I am working with an inundation model that produces, as output, a raster representing the extent of flooding (and its depth) in a study site. This is my 'modelled extent'. Then I have a polygon layer representing an observed extent, let's say a 'real extent'.
Now I want to compare the two and see where they overlay and where they don't. In a second part of my work I will then need to produce different 'modelled extents' and compare them and see which is the difference between their areas and where they overlap and where they don't.
What I have done so far for comparing the modelled extent and the real one (raster vs. polygon) is transforming the 'modelled extent' in a raster where inundated pixels have a value of 1 (using the Con tool in Raster Calculator), then I created for it an attribute table and used the 'Count' field to compute its area. Nevertheless, with this method I can't quantify the overlapping areas of the two layers, but only compare their area extent.
1) Is my approach reasonable? How could I estimate the overlap between the two?
2) Should I use two different approaches when I compare raster vs polygon and raster vs raster?
3) I had been suggested to digitize my 'modelled extent' but it seems to me very time consuming, especially if I need to do it several times.
I hope I was clear, thank you so much for your help,