Severe fragmentation from suitable-habitat patches and population density

Implements the habitat-and-density fragmentation method of the IUCN Red List guidelines (Santini et al. 2019), as used by sRedList (Cazalis et al. 2024): the suitable habitat (Area of Habitat) within the range is split into patches, patches within a dispersal / isolation distance of one another are grouped into isolated subpopulations, and each subpopulation's size in individuals is its habitat area times the population density. This is the guideline-compliant companion to assess_fragmentation (which uses occurrence counts as a rough proxy when no habitat model or density is available).

Usage

fragment_habitat(
  range,
  isolation_km,
  density,
  year = NULL,
  collection = NULL,
  initiative = "brazil",
  src = NULL,
  suitable_groups = "natural",
  suitable_codes = NULL,
  elev_min = NA,
  elev_max = NA,
  z = 9,
  elev_src = NULL,
  max_pixels = 4000,
  small_size = NULL
)

Arguments

range: An sf/sfc range polygon (e.g. the EOO hull).
isolation_km: Isolation / dispersal distance in km: patches further apart than this host isolated subpopulations. Required.
density: Population density in mature individuals per km^2 of suitable habitat. A length-2 vector gives a low/high range (a second population column pop_high is added). Required.
year, collection, initiative, src: Land-cover selection, as in calc_aoh / mb_raster_local.
suitable_groups, suitable_codes: Which land-cover classes count as suitable habitat (default: the "natural" group). See calc_aoh.
elev_min, elev_max, z, elev_src: Optional elevation band (metres) and DEM source, as in calc_aoh.
max_pixels: Display pixel budget for the habitat mask read (default 4000); larger ranges are read coarser, as sRedList aggregates them.
small_size: Optional size (individuals) the assessor considers "small"; when given, prop_in_small and severe_suggested are returned.

Value

A list with method = "habitat", n_subpop (occupied subpopulations), isolation_km, density, sizes (a data.frame of cluster, aoh_km2, pop and, for a density range, pop_high), clusters (subpopulation polygons as an sfc in WGS84), curve, median_subpop_size, max_subpop (C2a(i)), pct_largest (C2a(ii)), total_pop, and prop_in_small / severe_suggested. Values degrade to NA when the land cover cannot be read.

Details

Following Santini et al. (2019): contiguous suitable cells form patches; each patch is buffered by half the isolation distance and buffers that touch are merged, so patches closer than the isolation distance share a subpopulation; the number of individuals in a subpopulation is its total suitable area (km^2) times density (individuals km^-2), and a subpopulation of fewer than two individuals is treated as unoccupied. The curve then gives the proportion of the population in subpopulations no larger than a given size, whose median_subpop_size is the size below which half the population lives; if the assessor considers a subpopulation of that size "small", the population can be treated as severely fragmented (Criterion B sub-criterion a). The largest subpopulation's size (max_subpop, Criterion C2a(i)) and its share of the population (pct_largest, Criterion C2a(ii)) are also returned.

Use with care: use the result only when the mapped clusters make ecological sense as subpopulations, exactly as sRedList advises. Barriers (water, mountains, infrastructure) are not accounted for.

References

Santini L. et al. (2019) Applying habitat and population-density models to land-cover time series to inform IUCN Red List assessments. Conservation Biology 33:1084-1093. doi:10.1111/cobi.13279

Cazalis V. et al. (2024) Accelerating and standardising IUCN Red List assessments with sRedList. Biological Conservation 298:110761. doi:10.1016/j.biocon.2024.110761

Examples

# \donttest{
f <- system.file("extdata", "example_occurrences.csv", package = "mappingAS")
occ <- read_occurrences(f)
sp1 <- occ[occ$species == occ$species[1], ]
hull <- calc_eoo(sp1)$hull
# Reads land cover; density in mature individuals per km^2:
fr <- fragment_habitat(hull, isolation_km = 20, density = 5)
fr$n_subpop
#> [1] 1
fr$max_subpop      # C2a(i)
#> [1] 6850.165
# }