1

I have nine raster layers (.tif) and each needs to have the same extent, resolution and CRS in order to work in Maxent. I have tried converting each layer to the same CRS and translating them to .asc format in QGIS. After that I tried to resample the layers in R to match one of the layers, but this resulted in errors, such as that the extents do not overlap. My question is how do I match all these layers in order to proceed with Maxent and also to use the 'stack' function in R?

Here is the zip-file with the rasters: https://drive.google.com/file/d/1lle95SPdQ7FyQSbFoFvmAzyuO2HUt7-L/view?usp=sharing

Ben De Coensel
  • 29
  • 5
  • Do you know what the original crs is for each layer? – bischrob Feb 23 '22 at 12:03
  • The CRS for the wc2.1 rasters is: EPSG:4326 - WGS 84 - Geographic. The CRS for the SBPC and SPPC rasters is: IGNF:ETRS89LAEA - ETRS89 Lambert Azimutal Equal Area - Projected. The CRS for the CLC layer is: EPSG:3035 - ETRS89-extended / LAEA Europe - Projected and the CRS for the bedrock is an invalid projection. – Ben De Coensel Feb 23 '22 at 12:20
  • Thanks, there are a few problems to deal with, but I'm taking a look at solving them. – bischrob Feb 23 '22 at 15:01

1 Answers1

0

So the initial problem is to set the crs using the 'crs' function from the raster package (I haven't used the new terra package yet). Then you need to reproject into the same crs. The next step is to resample the rasters so they all have the same cell resolution and size. Last you can put them in a stack. I was in a rush, so I didn't comment very well, but let me know if you have questions. The last point is the bedrock file. You'll need to use QGIS or another program to georeference it first. Try to find a map with a known projection that looks similar to it.

library(raster)
ls = list.files(".",pattern ="tif")
ls = ls[-which(ls == "bedrock.tif")]
r  = lapply(ls,raster)
names(r) = ls
wgs84 = "+proj=longlat +datum=WGS84 +no_defs"
ETRS = "+proj=laea +lat_0=52 +lon_0=10 +x_0=4321000 +y_0=3210000 +ellps=GRS80 +units=m +no_defs"

crs(r$wc2.1_2.5m_bio_1.tif) = wgs84
crs(r$wc2.1_2.5m_bio_12.tif) = wgs84
crs(r$wc2.1_2.5m_elev.tif) = wgs84
crs(r$SBPC1.tif) = ETRS
crs(r$SBPC2.tif) = ETRS
crs(r$SPPC1.tif) = ETRS
crs(r$SPPC2.tif) = ETRS
crs(r$U2018_CLC2018_V2020_20u1.tif) = ETRS
# aggregate for faster processing -- you'll want to change this, but my machine couldn't process it
ra <- lapply(r,aggregate,fact=10, fun=max)
# not all need to be reprojected - this is me being lazy
rp = lapply(ra,projectRaster, crs = ETRS)

# resample rasters to match
sapply(rp,area)
rpr = lapply(rp,resample, y = rp$SBPC1.tif)
sapply(rpr,area)
rs = stack(rpr)
plot(rs)
bischrob
  • 544
  • 3
  • 10