Dear SAGA developers,
I am trying to use SAGA to calculate the slope gradient of the main channel in a watershed. I think I need to first identify (preferably by the software rather than visually) the main channel from the channel network after watershed delineation, and then calculate the slope for the channel based on DEM. I can't find the right tool(s) for this task. Would you please advice? Eventually I need to do this for a number of watersheds and I plan to do it in R with the Rsagacmd package, if I can figure out the right procedure.
Thanks for you reply. I can understand this procedure. The question is how I can determine the main channel (the longest path in the channel network) and the associated channel slope from the channel network.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I don't know if this can be achieved easily, as I doubt that the "main channel" of a catchment can be found simply by length. But maybe you find some procedure that is working for your data / use case. I'm missing the time to think about this is in detail, but maybe this could serve as a starting point:
Maximum Flow Path Length tool (direction of measurement: upstream); from the result you can extract channels above a certain length
maybe combine this with a certain Strahler order threshold
...
Best regards,
Volker
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
The Maximum Flow Path Length is a useful tool. I also looked at it. Is it possible to export the maximum flow path polyline vector from this tool? Looks like it does not do this at present.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
No, the tool does not create a polyline and there is no way to create such a line from the raster output.
Another possibility, though still not "stable" (in the sense that it will not work for every case), is to use the "Channel Network" tool. It creates a polyline output of the channel network, but the lines are split at the channel junctions. In order to get only the "main" streams, you can experiment to set the "Min. Segment Length" parameter to a quite high value. This way you should be able to get long channel segments.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Dear SAGA developers,
I am trying to use SAGA to calculate the slope gradient of the main channel in a watershed. I think I need to first identify (preferably by the software rather than visually) the main channel from the channel network after watershed delineation, and then calculate the slope for the channel based on DEM. I can't find the right tool(s) for this task. Would you please advice? Eventually I need to do this for a number of watersheds and I plan to do it in R with the Rsagacmd package, if I can figure out the right procedure.
Thanks,
Last edit: sand 2021-01-23
Hi,
I think I would calculate the slope for the whole DEM and then mask the slope grid with the channel network.
(1) Slope, Aspect, Curvature tool: use the "maximum slope" method to get the channel slope
(2) Channel Network and Drainage Basins tool: output the Strahler order and choose an appropriate (Strahler order) threshold
(3) Grid Masking tool: apply the Strahler Order grid as mask to your slope grid
For sure there are several alternative approaches.
Best regards,
Volker
Thanks for you reply. I can understand this procedure. The question is how I can determine the main channel (the longest path in the channel network) and the associated channel slope from the channel network.
Hi,
I don't know if this can be achieved easily, as I doubt that the "main channel" of a catchment can be found simply by length. But maybe you find some procedure that is working for your data / use case. I'm missing the time to think about this is in detail, but maybe this could serve as a starting point:
Best regards,
Volker
The Maximum Flow Path Length is a useful tool. I also looked at it. Is it possible to export the maximum flow path polyline vector from this tool? Looks like it does not do this at present.
No, the tool does not create a polyline and there is no way to create such a line from the raster output.
Another possibility, though still not "stable" (in the sense that it will not work for every case), is to use the "Channel Network" tool. It creates a polyline output of the channel network, but the lines are split at the channel junctions. In order to get only the "main" streams, you can experiment to set the "Min. Segment Length" parameter to a quite high value. This way you should be able to get long channel segments.