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 > 2. Marine GIS > 2.35 MGD77 Surveys

2.35 Managing & Visualizing MGD77 and XYZ Geologic Survey Data in Saga: NGDC

1.  Open the Trackline Data webpage, and take a few minutes to read about it.  It is still a work in progress, and possibly may differ slightly from this exercise at times.  Obviously this is a world map with lines and points indicating where the survey data have been obtained.

2. Use the MARINE SURVEYS tab on the right margin to see the available data types.  There are 9 at present, each representing a major geophysical sensing category.  Nearly all are global, but Side Scan Sonar is largely limited to US waters where it is required for offshore oil and gas sites.
3.  Use the LEGEND tab to see the temporal coverage of the NGDC datasets.  Almost 8 decades are included.
4.  Use the MORE INFORMATION tab to see background links, and to select the main background map.
5.  To find data, first go back to the MARINE SURVEYS tab.  Uncheck the ALL SURVEY TYPES box, and check a few types of interest.

Just for this exercise, the author suggests these two.  As soon as they are selected, the map updates automatically, as you see here.

6.  Now, to zoom in on your area of interest, find this tiny control (upper left corner of the map) and click on the XY tool.
  • XY - Area selection when you know the exact coordinates
  • i - Select area by dragging a box
7.  Enter the Liberia area coordinates.  Then click OK.
8.  The above order of steps must be used:  First select DATA TYPES and then select AREA.  This might be a small program bug.
9.  Now you will see a zoom box around the area, the lines for the available data, and a listing of the surveys that have been identified (5 in this case).  You can select any one with your cursor to see it highlighted on the map.


10.  This complicated data control window opens.  Take some time to go over it and try to see what every part means.

11.  In the space across the bottom you'll find a list of the 5 selected cruises.  Small icons on the right identify the data types actually in each survey.  There are 9 basic data types:
  • B - Bathymetry
  • G - Gravity
  • M - Magnetics
  • MC - Multi-Channel Seismics
  • SC - Single-Channel Seismics
  • SB - Subbottom Profile
  • SP - Shot-Point Navigation
  • SR - Seismic Refraction
  • SS - Side-Scan Sonar

Every survey, except the first one, contains SB or SR.  The presence of the first item is unexplained and has been referred to the publishers.

12.  To make choices about how we want to get the data, look at the REQUEST DATA section (upper right).

HINT:  You must click ADVANCED to see all these options.

13.  Here's what these FILE OPTIONS mean:
  • Single File: Data in Area - You get only one file, with data from all (5) surveys, limited to the area of interest box
  • Multiple Files: Data in Area - You get separate files for each survey, limited to the area of interest box
  • Multiple Files: Complete Surveys - Same as above, but the complete surveys (which can even be global)

Here's what these FORMAT options mean:

  • MGD77T - You get the data in one or more very easy to use tab-separated, ASCII spreadsheets.  Saga has no trouble working with tables that hold hundreds of thousands of records
  • XYZ - This option delivers tab-separated, ASCII tables that hold only LONGITUDE (i.e. X), LATITUDE (i.e. Y) and one selected variable (i.e. Z).  Not all the variables are provided, so check carefully before you begin to use this option.  Also, the table itself has no header line, so it is not entirely convenient to load it into Saga.

For right now, make the choices shown above.

14.  You can request microfilm copies of the original data, as shown here, which we will skip.

But to get the digital data, enter your email address as shown here, and click REQUEST.

15.  You will be guided through the process of downloading the resulting file, which contains the choices you made above.  In this case, navigate to the folder DATA > OCEAN > NGDC and save it with a name like ngdc_order_2014-08-16T19_44_00.309Z.tar.gz
16.  Unzip the TAR.GZ file, in place, and inspect the results.  You'll find files like these, plus some auxiliary files for further information
17.  The H77T file contains the ASCII metadata describing the survey.  It includes these items.  In the formats documentation for MGD77 you can research these for your own interest.

  •  FORMAT_77









  •  CHIEF














  •  LAT_TOP
























  •  G_ST_DEP_G

  •  G_ST_DEP

  •  G_ST_ARR_G

  •  G_ST_ARR

  •  IDS_10_NUM

  •  IDS_10DEG

  •  ADD_DOC

18.  The M77T file contains the actual data, outlined here.  The approach of the NGDC has been to provide fields that cover the 9 major data types, in the most flexible way.  You can also read about these in the documentation cited above.

