Hosted by "1PLs Agency"

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 > 4.5 Surface Plots

4.5 Creating Marine Data Surface Plots in ODV

  • Exercise Title:  Creating Marine Data Surface Plots in ODV

  • Abstract:  Surface plots are introduced, and methods to create them in ODV are demonstrated.  Surface plots involve estimates by ODV of any variable for every station where such data exist.  Surfaces can be defined by any variable, including depth (the first variable that comes to mind).  Thus this method is a powerful analysis mechanism for any variable, and is an important way to create depth-defined marine data analyses with ODV. 

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

    • N/A

  • Required Software:

  • Other Resources: 

    • ODV collection osd_all_liberia_wod.odv

  • Author:  Murray Brown

  • Version:  6-16-2014

1.  A surface plot is not necessarily a graphic portraying any variable at the surface of the sea, although it can be.  Surface plots portray any variable as it would be found on a 3-dimensional sheet within (or on, or just beneath) the sea, defined by a constant value of some other variable.  For example, consider a sheet defining a temperature of 15˚ C within the sea; globally this sheet would be quite deep near the equator, but would rise toward (and even intersect) the the surface near the poles.  A related surface plot could be, for example, the salinity found at all locations on this 15˚ degree sheet, specified as "Salinity at Temperature = 15˚ C".  This type of plot became known and widely used when it was early discovered that mixing in the sea tends to occur along "iso-density surfaces" or sheets of constant density.  Starting with density surfaces, such plots are now created and interpreted for many other variables. 
2.  Open the ODV collection osd_all_liberia_wod.odv
4.  This graphic usually appears, the default plot in ODV for the surface mode. 

Don't worry if your ODV shows something else.  We're only setting up now.

5.  Now we're going to narrow down the data to just one season (January-February-March).

Right-click on the small station map, and select STATION SELECTION CRITERIA.

6.  For SEASON, use the drop-down menus to select FROM JAN and TO MAR.  This range is annotated as JFM below.

Then click OK.

7.  That takes care of temporal selections.  Now we need to worry about data quality and depths.
8.  Right-click on the data figure and select SAMPLE SELECTION CRITERIA. 
9.  Here you have two tabs.  Select the QUALITY tab. 
  • Select 0: ACCEPTED VALUE (meaning good)

Then click OK.

10.  We need to see what variables are available now.  To do this, select VIEW > ISOSURFACE VARIABLES.
11.  This window opens, showing 4 default variables, none of which are needed by us.

NOTE:  "First" in the ODV syntax means the shallowest sample at the station, not the sea surface.  "Last" means the deepest sample at the station, not the sea floor.

12.  Select all of the "FIRST" isosurface variables and use DELETE to remove them.  They are not needed.
13.  Set the top line to read TEMPERATURE AT DEPTH [m] = 0.  Then click ADD to complete the addition.

Also add these other TEMPERATURE @ depths:

  • 1000 m
  • 2000 m
  • 3000 m
  • 4000 m
  • 5000 m

Then click OK when finished.

14.  Now, let's view the new surface variables we created. 
15.  Right-click on the graphic and select the Z VARIABLE.
16.  Then select TEMPERATURE @ DEPTH = 0.

NOTE:  Yes, this was already chosen, but you need to know how to make these choices later on.

Then click OK.

17.  This graphic appears, showing (by color) the 0-m temperatures.

NOTE:  There are some crazy Fahrenheit temperatures in this file, which you should not see if you chose only ACCEPTED = GOOD quality values above.  If your value range goes up to 80, then go back and start over.

18.  We need to grid the data in ODV, for better viewing, so right-click on the graphic and select PROPERTIES.
19.  Make these selections to get a good first gridding:
  • Select DIVA GRIDDING from the list of gridding methods
  • Set the X and Y scale-lengths to at least 100 (=10% of the map width).  This is a good first guess, but might need later adjustment
  • QUALITY LIMIT set to 2.5 standard deviations (from the mean)
  • Uncheck DRAW MARKS.

Then click OK. Expect to wait a few minutes, in most cases, due to the more thorough gridding with this method. 

20.  Take a moment to note that in making the above decisions about the "length scales" (set here at 100 per mille, or 10%) you have constrained the resulting products to an inherent resolution no smaller than this approximate size.  In the present case, the Liberia area is 20 degrees wide and 15 degrees tall.  10% of these dimensions is roughly 1-2 degrees.  Remember this when you are doing any further work with these products.
21.  This graphic appears. 

NOTE:  The image is rather boring, because the data range (which you can play with) is rather too large.

22.  Select VIEW > SAVE VIEW AS and save this view as temp_0m_jfm_liberia_wod
23.  Right-click on the small station map, and use STATION SELECTION CRITERIA to show only the data for July, August and September.

After a few minutes, this image appears.  Notice the very interesting small upwelling areas offshore Cote d'Ivoire and Ghana.  These are due to "lee waves" behind the promontories in an eastward-flowing current area.


24.  If you want to know more about the actual data values being displayed, right-click on the image and select EXTRAS > STATISTICS.
25.  Here's a huge display of available statistical tools.  Come back later and study these.

Click on the Z HISTOGRAM control.

26.  Here's the temperature values histogram.  This histogram will be different for each season (JFM or JAS) and for each surface (0m or deeper).

One thing that you can immediately see in this example histogram is a good estimate for the range of temperature values to match with a palette, roughly 20-30 degrees.  But of course, you'd have to look at histograms for all seasons (and depths) to get a good general-purpose range, based on the "highest high" and the "lowest low" you find. 

For good visual comparisons, all figures should use the same value ranges and palettes.  But sometimes surface and very deep analyses are so very different, that this is not strictly followed.

You can study all these concepts later on your own.  For now just agree with us that there's a lot of work involved in selecting the best color palette for data that varies in time and space (i.e. depth).

27.  If you decide what are the best value ranges to use, then here is how you set them:
  • Right-click on the image
  • Select SET RANGES
  • Then on the list you'll see, insert your value choices.

We won't do that now, but you can come back and work on it later.

28.  Select VIEW > SAVE VIEW AS and save this graphic as temp_0m_jas_liberia_wod
29.  Now we're going to make analyses deeper and deeper in this area, until we find something really interesting.  We'd like a big contrast with the 0-m analyses we already have.
30.  Set the Z for the 1000-m temperature (JFM).  Looks nice, but nothing special.
31.  Set the Z for the 2000-m temperature (JFM).  Looks nice, but nothing special.
32.  Set the Z for the 3000-m temperature (JFM).  The shallowest areas of the mid-Atlantic Ridge are beginning to show up as white, i.e. no data.
33.  Set the Z for the 4000-m temperature (JFM).  This is amazing.  You can see that the mid-Atlantic ridge (the nearly S-shaped white area in the center) cuts through our data.  It completely separates the Eastern Atlantic Bottom Water (warmer) from the Western Atlantic Bottom Water (cooler).  The difference is roughly 0.7 degree C, at that depth.
34.  So make and save these views of the 4000-m temperature surface, using all the same gridding settings:
  • temp_4000m_jas_liberia_wod
  • temp_4000m_jfm_liberia_wod

These should be identical, because you don't expect any annual variation at that depth.

35.  So now you have a powerful new tool for visualizing water mass structures in the sea.  Obviously this extremely simple introduction is just to tempt you to do your own work with data from other locations.  And don't be afraid to try experiments!!