Skip to contents

Distance-based Moran's eigenvector maps (dbMEM)

geo_mem.Rd

Builds spatial eigenvectors from a truncated geographic distance matrix (Dray, Legendre and Peres-Neto 2006). Implemented natively so callers stay in tidymapr; when adespatial is installed, tests compare the leading eigenvalues.

Usage

geo_mem(
  data,
  lon = lon,
  lat = lat,
  site = NULL,
  thresh = NULL,
  method = c("haversine", "euclidean"),
  positive_only = TRUE
)

Arguments

data

A data frame of sites with coordinates.

lon, lat

Coordinate columns (degrees for haversine).

site

Optional site id column retained in the output.

thresh

Truncation threshold in metres (haversine) or coordinate units (euclidean). When NULL, the longest edge of the minimum spanning tree is used.

method

"haversine" or "euclidean" for the distance matrix.

positive_only

Keep only eigenvectors with positive eigenvalues (the usual dbMEM set)?

Value

A tibble (tmr_geo_mem) with site columns and MEM1, MEM2, ... Eigenvalues are stored in attribute eigenvalues.

References

Dray, S., Legendre, P. and Peres-Neto, P. R. (2006). Spatial modelling: a comprehensive framework for principal coordinate analysis of neighbour matrices (PCNM). Ecological Modelling, 196, 483-493. doi:10.1016/j.ecolmodel.2006.02.015

Examples

set.seed(1)
g <- expand.grid(lon = 1:5, lat = 1:4)
g$site <- paste0("s", seq_len(nrow(g)))
mem <- geo_mem(g, lon, lat, site = site, method = "euclidean")
names(mem)[1:5]
#> [1] "site" "lon"  "lat"  "MEM1" "MEM2"