Apply calc_lake_morphometry to a named list of SpatRasters — batch_lake

Iterates calc_lake_morphometry() over a named list of bathymetric rasters and returns results as a single tidy data.frame with one row per lake.

Parallel execution is supported via the parallel package (base R). When parallel = TRUE the function creates a PSOCK cluster with n_cores workers, exports the required functions and packages to each worker, then distributes lakes across workers using parallel::parLapply(). The cluster is always stopped on exit — including on error — via on.exit(), so no manual cleanup is needed.

Choosing a backend

parallel = FALSE (default): single-threaded lapply(). Safe in all contexts including interactive sessions and RMarkdown.
parallel = TRUE: spawns n_cores PSOCK workers. Recommended when processing many lakes (roughly > 20) on a machine with multiple cores. Note that terra SpatRaster objects are serialised to each worker; very large rasters may make parallelisation slower than sequential processing due to inter-process data transfer overhead.

Usage

batch_lake_morphometry(
  raster_list,
  parallel = FALSE,
  n_cores = parallel::detectCores() - 1L,
  ...
)

Arguments

raster_list: A named list of terra::SpatRaster objects, one per lake. If unnamed, lakes are labelled lake_1, lake_2, etc.
parallel: Logical. If TRUE, processing is distributed across multiple CPU cores using parallel::makeCluster() / parallel::parLapply(). Default FALSE.
n_cores: Integer. Number of parallel workers to use when parallel = TRUE. Defaults to parallel::detectCores() - 1L, leaving one core free for the main R session. Silently capped at the number of lakes so no idle workers are created.
...: Additional arguments passed to calc_lake_morphometry().

Value

A data.frame with one row per successfully processed lake and one column per morphometric metric, plus a leading lake_id column. Lakes that raise an error are dropped with a warning() and do not appear in the output.

Examples

if (FALSE) { # \dontrun{
  files     <- list.files("bathymetry/", pattern = "\\.tif$", full.names = TRUE)
  rasters   <- stats::setNames(lapply(files, terra::rast),
                               tools::file_path_sans_ext(basename(files)))

  # Sequential
  results   <- batch_lake_morphometry(rasters)

  # Parallel — use 4 cores
  results_p <- batch_lake_morphometry(rasters, parallel = TRUE, n_cores = 4)
} # }