I'm trying to put together a layer representing horizontal visibility at ground level. I'm thinking of using the EVC (vegetation cover) layer from the landfire project (see http://www.landfire.gov/vegetation.php). However, I know that, for instance, 70% cover does not correspond directly to 30% horizontal visibility...I imagine there is a threshold at which the cover saturates any visibility, with visibility increasing in a non-linear way as percent cover decreases - probably calculated over some neighborhood (100 meters, maybe?). I'm thinking of cooking something up myself along these lines.

However, I'm wondering if there hasn't been work done in this realm, or some theory to exploit that anyone may be aware of. A lit/google search hasn't turned up much of interest.

  • 1
    A good model for visibility is an exponential decrease with distance. This follows from the assumption that any thin vertical layer of vegetation (or smoke or whatever) of thickness dx transmits an amount 1 - k*dx of the light: that is, the amount of light blocked is directly proportional to the thickness for small thicknesses. Integrating this over distance produces the exponential relationship. Approximations to this have been created for 3D computer games and visualizations (to create smoke, haze, and atmospheric effects). I am not aware of any GIS implementations. – whuber May 17 '12 at 15:46
  • 1
    Do a search for Prof. Ian Bishop. He's based somewhere in Australia and has done a lot of reasearch into landscape visibility. He might have published something that could help. Sorry I don't have a specific paper to cite. – MappaGnosis May 18 '12 at 12:35

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.