In this tutorial I will talk about the weighted sum overlay analysis in QGIS. Weighted sum overlay is usually done for the suitability analysis. It uses several weighted raster layers for the calculation. GIS uses Boolean logic to perform overlay analysis. This means it uses conditional statements on input layers to generate suitability map.
For each input layer we have to assign the weight value. This weight value is multiplied to the input layer. Then you have to sum all these weighted layers to generate the final output map. Let me give you simple example to identify the soil erosion potential area. The inputs are terrain slope angle and vegetated/not vegetated area. Result will be calculated as “if slope >5% and vegetation=No” then erosion potential = true.
Here we are using weighted overlay analysis to find out ground water potential zones in a river basin using open source QGIS software. Usually this kind of analysis is done using ArcGIS. Unlike ArcGIS, QGIS have certain limitations but still we can do the raster overlay analysis. The inputs are as follows:
- Land use/ land cover
Some of the layers are vector and they will be converted into raster images. Then all raster layers will be multiplied by the weight value. Following table shows the weight of each layer, total weight is 100. Our aim is to find out the ground water potential zones of the river basin.
land use / land cover
Process 1: Find out the Drainage Density and Lineament Density
a. Open the QGIS software and add vector layers or “Ctrl+shift+v “ (Key board shortcut)
Figure 1: vector files loaded in the QGIS
b. From the main menu select “processing” and click on “toolbox” to activate the processing tools in the QGIS window
c. From the “Grass algorithms” in the “processing toolbox” select the “V to points tool” for convert line into points. Because to find out lineament and drainage density there is no line density calculation tool in the QGIS, so we can convert them into points and can find the kernel density.
Figure 2: V to points tool from Grass algorithms
When you click on the tool a window will open
d. Set the input line layer as “drainage”
c. Set the maximum distance between the map units as “100”
e. Set the output file name and the folder and press “OK”
Figure 3: shows the points created along the line
g. Similarly generate the points for lineament also
Figure 4: Points generated with lineament lines
h. Click on the “Grass algorithms” and select the “V Kernel tool” to find out the density map
Figure 5: “V Kernel tool box”
i. V kernel tool window will open, set the Input layer name as “lineament_points” and set the “output file name and folder”- press “OK”
Figure 6: Lineament density
j. Similarly find the drainage density also
Figure 7: drainage density
Process 2: Convert the Vector File into Raster
a. From the main menu select “Raster” –Select “Conversion” and click on “Rasterize (Vector to raster)
Figure 8: Rasterize window
Rasterize window will open
b. Set the Input file “Gemorphology.shp”
c. Set the desired attribute field
d. Set the output file name and folder
e. Click “OK”
Figure 9: Geomorphology raster is created
f. Similarly rasterize the other themes (Soil, Lithology and land use / land cover)
Figure 10: Land use Raster
Figure 11: Lithology Raster
Figure 12: Soil raster
Figure 13: Slope map (Prepared from DEM)
Process 3: Ordered Weighted Averaging
a. Select “SAGA geo algorithm” from the Processing tools and click on “Grid Analysis and select Ordered weighted averaging”
( note: if you do not see the tool then make sure you switch to “Advanced Interface” button located at the bottom from “simplified interface“)
Figure 14: Ordered weighted averaging tool box
b. Click on the “input grids and set the themes example “Soil”
c. Set the weights as per the table we are prepared for soil its 10
d. Input rest of the numbers from the above table
d. Set the output file name and folder
e. Press “OK”
Figure 15 and 16: shows the Input grid selection and assigning weight
Figure 17: raster overlay analysis result
We found the ground water potential zones
Green most suitable
Light green moderate suitable
Brownish yellow less suitable
This Post Has 17 Comments
UDAY12 Dec 2016
where did u find that data used in this tutorial ?
admin12 Dec 2016
Was written by our guest, may know why?
choudharipp17 Jan 2017
where will i get this sample data which is used in this tutorial. please send or send the link for practice purposes.
admin18 Jan 2017
Thanks, I will…
Harinda23 May 2019
Hi sir, Where can I download this data set
Heisenberg13 Apr 2017
Which version of QGIS was used to perform the analysis. Some of the processing tools are missing in 2.14.1 (LTR)
admin16 Apr 2017
Hi, I forget the version but was one and half year ago. Hope that helps
Heisenberg3 May 2017
I have a few doubts I’d like to clarify. Can I get your mail or something? It would be really helpful.
Heisenberg23 Jun 2017
I have few doubts, a response would be greatly appreciated
Raju Sukhaye13 Jul 2017
Can I please get this sample data which is used in this tutorial for the practice purpose and also steps for the creation of lithological and geomorphological maps. It would be kind of you if you email me the data and steps to my ID.
admin14 Jul 2017
I will see if I still have data available.
Anu raadha16 Aug 2017
I am unable to perform weighted average in SAGA. It says error. Please help. I am doing analysis of ground water potential zone.
Anu raadha16 Aug 2017
Unable to use weighted overlay in SAGA. Gives an error of type_raster. Please help.
Kirthiga9 Oct 2017
Hi, good tutorial!
How can I give ranks for categorical data. For example, if each of my soil class should be ranked before I assign theme weights for the layer. How to do that?
admin12 Oct 2017
Use number to assigns Category.
Kirthiga SM15 Oct 2017
Thanks for the reply. How to input weights for more than three rasters? Only a fixed table of 1×3 is given for inputting weights.
Justine7 Jun 2018
I’m trying to use you tutorial to analyse suitability for tourism development. Got an error message saying “”Input layer has more than one band. Multiband layers are not supported by SAGA”. Why did this didn’t happen to you ? Struggling a bit with the logic behind the commands, thank you for your response ! 🙂