One has to know the data sources from which they can either download the data directly if the data provider offers the data in public domain or registration based data access. It has to be noted that most of the data source mentioned here are mostly of public domain type, which are mostly offered by established research institutions, and includes collaborative authorship sources too. Since we here planned to work with terrain/topography data, the following data sources can be used :

TYPE	ACCESS	SOURCE	RANGE	COVERAGE
Elevation	Public Domain	SRTM	60°N - 56°S	80% Landmass
Elevation	Public Domain	ASTER	83°N - 83°S	99% Landmass
Physical & Cultural	Public Domain	Natural Earth	-	-

If you have decided to use the popular SRTM data, as the primary data source, one shall use SRTM Tile Grabber by Derek Watkins, one of those inspiring designers in Internet. In Step 5, i have tried to explain how i have done it manually.

NOTE:

It has to be noted that since we are focusing on some degree of specificity towards terrain hacking, not all exhaustive sources are listed. If one is interested in knowing them, please refer to my presentation listed at links and downloads section of this page. The data source nowadays can not only be acquired from Satellite, Aerial, UAV, RADAR, LIDAR, Photogrammetry, sources, but also include 3D scanning, SONAR with unmanned Under Water Vehicles/Robots for localized mapping. This breaks the surface mapping limitation, and thus can be used in archeological studies too. In such cases, the location is measured by position measurement methods such as dead reckoning, etc. Most probably Simultaneous Localization and Mapping (SLAM) are developed and used to aid data acquisition.

With data sources identified, and being downloaded with your favourite download manager, one has to know what are the FOSS tools available to play with. Remember not to mirror the website when the service is offered by plain FTP. That would load the server and would eventually lead the administrators to shutdown for good. I have just listed the popular tools in FOSS realm - which offer a wide variety of data acquisition, analysis, visualization and storage workflows. However, apart from specific differences, they tend to embrace a common & overlapped workflow.

• GRASS (or)
• QGIS (or)
• SAGA (or)
• UDIG (or)
• gvSIG (or)
• LSD Topo Tools (or)
• OSGEO live distro.
• ...

However, i am not to list exhausting set of FOSS based GIS tools out there, and for those who are interested can see my presentation listed at the end of the page in links and downloads section. It is wise to note, that almost all of the FOSS - GIS tool stacks work upon the fundamental GDAL/OGR implementation of open geospatial standards.

Now what to do with the Data & Tools ?

One can play

To start working with anyone of the above tools, it is necessary that one understands some degree of projection systems, cartography, raster & vector data formats to start with. However, if one is interested in exploring the rabbit hole, all beautiful math & statistical world is waiting where cool computational, algorithms crank and churn out numbers and commands to enable us to do crazy things we wish to do !

Understanding an abstract organization of layers of information while creating a map enables easier execution of cartography workflow.

 @
/.\ ----+---------------------------------+
/ \     | Shall we clear the confusion by |
  	| illustrating the overlaying  ?  |
	+---------------------------------+
							+-------------+---( me )
  							|   Yeah !!!  |
  							+-------------+

Any map requires a reference layer of grid, especially when the data is a field representing a phenomenon. This field data is typically available in raster format. Thus to play with the SRTM data, one has to decide the local region falling under the administrative boundaries of a state/province; country; continent.

→ Add Graticules:

Download Graticule file from Natural Earth & apply label properties. XKCD's finding your graticule would be of good help to understand geohashing in general. Starting with a wide geographical area/territory, we can pin down the required locality or region through geohashing resulting in the required graticule of the desired region.

→ Add Basemap:

Download files from Natural Earth & apply country, state filters.

→ Add Local Region & Find Graticues:

Query OSM using Overpass & filter the required region. Your region of interest might vary... and wouldn't it be interesting to select your locality :D ?

LONGITUDE	LATITUDE	FILE TO ADD
78	12	lon[78] + lat[12] = N12E078.hgt
79	13	lon[79] + lat[13] = N13E079.hgt

Manual: In this case, the map shown above is annotated with coordinates which are bottom-left corners (red markers) of the graticules that covers the region (yellow) required. Please do read the SRTM documentation to know why is it so.

In SRTM case, the raster data represents the height, in which the files represent the elevation profile of an area limited by the graticule formed by cardinal latitudes & longitudes. The above table clarifies how to find the required files. Thus the following 4 files are required...

• N12E078.hgt
• N12E079.hgt
• N13E078.hgt
• N13E079.hgt

With Help: One shall use SRTM Tile Grabber web tool designed based on D3 by Derek Watkins, one of those inspiring designers in Internet.

