I'm trying to compute a flowline path with r.flow so as to create a
vector drainage pattern. However I'm not getting good results as r.flow
is giving me a vector file in which the the flowline are very short and
do not connect in the valleys. The valleys actually seem to have no
flowlines while only very steep slopes do have flowlines.
My DEM does have flat areas (r.stats reports 500.000/2.800.000 zero
degree cells) so I presume it's this that stops the flowlines.
I must say though, that my DEM was built using only contour lines and no
quoted points. I presume this is the cause of my 500.000/2.800.000 zero
cells .....?
I built my DEM using r.surf.contour on a 7,5 meter square cell
resolution for an area of 11,5 Km x 14 Km (total of 1800x1550 celled
grid). Do you think my resolution is too low to get a valid flowline
raster map? I read in the r.flow man page that centimeter resolution
should be used. But if I use a centimeter resolution my PC would take
1000 light years to compute my DEM ....
I'm sure I missing something but I'm coming up to Gis technologies only
recently so ...... not good at it yet.
Your DEM may have little pits (closed basins) that stop drainage. Usually, these pits are artifacts of the DEM so you can use a program like r.fill.dir to "fill" the pits and created a "filled" dem.
Use the "filled" dem in r.flow and see if the results are better. r.flow is pretty picky about flat areas (see the notes in the manual). You may have better luck with r.watershed and r.terraflow for extracting the river networks.
David Finlayson
Antonio G. - Geotronix wrote:
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Hi.
I'm trying to compute a flowline path with r.flow so as to create a
vector drainage pattern. However I'm not getting good results as r.flow
is giving me a vector file in which the the flowline are very short and
do not connect in the valleys. The valleys actually seem to have no
flowlines while only very steep slopes do have flowlines. My DEM does have flat areas (r.stats reports 500.000/2.800.000 zero degree cells) so I presume it's this that stops the flowlines.
I must say though, that my DEM was built using only contour lines and no
quoted points. I presume this is the cause of my 500.000/2.800.000 zero
cells .....?
I built my DEM using r.surf.contour on a 7,5 meter square cell
resolution for an area of 11,5 Km x 14 Km (total of 1800x1550 celled
grid). Do you think my resolution is too low to get a valid flowline
raster map? I read in the r.flow man page that centimeter resolution
should be used. But if I use a centimeter resolution my PC would take
1000 light years to compute my DEM ....
I'm sure I missing something but I'm coming up to Gis technologies only
recently so ...... not good at it yet.
* giovedì 03 luglio 2003, alle 11:42, David Finlayson scrive:
Your DEM may have little pits (closed basins) that stop drainage.
Usually, these pits are artifacts of the DEM so you can use a program
like r.fill.dir to "fill" the pits and created a "filled" dem.
Yes, i realized that r.flow was rather nasty about pits. I read it in
the man page.
Use the "filled" dem in r.flow and see if the results are better. r.flow
is pretty picky about flat areas (see the notes in the manual). You may
have better luck with r.watershed and r.terraflow for extracting the
river networks.
Yes, Markus has already told me about r.terraflow. I'll give it a try
but first I have to learn how to use CVS on my Debian box as I seem to
understand that r.terraflow is available only in the source tree.
(howcome it's not available in the weekly snapshots?)
Thanks for the "filled" option which I didn't consider. I'll give it a
try.
I'll give it a try but first I have to learn how to use CVS on my
Debian box as I seem to understand that r.terraflow is available only
in the source tree.
(howcome it's not available in the weekly snapshots?)
