[GRASS-user] Comparing 2 large Corine Land Cover maps

[Hermann Peifer]

Thanks to Paulo and S. for pointing to r.cross. As suggested, I will go for a sequence of r.cross > r.category > awk > r.reclass

r.cross is pretty fast and generates the output map in 16 minutes (r.mapcalc wasn't done after 16 hours..). 

Hermann

-------- Original-Nachricht --------
> Datum: Sun, 24 Jul 2011 00:15:29 +0200
> Von: Paulo van Breugel <p.vanbreugel at gmail.com>
> An: Hermann Peifer <peifer at gmx.eu>
> CC: grass-user <grass-user at lists.osgeo.org>
> Betreff: Re: [GRASS-user] Comparing 2 large Corine Land Cover maps

> Maybe r.cross can be of help. It creates a cross product of the category
> variables of a layer, i.e., a map with separate categories for each unique
> combination of the two CLC maps (the category labels gives you the
> combination of input maps for each new category).
> 
> You can use this new map to create a reclass map with the new values from
> your decision matrix. To do this, you could for example follow the
> following
> steps:
> 
> 1) Run r.category to get a list with the categories and category labels of
> the cross product map you just created. You can output this as a text file
> 2) Use a spreadsheet or something like awk to create a textfile with three
> columns, one with the new category, and two with the clc2000 and clc2006
> categories
> 3) Find a way to link the CLC combinations in your text file to the new
> classes given in your decision matrix (use e.g., spreadsheet or database)
> 4) Use this new table to create the reclassification rules you need in the
> next step (again, if you know awk this will probably be very easy, but
> also
> in R or in a spreadsheet this is easy to do).
> 5) Create a text file with reclassification rules (see r.reclass how these
> should look like) and use r.reclass to reclassify the map you just created
> with r.cross.
> 
> I guess there must be easier ways. But in any case, these steps are not
> very
> difficult to implement. And in case you have to repeat this type of
> analysis
> more often, once you have set up a workflow, it will be easy to repeat
> (especially if you manage to implement all steps as a script).
> 
> 
> 
> On Fri, Jul 22, 2011 at 12:41 PM, Hermann Peifer <peifer at gmx.eu> wrote:
> 
> > Hi,
> >
> > I have 2 Corine Land Cover (CLC) raster maps, from 2000 and 2006. Both
> maps
> > have the same extent, the same resolution (100m) and use the same 44 CLC
> > categories. The maps are pretty large: 67 000 cols x 58 000 rows = 3 886
> 000
> > 000 cells.
> >
> > I'd like to categorise the changes between the 2 layers, according to a
> > decision matrix which I pasted below [1]. Row headers indicate the
> clc2000
> > category, column headers represent the clc2006 category. I only pasted
> the
> > "left half" of the table in order to avoid ugly line breaks.
> >
> >
> > Question:
> > Is there a better (and faster) way than doing this comparison by using
> > r.mapcalc with a (nearly) endless if/else chain [2] ?
> >
> > I also noted that in 99+ % of the cases, both maps are NULL or have an
> > identical CLC category, so I wouldn't need the if/else chain at all. I
> guess
> > I can use this fact to shortcut the comparison?
> >
> >
> > Thanks in advance, Hermann
> > (using GRASS 6.4 on 64-bit Debian Linux)
> >
> > [1]
> >
> > ##  1  2  3  4  5  6  7  8  9 10 11 12 13 14 15 16 17 18 19 20 21 22 ..
> >  1 NA 12 31 32 33 34 35 36 37 13 38 56 56 56 56 56 56 56 56 56 56 56 ..
> >  2 11 NA 31 32 33 34 35 36 37 13 38 56 56 56 56 56 56 56 56 56 56 56 ..
> >  3 12 12 NA 32 33 34 35 36 37 13 38 56 56 56 56 56 56 56 56 56 56 56 ..
> >  4 12 12 31 NA 33 34 35 36 37 13 38 56 56 56 56 56 56 56 56 56 56 56 ..
> >  5 12 12 31 32 NA 34 35 36 37 13 38 56 56 56 56 56 56 56 56 56 56 56 ..
> >  6 12 12 31 32 33 NA 35 36 37 13 38 56 56 56 56 56 56 56 56 56 56 56 ..
> >  7 12 12 31 32 33 34 NA 36 37 13 38 56 56 56 56 56 56 56 56 56 56 56 ..
