r.watershed

de584438 at longs.lance.colostate.edu de584438 at longs.lance.colostate.edu
Sun May 10 16:41:47 EDT 1992


I am running GRASS4.0 on a DEC 5100.  The watershed of concern is very
large - the elev map is roughly 1700 x 750 cells.  In running
r.watershed for basin and stream delineation, I dont have enough memory
to run up front.  As a result, the program resorts to swapping to disk.
 I managed to free up enough disk space to allow it to do this. 
However, after approx 3 days of running, the program is only 11% done
with Section 1b- Determining Offmap flow

SECTION 1b (of 5): Determining Offmap Flow. Percent Complete:
	11%

At this rate, the program might terminate in 2-3 weeks.  

I am running using a min. outside sub-basin area of 150 km2.  The map
is at roughly 90 meter resolution.

Any suggestions on how to speed this up ?  Or, if I rescale the
elevation data to 270 meters ( for example), any ideas on how much
information will be lost ?  At what resolution will the delination
algorithm not adequately describe the topography ?  Will a 600 X 250
map run significantly faster ?  Does anyone have experince with scale
and size in r.watershed ?

I would appreciate any/all input.  I will summarize and post if there
is enough feedback.

Dan Epstein



More information about the grass-user mailing list