Home - Océanographie - Calibration

TSG-QC: A tool for interactive quality control of sea surface temperature and salinity

Issued Tuesday 18 May 2010.

TSG-QC is a software for interactive analysis and validation of underway SST / SSS (Sea Surface Temperature and Sea Surface Salinity) measurements from a SeaBird Thermosalinograph (TSG). It has been developed under Matlab.

Updated on November 29, 2021


TSG-QC is a tool for analysis and validation of surface temperature and salinity data acquired from a Thermosalinograph (TSG) installed on research or commercial ships.

It allows:

  1. Visualization of TSG variables: Temperature, salinity, and ship speed
  2. Interactive comparison with climatological values (WOA and ISAS)
  3. Automatic quality control using selected threshold criteria
  4. Data validation and correction with external “bucket” measurements (water samples usually collected once a day)
  5. Quantitative estimation of sensor drift

The software can deal with different input data formats: ASCII, Labview, "Real-time" Coriolis NetCDF, Seabird (*.cnv), ...


The development of the software is now hosted on the IRD Gitlab platform under:

If you have a Matlab license, there are 2 options:

  1. Retrieve and install the source code of the latest release https://forge.ird.fr/us191/TSG-QC/-/tags/v1.50
    Choose the type of archive to download (.zip or .tgz) by clicking on the "Source Code" drop-down menu.
    Once the archive is uncompressed on the computer, launch Matlab and add the installation directory of tsgqc on the Matlab path and then launch the software with the command:
  2. Retrieve the sources of the latest version in developmentwith the Git software. If you are running Windows or Linux / Mac, you will need to install the Git software first.

Installing Git:

Clone your local copy of the Git repository:

Go to the TSG-QC Git page https://forge.ird.fr/us191/TSG-QC and copy the HTTPS URL as shown below:


In the directory of your choice, for example (c:\gitlab), right-click to open the context menu and choose "Git clone"

then click OK, the cloning of the repository should begin.

Update Deposit: When you’ve been notified by email of adding new features or bug fixes, from the pop-up menu, choose "Git pull" to update your local copy.


Clone your local copy of the Git repository on Linux from the command line, replace with the directory of your choice:

Update your local copy:


The latest version 1.50, updated November 25, 2020 works with the latest versions of Matlab and has been tested up to R2021b.

Read the "Release Notes" file for the latest changes.

Follow the instructions of the files README and INSTALL to modify the paths under Matlab and install the toolboxes or "toolbox" necessary for the application.

Compiled Version:

If you do not have a Matlab license, a compiled version is available by clicking the following links using the ftp client program Filezilla:

Read the file readme.txt.

Using tsgqc.exe requires installing the Matlab runtime. You need to get it from the same link and install it.

To obtain a version compiled under Linux (Ubuntu), send me a direct request by mail.


Start Matlab, go to the tsg_data subdirectory and type tsgqc to run the program. Use the File / Open menu to open a data file.


A user’s guide is available here

Standards and best practices from GOSUD


The reference document CORTSG_format_gosud.pdf version 3.0 describing the output NetCDF format is available online from the subversion repository.



In case of technical problems, you can contact me at the following address: hotline-tsgqc (order to prevent spam, replace _at_ with @ in the field to:). Please report the following information:

under Matlab, the results of the following commands: