how to spilt one classification Map into different tiles，and then calcaulate all tiles's landscape values
Closed this issue · 4 comments
Hey,
One approach would be to split the raster before calculating any metrics. There are several packages that allow to do so (e.g. SpaDES).
Also, it would be possible to use the sample_lsm() function if sample plots are basically a grid on top of the RasterLayer. Here is one way to do that:
library(landscapemetrics)
library(raster)
#> Loading required package: sp
library(sp)
# create grid with 9 cells which are 10x10
poly_grid <- sp::GridTopology(cellcentre.offset = c(5, 5),
cellsize = c(10, 10),
cells.dim = c(3, 3))
# convert to SpatialPolygons
poly_grid <- as(poly_grid, "SpatialPolygons")
# plot result
plot(landscape)
plot(poly_grid, add = TRUE)
# "sample" metrics in each grid cell
result <- sample_lsm(landscape = landscape, y = poly_grid,
level = "landscape", verbose = FALSE)
# results
result
#> # A tibble: 585 x 8
#> layer level class id metric value plot_id percentage_inside
#> <int> <chr> <int> <int> <chr> <dbl> <int> <dbl>
#> 1 1 landscape NA NA ai 87.4 1 100
#> 2 1 landscape NA NA area_cv 110. 1 100
#> 3 1 landscape NA NA area_mn 0.002 1 100
#> 4 1 landscape NA NA area_sd 0.00219 1 100
#> 5 1 landscape NA NA cai_cv 140. 1 100
#> 6 1 landscape NA NA cai_mn 14.9 1 100
#> 7 1 landscape NA NA cai_sd 20.8 1 100
#> 8 1 landscape NA NA circle_cv 56.2 1 100
#> 9 1 landscape NA NA circle_mn 0.467 1 100
#> 10 1 landscape NA NA circle_sd 0.262 1 100
#> # … with 575 more rowsCreated on 2019-12-12 by the reprex package (v0.3.0)
Hey,
One approach would be to split the raster before calculating any metrics. There are several packages that allow to do so (e.g.
SpaDES).Also, it would be possible to use the
sample_lsm()function if sample plots are basically a grid on top of theRasterLayer. Here is one way to do that:library(landscapemetrics) library(raster) #> Loading required package: sp library(sp) # create grid with 9 cells which are 10x10 poly_grid <- sp::GridTopology(cellcentre.offset = c(5, 5), cellsize = c(10, 10), cells.dim = c(3, 3)) # convert to SpatialPolygons poly_grid <- as(poly_grid, "SpatialPolygons") # plot result plot(landscape) plot(poly_grid, add = TRUE)
# "sample" metrics in each grid cell result <- sample_lsm(landscape = landscape, y = poly_grid, level = "landscape", verbose = FALSE) # results result #> # A tibble: 585 x 8 #> layer level class id metric value plot_id percentage_inside #> <int> <chr> <int> <int> <chr> <dbl> <int> <dbl> #> 1 1 landscape NA NA ai 87.4 1 100 #> 2 1 landscape NA NA area_cv 110. 1 100 #> 3 1 landscape NA NA area_mn 0.002 1 100 #> 4 1 landscape NA NA area_sd 0.00219 1 100 #> 5 1 landscape NA NA cai_cv 140. 1 100 #> 6 1 landscape NA NA cai_mn 14.9 1 100 #> 7 1 landscape NA NA cai_sd 20.8 1 100 #> 8 1 landscape NA NA circle_cv 56.2 1 100 #> 9 1 landscape NA NA circle_mn 0.467 1 100 #> 10 1 landscape NA NA circle_sd 0.262 1 100 #> # … with 575 more rowsCreated on 2019-12-12 by the reprex package (v0.3.0)
Thanks,but I still met some problem about how to change the values of GridTopology().
Here are one example,
create grid with 9 cells which are 10x10
poly_grid <- sp::GridTopology(cellsize = c(5,5),
cellcentre.offset = c(5,5),
cells.dim = c(6,6))
I hope to split the raster to 6*6 grids,and can every tiles export with the coordinates?
You need to adjust the cellcentre.offset argument to be 1/2 of the cellsize argument.
If you set return_raster = TRUE in the sample_lsm() function, all tiles will be returned as additional column.
Should be solved. Please feel free to re-open if nessesaire.