Dear friends,
I am working with the r.watershed module in order to compute the LS Faktor for (R)evised-Universal-Soil-Loss-Equation. For this option in the module, I have to set the threshold for the minimum size of exterior watersheds. There are many options out there to compute the LS-factor, but I could not find the paper/report/… that describes the algorithm used in GRASS.
Revised Universal Soil Loss Equation for Western Rangelands (Weltz et al. 1987).
I want to understand why I need to set this threshold if I want r.watershed to compute the LS -factor. So if somebody could help me with:
a) the reference, so I can check by my self.
or
b) a short explanation why this parameter must be set.
By the way, I tested r.watershed with several thresholds (resulting in 1 big subbasin up to several hundreds) but the results for the Ls factor hardĺy changed.
Thanks in advance,
Stefan
On Wed, Dec 12, 2012 at 10:16 AM, Stefan Luedtke
<sluedtke@gfz-potsdam.de> wrote:
Dear friends,
I am working with the r.watershed module in order to compute the LS Faktor
for (R)evised-Universal-Soil-Loss-Equation. For this option in the module, I
have to set the threshold for the minimum size of exterior watersheds. There
are many options out there to compute the LS-factor, but I could not find
the paper/report/.. that describes the algorithm used in GRASS.
Revised Universal Soil Loss Equation for Western Rangelands (Weltz et al.
1987).
I want to understand why I need to set this threshold if I want r.watershed
to compute the LS -factor. So if somebody could help me with:
a) the reference, so I can check by my self.
or
b) a short explanation why this parameter must be set.
The LS and S factor calculation is only done for cells that are not
streams. I guess the reason is that soil loss does not apply to
streams (rivers). Streams in turn are determined using the threshold
option: all cells receiving more than threshold accumulated surface
flow are regarded as streams. A reasonable threshold depends on 1) the
resolution of the DEM, 2) the characteristics of the area under
consideration, 3) the context of the analysis.
By the way, I tested r.watershed with several thresholds (resulting in 1 big
subbasin up to several hundreds) but the results for the Ls factor hardĺy
changed.
Because it should have changed only for cells that are streams / not
streams depending on the threshold.
HTH,
Markus M
Thanks in advance,
Stefan
_______________________________________________
grass-user mailing list
grass-user@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/grass-user
Thanks a lot, that helped and made my day.
Cheers,
Stefan
On Wed, 2012-12-12 at 21:52 +0100, Markus Metz wrote:
On Wed, Dec 12, 2012 at 10:16 AM, Stefan Luedtke
<[sluedtke@gfz-potsdam.de](mailto:sluedtke@gfz-potsdam.de)> wrote:
> Dear friends,
>
> I am working with the r.watershed module in order to compute the LS Faktor
> for (R)evised-Universal-Soil-Loss-Equation. For this option in the module, I
> have to set the threshold for the minimum size of exterior watersheds. There
> are many options out there to compute the LS-factor, but I could not find
> the paper/report/.. that describes the algorithm used in GRASS.
>
> Revised Universal Soil Loss Equation for Western Rangelands (Weltz et al.
> 1987).
>
> I want to understand why I need to set this threshold if I want r.watershed
> to compute the LS -factor. So if somebody could help me with:
>
> a) the reference, so I can check by my self.
> or
> b) a short explanation why this parameter must be set.
The LS and S factor calculation is only done for cells that are not
streams. I guess the reason is that soil loss does not apply to
streams (rivers). Streams in turn are determined using the threshold
option: all cells receiving more than threshold accumulated surface
flow are regarded as streams. A reasonable threshold depends on 1) the
resolution of the DEM, 2) the characteristics of the area under
consideration, 3) the context of the analysis.
>
> By the way, I tested r.watershed with several thresholds (resulting in 1 big
> subbasin up to several hundreds) but the results for the Ls factor hardĺy
> changed.
Because it should have changed only for cells that are streams / not
streams depending on the threshold.
HTH,
Markus M
>
> Thanks in advance,
>
> Stefan
>
>
> _______________________________________________
> grass-user mailing list
> [grass-user@lists.osgeo.org](mailto:grass-user@lists.osgeo.org)
> [http://lists.osgeo.org/mailman/listinfo/grass-user](http://lists.osgeo.org/mailman/listinfo/grass-user)
>