[GRASS-dev] r.sun artifacts

[Markus Neteler]

On Fri, Oct 21, 2011 at 7:08 PM, Sergio Gollino <gollinosergio at gmail.com> wrote:
> Hi, this is my first post and I hope someone could help me.

Welcome here!

> I'm using r.sun for computing a solarmap of the alpine arch in Friuli
> Venezia Giulia (IT).
> I'm using this command:
>
> r.sun -s elevin=dem_40_fvg_wgs84 aspin=asp_fvg_wgs84 slopein=slo_fvg_wgs84

try to avoid aspin= and slopein= since r.sun can also calculate
those internally.

> lin=1.5 alb=0.2 horizonstep=10 beam_rad=irr_fvg1   insol_time=ins_fvg1
> diff_rad=diffuse_irr__fvg1   refl_rad=ground_refl_fvg1
> glob_rad=total_irr_fvg1   day=1   step=0.25 dist=1.0 numpartitions=1
> --overwrite

... this part looks ok.

> But the resultant map have a lot of artifacts (lines of pixels with wrong
> values, see the attachment)
> http://osgeo-org.1803224.n2.nabble.com/file/n6917645/r.sun.jpg
>
> I've tried to delete "aspin" and "slopein" parameters and i've changed
> "step" and "dist" with no results.

Mhh. So please check the DEM then bu calculating a shaded map with
r.shaded.relief. Perhaps the artifacts are therein?

> I'm using Grass6.4.1. My OS is W7 64bit.

I had no such problems in a close-by area with the same version (but Linux
which should not matter here).

Markus