Calculate the Topographic Wetness Index in QGIS

This video shows how to calculate the Topographic Wetness Index (TWI) from a DTM using QGIS.
You can use this procedure for catchments, subcatchments or parcels.
This video is part of IHE Delft OpenCourseWare. More free materials are available at www.gisopencourseware.org
For an overview of our GIS courses and info on tailor made trainings and advise see our folder: www.un-ihe.org/sites/default/...
More information about the book "QGIS for Hydrological Applications - Recipes for Catchment Hydrology and Water Management" by Kurt Menke and Hans van der Kwast can be found at the website of Locate Press: www.locatepress.com/hyd
IHE Delft and Nieuwland Geoinformatie offer an online course. It includes support from experienced trainers and you'll get the official QGIS certificate after completion. More info at: www.learning-gis.com
Subscribe to the KZbin Channel for updates!

[@HansvanderKwast]

The last equation contains an error. The upslope area is calculated by the SAGA tool in pixels. So it needs to be multiplied by the area of the pixels. In this case 5 x 5. Also maybe not to clear why we have to modify the slope map. Because we take the natural logarithm, we can't have 0 or negative slopes. We make those cells 1, because ln(1) results in 0.

[@hannahwhite9249]

Why do you add one? Would the corrected equation be.... ln(upslope_area + 1 * 5*5) / tan(slope_radians) ?

[@HansvanderKwast]

@@hannahwhite9249 The one is not needed. So the correct equation is ln((upslope_area * 5 * 5) / tan (slope_radians))

[@maggienguyen8676]

hello. how can I determine the area of a pixel?

[@HansvanderKwast]

In the Layer properties you can find the cell size as length in x direction and length in y direction. If you multiply them you have the area.

[@maggienguyen8676]

@@HansvanderKwast Thank you so much for your reply. I cannot add the image of my layer property. Is this true that the "Pixel size" in the "Information" of the "Layer Properties" is the area of a pixel? Because in my case it is 30, -30. I am confused about the negative value. I hope to receive your feedback

[@Ashrumochan]

Best for hydrology

[@Mp3Robbo]

This is good, do you know of a way to easily do this stochastically?

[@johnmaskell5124]

Does the pixel area have to be in metres squared or can it be kept in degrees? Is it best to reproject the DEM so that it is metres?

[@user-xl7yb6dl2x]

Hi! Thanks for this video. But the upslope area to be used is the total catchment area multiplied by the area of the pixel or the specific catchment area?

[@user-xl7yb6dl2x]

Or is it the "contributing area" calculated by the flow accumulation, already an SCA?

[@schrockandroll]

Hi Hans, I successfully made a TWI map of my study area, and have since done a series of 120 cm deep soil core observations. In hilly areas the TWI corresponded quite well with soil wettness of my field observations, but in the flat areas I observed much wetter soils than the TWI model predicted. Is there a better model for predicting wetlands in flat terrain?

[@HansvanderKwast]

TWI is indeed not suitable to detect wetlands on plateaus, because it only takes into account the contributing area and the slope. I don't think there are specific indices for those cases, but you can create an empirical model when you know the conditions that these wetlands exist, for example by adding data on permeability.

[@RK-kv3yo]

Dear Hans, Coukd you please guide me how to calculate TWI with SRTM filled void? There is no DTM for the area I want to calculate it. Thanks 🙏

[@HansvanderKwast]

ocw.un-ihe.org/course/view.php?id=11&section=10

[@omarabdelwahab3722]

Hi, Why did you use the DTM for calculating the Slope instead of filling sinks and using the filled DTM?

[@HansvanderKwast]

This was a filled DTM, but filling was not part of the video. The scope was to show the TWI calculation. Filling is in another video.

Hi Hans, thanks for sharing this methodology. Following your approach, I am getting a TWI with negative values. Before I made those slope values of 0 equal to 1 so I tried to avoid the issue with natural logarithm. Do you have any comment on this? I would appreciate it!!!

[@HansvanderKwast]

Hi Cristo. Not sure why you get negative values. The detailed steps are in the new tutorial on this page: ocw.un-ihe.org/course/view.php?id=11&section=10

[@sadiemoland2157]

Hey I also got negative values! I know your comment is from a year ago, but did you ever figure out why you got this result? Thanks!

[@schrockandroll]

Hi Hans, great video. I'm having trouble with the Flow Accumulation (qm of esp) tool in QGIS. If I use the option "fill sinks," temporarily or permanently, the operation never finishes. It just adds "(not responding)" to the top of the tool. Is fill sinks really necessary?

[@HansvanderKwast]

Fills sinks gives problems in the following cases: 1. DEM too flat 2. DEM too large for memory 3. Geographic Coordinate System used instead of projection. 4. Spaces in folder and file names Fill sinks is necessary in hydrological analysis, otherwise water gets trapped in artificial sinks and will not flow to the outlet of the catchment.

[@hvzxjgqnqmz]

@@HansvanderKwast Can you reccomend another tool for filling sinks? I also have problems with this tool working with big DEMs

[@HansvanderKwast]

@@hvzxjgqnqmz Did you try Wang & Liu XXL? That's for big DEMs.

[@Ga3rich]

Hello! I wanted to ask why you do that from the DTM and not the DEM. Would it work if we start with DEM data? Secondly, when I created the upslope area. It is created in another coordinate system. So the results are sent to the middle of nothing. Instead of the place where my study area is.

[@HansvanderKwast]

Check this video for the definition of DEM, DSM and DTM: kzbin.info/www/bejne/pp-klZSppq2rfc0. A DEM is the general term. DTM is when the objects (human and natural) are removed and we only have the elevation of the terrain. For the algorithm it doesn't matter what you use, but if you use a DSM, the objects will be interfering with the general drainage. If your upslope area is in another projection, you need to reproject it to the same as the other layers in the project. With raster calculations also all rasters need to be alligned, using the Allign rasters tool.

[@RK-kv3yo]

Dear Hans van der Kwast how can I create DTM as shown in this video?

[@HansvanderKwast]

This DTM was created with special equipment (LIDAR) on the ground, it's not possible to get this as open data for the whole world. If you have LIDAR, you need to interpolate the point cloud to create a DTM like this.

[@spaciandd]

@@HansvanderKwast Can we use DEM instead of DEM ?

[@hannahwhite9249]

@@spaciandd Yes, both DEM and DTM are bare-earth elevation models (often used synonymously).

[@carolinesrensen4080]

I can't seem to make the flow acumulation, QGIS shuts down everytime. Can you help with this?

[@HansvanderKwast]

Hi. Which version are you using? Is your DEM projected? Which projection?

[@carolinesrensen4080]

@@HansvanderKwast Version of QGIS? 3.10.3 When I look at the properties, it says it's projected, "EPSG:25832 - ETRS89 / UTM zone 32N - Projected"

[@HansvanderKwast]

That seems to be okay. Can you please check if file or folder names have spaces or special characters. That is also often a problem. Which operating system are you using?

[@carolinesrensen4080]

@@HansvanderKwast the path to the file is: C:\Qgis\Bachelor\DTM_.tif so there are no spaces or different letters. I use windows 8, would that have any effect?

[@HansvanderKwast]

Still looks okay. Can you e-mail me screenshots of the dialogue screens before it crashes?

[@giang2u]

I thought you might have a mistake when writing the formula for calculation (at least at the last step)

[@HansvanderKwast]

The mistakes are discussed in the comments.