[R-sig-Geo] Creating a maximum raster layer from RasterStack

[Agustin Lobo]

oops, copied the wrong line,
it must be
mx <- stackApply(s,indices=c(1,1),max)

Agus


On Mon, Oct 7, 2013 at 8:18 PM, Agustin Lobo <alobolistas at gmail.com> wrote:
> What about this:
> r1 <- raster(matrix(rnorm(100,5,10),10,10))
> r2 <- raster(matrix(rnorm(100,2,30),10,10))
> s <- stack(r1,r2)
> mx <- stackApply(s,max)
>
> which.max() fails because raster requires the na.rm argument (I think, Robert?)
>
> mxd <- stackApply(s,indices=c(1,1),which.max)
> Error in FUN(newX[, i], ...) : unused argument (na.rm = TRUE)
>
> so make your own function icluding the na.rm argument:
>
> miwhichmax <- function(x,na.rm=TRUE) which.max(x)
> mxd <- stackApply(s,indices=c(1,1),miwhichmax)
>
> Agus
>
>
> On Thu, Oct 3, 2013 at 6:55 PM, Eddie Smith <eddieatr at gmail.com> wrote:
>> Dear list,
>>
>> I have a RasterLayer that contains 365 bands with a dimension of 1032,
>> 1656, 1708992  (nrow, ncol, ncell).
>>
>> library(raster)
>> img <- list.files(pattern='\\.img$')
>> stack <- stack(img)
>> data2002 <- writeRaster(stack, filename="2002.tif", format="GTiff",
>> overwrite=TRUE)
>>
>> 1. How could I produce a raster layer that contains a maximum value from
>> 365 bands for each pixel?
>> I tried maxValue() but that only give me a maximum value for each band.
>>
>> 2. If I manage to do step 1, is it possible to know from which band is
>> actually the maximum value originate?
>>
>>         [[alternative HTML version deleted]]
>>
>> _______________________________________________
>> R-sig-Geo mailing list
>> R-sig-Geo at r-project.org
>> https://stat.ethz.ch/mailman/listinfo/r-sig-geo