Viewshed Tool to Identify Visible Locations FROM a Set of Observer Features?

CassandraHiggins

I am looking for something similar to the Viewshed tool, that determines the raster surface locations visible from a set of observer features. It appears that the available Viewshed and other Visibility/Observer Point tools only identify areas that are visible to a set of observer features.

To give specifics, I have a point layer containing mountain peaks within an AOI. I want to determine the areas on the raster surface from which the mountain peaks are visible. I developed a work-around to create rudimentary observer point locations within the AOI and ran the viewshed tool from those point locations, but those viewsheds are not directly related to the mountain peaks. So I'm wondering if there is a tool that exists where I can run it with the mountain peaks as the input. Is this possible? 

Anonymous User

Hi Cassandra:

I want to determine the areas on the raster surface from which the mountain peaks are visible

The mountain peaks *are* your observers. 🙂

The raster cells that can be seen from the mountain peaks are also the ones that can see the mountain peaks (visibility is two way: If I can see you, you can see me). Of course, the non-peak cells can also see other stuff, but it doesn't sound like you're asking for that.

Here are some details on how to use Geodesic Viewshed for this.

Your input observers will be points or polylines representing peaks /ridges. You will need to create that input somehow.

Your 'analysis type' parameter will be 'frequency'.

The output will be a raster where the DATA cells can see at least one peak cell. The value in each cell will be the number of distinct input observer cells (peak cells) that it can see. Unfortunately, with this output, you won't know exactly which peaks a given DATA cell can see, only the number of peaks.

To get more specific information, you have two options:

1. you can run geodesic viewshed for each peak separately

2. if you have less than 32 peak cells, you can use an 'analysis type' of 'Observers' instead of 'Frequency',  and then use the output 'observer-region relationship table' to figure out which peaks are visible from each non-peak DATA cell.

Let me know if this helps.

Regards,

-Jim TenBrink

spatial analyst team