runCS.R

R function to run CIRCUITSCAPE from an R console on Mac/Linux

Description

This R function will return a "resistance" matrix as a dist object provided a raster and a set of points are specfied. The raster needs to be a RasterLayer object read using the R package raster. The points can be given as matrix objects with "x" and "y" coordinates or as a SpatialPoints* object (see R package sp). The script will not check if the RasterLayer and the SpatialPoints objects are in the same coordinate system, so please ensure that they are.

Dependencies

This script has only been tested on Mac/linux and was designed to run on a Unix-based CIRCUITSCAPE installation. This means that the csrun.py program should be in directory found in the $PATH variable; for example, /usr/local/bin. The script also requires that the raster and sp packages are installed.

Install package dependencies:

install.packages(c("raster","sp"))

Source the code

source("https://raw.githubusercontent.com/santiagosnchez/runCS/master/runCS.R")
# the R code will need to be sourced every time a new session is opened.

Run the code

resistance <- runCS(layer, points)
# "layer" needs to be a RasterLayer object and 
# "points" a xy matrix or a SpatialPoints object

Suggestions

• Make sure that cell sizes in the raster have the same width/length ratio. For example, if working on a UTM system the resolution should be something like c(1000,1000).

  r <- layer
  res(r) <- 1000
  layer2 <- resample(layer, r)
  

• Make sure that you only have one coordinate point per cell. So if you have multiple points per cell, choose only one.

  r <- rasterize(layer, points)
  xy <- rasterToPoints(r)
  xy <- xy[ order(xy[,3]), ]
  rownames(xy) <- xy[,3]
  xy <- xy[,-3]