The SURVEY_ID field is the link between the H77T and M77T files.



  •  DATE

  •  TIME

  •  LAT

  •  LON








  •  MAG_TOT

  •  MAG_TOT2

  •  MAG_RES





  •  GRA_OBS






19.  The GEODAS-NG software provided by NGDC includes a module that converts the M77T spreadsheet to shapefiles for use in GIS.  This author find that direct use of the spreadsheet in Saga is so easy that this isn't really needed.  We'll continue below with a method that lets you have complete and explicit control of the tables and shapes that you make from the M77T file.
20.  Run the latest version of Saga.
21.  Use FILE > TABLE > LOAD to find and load the M77T table you just made. 

HINT:  Make sure to set the loading module to "ALL TYPES" so you can see the M77T file...Saga expects TXT.

22.  You can open the table to see its contents, just to make sure everything is OK.
23.  To use the table, we must first convert it to a point shape.  Select TOOLS > SHAPES-POINTS > CONVERT TABLE TO POINTS and make these settings.

Leave Z at "<not set>" to select all fields.

Then click OK.

24.  If everything went OK, then you should have at least these 2 objects in Saga.
25.  Here we've used ADD TO MAP to see if all looks OK.  The map is ugly, but it seems to match with the early visualization in the panels above.
26.  From experience with these plots, we know that the settings shown here will improve the map.
  • OUTLINE - Not Checked
  • COLORS > TYPE - Graduated Colors
  • ATTRIBUTE - Your choice, but we suggest CORR_DEPTH (for corrected depth)

You are invited to explore other options.  Then click APPLY (bottom of the panel)

27.  And here are your corrected depth data, with an explanatory LEGEND.

HINT:  You can find the LEGEND tab at the top of the properties panel, but the desired variable and map must also be selected.

28.  And if you get exploratory, you can grid the point shape to see if there's enough data to make good sense.  The World Borders map has been added for orientation.

You can see the middle and coastal parts look acceptable, but there isn't enough data in the lower right to do anything useful.

29.  Unless your data extraction is too large for Saga, the author strongly recommends that you always use the M77T format option for exporting the data.  If, however, you have an extremely large extract or you have other good reasons, then you can use the XYZ format for the data.  In the few panels below, we'll see what we can get.
30.  Here we have use he option ALL SURVEY TYPES for the Liberia area of interest.  And the format is XYZ.

NOTE:  The XYZ header is not currently supplied by the NGDC server, so you must add it.  Make sure that it has tabs between X and Y, and between Y and Z.  This is a tricky step with many ASCII editors.

31.  And here's all 280,983 point values in the XYZ file for all the survey data in this area.

The east-west features are real; they represents the fractures across the mid-Atlantic Ridge.

32.  And here is a very quick gridding job with these same data, using the NEAREST NEIGHBOR algorithm, which is the simplest method available.  There are obviously some areas of data problems, but in general this figures indicates we are on the right track with good data.

33.  At this point the author must confess a little sin.  The original NGDC datasets use negative numbers for depths, whereas many of the relief files already used by us in Saga have used positive values.  To insure physical compliance, at this point that author use TOOLS > GRID-CALCULUS > GRID CALCULATOR to multiply all the NGDC depth values by -1, so you'll see "CALCULATION [-g1]" as the name for the grid variable below.

34.  And here we've combined some of the GEBCO expert-drawn relief contours from 2.6 Adding GEBCO Depth Contours to a Project Map in Saga and some contours of those same depths we've made from our M77T grid using the method in 7.2 Creating Bathymetric Contours from Grids: GEBCO 08 Shelf Estimations.  

  • The background is the M77T grid itself. 

  • The depth contours are 1000 m, 2000 m, 3000 m and 4000 m (east to west). 

In view of the fast process used here, the agreement is quite good.  You can consider the Saga contours to be a "quick look" draft product that would be a good place for experts to begin making their refined products.  In areas where there are no available experts, then a Saga analysis of the contours might be an acceptable temporary product for scientific purposes; it would not under any conditions be acceptable for navigation or any operational activities.