Identifies the cell numbers of all cells within a ring defined by minimum
and maximum distances from focal cells.
Uses `spread`

under the hood, with specific values set.
Under many situations, this will be faster than using `rgeos::gBuffer`

twice (once for smaller ring and once for larger ring, then removing the
smaller ring cells).

rings(landscape, loci = NA_real_, id = FALSE, minRadius = 2, maxRadius = 5, allowOverlap = FALSE, returnIndices = FALSE, returnDistances = TRUE, ...) # S4 method for RasterLayer rings(landscape, loci = NA_real_, id = FALSE, minRadius = 2, maxRadius = 5, allowOverlap = FALSE, returnIndices = FALSE, returnDistances = TRUE, ...)

landscape | A |
---|---|

loci | A vector of locations in |

id | Logical. If |

minRadius | Numeric. Minimum radius to be included in the ring.
Note: this is inclusive, i.e., |

maxRadius | Numeric. Maximum radius to be included in the ring.
Note: this is inclusive, i.e., |

allowOverlap | Logical. If |

returnIndices | Logical. Should the function return a |

returnDistances | Logical. Should the function include a column with the
individual cell distances from the locus where that event
started. Default is |

... | Any other argument passed to |

This will return a `data.table`

with columns as described in
`spread`

when `returnIndices = TRUE`

.

`cir`

which uses a different algorithm.
`cir`

tends to be faster when there are few starting points, `rings`

tends to be faster when there are many starting points. Another difference
between the two functions is that `rings`

takes the centre of the pixel
as the centre of a circle, whereas `cir`

takes the exact coordinates.
See example.

library(raster) library(quickPlot) # Make random forest cover map emptyRas <- raster(extent(0, 1e2, 0, 1e2), res = 1) # start from two cells near middle loci <- (ncell(emptyRas) / 2 - ncol(emptyRas)) / 2 + c(-3, 3) # Allow overlap emptyRas[] <- 0 rngs <- rings(emptyRas, loci = loci, allowOverlap = TRUE, returnIndices = TRUE) # Make a raster that adds together all id in a cell wOverlap <- rngs[, list(sumEventID = sum(id)), by = "indices"] emptyRas[wOverlap$indices] <- wOverlap$sumEventID if (interactive()) { clearPlot() Plot(emptyRas) } # No overlap is default, occurs randomly emptyRas[] <- 0 rngs <- rings(emptyRas, loci = loci, minRadius = 7, maxRadius = 9, returnIndices = TRUE) emptyRas[rngs$indices] <- rngs$id if (interactive()) { clearPlot() Plot(emptyRas) } # Variable ring widths, including centre cell for smaller one emptyRas[] <- 0 rngs <- rings(emptyRas, loci = loci, minRadius = c(0, 7), maxRadius = c(8, 18), returnIndices = TRUE) emptyRas[rngs$indices] <- rngs$id if (interactive()) { clearPlot() Plot(emptyRas) }