Subscribe to this thread
Home - General / All posts - visibility analysis
YorkSteve5 post(s)
#22-Sep-14 14:51

I'm trying to do a visibility analysis on a large surface. The basic question I'm trying to solve is: from where on the surface is a 60m high tower visible to another tower that is 15m above a fixed point. I can get an approximation by doing a visibility analysis from the second tower, fixing its height at 75m and plotting the area on the surface. However, when I subsequently check the result by generating a profile, I'm finding that the problem isn't always solved, ie the visibility of point A that is 60m above the ground from a point B that is 15m above the ground isn't the same as the visibility of point B at 75m. Point A can be anywhere on the surface but B is fixed. Any help would be appreciated.

KlausDE

6,410 post(s)
#22-Sep-14 17:19

My be this thread helps if distance is so far that curvature shouldn't be neglected.

I guess the simple aproach would be to add the higth of the tower's at point A to that pixel of the surface.


Do you really want to ruin economy only to save the planet?

YorkSteve5 post(s)
#23-Sep-14 10:50

Thanks for the response, Klaus. The proposal to increase the pixel height at point A, doesn't appear to solve things, however. Point B is fixed and at 15m above the surface. From this point, I'm trying to doing a visibility analysis to find where on the surface I can see a 60m high tower, ie Point A can be anywhere on the surface. I'll keep scratching my head!

tjhb
10,094 post(s)
#23-Sep-14 10:59

So swap points A and B?

Increase the pixel height at point B to 60m, and do a visibility analysis for the set of points A at height 15m.

YorkSteve5 post(s)
#23-Sep-14 11:43

I'm thinking that the visibility of a 60m high tower from a 15m high tower is the same both ways, irrespective of which is which, so swapping the points might not solve things. The "Height" selection in the visibility dialogue box refers to objects from where visibility is being calculated (my point B), not to what is being seen (my points A). If it was possible to use the height box to refer to the surface then that would be problem solved. I think I'm resigned to adding the heights of the two towers at B and taking the resultant visibility map as an approximation only. But thanks anyway.

KlausDE

6,410 post(s)
#23-Sep-14 22:41

Do you know that you can overlay a surface with a drawing in the terrain and that this drawing can extrude objects over the surface. So if you manage to locate the camera at point B you could look around and just see the towers over the horizon of nearby hills or not.


Do you really want to ruin economy only to save the planet?

tjhb
10,094 post(s)
#24-Sep-14 00:18

I think I confused myself over a misplaced "from" in your first post.

The basic question I'm trying to solve is: from where on the surface is a 60m high tower visible to [from] another tower that is 15m above a fixed point

Sorry about that. Anyway I think you are right:

I think I'm resigned to adding the heights of the two towers at B and taking the resultant visibility map as an approximation only.

The reason it's only an approximation, I think, is that raising the viewing height stretches the vector (ray) between the viewpoint and any given pixel of the surface in the vertical direction (only). This means that a ray between B and some A may just clear an intervening hill when B is 60 and A is 15, but fail to clear it (be obscured by the hill) when B is 75 and A is 0.

After the stretch, all hills will appear slightly higher than they really are, so visibility will be systematically underestimated.

[Edit: that's wrong, I think. I don't understand why the approximation is only an approximation after all (yet). For now skip a few pointless paragraphs to "Alternatively..."]

Can this be corrected for? I think so if, as well as stretching the vector between B and every pixel in the vertical direction, we also stretch it in the horizontal (i.e. radial) direction by the same proportion. This would preserve the angle of incidence for every ray, thus ensuring the same visibility pattern, pixel for pixel.

This would involve a custom resampling of the surface, spreading all pixels (except the centre) outwards from B by 75 / 60 * their current radius.

That's not exactly trivial, but doable. It would also be necessary to shift the pixels back again after drawing visible areas, though this could be done by a look-up table.

Alternatively though, how about using Global Mapper, which allows explicitly for tower height at both ends? If you don't have Global Mapper and want to see the difference from the approximation in Manifold, send me some data (privately if necessary) and I'll send you back the result.

tjhb
10,094 post(s)
#24-Sep-14 01:16

[Edit--sort of.]

