Marine Data Literacy 2.0

Providing instruction for managing, converting, analyzing and displaying oceanographic station data, marine meteorological data, GIS-compatible marine and coastal data or model simulations, and mapped remote sensing imagery





Home > 4. Ocean Data View > 4.17 GTSPP "Best NC"

4.17 Importing Global Temperature-Salinity Profile Program (GTSPP) "Best Copy" NetCDF Data

  • Exercise Title:  Importing Global Temperature-Salinity Profile Program (GTSPP) "Best Copy" NetCDF Data

  • Abstract:  In this exercise you'll learn how to go directly to the GTSPP archive for their monthly global products.  They can be downloaded in a compressed format that goes directly into ODV without any further processing or uncompressing.  One month of data, demonstrated here, allows some usual ODV graphical displays.

  • Preliminary Reading (in OceanTeacher, unless otherwise indicated):

  • Required Software:

  • Other Resources: 

  • Author:  Murray Brown

  • Version:  March 2012

1.  There are several major sources of GTSPP data, including the Coriolis website (Obtaining NetCDF Operational Marine Data (Argo Profilers/GTSPP/SOOP) from Coriolis) and the Ocean Data Portal, the method below is arguably the simplest and quickest.  The drawback to this method is that you are limited to whole ocean basins (Atlantic/Arctic, Pacific, Indian) and to whole months.  The ease of this method is due to the presence of an existing GTSPP template in ODV.  A planned upgrade to the GTSPP website may include user-defined data products that are compliant with this template.
2.  Open the GTSPP Website, and take a little time to read through the main materials.
3.  This map of the data "regions" is provided by the GTSPP User's Manual (link above).  Note that the Arctic is combined with the Atlantic.
4.  In the left-hand margin are three data products.  The first one is User-Defined Data Sets.  Currently these are provided only in the MEDS GTSPP format, which is not compatible with ODV.
5.  The second possibility would be Real-Time Data Sets.  These are also in the MEDS format.
6.  The final possibility would be these Best Copy Data Sets, available from numerous sources.  We'll go for the FTP source in the steps below.

NOTE:   The correct THREDDS catalog URL ends with .xml, not the .html you can see in the fourth item.

7.  Select the FTP link above, and this directory opens.
8.  Select the BEST_NC link in the directory.  This long list of compressed files opens.  They are sorted by ocean basic, year, month and compression type, in that order.

NOTE 1:  The compressed files have YYYYMM year-month codes in their filenames.  Plus these regional codes:

  • at = Atlantic Ocean
  • in = Indian Ocean
  • pa = Pacific Ocean

NOTE 2:  The much smaller size of the TGZ files, compared to the ZIP files, is impressive.  That's why we'll select TGZ below.

9.  Scroll down to the very latest Atlantic TGZ file, which in this case is for January 2012.

Right-click on the file, and select SAVE LINK AS

10.  Navigate to the folder DATA > OCEAN > GTSPP > BESTNC and save the file with the existing filename gtspp_at201201.tgz.  That's all you need to do here, for the moment.  You can close the website if you want, or return later to download other data.
11.  Run Ocean Data View.
12.  Select FILE > NEW and navigate to the folder PRODUCTS > ODV > COLLECTIONS and enter the new collection name gtspp_at201201
13.  When you see this window, select the GTSPP template for the structure of your collection.
14.  This empty map opens.
16.  Navigate to DATA > OCEAN > GTSPP > BESTNC and select the file gtspp_at201201.tgz
17.  Because a GTSPP template was used, this window shows all SOURCE variables are already correctly mathced with COLLECTION variables.  Click OK.
18.  The file is rather large, and the import process involves opening many individual NetCDF profiles, so it will take some minutes.  Expect a 10-20 minute wait here.
19.  Finally, this message appears.  Click OK.
20.  Click YES.
21.  And here is your collection.  Most of the points are almost random, but a few linear patterns indicate frequently-traveled commercial lines.

This completes the collection and import processes.  We'll take a few minutes below to view some common graphics in ODV to get an idea of what data are included in this single month of the GTSPP program.

23.  This view of temperature versus depth should open as the default.
24.  The appearance of several constant-temperature profiles, going to great depths, opens all kinds of questions about quality control, data sources, etc.  The fist thing we need to do is take advantage of any data quality flags that have already been assigned by the GTSPP system.
25.  Right-click on the station map and select SAMPLE SELECTION CRITERIA.  Then select QC WAS PERFORMED; GOOD DATA.  Then click APPLY TO ALL VARIABLES.
26.  The selection of only good data does clean up the figure somewhat.
27.  Now right-click on the scatter plot and set X VARIABLE to salinity, and Y VARIABLE to temperature.  This is the well-known T-S Plot.

But this is a very strange T-S Plot, probably due to the huge area covered, i.e. many different water masses.  You can even see the very different deep-water characteristics of the Arctic, at the bottom right corner.

This author was very curious about the huge mass of points between salinities 5-30, and temperature 10-25.  Many of the points appeared be in the northern Gulf of Mexico, which would not be expected.

28.  Using ODV's zoom function, most of these points can manually be located in and around the mouth of Mobile Bay in Alabama, USA.  You can see these points isolated here.

This is a good object lesson in how vigorous local sampling programs can supply so much data that care must be taken NOT to use the data without understanding where they came from, and under what conditions.  A major local river, the Mobile River, has obviously affected these data.

29.  Here the SAMPLE SELECTION CRITERIA have been relaxed, as well as the geographic limits, so you can see the entire collection again.

This would be a good time to select VIEW > SAVE VIEW so you can always return to this exact place.

30.  So in fact only 21 panels of this exercise were needed to obtain the GTSPP data and make the ODV collection.  This is quite efficient.  Although the data are not exactly "near-real-time" this is an effective way to obtain recent data for ODV analyses.  Upgrades to the GTSPP website are greatly needed to allow closer user-specification to focus in on times and places of interest, so long as the existing ODV template for GTSPP is observed.