[GRASS-user] Assess land cover change

[Moritz Lennert]

On 17/10/12 16:28, Luis Lisboa wrote:
> Thank you Moritz
> Just one last question: is it possible to produce a raster product that
> attributes a code/color to a specific land cover change?

One way I can think of immediately would be to code the land covers of 
the two dates in different order of magnitudes and then sum the two, i.e.

year t:
100 forest
200 built-up
300 agriculture

year t+n

1 forst
2 built-up
3 agriculture

year t + yeart+n (you can do that with r.mapcalc):

101 : forest that remains forest
102 : forest that has become built-up land
301 : agricultural land that has become forest

etc

But, as always, there are probably many other ways :-).

Moritz