[GRASS-user] Re: Defining Partial Drainage Basin

[Micha Silver]

Rich Shepard wrote:

> On Wed, 10 Feb 2010, Micha Silver wrote:
>
>> A quick tip regarding r.water.outlet (you may have noticed this already)
>> The module creates a new raster covering the whole analysis region with
>> two possible values: 1 for all cells draining into the outlet, and 0
>> everywhere else. If you do r.to.vect straightaway, you?l get a vector
>> containing two polygons... I always run r.null setnull=0 on the
>> r.water.outlet result first. This insures that r.to.vect leaves you with
>> just the drainage area.
>
> Micha,
>
>   I'm under a deadline to finish the current project, and nothing 
> seems to
> be working correctly for me.
>
>   I just ran r.water.outlet using the drainage (aspect) map created by
> r.watershed. The output is a yellow rectangle that the legend tells me
> represents '0'; that is, there're no cells in the calculated basin.

The r.water.outlet module takes two parameters: first is the drainage 
*direction* raster, which I usually create first with r.watershed, like:
 >r.watershed elev=dem drain=flow_dir accum=flow_acc basin=basin 
stream=str thresh=xxxx
(At this stage you can go and make popcorn...)
Next I do r.thin on the streams raster, and convert the thinned raster 
to a line vector:
 >r.thin str out=str_thin
 >r.to.vect -s str_thin out=streams feature=line

Now you need the pour point which, as others have made abundantly clear 
( ;-) ), must be exactly on a stream line.  So zoom in *very* close to 
the flooded house, and identify the X-Y coords on the stream just below 
the house.
Now try:
 >r.water.outlet drain=flow_dir basin=house_catchment east=xxx north=yyy

And finally the above trick to get your final vector drainage basin with:
 >r.null house_catchment setnull=0
and
 >r.to.vect -s house_catchment out=house_catchment feature=area

Good luck...

>
>   g.region -p reports:
>
> projection: 99 (Lambert Conformal Conic)
> zone:       0
> datum:      nad83
> ellipsoid:  grs80
> north:      1334420
> south:      1279150
> west:       769190
> east:       819260
> nsres:      10
> ewres:      10
> rows:       5527
> cols:       5007
> cells:      27673689
>
>   The command line was:
>
> r.water.outlet drain=aber10m.drain basin=lockie e=795542.95 
> n=1308323.52 --o
>
>   I can send copies of aber10m.drain and lockie. It makes no sense to me
> that such a simple run fails to define the partial drainage basin. Any 
> help
> or ideas you offer will be much appreciated.
>
> Rich
> _______________________________________________
> grass-user mailing list
> grass-user at lists.osgeo.org
> http://lists.osgeo.org/mailman/listinfo/grass-user


-- 
Micha Silver
Arava Development Co. +972-52-3665918
http://surfaces.co.il