bgDistance: Randomly sample pseudo-absence points from within a given...

In SEEG-Oxford/seegSDM: Streamlined Functions for Species Distribution Modelling in the SEEG Research Group

Description

Sample n background points from within a region within distance of points (a SpatialPoints* object)and in areas where raster is not NA.

The dots argument can be used to pass additional options to bgSample. To draw a biased sample (pseudo-absences more likely to be drawn from some areas within the distance buffer than others) set raster as the bias grid and add prob = TRUE via the dots argument (see examples for details).

Usage

bgDistance(n, points, raster, distance, ...)

Arguments

n	The number of pseudo-absence points to sample.
points	A SpatialPoints* object from which to draw the buffer zone.
raster	A template raster giving the regions from which to sample (non-NA areas). This can optionally be a bias grid (see examples).
distance	The distance in map units defining the buffer zone around points from which to sample.
...	Additional arguments to pass to bgSample.

Value

A SpatialPoints object with n pseudo-absence points.

See Also

bgSample, buffer, biasGrid, SpatialPoints

Examples

rm(list = ls())

# load a test raster
r <- raster(system.file("external/test.grd", package="raster"))

# find the extent
e <- extent(r)

# define some occurrence points
pts <- matrix(c(179000, 330000,
                181000, 333000,
                180000, 332000),
              ncol = 2,
              byrow = TRUE)

# build a SpatialPoints object with them
pts <- SpatialPoints(pts, r@crs)



# randomly sample 1000 background points from within 500m of presence points
# *without* replacement
pabs <- bgDistance(1000, pts, r, 500)

# plot the raster
plot(r)
# the pseudo-absences (with transparency)
plot(pabs, add = TRUE, pch = 16, cex = 0.8, col = rgb(0, 0, 0, 0.2))
# and the presences
plot(pts, add = TRUE, pch = 16, cex = 1.5, col = 'blue')



# this time sample with replacement
pabs <- bgDistance(1000, pts, r, 500, replace = TRUE)

plot(r)
plot(pabs, add = TRUE, pch = 16, cex = 0.8, col = rgb(0, 0, 0, 0.2))
plot(pts, add = TRUE, pch = 16, cex = 1.5, col = 'blue')



# now with replacement and biased by the values of the raster
pabs <- bgDistance(1000, pts, r ^ 2, 500, prob = TRUE, replace = TRUE)

plot(r)
plot(pabs, add = TRUE, pch = 16, cex = 0.8, col = rgb(0, 0, 0, 0.2))
plot(pts, add = TRUE, pch = 16, cex = 1.5, col = 'blue')

SEEG-Oxford/seegSDM documentation built on May 9, 2019, 11:08 a.m.