[GRASS-user] Calculate average of 3 highest values for a set of raster maps

[Paulo van Breugel]

Hi Miltinho

I assume you want the mean of the three highest values for each raster 
cell? If so, you can do this using another step with r.series

1) If your 7 maps are a1 .. a7:

mapcalc "a1 = 1"; r.mapcalc "a2 = 2"; r.mapcalc "a3 = 3"; r.mapcalc "a4 
= 4"; r.mapcalc "a5 = 5"; r.mapcalc "a6 = 6"; r.mapcalc "a7 = 7"

2) calculate the 4/7 quantile. Actually, it should be a little bit 
smaller to make sure you include your 3rd largest layer. In this 
example below, 0.57 as approximation is fine

r.series --overwrite input=a1,a2,a3,a4,a5,a6,a7 quantile=0.57 
output=res1 method=quantile

2) Next, for all 7 layers, set all values < 4/7 quantile to 0

r.mapcalc "b1 = if(a1>res1,a1,null())"
r.mapcalc "b2 = if(a2>res1,a2,null())"
r.mapcalc "b3 = if(a3>res1,a3,null())"
r.mapcalc "b4 = if(a4>res1,a4,null())"
r.mapcalc "b5 = if(a5>res1,a5,null())"
r.mapcalc "b6 = if(a6>res1,a6,null())"
r.mapcalc "b7 = if(a7>res1,a7,null())"

3) Now, calculate the sum of the 7 new layers

r.series input=b1,b2,b3,b4,b5,b6,b7 method=average output=res2  
--overwrite

Your output res2 should have a value 6 for all raster cells.

Cheers,

Paulo






On Fri 15 Mar 2013 03:57:35 AM CET, Milton Cezar Ribeiro wrote:
> I have a set of 7 raster maps.
> I am able to calculate several stats with these mapas using r.series.
> Now I need to calculate the mean of the tree highest values of my 7 maps.
> Any idea of how can I do that?
>
> cheers