Viewshed irregularities

923
4
04-12-2020 08:40 AM
Oliver_Burdekin
Occasional Contributor II

I'm trying to find carry out a viewshed analysis but am not able to get consistent results.

Having tried and failed with viewshed and viewshed2 I discovered interactive viewshed analysis in ArcGIS Pro 2.5.0. This appeared promising but having input an observer offset of 10m and using the Ordnance Survey's 50cm DSM as the ground layer I was surprised to see the viewshed change as I zoomed the camera out (using the "C" shortcut to change the cursor to explorer mode). The ground layer uses EPSG: 27700 for the coordinate system and EPSG: 5701 for the vertical coordinate system. I'd like to know why this is happening as it makes any output from this tool unusable. 

0 Kudos
4 Replies
DanPatterson_Retired
MVP Emeritus

lots of options here, any particular settings you are using?

Interactive viewshed basics—ArcGIS Pro | Documentation 

and at the very bottom there is this note...

Visual inaccuracies can appear as the viewshed's analysis range ratio increases.

0 Kudos
Oliver_Burdekin
Occasional Contributor II

Thanks for the quick response Dan. The settings I'm using are below:

It seems that the tool is working off the pyramids rendered rather than the actual data values.

0 Kudos
DanPatterson_Retired
MVP Emeritus

Other than the statement, that it takes "a raster approach", there is nothing I can find, other than the exploratory videos... sorry

Exploratory analysis how-to videos—ArcGIS Pro | Documentation 

0 Kudos
Oliver_Burdekin
Occasional Contributor II

Thanks for trying Dan. I'm struggling to think of a use case for a "sometimes" accurate analysis tool. Do you have any experience with the viewshed or viewshed 2 geoprocessing tools? I tried both prior to using the interactive one but had very mixed results.

0 Kudos