Hosted by "1PLs Agency"

Home > 7. Bathy/Topo Data > 7.3 Topo Contours

7.3 Viewing Relief and Estimating Topographic Contours in Saga:  DIVA 30-sec SRTM

  • Exercise Title:  Viewing Relief and Estimating Topographic Contours in Saga: DIVA 30-sec Shuttle Radar Topography Mission (SRTM) Data

  • Abstract:  In this exercise you'll learn how to view regional (land) relief and make topographic contours in much the same way you did with the bathymetry contours from GEBCO.  The dataset here, though, is a 30-second land topography derived from the Shuttle Radar Topography Mission (SRTM) datasets.

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

  • Required Software:

  • Other Resources: 

    Author:  Murray Brown & Lilian Krug

  • Version:  31-Jan-2019 (with SAGA 7.1.0)

1.  Open the DIVA website, and you'll find a nice interface to some basic but necessary GIS elements for national mapping projects.  Many of these data are actually derived from other sources, as cited.
2.  Below the above controls is this list of data layers.  You should take some time to go through these to see what's available.

We're going to use the ELEVATION data, in a 30-second grid derived from the CGIAR SRTM source (3-second resolution).


NOTE:  The "ELEVATION (COUNTRY MASK)" option, which omits all data around the Liberia polygon, is not suitable for this exercise, because it appears to the GIS software to have a continuous "coastline" all round.

Then click OK.

4.  Click DOWNLOAD.
5. Navigate to the folder DATA > BASEMAP > RELIEF > DIVA and save the file with the name
6.  Unzip the file in place, and you'll find these files:
  • LBR_alt.grd
  • LBR_alt.gri
  • LBR_alt.vrt

Based on experience, we would expect that the largest file (*.GRI) is either an ASCII grid or a binary grid, and the smaller files contains auxiliary information.  The token "alt" in the names stands for "altitude".

7.  Examine the GRD file in an ASCII editor, and you'll see it is indeed a header file.

These items address nearly all of the items called the "13 Questions" in the above resource article on Gridded Data.  Data managers are constantly faced with unknown grids (or at best, poorly known grids), where the answers to these questions are critical to using the files. 

NOTE:  If you are faced with a grid that cannot be read directly by Saga, then you could use the IMPORT BINARY RAW data module, which requires you to supply any of these items individually.

8.  But, fortunately, we can load and work with this grid directly, because the GDAL/OGR module recognizes the trio of GRD/GRI/VRT files.

NOTE:  This module can recognize and load dozens of formats.  No matter what, if you have trouble loading a file, always try to see if this module works.

You can experiment with these settings, but the author always tries them first to see what works:

  • For FILES select the VRT file (previous experiments with the GRI and the GRD files failed)
  • For SELECT FROM MULTIPLE BANDS, check it for your first attempt.  Experience may convince you to do otherwise.
  • For TRANSFORMATION, check it for your first attempt.  Experience may convince you to do otherwise.
  • For INTERPOLATION, select NEAREST NEIGHBOR for a first attempt.  Experience may convince you to do otherwise.

Then click OK.

10.  The raster (grid) will load, as you see here.  Right-click on it and select ADD TO MAP.
11.  And here is the grid.  You can see how detailed even this 30-second grid will be, when we fix the colors.
12.  Take a minute to use SAVE AS to save this grid in the folder PRODUCTS > SAGA > GRIDS with the name topo_grid_liberia_srtm_cgiar_diva_saga_30sec
13.  To fix the colors, look on the properties panel, and select
  • COLORS = 100

Then choose SETTINGS > APPLY, just below..

14.  Here is the colored grid.  Look closely, and you can see an "island" of higher elevation near the left edge of the coastal section.
15.  Right-click on the grid object and select HISTOGRAM to produce this "hypsographic curve" showing the distribution of elevations within the raster.

The hypsographic curve shows that much of this part of Africa consists of these features:

  • Coastal plain 0-250 m (purple to dark blue)
  • Piedmont ~250 m (medium blue)
  • Upper plain 300-550 m (light blue)
  • Mountains > 600 m (green)


16.  Now, we're going to create and add a specific contour line to the map.  From experience, we know that it's often best to make special contours like this one separately.  Arbitrarily, we've chosen a topographic contour of 8 m, after some reports that the Banda Aceh tsunami (12-26-2004) reached land with that elevation in some places.  You be the judge of which contour(s) to make.

Set the GRID SYSTEM to the SRTM system, and set the GRID to the LBR_alt grid

Make sure CONTOUR LINES is set to CREATE.

Then set the MINIMUM and MAXIMUM to the same value (8 m), and set EQUIDISTANCE to 0.  This forces Saga to make a single line.

Click OK.

18.  Notice that a new shape has been added, and it is named after its grid source.  Right-click on it and select ADD TO MAP.
19.  Here you can see the new contour line on the grid.  There are many areas, particularly around inlets/estuaries, where the 8-m contour line goes far inland.  This is a good indication that low-land flooding is probably a serious issue in this region.

This map is only a gross estimate of the 8-m contour location, provided here only for training purposes.  For actual hazard preparation work, you should use much higher resolution local elevation data.  Do not depend on this exercise or its products shown here for any actual operations or preparations.

20.  Select the MEASURE DISTANCE tool at the top of the Saga window, and drag the cursor from the shore to the upland side of one of the largest incursions (center of this image).  You'll see that it measures 0.04 degrees ("D" in the scale below,) or about 4 km at this latitude.
21.  Finally, right-click on the new 8-m contour shape and select SAVE SHAPES AS.  Navigate to PRODUCTS > SAGA > VECTORS and save the shapefile as topo_estim_cons_8m_liberia_srtm_cgiar_diva_saga_30sec.

We must repeat again here that this method is approximate.  Use of the higher resolution versions of the SRTM data would yield even better products.