Converting species occurences into Raster stack

Question

I have a dataframe with the occurrence of different species. I want to create a raster for different species in the species column but then I want to stack the raster_list into a raster stack but I get error message that rasters have different extent

library(terra)

Occurrence_data <- data.frame(
    lon = c(11.002470, 10.733250, 11.135831, 6.003845, 5.073000, 8.859500, 10.740000),
    lat = c(59.05563, 63.57087, 60.15113, 62.40066, 60.11600, 58.62880, 59.95000),
    species = c("B.terrestris", "B.pascuorum", "B.hortorum", "B.pratorum", "B.terrestris", "B.pascuorum", "B.hortorum")
)

###### Get unique species from the 'species' column
species_list <- as.factor(unique(Occurrence_data$species))


create_species_dataframes <- function(DF) {
  # Get unique species from the 'species' column
  species_list <- unique(DF$species)

  # Create an empty list to store data frames
  species_dataframes <- list()

  # Loop through each unique species
  for (species in species_list) {
    # Filter the DataFrame for the current species
    species_df <- DF[DF$species == species, ]

    # Store the filtered DataFrame in the list
    species_dataframes[[species]] <- species_df
  }

  # Returning the list of data frames
  return(species_dataframes)
}
#call function to create individual dataframes for each species 
Models<-create_species_dataframes(Occurrence_data)

# Function to convert data frame to raster
convert_to_raster <- function(df) {
  # Convert the data frame to a raster
  raster_data <- rast(df, type = "xy", crs = "EPSG:4326")

  return(raster_data)
}

# Convert each data frame in the Models list to a raster
raster_list <- lapply(Models, convert_to_raster)
# Then, we combine all these rasters into stack. 
for(x in raster_list ){
   stackrasters<-stack(stackrasters,raster(x))
 }

Error in compareRaster(x): different extent

Answer 1

The process can be simplified. And to create raster(s) with species occurence, you should rasterize the coordinates.

Occurrence_data <- data.frame(
  lon = c(11.002470, 10.733250, 11.135831, 6.003845, 5.073000, 8.859500, 10.740000),
  lat = c(59.05563, 63.57087, 60.15113, 62.40066, 60.11600, 58.62880, 59.95000),
  species = c("B.terrestris", "B.pascuorum", "B.hortorum", "B.pratorum", "B.terrestris", "B.pascuorum", "B.hortorum")
) |>
  dplyr::arrange(species)

species_list <- unique(Occurrence_data$species)

Now, let's create spatial vector of the species and simple raster with extent covering all species. Note: you can manage the resolution by changing nrows (number of rows) and ncols (number of columns in raster). Or, you can provide resolution parameter to terra::rast() function. Extent in longitude divided by number of columns gives the resolution on longitudinal axis, and the same for latitude: "vertical" extent / nrows = resolution. Instead of number of rows/cols you can provide resolution parameter => nrows and ncols will be created appropriately. For details please check help to ?terra::rast()

v <- terra::vect(Occurrence_data)
r <- terra::rast(nrows = 12, ncols = 12, crs = "EPSG:4326", ext = terra::ext(v))

And let's rasterize the points:

x <- terra::rasterize(v, r, fun = "count", by = "species")

names(x) <- species_list

x
#> class       : SpatRaster 
#> dimensions  : 12, 12, 4  (nrow, ncol, nlyr)
#> resolution  : 0.5052359, 0.4118392  (x, y)
#> extent      : 5.073, 11.13583, 58.6288, 63.57087  (xmin, xmax, ymin, ymax)
#> coord. ref. : lon/lat WGS 84 (EPSG:4326) 
#> source(s)   : memory
#> names       : B.hortorum, B.pascuorum, B.pratorum, B.terrestris 
#> min values  :          2,           1,          1,            1 
#> max values  :          2,           1,          1,            1

terra::plot(x)

^{Created on 2024-02-15 with reprex v2.1.0}