In Pakillo/rSDM: Species distribution and niche modelling in R

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>"
)

library(rSDM)

rSDM includes a handy function to move points to the nearest raster cell with data (i.e. not NA).

Example data

Let's generate example point coordinates:

locs <- data.frame(lon = c(1, 2, 1, 2, 2.2), lat = c(1.2, 1, 2.3, 3, 2))
locs.sf <- locs2sf(locs)
locs.sf

Now let's generate an example raster:

library(terra)
ras <- rast(nrows = 2, ncols = 2, xmin = 0.5, xmax = 3.5, ymin = 0.5, ymax = 3.5,
 resolution = 1, vals = c(NA, 1, 1, NA, NA, 1, NA, 1, 1))

As we can see in the map below, some points fall outside raster cells with data:

occmap(locs.sf, ras, pcol = "black", psize = 3)

Move point coordinates to nearest raster cell

moved <- points2nearestcell(locs.sf, ras)

Note you can choose the format of the output map, e.g. now using ggplot (interactive leaflet maps are available too):

points2nearestcell(locs.sf, ras, map = "ggplot")

The function returns a spatial object where the coordinates of the points falling outside the raster have been moved:

moved

Let's compare the original and new coordinates:

sf::st_coordinates(locs.sf)
sf::st_coordinates(moved)

If you don't want to change coordinates but only check which would the nearest raster cells for each point, use move = FALSE.

Move points to nearest raster cell within a given distance

In case you want to move points only if the nearest raster cell is within a given distance, you could use the distance argument to set a threshold:

moved <- points2nearestcell(locs.sf, ras, distance = 100000)

Pakillo/rSDM documentation built on March 1, 2025, 12:30 a.m.