[GRASS-user] r.sim.water and Spearfish60 sample data

GRASS GRASS user
1 
Anyway, I decided to install grass 6.3.0 on my pc but i have some problems (i will open another thread for this).
I think I will install linux ubuntu on my nb too, so can
use grass without cygwin.
About what Hamish said, the problem is in the elevation resolution?? Could use
the 6.3.3 script usinf elevation.dem instead of elevation.10m?

Thank you,
Roberto

Markus Neteler wrote:
> *I made a test, it works (Linux box):*
>
> *GRASS 6.3.svn (spearfish60):~/grass63 > sh*
> *raster/simwe/r.sim.water/spearfish.sh*
> *projection: 1 (UTM)*
> *...*
> *cells:      294978*
> *Preparing input maps...*
> *Percent complete:*
> *100%*
> *Creating support files...*
> *Elevation products for mapset [neteler] in [spearfish60]*
> *E-W SLOPE [elevation.dem_dx] COMPLETE*
><i> N-S SLOPE [elevation.dem_dy]
 COMPLETE</i>
> *100%*
> *100%*
> *100%*
> *r.sim.water --o elevin=elevation.dem dxin=elevation.dem_dx*
> *dyin=elevation.dem_dy       rain=elevation.dem_rain*
> *manin=elevation.dem_manin infil=elevation.dem_infil*
> *depth=elevation.dem_depth disch=elevation.dem_disch*
> *err=elevation.dem_err*
> *Running MAY 10 version*
>
> *zmin,zmax 1066.000000 1840.000000*
> *simean,vmean,chmean,deltap,deltaw 0.000001 8.428620 19.670823*
> *0.889825*
> *1000000000000.000000*
> *MITER, timec 0 4.000000*
> *100%*
> *min=0.000155*
> *max=0.316380*
> *min=0.000000*
> *max=25.006544*
> *min=0.000155*
> *max=0.316380*
> *Written:*
> *Output water depth raster file:     elevation.dem_depth*
><i>  Output water discharge
 raster file: elevation.dem_disch</i>
> *Output simulation error raster file: elevation.dem_err*
>
> *GRASS 6.3.svn (spearfish60):~/grass63 >*
>
> *Since you used GRASS6.2.3, maybe give 6.3 a try?*
> *Possibly some bugfixes have not been backported.*

I just tried the raster/simwe/r.sim.water/spearfish.sh script from
GRASS 6.2.3 [in 6.2.3], and it worked ok (Linux). It took an hour+ to
finish, but it got there.

The "segfault on line 32" refers to the line in the script which
r.sim.water is called.

I notice it used a lot of memory, here is what top reported while the
process was running:
   VIRT  RES  SHR
  1115m 290m 5756

Maybe that is the problem?

the script in 6.2.3 sets the region to match elevation.10m, while the
one in 6.3svn uses elevation.dem (30m). So the 6.3 script will use much
less memory and be
 faster.

Hamish

L’email della prossima generazione? Puoi averla con la nuova Yahoo! Mail

2

roberto caselli wrote:

About what Hamish said, the problem is in the elevation resolution??
Could use the 6.3.3 script usinf elevation.dem instead of
elevation.10m?

In the 6.2.3 version of spearfish.sh, just edit the line at the top to
be elevation.dem instead of elevation.10m. elevation.dem is at 30m
resolution so will be 9 times smaller and need that much less RAM/time.
With most of the raster modules you can get away without very much RAM,
but apparently not r.sim.water.

Assuming that's the problem- in that case I would have thought that
G_malloc() would give you an "out of memory" error instead of a
SegFault, but I am not very familiar with the cygwin stuff.

Hamish

      ____________________________________________________________________________________
Never miss a thing. Make Yahoo your home page.
http://www.yahoo.com/r/hs