landconn

This package contains functions to calculate landscape connectivity. These functions have been developed for personal teaching and research purposes and are subject to update/change as needed for these purposes. Documentation and examples may be limited.

Installation

You can install the development version of landconn like so:

devtools::install_github('wpeterman/landconn')

Example: incidence function connectivity

library(landconn)

set.seed(123)
## Create random coordinates
loc <- data.frame(x = runif(10, 0, 10),
                  y = runif(10, 0, 10))

## Create a pairwise distance matrix (a `land_dist` object)
d_mat <- dist_mat(loc)

## Set alpha (average dispersal distance) and indicate occupied sites
alpha <- 3
occ_sites <- c(0, 0, 1, 1, 0, 1, 0, 0, 1, 1)

## Patch areas
p_a <- runif(10, 5, 50)

## Model 1: distance only
c1 <- ifc(alpha, d_mat, model = 1)

## Model 1, scaled to a maximum of 1
c1s <- ifc(alpha, d_mat, model = 1, scale = TRUE)

## Model 2: weight by contributing patch area
c2 <- ifc(alpha, d_mat, model = 2, patch_area = p_a)

## Model 3: weight by focal patch area
c3 <- ifc(alpha, d_mat, model = 3, patch_area = p_a)

## Model 4: weight by both
c4 <- ifc(alpha, d_mat, model = 4, patch_area = p_a)

## Model 4, occupied sites only
c4b <- ifc(alpha, d_mat, model = 4, occ_sites = occ_sites, patch_area = p_a)

## The result is an `ifc` object with print, summary, and plot methods
c4
summary(c4)
plot(c4)

As of version 0.3.0, Models 2 to 4 use patch area directly (the standard incidence function formulation). To reproduce the earlier log-area behavior, pass area_transform = "log".

Circuitscape

landconn can also run Circuitscape from R through JuliaConnectoR (functions run_cs() and omni_cs()). These require a local Julia installation and are demonstrated in vignette("getting-started", package = "landconn").