[GRASS-user] r.flow: units for flow accumulation

[Rich Shepard]

On Sat, 20 Oct 2018, Micha Silver wrote:

> I have always found that the best way to do watershed delineation, to get
> streams and flow accumulation is to set the computational region a *bit
> larger* than the basin you are examining, and use a DEM that covers the
> full region. i.e. don't second guess the module. Then all pixels within
> the basin will have positive flow accum values. (Only pixels outside the
> main basin, which are in partially covered, adjacent basins will have
> negative values).
> 
> The final step could be to mask out all areas outside the basin of interest.

Micha,

   Thank you. It's been many years since I used r.watershed so this is a
valuable reminder.

> Depending on what results you need, you might apply a "depression" raster
> if there are true depressions in you region (like the Dead Sea in my
> case), but I don't think you have anything like that.  And of course the
> threshold parameter is very important - determines how dense the final
> stream network will be.

   The hydrography on the map I appended to my message was not generated by
r.watershed but is part of the national hydrography data set. I'll let
r.watershed determine create the streams.

Regards,

Rich