Generate random points that can be used to extract background values ("random-absence"). The points are sampled (without replacement) from the cells that are not '`NA`

' in raster '`mask`

'.

If the coordinate reference system (of `mask`

) is longitude/latitude, sampling is weighted by the size of the cells. That is, because cells close to the equator are larger than cells closer to the poles, equatorial cells have a higher probability of being selected.

```
randomPoints(mask, n, p, ext=NULL, extf=1.1, excludep=TRUE, prob=FALSE,
cellnumbers=FALSE, tryf=3, warn=2, lonlatCorrection=TRUE)
```

matrix with coordinates, or, if `cellnumbers=TRUE`

, a vector with cell numbers.

- mask
Raster* object. If the object has cell values, cells with

`NA`

are excluded (of the first layer of the object if there are multiple layers)- n
integer. Number of points

- p
Presence points (if provided, random points won't be in the same cells (as defined by mask)

- ext
Extent object. Can be used to restrict sampling to a spatial extent

- extf
numeric. Multiplyer to adjust the size of extent 'ext'. The default increases of 1.1 increases the extent a little (5% at each side of the extent)

- excludep
logical. If

`TRUE`

, presence points are exluded from background- prob
logical. If

`TRUE`

the values in`mask`

are interpreted as probability weights (and the values should be positive numbers (or NA)). NOTE: this currently only works for rasters of a relatively modest size (that can be loaded into RAM)- cellnumbers
logical. If

`TRUE`

, cell numbers for`mask`

are returned rather than coordinates- tryf
numeric > 1. Multiplyer used for initial sample size from which the requested sample size is extracted after removing NA points (outside of mask)

- warn
integer. 2 or higher gives most warnings. 0 or lower gives no warnings if sample size

`n`

is not reached- lonlatCorrection
logical. If

`TRUE`

then correct for the fact that longitude/latitude is not a planar coordinate system

Robert J. Hijmans