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 > 4.18 Multi-Depth Currents

4.18 Visualizing ADCP-Type Multi-Depth Currents in ODV

  • Exercise Title:  Visualizing ADCP-Type Multi-Depth Currents in ODV

  • Abstract:  ODV not only does a marvelous job with static station data, but it can display vector arrows from typical U and V current component records.  This exercise not only demonstrates that basic ability, but it works with a beautiful multi-depth time-series, similar to those obtained from Acoustic Dopper Current Profilers, which return currents from discrete depths at selected time intervals.  ADCPs can be located at the surface or bottom, or can be mounted on vessel hulls.  A very clever quasi-3-dimensional data display method is invoked to show the results. A preliminary data preparation step (in Excel) is an ordinary process, sometimes necessary to get existing spreadsheets ready for import into ODV.

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

  • Required Software:

  • Other Resources: 

    • currents_multidepth_adriatic_201207070030_201207130000.xlsx - Current meter data at an unspecified location in the Adriatic Sea with a total water depth of 22 m.  The instrument is probably upward looking, because the data depths are given as bin heights above the bottom, e.g. a datum height of 2 m means a water depth of 20 m.  [Source and site are withheld here to protect the original researchers' rights in the data.]

  • Version:  August 2012

  • Author:  Murray Brown

1.  Download the current meter data (an Excel spreadsheet) to the folder PROJECTS > MISC
2.  If you open the spreadsheet this is what you'll see. 
  • The measurement times are given as "start" and "end" times
  • As expected the depths are given as height-above-bottom
  • No exact location is given, but these are required by ODV
3.  If you check the bottom of the file, you'll find it goes to July 13.  So there are about 6 days of data, at 11 different depths, taken every half hour.  Scroll through the data to make sure you understand the structure and contents.
4.  We need to do a lot of housekeeping with these data before they can be used:
  • Each time period must be identified by a single time, the mid-point between start and end
  • The data/time must be given in the exact format expected by ODV for easy use (yyyy-mm-dd hh:mm:ss.sss)
  • A fictitious latitude and longitude will be supplied, just for ODV's sake, but not for any scientific purpose
  • A depth column will be added, based on the height-above-bottom values

We can do all of this in Excel (or any good spreadsheet program), as you see below.

5.  Create a new column, with an appropriate name, for the new time mid-point.  Here it's called "mid_date".
6.  Insert a formula in the top cell, that simply averages the start and end times.  The result is the exact minute value at the mid-point.
7.  Duplicate this formula all the way to the bottom of the spreadsheet.
8.  Now block the new time column, right-click it, and select FORMAT CELLS.

Select NUMBER > CUSTOM and enter the formula you see here (if it doesn't already exist).

Then click OK.  You can see the how the new format looks in the next panel.

9.  In the new depth column, put the formula shown here into the top cell.  Then copy the formula all the way to the bottom of the column.
10.  Now add 2 new columns for latitude and longitude, as shown.  Fill both with zeros, a safe, known location in the Gulf of Guinea, that won't cause any "on land" problems.
11.  Here's what your final spreadsheet, with the new columns, would look like.
12.  Now, save your spreadsheet 2 different ways in the folder PROJECTS > MISC:
  • currents_multidepth_adriatic_201207070030_201207130000.xlsx - The binary spreadsheet, with the formulas; for corrections and changes
  • currents_multidepth_adriatic_201207070030_201207130000.txt - The ASCII, tab-separated version for use in ODV
13.  Run ODV.
14.  Select FILE > NEW > COLLECTION and navigate to PROJECTS > MISC and enter the name currents_multidepth_adriatic_201207080030_201207090000

NOTE:  We're going to do everything in the MISC folder, because this short exercise doesn't require much space or any specific structure..

15.  We're going to use the spreadsheet itself as the template for the ODV collection, so select the USE TXT option.  Then click OK.
16.  Navigate to the folder MISC and select the spreadsheet currents_multidepth_adriatic_201207080030_201207090000.txt as the template for the collection.  (In other words the file will be the template for its own collection in ODV.)
17.  This window is provided to let you adjust how ODV "reads" the spreadsheet.  The observance of TAB as the separator results in correct reading of the COLUMN LABELS.  Also, the data begins, as you see, in LINE 2.

Make these selections and click OK.

18.  Here we are informing ODV about which columns provide META VARIABLES (about the cruise, stations, etc.) and which columns provide the COLLECTION VARIABLES (i.e. the data, including sample depths).

The META VARIABLES are those always expected by ODV, and cannot be deleted.

