Apply a climatekit Index Function Over a SpatRaster

Compute a daily climatekit index function for every cell in a SpatRaster x whose layers represent successive days, and return the per-period results as a SpatRaster. The supplied function fun must accept a numeric vector and a Date vector and return a data frame with period and value columns (the standard climatekit shape).

Usage

ck_apply_grid(x, fun, dates, ...)

Arguments

x

A SpatRaster. Layers correspond one-to-one with dates.

fun

A ck_* function (or any function with the same signature: numeric vector + Date vector + optional named arguments, returning a data frame with period and value).

dates

Date vector of length terra::nlyr(x).

...

Additional named arguments forwarded to fun (for example period = "annual", ref_start, ref_end).

Value

A SpatRaster with one layer per output period (layer names are the period labels).

Details

All cells must share the same dates. Cells that are entirely NA are returned as NA. Run-time scales linearly with the number of cells; for very large grids consider sub-setting first.

Examples

# \donttest{
  if (requireNamespace("terra", quietly = TRUE)) {
    dates <- seq(as.Date("2024-01-01"), as.Date("2024-12-31"), by = "day")
    n <- length(dates)
    # Tiny 2x2 SpatRaster of synthetic daily Tmax
    r <- terra::rast(nrows = 2, ncols = 2, nlyrs = n,
                     xmin = 0, xmax = 2, ymin = 0, ymax = 2)
    set.seed(1)
    for (i in seq_len(n)) {
      terra::values(r[[i]]) <- rnorm(4, 15, 5)
    }
    txx_r <- ck_apply_grid(r, ck_txx, dates = dates, period = "annual")
    terra::nlyr(txx_r)
  }
#> [1] 1
# }