Due to SPAM attacks, new members must be approved before posting.  Please email when registering and your account will be approved.

Main Menu

identifying areas with significant changes

Started by chayek, April 06, 2010, 09:21:02 AM

Previous topic - Next topic


I'm modeling a relatively large area and I was wondering if there is an automated way to identify areas with significant changes or if I should rely soley on visual inspection.


Jonathan S. Clough

Well, you can produce as many rectangular "output sites" as you'd like and the model will produce tables of predicted changes for each "output site."  See the Users Manual for more information as to how to do this.

Other than that, there is no automated manner of identifying geographical areas where the most changes occur.  I would recommend the following "tricks" for visual inspection.

1. Load output files into GIS software and move back and forth between layers so that you can easily see the flickering portions of the map where predicted changes have occurred.
2. Produce identically sized images in Microsoft Word and view results in "full-screen Reading" mode.  Page up and page down through the maps and you can get an animation of predicted changes over time.
3. If your site is too large to inspect visually in one image, produce "output-sites" and look for changes in those MSWord maps.

I'm not sure what automated procedure to look for changes would be appropriate.  Breaking the map up into grids and then identifying the percentage change by grid cell?  (That sounds time-consuming to implement and of limited use compared to simple visual inspection.) 

Good luck.


I figured out a way to do it and thought I would share with others:

If you load the output back into GIS, you can use the "raster math" toolset to subtract the output from the input, giving you a rster in which all non-zero values represent areas that experience change.  Those areas can then be inspected in greater detail.