Best Practice question about Raster Viewshed Analyses

Discussion created by jhanyoho on Mar 27, 2012
Latest reply on Mar 28, 2012 by JSwain-esristaff

I had a question about what should be the best approach when it comes to doing Viewshed Analyses.

Premise of project: Where are the towers visible? (Point Shapefile with hundreds of discrete points representing towers along a linear route). OFFSETA = tower height and OFFSETB = person height who will see the tower.

I was able to download elevation data with resolution of approx 10meters.  Instead of one large file for the extent, I was given 6 smaller chunks (due to size). 

My dilemma lies in the viewshed analyses.  The point shapefile lies on all 6 small rasters.  However, I noticed that, let's say tower 1 that lies on raster A, cannot be processed on raster B because it does not lie on raster B (or this is my conclusion because when I try to run a viewshed of a selection of points that do not lie on top of the raster, it fails). 

My solution was to automatically combine all the rasters using a mosaic dataset.  However, I understand that now the problem becomes processing capacity. 

Is there a way around this problem?  Do I just need to combine all 6 rasters to 1 to get an accurate viewshed of all the discrete points in the shapefile?  Is there a best practice approach to this issue?

Thank you!