19.  Remove the DATESTART, DATEEND and HEIGHT items from the collection.  We will be using the new columns instead (MID_DATE and DEPTH).

To remove them, select them and use the << control.

Then click OK.

20.  Make these COLLECTION PROPERTIES choices.

NOTE:  We have not done many experiments with other choices here, but there seems to be much flexibility.  We do know that the PRIMARY VARIABLE must be DEPTH.

21.  The collection is created, as shown by this new, empty map.
22.  Select IMPORT > ODV SPREADSHEET.

 

23.  This is the same window we saw above (Panel 17), which checks to make sure that we are reading the spreadsheet correctly.  Make the same selections, and click OK.
24.  Now this window appears, where you can make the specific associations between the incoming spreadsheet and the collection structure.
25.  Select LAT/LATITUDE on both sides and click ASSOCIATE.

Then select LON/LONGITUDE on both sides and click ASSOCIATE.

You can see the successful associations have been made by the appearance of *'s beside the variable names.

26. Now click MID-DATE on the left, and YEAR on the right.  Then click CONVERT.
27.  Drop down the menu of conversion formulas, and select the formula you see here.
28.  When you make this association, ODV correctly notes that YEAR, MONTH, DAY, HOUR, MINUTE and SECOND have been accounted for.
29.  We don't have any CRUISE, STATION or TYPE data columns in the spreadsheet, so we'll leave them unassociated. 

Just click OK

30.  Now, you can see that ODV has correctly made the remaining VARIABLE ASSOCIATIONS correctly.

This is because we "removed" the unneeded variables in Panel 19, above.

Click OK.

31.  This message tells us that >300 stations have been imported.  ODV is recognizing each time as a station.

Click OK.

32.  This map appears, showing that all the stations are in the one place.
33.  Before we can continue to analyze the currents, we need to add a special variable to the collection, based on the dates/times.  ODV cannot work with the dates/times in their imported format, but almost always need this special conversion.
34.  Select VIEW > DERIVED VARIABLES > TIME > STATION DATE&TIME
35.  This epoch selection window opens, to allow you to set the beginning point for the analysis timeline.  The one entered here is the one often used by satellite data workers, because it pre-dates all satellites.

There are various beginning time points required (or used by) many data systems and formats, e.g. NetCDF.

Make an entry to you are certain precedes the actual dataset.  Then click OK.

36.  Click ADD to move the new variable to the ALREADY DEFINED list.  Then click OK.
37.  Select VIEW > LAYOUT TEMPLATES > 1 SCATTER WINDOW
38.  You may see something like this.
39.  Right-click on the graphic and use the X-VARIABLE and Y-VARIABLE controls to make these selections:\
  • X-VARIABLE - The new time variable
  • Y-VARIABLE - Depth
40.  This is the new graphic.  It looks terrible!  That is because it contains so many data points that they run together like heavy black lines.
41.  Right-click on the graphic and select ZOOM.  Then zoom into a space near the top, as you see here.

Double-click to see the enlarged part of the graphic.

42.  Now you can see the separate times and depths for the measurements.
43.  Right-click on the graphic and select PROPERTIES > DISPLAY STYLE > ORIGINAL DATA > ARROWS.
44.  We don't really know all the information yet, but you can make these selections for X and Y and then click OK.
45.  Here is the graphic with arrows, and obviously they are too large.
46.  Repeat Panel 43 and 44, and enter SCALE 25 PER CM.  This gives you a much better figure.

Working with your own data may require completely different settings.  Do some experiments to find out.

Now you can see the current arrows, pointing in geographic directions.  In the first two measurement periods, for example, the currents flow strongly toward the north (not upwards, out of the sea!).  Later on, the surface currents flow mainly toward the west-southwest.  Etc.

47.  Although it took many steps to get here, the actual creation of the current arrows is quite easy.  The real issue is how to deal with so much data.  In the above example we narrowed down the final graphic to only about 10 hours and 5 depths, only a tiny percentage of the total original data.  You will have many choices to make in how to simplify these results for actual research.  Further averaging is possible, but you don't want to lose the real-world structure of the (probably) tidal-driven currents, in time and space.  These difficult choices have always been faced by physical oceanographers.  Visual graphics are nice, but you can see now why so many large data collections are finally portrayed as tabular summaries, spectra and statistics.
48.  Select VIEW > SAVE VIEW AS and save this analysis of the data as current_arrows_surface.  You can always use VIEW > LOAD VIEW to get back to this graphic for this collection.