Estimate the effective sampling frame for quadrat centres

For an axis-aligned rectangle (quadrat) to lie fully within an irregular polygon domain, the quadrat centre must lie inside a shrunk version of the domain.

Exactly, this shrink is an erosion in an L-infinity metric (half-width and half-height constraints). For simplicity and robustness in teaching/method testing, this helper uses a conservative Euclidean buffer based on the half-diagonal of the quadrat.

This provides an interpretable, single-number summary of boundary constraints:

• safe_area_fraction = area(safe_domain) / area(domain)

Values close to 1 mean boundary exclusion is mild; values near 0 mean that only a small interior region can host fully-contained quadrats.

Usage

estimate_sampling_frame(domain, quadrat_size)

Arguments

domain

An sf polygon/multipolygon.

quadrat_size

Numeric length-2, c(width, height), in the same units as the domain CRS.

Value

A list with safe_domain (sf), safe_area_fraction (numeric), and buffer_dist (numeric half-diagonal).