The part crossed out above is not completely wrong. Here's a correction.

This means that a ray between B and some A may just clear an intervening hill when B is 60 and A is 15, but fail to clear it (be obscured by the hill) when B is 75 and A is 0. Or the reverse, i.e. fail to clear a hill in the first case but clear it in the second.

After the stretch, hills closer than halfway between B and a given target pixel will appear lower higher than they really are. Hills further than halfway between B and a target pixel will appear higher than they are. (Hills exactly halfway will appear their correct height, proportionately.)

So the visibility of nearby pixels will be overestimated, while that of relatively far pixels will be underestimated.

I'll draw a picture.

The paragraphs from "Can this be corrected for?" onward are correct, I think.

tjhb
10,094 post(s)
#24-Sep-14 04:35

I just noticed there's a stray "higher" in the second corrected paragraph above. It should read not "lower higher" of course but just "lower".

tjhb
10,094 post(s)
#24-Sep-14 01:39

Picture.

Attachments:
Example.png

YorkSteve5 post(s)
#24-Sep-14 10:35

Link to Map file..

https://onedrive.live.com/redir?resid=E124ACDD786CA19B!41657&authkey=!ADOKvjuHZins9QQ&ithint=file%2cmap

Presumably, though, when Manifold does a visibility scan, it does so on an true surface with true heights for hills (and without earth curvature it seems!)? In any event, here is the Map I've used. Just to put the problem into context, I'm helping to identify sites across Yorkshire, UK, where community wind turbines 62m high can be placed without interfering with several radar stations 15m high, across the region. This would make it easier to assess potential sites and it cuts out a lot of heart-ache further down the line. I'm using the resultant areas on a website I've created so that landowners can see clearly if they would have a problem or not before any funds are spent preparing planning applications. You can see the (very) rough website here, which works best in Firefox:

http://www.originenergy.org/radar.html

The Map I've enclosed above shows one site, Temple Farm, that is outside a visibility splay from DTA (an airport radar) yet if you look at the profile I've generated, you will see that there is a problem and that the turbine would be visible. The splay is from DTA at a height of 62+15. You need to imagine the two towers, the 15m high radar is at the left side and the 62m turbine is at the right. The same problem does not occur on other sites, which is leading me to believe that the theory of adding the two tower heights together is flawed. You need to imagine the two towers, the 15m high radar is at the left side and the 62m turbine is at the right.

tjhb
10,094 post(s)
#24-Sep-14 10:38

Thanks!

By the way I like your way of working, how you describe the process and where the analysis fits in.

Graeme

990 post(s)
#25-Sep-14 07:52

If I understand your objective correctly, you need to add the radar height (15) to the radar site. Next transform the surface so that the height of each pixel on which a proposed tower may be sited is increased by the height of the tower (62). You can do this by selecting relevant tower points, transfer selection to the surface then in the surface transform dialogue, "scope - selected pixels / formula [Nz80 Image] + 62" good idea to check save as new component.

Running visible areas from your nominated radar site on the modified surface does indeed show your Temple Farm site as visible, but not on the original surface - see screenshots.

Profile also seems reasonable - your "overlay line for the profile wasn't snapped to either point, though I doubt this made any difference.

It will be good to have earth curvature implemented in a future edition, but it is not significant for distances under 30 km or so. Welcome to the Forum Steve!

Attachments:
Profile result from modified surface.PNG
Vis areas on original and modified surface zoom.PNG
Vis areas on original and modified surface.PNG

YorkSteve5 post(s)
#25-Sep-14 09:08

Thanks for such considerable help, Graeme. I understand what you've done and it certainly helps with assessing the site.

KlausDE

6,410 post(s)
#30-Sep-14 06:01

For radar and a safe assumption add radius of the radar antenna and the fans to each hight.


Do you really want to ruin economy only to save the planet?

Manifold User Community Use Agreement Copyright (C) 2007-2021 Manifold Software Limited. All rights reserved.