> >  8 12 12 31 32 33 34 35 NA 37 13 38 56 56 56 56 56 56 56 56 56 56 56 ..
> >  9 12 12 31 32 33 34 35 36 NA 13 38 56 56 56 56 56 56 56 56 56 56 56 ..
> > 10 11 11 31 32 33 34 35 36 37 NA 38 56 56 56 56 56 56 56 56 56 56 56 ..
> > 11 11 11 31 32 33 34 35 36 37 13 NA 56 56 56 56 56 56 56 56 56 56 56 ..
> > 12 21 22 31 32 33 34 35 36 37 13 38 NA 45 43 42 42 42 41 42 41 83 61 ..
> > 13 21 22 31 32 33 34 35 36 37 13 38 43 NA 43 42 42 42 41 42 41 83 61 ..
> > 14 21 22 31 32 33 34 35 36 37 13 38 43 43 NA 42 42 42 41 42 41 83 61 ..
> > 15 21 22 31 32 33 34 35 36 37 13 38 45 45 45 NA 44 44 41 45 45 83 61 ..
> > 16 21 22 31 32 33 34 35 36 37 13 38 45 45 45 44 NA 44 41 45 45 83 61 ..
> > 17 21 22 31 32 33 34 35 36 37 13 38 45 45 45 44 44 NA 41 45 45 83 61 ..
> > 18 21 22 31 32 33 34 35 36 37 13 38 45 45 45 42 42 42 NA 45 45 83 61 ..
> > 19 21 22 31 32 33 34 35 36 37 13 38 45 45 45 42 42 42 41 NA 41 83 61 ..
> > 20 21 22 31 32 33 34 35 36 37 13 38 45 45 45 42 42 42 41 45 NA 83 61 ..
> > 21 21 22 31 32 33 34 35 36 37 13 38 52 52 52 52 52 52 41 52 54 NA 54 ..
> > 22 21 22 31 32 33 34 35 36 37 13 38 45 45 45 42 42 42 41 45 45 83 NA ..
> > 23 21 22 31 32 33 34 35 36 37 13 38 51 51 51 51 51 51 53 51 53 53 53 ..
> > 24 21 22 31 32 33 34 35 36 37 13 38 51 51 51 51 51 51 53 51 53 53 53 ..
> > 25 21 22 31 32 33 34 35 36 37 13 38 51 51 51 51 51 51 53 51 53 53 53 ..
> > 26 21 22 31 32 33 34 35 36 37 13 38 52 52 52 52 52 52 54 52 54 54 54 ..
> > 27 21 22 31 32 33 34 35 36 37 13 38 52 52 52 52 52 52 54 52 54 54 54 ..
> > 28 21 22 31 32 33 34 35 36 37 13 38 52 52 52 52 52 52 54 52 54 54 54 ..
> > 29 21 22 31 32 33 34 35 36 37 13 38 52 52 52 52 52 52 54 52 54 54 54 ..
> > 30 21 22 31 32 33 34 35 36 37 13 38 52 52 52 52 52 52 54 52 54 54 53 ..
> > 31 21 22 31 32 33 34 35 36 37 13 38 52 52 52 52 52 52 54 52 54 54 53 ..
> > 32 21 22 31 32 33 34 35 36 37 13 38 52 52 52 52 52 52 54 52 54 54 53 ..
> > 33 21 22 31 32 33 34 35 36 37 13 38 52 52 52 52 52 52 54 52 54 54 53 ..
> > 34 NA NA NA NA NA NA NA NA NA NA 38 NA NA NA NA NA NA NA NA NA NA NA ..
> > 35 21 22 31 32 33 34 35 36 37 13 38 55 55 55 55 55 55 55 55 55 55 55 ..
> > 36 21 22 31 32 33 34 35 36 37 13 38 55 55 55 55 55 55 55 55 55 55 55 ..
> > 37 21 22 31 32 33 34 35 36 37 13 38 55 55 55 55 55 55 55 55 55 55 55 ..
> > 38 21 22 31 32 33 34 35 36 37 13 38 55 55 55 55 55 55 55 55 55 55 55 ..
> > 39 21 22 31 32 33 34 35 36 37 13 38 55 55 55 55 55 55 55 55 55 55 55 ..
> > 40 21 22 31 32 33 34 35 36 37 13 38 55 55 55 55 55 55 55 55 55 55 55 ..
> > 41 21 22 31 32 33 34 35 36 37 13 38 55 55 55 55 55 55 55 55 55 55 55 ..
> > 42 21 22 31 32 33 34 35 36 37 13 38 55 55 55 55 55 55 55 55 55 55 55 ..
> > 43 21 22 31 32 33 34 35 36 37 13 38 55 55 55 55 55 55 55 55 55 55 55 ..
> > 44 21 22 31 32 33 34 35 36 37 13 38 55 55 55 55 55 55 55 55 55 55 55 ..
> >
> >
> > [2]
> >
> > change = \
> > clc2000 == 1 && clc2006 == 2 ? 12 : \
> > clc2000 == 1 && clc2006 == 3 ? 31 : \
> > clc2000 == 1 && clc2006 == 4 ? 32 : \
> > clc2000 == 1 && clc2006 == 5 ? 33 : \
> > clc2000 == 1 && clc2006 == 6 ? 34 : \
> > clc2000 == 1 && clc2006 == 7 ? 35 : \
> > clc2000 == 1 && clc2006 == 8 ? 36 : \
> > clc2000 == 1 && clc2006 == 9 ? 37 : \
> > clc2000 == 1 && clc2006 == 10 ? 13 : \
> >
> > ... (snip away some 1700 if/else rules)
> >
> > clc2000 == 44 && clc2006 == 30 ? 89 : \
> > clc2000 == 44 && clc2006 == 31 ? 89 : \
> > clc2000 == 44 && clc2006 == 32 ? 89 : \
> > clc2000 == 44 && clc2006 == 35 ? 82 : \
> > clc2000 == 44 && clc2006 == 36 ? 82 : \
> > clc2000 == 44 && clc2006 == 37 ? 82 : \
> > clc2000 == 44 && clc2006 == 38 ? 89 : \
> > clc2000 == 44 && clc2006 == 39 ? 89 : \
> > clc2000 == 44 && clc2006 == 42 ? 82 : \
> > clc2000 == 44 && clc2006 == 43 ? 82 : 0
> > ______________________________**_________________
> > grass-user mailing list
> > grass-user at lists.osgeo.org
> >
> http://lists.osgeo.org/**mailman/listinfo/grass-user<http://lists.osgeo.org/mailman/listinfo/grass-user>
> >