treeslabgrid

The goal of treeslabgrid is to help users to aggregate spatial data using regular grids. To achieve this, treeslabgrid extends the sf and terra packages, providing additional functions to create grids and aggregate data using many aggregation functions simultaneously.

Installation

You can install the development version of treeslabgrid from GitHub with:

# install.packages("pak")
pak::pak("albhasan/treeslabgrid")

Example

This is a basic example which shows you how to create a grid and use it to aggregate a raster.

First, let’s load some packages:

library(treeslabgrid)
library(sf)
#> Linking to GEOS 3.13.1, GDAL 3.10.3, PROJ 9.6.0; sf_use_s2() is TRUE
library(terra)
#> terra 1.8.93

Now, let’s get a raster from the examples in the terra package:

# Get a raster.
r <- rast(system.file("ex/meuse.tif", package = "terra"))

treeslabgrid offers many ways to create a grid that covers spatial data. In our example, we can center our grid in the middle of the raster’s extent:

# Create a grid over that covers the raster.
g <- make_grid_origin_res(
  xy_origin = get_center(r),
  xy_min = get_min(r),
  xy_max = get_max(r),
  cell_size = 500,
  crs = terra::crs(r),
  id_col = "grid_id"
)

We can plot what we just did:

plot(r, reset = FALSE)
plot(st_cast(g[get_geom_colname(g)], "LINESTRING"), add = TRUE)

The grid seems OK; so, we can now aggregate the raster data using it:

meuse_agg <- aggregate_raster(
  x = r,
  by = NA_character_,
  grid = g,
  grid_id = "grid_id",
  funs = c("sum", "min", "mean", "max"),
  na.rm = TRUE
)

Note we use many aggregation functions at the same time, so, the results present the results for each one:

plot(meuse_agg["sum.meuse.1"])

plot(meuse_agg["min.meuse.1"])

plot(meuse_agg["mean.meuse.1"])

plot(meuse_agg["max.meuse.1"])