1

How can I get the the total area of my land cover classes 1,2,3,4,5 given the following example:

library("raster")
r <- raster(nrow=10, ncol=10)
r<-setValues(r,c(rep(1,20),rep(2,20),rep(3,20),rep(4,20),rep(5,20)))

One approach might be to subset the raster based on these values, however somethig like a <- area(r[getValues(r)==1]) did not work neither did a <- area(r[r==1]).

There is a solution provided on this website, however it requires creating new raster layers for each value to be analyzed. I would prefer rather not to do so since my original raster contains lots of different values and is very large. A similar approach is presented here, however it works only for small regions.

Community
  • 1
  • 1
joaoal
  • 1,892
  • 4
  • 19
  • 29

1 Answers1

2

You can use the base function aggregate on the areas grouped by the values in r and sum them up. 

aggregate(getValues(area(r, weights=FALSE)), by=list(getValues(r)), sum)
  Group.1         x
1       1  48166136
2       2 126933351
3       3 320336528
G5W
  • 36,531
  • 10
  • 47
  • 80
  • 1
    Does this method take into account the differences in latitudinal resolution if large areas are mapped? – joaoal Apr 26 '17 at 13:32
  • I believe so. Your example code has somewhat odd coordinates as lat-lons, but it did produce different areas for different grid boxes. – G5W Apr 26 '17 at 13:59
  • yes it did. I edited my example code slightly to be more comprehensive for that purpose. Would be helpfull if you change the results in your response. thx. – joaoal Apr 26 '17 at 14:19