3

I am trying to resample a raster to a finer resolution, matching the resolution of my raster of interest. The problem is, my raster I'm resampling is a population raster. When I resample, the total values of all of the cells increases in proportion to the increase in resolution. Is there a way I can resample my raster and keep the sum of all cells equal?

Example code:

r1 <- raster(ncol=10,nrow=10)
r2 <- raster(ncol=4,nrow=4)
values(r1) <- runif(ncell(r1))
values(r2) <- runif(ncell(r2))
r2_res <- resample(r2,r1,method='bilinear')
#sum(r2_res) should equal sum(r2)

cellStats(r2_res,sum,na.rm=T) - cellStats(r2,sum,na.rm=T)==0

[1] FALSE

1 Answer 1

1

You are assuming population is equally distributed in each cell, thus, we shall divide the original cell population by the number of finer resolution cells that fit in the original:

library(terra)
r <- rast(nrows=8, ncols=6, xmin=0, xmax=10, ymin=0, ymax=10) # finer
pop <- rast(nrows=4, ncols=3, xmin=0, xmax=10, ymin=0, ymax=10) # coarser; the population rast
values(pop) = 1:ncell(pop)

x = resample(pop, r, method = "near" ) # we pass raw values to finer raster
wr = x*(prod(res(x)/res(pop)))         # we divide the original cell by the resolution

sum(wr[]) == sum(s[])
# TRUE
res(x) == res(r)
# TRUE

plot(pop)
text(pop)
text(wr, col = "red", digits = 3)

enter image description here

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.