Skip to contents

Open + collect a per-scenario biomass_snapshots Arrow dataset

read_biomass_c_snapshots_for_scenario.Rd

Convenience wrapper that opens <scenario_dir>/<subdir> as an Arrow dataset partitioned by replicate, collects the full contents into a tibble, and returns NULL for missing directories or empty datasets so callers can short-circuit gracefully.

Usage

read_biomass_c_snapshots_for_scenario(
  scenario_dir,
  subdir = "_aggregates/biomass_snapshots"
)

Arguments

scenario_dir

Path to a scenario's root directory (e.g. "LANDIS-II/ForCS_only").

subdir

Path within scenario_dir to the parquet dataset root (default "_aggregates/biomass_snapshots").

Value

A tibble with columns scenario, replicate, Time, row, column, ecoregion, species, biomass (matching read_biomass_c_snapshots()'s output), or NULL.

Details

Isolated from the rest of the read_biomass_c_snapshots() pipeline so its body stays stable across project-side refactors: {targets} does not track installed-package function bodies, so consuming targets are not invalidated by cosmetic changes to how the collect happens under the hood. Callers that want that invalidation guarantee should call this function from their target commands rather than inlining the arrow::open_dataset() |> collect() chain.

The default subdir = "_aggregates/biomass_snapshots" matches the layout produced by write_biomass_c_snapshots_parquet()-style writers in the FOR-CAST post-processing pipeline; override for other layouts.

See also

Other Vegetation transition helpers: biomass_landscape_summary(), community_label(), leading_species(), plot_species_biomass(), plot_transitions(), read_biomass_c_snapshots(), read_biomass_output_rasters(), transition_data(), write_biomass_c_snapshots_parquet()