NOTE: One has to be careful when working in a disputed area. Make sure to work by including the disputed area as a part of your residing country. That way map would be protected by the local law, and can immune the individual from foregin complaints. Do remember, that when sharing maps in Internet, the map comes in public view without any geographical limit to access it, and thus be prepared to take the heat from attacks and do not forgot to provide disclaimer while doing so !

→ Add SRTM - hgt files:

One might easily note discontinuity in the SRTM rasters on the selected region. This occurs irrespective of the size of the region of interest. To eradicate this discontinuity, one can merge the rasters and render into a single raster.

We shall disable other layers that acted as reference & helped us selecting required SRTM files.

One might learn that, after following these steps, that there is no need to access natural earth data at all for working with a specific locality. But i have included it for learners, and might help learners who are interested in doing this stuff for the whole state or country itself !!

OK, with the completion of acquiring required data - accessing files from public domain remote sensing websites, and projecting them into the canvas of the GIS tool (GRASS/QGIS), we shall see how to pre-process the data, and apply required/interested raster/vector analysis. We shall call the available elevation data as digital elevation model (DEM).

→ Fill voids in the SRTM raster:

SRTM files are collected by RADAR sensing method, often comes with voids when the terrain is so much rugged (high variance/high slopes). These voids can be filled out using the GIS tool.

→ Terrain Analysis:

Making sense of the topographical data in of elevation layers (DEM) is vital, as it can drive the impetus to apply suitable analysis upon interpreting the terrain based on its statistical variations.

• Slope, Aspect, Derivatives & Curvature

  

  

  

• Shaded Relief

  

• GDAL DEM package can be used to analyze and visualize DEM raster data:
  • hillshade
  • slope
  • aspect
  • color-relief
  • Terrain Ruggedness index
  • Terrain position index
  • Roughness

→ Vectorization:

After generating the DEM, and applying suitable terrain analysis, the above resultant terrain models can be used to extract the contours. The following contour lines are generated from the void filled shaded relief raster obtained above.

→ Overlaying:

Further, the contour lines can be beautified by labelling each contour line with their level value. The following zoomed in view of contour lines with different step levels overlayed on the shaded relief layer provides good interpretation of elevation about the terrain of the selected region.

Whatever might be the analysis implemented on the data, the results along with the original data, can be visualized in number of ways. 3D visualization is one of the way that modern webgl & js enables one to simply use modern web browesers to communicate the data effectively.

Visualization, Interpretation & Communication plays vital role in enabling the user to properly interpret the information that the map is designed to convey. Level of detail varies upon the information carried by the map itself defined by the application. With more information the resolution and interpretation would be improved, while craving for computing power. Try it out for yourself !!!

The 3D rendering can also be awesomely done using GRASS-Nviz tool which offers wider controls and visualization options. For instance, take a look at what Open Research + Open Data + FOSS Hardware & Software + Open Design + Open Geomatics + Open Archeology can do.

With the widespread usage of Internet, and accelerated usage of client server model throughout the web mapping services like Open street map; Open cycle map; Open topo map; ... etc, have enabled commons to easily contribute as a collaborative effort. However, with transition towards grassroots based internet, the ability to serve the content/information locally to the locality in a decentralized way or a flat p2p way is a better approach than the centralized hierarchically controlled internet based business that restricts access to web map data.

★★★★★ : This is where the federated free software projects like tileserver-light - by osmtovector tiles project comes to play in place of traditional mapnik based image(png) based web map services with space and presentation efficiency that effectively reduces the delay and offers information capacity by hacking the nature of scalable vector graphics and the ability of modern web technologies. The end result is a scalable and federated approach to access map, which necessarily does not even require continuous internet connection to serve the map, to update the user screen - until and unless the new data is itself updated in the collaborative map servers. Already fully p2p capable map sharing applications are being developed and would more likely serve in the space of hard to reach & slow internet places, eliminating the digital divide in the Collaborative Web Mapping & possibly in GIS.

The above screenshot represents the xray-view rendered by the tilerserver-light that takes the mapbox-tiles file generated from the geojson files using tippecanoe. The required geojson files are generated from the shapefiles/xml files using GRASS/QGIS. Here, the vector layers such as contours, OSM based points, lines, multiline-strings, polygons, admin layers.. are used to generate the respective geojsons. The raw geojson can be easily styled and can be used to produce very attractive web maps. Styling can be done locally by Maputnik either using the web version or by deploying the desktop version. :D...