[GRASS-user] r.horizon and negative tilt

GRASS GRASS user
1

Dear all,

It seems that r.horizon in does not return negative heights. How it's possible in case of the given point from the calculation is done is located on mountain and the horizon around is below zero ??. I miss something may be...

Thank you for your explanations

Regards

Frank

2

Hi Frank,
just my opinion but r.horizon I suppose is targeted to solar radiation modules and because of that returns only the line-of-sight "obstacles" above the ground plane, where direct sunlight may be obstructed by obstacles.
best regards,

Andrea Balotti

Il 28/07/2018 15:59, Frank David ha scritto:

Dear all,

It seems that r.horizon in does not return negative heights. How it's possible in case of the given point from the calculation is done is located on mountain and the horizon around is below zero ??. I miss something may be...

Thank you for your explanations

Regards

Frank

_______________________________________________
grass-user mailing list
grass-user@lists.osgeo.org
https://lists.osgeo.org/mailman/listinfo/grass-user

3

Hi Andrea,
Thanks for your message. But what is the ground plane ? r.horizon use the topography, so nothing is plane (and the earth curvature tends to lower the ground). I guess as the calculation is done at ground level, the minimum viewable horizon cannot be lower than zero. The calculation should be done higher than ground level to “see” negatives horizon. Never mind I will try to create my own r.horizon with r.transect or r.profile.
Best regards,
Frank

···

Le 30/07/2018 à 08:47, balotti.and@gmail.com a écrit :

Hi Frank,
just my opinion but r.horizon I suppose is targeted to solar radiation modules and because of that returns only the line-of-sight “obstacles” above the ground plane, where direct sunlight may be obstructed by obstacles.
best regards,

Andrea Balotti

Il 28/07/2018 15:59, Frank David ha scritto:

Dear all,

It seems that r.horizon in does not return negative heights. How it’s possible in case of the given point from the calculation is done is located on mountain and the horizon around is below zero ??. I miss something may be…

Thank you for your explanations

Regards

Frank

grass-user mailing list
grass-user@lists.osgeo.org
https://lists.osgeo.org/mailman/listinfo/grass-user

grass-user mailing list
grass-user@lists.osgeo.org
https://lists.osgeo.org/mailman/listinfo/grass-user

4

Hi Andrea,
Thanks for your message. But what is the ground plane ? r.horizon use the topography, so nothing is plane (and the earth curvature tends to lower the ground).

Yes, what you explain below is what I mean with “ground plane”.

I guess as the calculation is done at ground level, the minimum viewable horizon cannot be lower than zero. The calculation should be done higher than ground level to “see” negatives horizon.

Have you taken a look to r.viewshed? I never tried it but maybe is useful for your purposes.

Never mind I will try to create my own r.horizon with r.transect or r.profile.
Best regards,
Frank

Best regards,
Andrea

···

Le 30/07/2018 à 08:47, balotti.and@gmail.com a écrit :

Hi Frank,
just my opinion but r.horizon I suppose is targeted to solar radiation modules and because of that returns only the line-of-sight “obstacles” above the ground plane, where direct sunlight may be obstructed by obstacles.
best regards,

Andrea Balotti

Il 28/07/2018 15:59, Frank David ha scritto:

Dear all,

It seems that r.horizon in does not return negative heights. How it’s possible in case of the given point from the calculation is done is located on mountain and the horizon around is below zero ??. I miss something may be…

Thank you for your explanations

Regards

Frank

5

Ho yes, why not. I could use r.viewshed to find the farest part of the topography able to see my point and get it’s angle mode. The trick to find is how to get that cell ?
Best regards,
frank

···

Le 30/07/2018 à 11:18, balotti.and@gmail.com a écrit :

Il 30/07/2018 09:24, Frank David ha scritto:

Hi Andrea,
Thanks for your message. But what is the ground plane ? r.horizon use the topography, so nothing is plane (and the earth curvature tends to lower the ground).

Yes, what you explain below is what I mean with “ground plane”.

I guess as the calculation is done at ground level, the minimum viewable horizon cannot be lower than zero. The calculation should be done higher than ground level to “see” negatives horizon.

Have you taken a look to r.viewshed? I never tried it but maybe is useful for your purposes.

Never mind I will try to create my own r.horizon with r.transect or r.profile.
Best regards,
Frank

Best regards,
Andrea

_______________________________________________
grass-user mailing list
[grass-user@lists.osgeo.org](mailto:grass-user@lists.osgeo.org)
[https://lists.osgeo.org/mailman/listinfo/grass-user](https://lists.osgeo.org/mailman/listinfo/grass-user)

Le 30/07/2018 à 08:47, balotti.and@gmail.com a écrit :

Hi Frank,
just my opinion but r.horizon I suppose is targeted to solar radiation modules and because of that returns only the line-of-sight “obstacles” above the ground plane, where direct sunlight may be obstructed by obstacles.
best regards,

Andrea Balotti

Il 28/07/2018 15:59, Frank David ha scritto:

Dear all,

It seems that r.horizon in does not return negative heights. How it’s possible in case of the given point from the calculation is done is located on mountain and the horizon around is below zero ??. I miss something may be…

Thank you for your explanations

Regards

Frank

6

On Sat, Jul 28, 2018 at 3:59 PM, Frank David <frank.david@geophom.fr> wrote:

Dear all,

It seems that r.horizon in does not return negative heights. How it’s possible in case of the given point from the calculation is done is located on mountain and the horizon around is below zero ??. I miss something may be…

r.horizon is indeed ignoring points that are lower than the given point, see
https://trac.osgeo.org/grass/browser/grass/trunk/raster/r.horizon/main.c#L1037
and following lines

I agree that this behaviour is questionable

Markus M

Thank you for your explanations

Regards

Frank

grass-user mailing list
grass-user@lists.osgeo.org
https://lists.osgeo.org/mailman/listinfo/grass-user

7

Hello Markus,

is it possible to add an option to evaluate the negatives hight in a future release ? (I don’t know C)

···

Best regards,
Frank

Le 30/07/2018 à 16:19, Markus Metz a écrit :

On Sat, Jul 28, 2018 at 3:59 PM, Frank David <frank.david@geophom.fr> wrote:

Dear all,

It seems that r.horizon in does not return negative heights. How it’s possible in case of the given point from the calculation is done is located on mountain and the horizon around is below zero ??. I miss something may be…

r.horizon is indeed ignoring points that are lower than the given point, see
https://trac.osgeo.org/grass/browser/grass/trunk/raster/r.horizon/main.c#L1037
and following lines

I agree that this behaviour is questionable

Markus M

Thank you for your explanations

Regards

Frank

grass-user mailing list
grass-user@lists.osgeo.org
https://lists.osgeo.org/mailman/listinfo/grass-user

8

Frank,

Some years ago I wrote a module (also called r.horizon) which I think does what you want. I never got around to formally contributing it (must compare with the existing r.horizon and do so some time). Anyhow, have updated for GRASS 7 and will send you the code direct - it compiles as r.horizon2 to avoid collision with the existing module.

Regards,

Mark

Frank David <frank.david@geophom.fr> writes:

Ho yes, why not. I could use r.viewshed to find the farest part of the
topography able to see my point and get it's angle mode. The trick to
find is how to get that cell ?
Best regards,
frank

Le 30/07/2018 à 11:18, balotti.and@gmail.com a écrit:

Il 30/07/2018 09:24, Frank David ha scritto:

Hi Andrea,
Thanks for your message. But what is the ground plane ? r.horizon use
the topography, so nothing is plane (and the earth curvature tends to
lower the ground).

Yes, what you explain below is what I mean with "ground plane".

I guess as the calculation is done at ground level, the minimum
viewable horizon cannot be lower than zero. The calculation should be
done higher than ground level to "see" negatives horizon.

Have you taken a look to r.viewshed? I never tried it but maybe is
useful for your purposes.

Never mind I will try to create my own r.horizon with r.transect or
r.profile.
Best regards,
Frank

Best regards,
Andrea

Le 30/07/2018 à 08:47, balotti.and@gmail.com a écrit:

Hi Frank,
just my opinion but r.horizon I suppose is targeted to solar
radiation modules and because of that returns only the line-of-sight
"obstacles" above the ground plane, where direct sunlight may be
obstructed by obstacles.
best regards,

Andrea Balotti

Il 28/07/2018 15:59, Frank David ha scritto:

Dear all,

It seems that r.horizon in does not return negative heights. How
it's possible in case of the given point from the calculation is
done is located on mountain and the horizon around is below zero
??. I miss something may be...

Thank you for your explanations

Regards

Frank

_______________________________________________
grass-user mailing list
grass-user@lists.osgeo.org
https://lists.osgeo.org/mailman/listinfo/grass-user

_______________________________________________
grass-user mailing list
grass-user@lists.osgeo.org
https://lists.osgeo.org/mailman/listinfo/grass-user

--
Dr Mark Lake
Senior Lecturer in Archaeology

UCL Institute of Archaeology
31-34 Gordon Square
London WC1H 0PY
UK

Tel: +44 (0)207 679 7495
Fax: +44 (0)207 383 2572

9

Dear Mark,

El vie., 3 ago. 2018 a las 19:05, Mark Lake (<mark.lake@ucl.ac.uk>) escribió:

Frank,

Some years ago I wrote a module (also called r.horizon) which I
think does what you want. I never got around to formally
contributing it (must compare with the existing r.horizon and do
so some time).

yes! please do :slight_smile:
Cheers,
Vero

10

On Mon, Jul 30, 2018 at 4:55 PM, Frank David <frank.david@geophom.fr> wrote:

Hello Markus,

is it possible to add an option to evaluate the negatives hight in a future release ? (I don’t know C)

Please try trunk r73049. Negative angles are now allowed by default.

Markus M

Best regards,
Frank
Le 30/07/2018 à 16:19, Markus Metz a écrit :

On Sat, Jul 28, 2018 at 3:59 PM, Frank David <frank.david@geophom.fr> wrote:

Dear all,

It seems that r.horizon in does not return negative heights. How it’s possible in case of the given point from the calculation is done is located on mountain and the horizon around is below zero ??. I miss something may be…

r.horizon is indeed ignoring points that are lower than the given point, see
https://trac.osgeo.org/grass/browser/grass/trunk/raster/r.horizon/main.c#L1037
and following lines

I agree that this behaviour is questionable

Markus M

Thank you for your explanations

Regards

Frank

grass-user mailing list
grass-user@lists.osgeo.org
https://lists.osgeo.org/mailman/listinfo/grass-user