Skip to contents

One size per ignition point. Each point keeps its own SIZE_HA unless a perimeter polygon from the SAME calendar year contains it, in which case the polygon's SIZE_HA replaces it. For NFDB points and NBAC perimeters, this keeps NFDB's full sample (pre-1972 fires, and small fires NBAC does not map) while taking NBAC's satellite-derived area wherever a fire was mapped.

Usage

observed_fire_sizes(points, polys = NULL, min_size_ha = 0)

Arguments

points

SpatVector. Ignition points with SIZE_HA and YEAR columns.

polys

SpatVector or NULL. Perimeter polygons with SIZE_HA and YEAR columns. NULL keeps every point's own size.

min_size_ha

Numeric scalar. Sizes below this, and missing sizes, are dropped. Default 0 keeps every positive and zero size.

Value

Numeric vector of sizes (ha), sorted ascending; at most one element per point.

Details

This is the fire-size rule behind save_observed_fire_targets()'s fire_sizes_ha. It is exported so that anything else derived from the same fire record – such as a fitted fire-size distribution – uses the same sizes as the calibration's size target, rather than a second rule that can drift from it. Binding points and polygons as separate rows instead would count every mapped fire twice.