Skip to contents

Mock simulator backend for testing the calibration driver without LANDIS-II

sim_mock.Rd

Returns plausibly-shaped parse_dynamic_fire_logs() output without invoking the real simulator. The output varies with par_vec so DEoptim sees a non-trivial loss surface (a few of the calibrated parameters bias the mock's fire count and size distribution; this is illustrative, not biophysical).

Usage

sim_mock(
  par_vec,
  par_names = NULL,
  paths = NULL,
  sim_years = 10L,
  base_seed = 1L,
  ...
)

Arguments

par_vec

Numeric. Named candidate parameter vector.

par_names

Character. Names in canonical order (calibration_par_names()).

paths

Named list. Currently unused; accepted for sim_landis() signature parity.

sim_years

Integer. Number of simulated years.

base_seed

Integer. RNG seed for deterministic mock output.

...

Ignored. Lets callers pass pool, pool_idx, method, etc.

Value

A list matching the shape of parse_dynamic_fire_logs() output.

Details

Use this in unit tests of calibrate_dynamic_fire() when Docker is not available; do NOT use for actual calibration.

See also

Other Dynamic Fire calibration helpers: apply_calibrated_hi_prop(), apply_calibrated_ignprob(), bc_fuel_code_to_base(), build_calibration_scenario_template(), build_calibration_spinup_scenario(), calibrate_dynamic_fire(), calibration_par_names(), default_severity_prior_sturtevant2009(), loss_from_stats(), parse_dynamic_fire_logs(), patch_fire_config(), run_calibration_spinup(), save_observed_fire_targets(), sim_landis(), sim_r_reimpl()