[GRASSLIST:5644] Re: r.flow and flowpath length

[Jaro Hofierka]

Marco,

r.fill.dir is based on D8 algorithm, while r.flow is based on D-infinity
algorithm. Therefore r.fill.dir cannot create a fully depression-free 
DEM for r.flow. This explains at least part of your problems.

A short example of how to fix the problem and create hydrologically 
correct DEMs can be found in this paper available online:

Cebecauer, T., Hofierka, J., Šúri, M. (2002). Processing digital terrain 
models by regularized spline with tension: tuning interpolation 
parameters for different input datasets. Proceeding of conference "Open 
source GIS - GRASS users conference 2002", Trento (Italy), 11.-13. 9. 2002.
http://www.ing.unitn.it/~grass/conferences/GRASS2002/proceedings/pdfs/Cebecauer_Tomas.pdf

Kind regards,

Jaro

marco castrogiovanni wrote:
> hi,
> I'm currently building up an hydrologic model based on raindrop method.
> I need to calculate the lengths of flowlines. I've tried with r.flow by
> the following syntax:
> 
> r.flow -3 elevin=bas.dem aspin=bas.asp flout= bas.flow lgout=bas.len
> 
> where bas.dem is a raster map (filled with r.fill.dir) with resolution
> of 1920x2100 cells, with cell-dimensions 20x20 meters and with high
> precision quota value (six decimal numbers).
> I haven't obtained any acceptable results yet, 'cause all flowlines
> suddenly stop in unexpected points and they don't reach the outlet.
> Raster map of distances from outlet that I get after running r.flow
> have a lot of zero-value cells, but their corresponding quota and 
> aspect value wouldn't allow these results.
> Any suggestions would be appreciated.
> 
> Regards, Marco.
> 
> 
> 
> 
> --
> Prendi GRATIS l'email universale che... risparmia: http://www.email.it/f
> 
> Sponsor:
>  nuovissimo Sony Ericsson T68i  marchiato TIM
> Clicca qui: http://adv.email.it/cgi-bin/foclick.cgi?mid=552&d=22-2
> 
>