Terrian Analysis in Hills

Nepal has a diverse geography that includes flat land in the south; high rise mountains in the north, and valleys surrounded by hills in between these flat lands of south and mountains in the north. The variation in elevation throughout the country has always been a challenge for the development of infrastructure.

To identify how can a terrain-based analysis can help in knowing the surface of the earth, I choose to work in the hilly region of Nepal. Since elevation information is the key to the process, I used ALOS PALSAR 12.5 meter Digital Elevation Model (DEM) provided by NASA Earth Data (url:  https://search.asf.alaska.edu/#/ )

Construction of Dam, roadway or even airway is a difficult task for these areas where elevation changes in less than a meter. Study of the change in elevation needs to be taken into account before making plans for such kind of construction. In simple words, this change in elevations answers us of how much steep a surface is or how high is the place I am standing now to that I was standing? Mathematically, the answer is the slope.

Slope was calculated in terms of degree, referring to that the lower the slope value, the flatter is the terrain and the higher the value, the steeper the terrain. In my case, the slope output was from 0 degrees and it extended up to 79 degrees with a mean variation of 26.28 degrees when keeping the z-factor as 1 since the unit of elevation was same to the linear measurement unit. The slope information allows us to identify the suitable location for the construction of infrastructure if the only slope was to be considered.

Following the statement of UNISDR, "Creating a culture of prevention not just a culture of reaction". When building an infrastructure knowledge of surface curvature is required else there is high chance of an event happening when construction is done without taking into consideration of its curved surface. There can be a situation where the surface should neither be too much concave nor too convex.

As the slope affects the overall rate of movement downslope. Aspect defines the direction of flow. The profile curvature affects the acceleration and deceleration of flow and, therefore, influences erosion and deposition. The planform curvature influences convergence and divergence of flow.The curvature can be obtained by second derivative of the slope.

(url:  http://help.arcgis.com/en/arcgisdesktop/10.0/help/index.html#//00q90000000t000000 )

Mathematically, Curvature = -2(D + E) * 100; where D and E are the rasters generated by using low ppass filters based on the horizonal and vertical kernels respectively to enhance the DEM by specifying a weight of 1/(12.5^2) as 12.5 being the resolution of the raster.

Then 2 seperate raster obtained from convolution using two different kernels were fed as an input to the formula mentioned above in a raster calculator. A new raster was obtained that curvature of the slope as shown in the curvature map above. The posivite value represented the convex surface whereas the negative represented the concave.

For calculating the statistics on a cell value of a raster within a certain zone or region can be done with another dataset. This method allows us to summarize raster datasets based on the vector geometries. In this example, I wish in finding the statistics for Slope per Elevation zone. I reclassified the 12.5m DEM into 6 classes with 500m height difference. An exception is the first class which includes all values below 400 m elevation, so that the total number of classes remains clear. 

For identifying the frequecy distribution of the value across zones, histogram can be of great help. It will show the cell values on the value input for each unique zone. The graph shows the percent of slope in 6 classified elevation zone from 400m to 3400m with an interval of 500m.

The statistics for the zone was seprately computed when visualizing it in the form of map and graphs. A summary table is considered of great importatance and ease for the study of the disturbution of the values across the zone.

When the similiar method was adopted with a more coarse dataset difference in the statistic was observered. In newely created elevation zone shown in fig 6, it is noticeable that the data in Slope per elevation zone are more coarse which may be due to increase in pixel size. Though it seems that the Slope per elevation zone have change increased from 29,18 degrees to 31.87 degrees.This might have been due to the increse in spatial extent of the pixel.

For knowing the geographical surface terrain analysis is done with the use of elevation data. This process includes several parameters that could be vital in the process of performing analysis such as slope, aspect or curvature of the surface. Since a slope indicates an area with a mix of flat and hilly terrain, it is a determining factor for identifying the potential sites and can also prevent erosion control programs. A technique of zonal statistics makes the work much better and easy in calculating statics for each zone in the terrain analysis process.