Scan many sites in memory-bounded chunks — grid

[scan_untested_envs()] allocates an `ndraws x g x m` array, which is infeasible for raster-sized `m`. This wrapper splits `W_new` into chunks, scans each, and stitches the per-site results. Quantities that are defined ACROSS new sites – `vcov` and `cor` – cannot be recovered from chunks and are returned as `NULL` with a warning if requested.

Usage

grid_scan(
  object,
  W_train,
  W_new,
  K_G = NULL,
  chunk_size = 500,
  max_draws = 200,
  verbose = TRUE,
  ...
)

Arguments

object, W_train, K_G: as in [scan_untested_envs()].
W_new: covariates for every grid cell / site (m x p).
chunk_size: sites per chunk.
max_draws: MCMC draws retained per chunk. Lower than the scan default because mapping needs the posterior mean far more than a tight interval.
verbose: report chunk progress.
...: passed to [scan_untested_envs()].

Value

list with `yHat_matrix`, `yHat_lower`, `yHat_upper`, `yHat_prob_win`, `genetic_var`, `interpolation`, `exceedance_detail`, `meta`. Class `scan_untested_envs` so downstream functions accept it, but with `vcov` and `cor` set to `NULL`.