A demo on the use of hydrosheds (hydrological data & maps based on shuttle elevation derivatives) at multiple scales for Orissa

The national hydrographic dataset can be used to delineate watersheds and mark all streams and channels that contribute to the flow at the outlets of these watersheds. Even though I was able to find out pdfs created by the National Information Center, I felt we need something better than the pdfs.

So if you don’t have what you need, the only solution is to build it. In order to make flow lines for a region all you need is the digital elevation model of the region and after following a sequence of steps, you will have your flow lines and watersheds delineated.

Searching for the digital elevation models DEM for India, I came across a lot of sites online, and to my surprise India didn’t have a DEM hosted on any of the government owned websites (correct me if I am wrong!).

Hydrosheds (Hydrological data and maps based on SHuttle Elevation Derivatives at multiple Scales) is a worldwide program that collects hydrographic information for regional and global scale applications in a consistent format. Hydrosheds is based on high-resolution elevation data obtained during a space shuttle flight for NASA's Shuttle Radar Topography Mission (SRTM). Hydrosheds provide vector and raster data at various scales. More information can be found at: http://hydrosheds.cr.usgs.gov/data.php. The data is available at scales of 90m, 500m, 1km and 10 km resolutions.

Below are the sequence of steps, I followed to create the flow lines and watershed boundaries for Orissa India:

1. I downloaded the 3 arc-sec resolution data for India by using the viewer tool.
2. Used ArcGIS to create a Mosaic database and added the relevant DEMs.
3. I downloaded the shape file for India from http://www.vdstech.com/map_data.htm.
4. Used the select by attribute option to locate the state Orissa and exported data for selected attributes, in this case Orissa.
5. Clipped the raster for the state of Orissa, India.
6. Used the Arc tool box --> Spatial Analyst --> Hydrology.
   1.  Open the fill tool and give the clipped Orissa raster as Input and create Fill Ori DEM.
   2.  Create a flow direction grid by using Flow Direction tool box.
   3.  Create the flow accumulation raster by using the Flow Accumulation tool box.
   4.  Change the symbology method to classify and have only two classes and alter the break value to 3000.
   5.  Now you will be able to see the flow lines appearing on the map.
   6.  Use the raster calculator to make a new raster from the Flow Accumulation grid with values greater than 3000.
   7.  Use the raster to polyline tool to create polylines by selecting the background value to ZERO.
   8.  Now you have the flow lines for the Area.

Since I used the 90m resolution data, it is possible that the flow lines will not match exactly at higher resolutions.

The next step was to delineate the watersheds for the streams. I used the basins tool to delineate the watersheds from the flow line features that I created before. I have attached screenshots of the final results. Since the shape files were large in size, I have attached the shape files on my blog site and you can download it from https://sites.google.com/site/harish2rb/home/tech-blogs.

Since the shape files have been downloaded from a freely available source, I don't think the boundaries of the state are accurate. I could myself see the differences when I overlay the shape file on the base map in ArcGIS. The boundaries of the watersheds could also be messed up since I clipped them to the state boundaries. So if someone is using this, he/she better be sure that the data is at 90m resolution and the boundaries of watersheds and flow lines are clipped to the state boundary. I was not able to add the exact names of the basins to the attributes table. Please refer to http://gis.ori.nic.in/ for accurate information.

The idea behind this demonstration was to use hydrosheds and show that we can create a flow line feature class and use it to analyze hydrological behavior for regions that don't have the capability to spend money to gather high resolution data. This kind of data helps in watershed management and other hydrological analysis.