new position on the Rottnest Island Ferry (PV SeaFlyte) to improve the previously poor quality SST data stream (see "Report on Rottnest Ferry SST Validation Tests" at http://imos.org.au/soopdoc.html). New SST data from PV SeaFlyte became available from 23 October 2012 and appear of sufficient accuracy to be used for satellite SST and ocean model validation.

Reports

Paper describing the new IMOS SST Sensor products: Beggs H., R. Verein, G. Paltoglou, H. Kippo and M. Underwood (2012) Enhancing ship of opportunity sea surface temperature observations in the Australian region, Journal of Operational Oceanography, (ISSN: 1755-8778), 5, 59-73.  PDF available from www.ingentaconnect.com/content/tandf/joo/2012/00000005/00000001/art00006 or by emailing h.beggs(at)bom.gov.au.

Report on Rottnest Ferry SST Validations Tests: A short technical report describing the quality of the SST data from the Rottnest Island Ferry (PV SeaFlyte) and tests performed to determine the sources of the SST errors.