[GRASS-user] understanding r.watershed

[Markus Metz]

Georg Kaspar wrote:
> Hi,
> when providing a binary depression layer for r.watershed, areas around
> those depressions will be left out in the resulting basin map.
Maybe your basin threshold was too high, try with a lower value. Basins 
are only calculated if there is a stream exceeding the threshold.
>  Are
> those areas the watersheds of the given depressions?
> According to Markus Neteler's book "Open Source GIS", r.watershed
> "does not require filling of depressions in DEM prior to it's
> application [...]". What will be the difference if I use a
> depressionless DEM (with/without binary depression layer)?
>   
A depressionsless DEM is a DEM filled with r.fill.dir. Water will flow 
through where there was once a depression, i.e. in and out.
A binary depression layer as additional input to r.watershed indicates 
real depressions, like Mark said, water will flow into these depressions 
but not out. This is a substantial difference to a depressionless DEM.

Markus M

> Thank's for your help. I'm new to this field so please excuse any
> missing base knowledge!
> Best regards,
> Georg Kaspar
> _______________________________________________
> grass-user mailing list
> grass-user at lists.osgeo.org
> http://lists.osgeo.org/mailman/listinfo/grass-user
>
>