Visibility/Viewshed Analysis Question

1336
5
09-15-2014 03:28 PM
KyleGallagher
New Contributor III

I am looking at a visibility analysis for a proposed facility layout to be built into fairly steep terrain.  I have taken the outlines of my facilities and converted them to 3d points representing the roof lines, and these will be used as the observer locations.  It is my understanding that running the Visibility analysis with these 3d points and using the Frequency method will help answer whether "I can see you, and you can see me" is true.  This gives me Output1.

In addition, I have 16 Key Observation Points (KOPs) and need to figure out what is visible from those locations.  If I run the Visibility analysis on these 16 points using Frequency method then I will get what is visible 360 degrees around that location.  This gives me Output2.  My logic here is that if I combine the results of Output1 and Output2, ensuring to account for all areas that are visible and limiting the view angle of my KOPs towards the proposed facilities, then I will have an aggregate viewshed.  Does this sound right?

Thanks for any help!

0 Kudos
5 Replies
DanPatterson_Retired
MVP Emeritus

hmmm got any images?

0 Kudos
KyleGallagher
New Contributor III

Can't provide any visuals, sorry.

0 Kudos
KyleGallagher
New Contributor III

Ok, here are some images illustrating my question.  One shows the viewshed from 39 different locations (Figure 3), while the other viewshed is based on a project site.  As you can see there are two different outputs and some locations are green on one viewshed and red on the other.  My question is shouldn't we account for all areas in green on both viewsheds to make one aggregate viewshed? That way it not only accounts for what you can see from the project site but also whether the project site can be seen from multiple locations farther awayvisib_KOPs.jpg?

0 Kudos
KyleGallagher
New Contributor III

visib_prjSite.jpg

0 Kudos
JustinBousquin
New Contributor

If I understand your question correctly, the short answer is yes, viewshed rasters can be added (SA-Math-Plus) to get an aggregated visibility for the multiple observers. This effectively just gives you an intensity of visibility like you already have in Figure 4. Combing output1 and output2 gives you what can be viewed from your project points and points of interest. An example of a case where you would want to combine the viewsheds would be if the project points were trying to optimize the total area that can be viewed along with the existing points of interest; like if the points were towers to see fires.

The question of if that is correct for your particular application is harder. If you just want to know if the project site is visible from a particular set of observation points (equivalent to if those sites are visible from the project sites). I would do this with a viewshed for the project points (this allows you to distance the observation points above the DEM as well) and then extracting the values (1=visible/ 0=not visible) by the points of interest. Alternatively, you can do this with a viewshed for the points of interest, where you are able to see how many points total can see each project point by extracting the values from the viewshed (value = number of points that can see the point you extract by). Adding the viewsheds will not tell you anything additional about the visibility of the project points.

0 Kudos