dHabitatMask: Ones trick distribution for irregular habitat shapes

In nimbleSCR: Spatial Capture-Recapture (SCR) Methods Using 'nimble'

Ones trick distribution for irregular habitat shapes

Description

The dHabitatMask distribution checks and ensures that the proposed activity center location (s) falls within the suitable habitat (defined in the binary matrix habitatMask).

Usage

dHabitatMask(x, s, xmax, xmin, ymax, ymin, habitatMask, log = 0)

rHabitatMask(n, s, xmax, xmin, ymax, ymin, habitatMask)

Arguments

x	Ones trick data.
s	Bivariate activity center coordinates.
xmax	Maximum of trap location x-coordinates.
xmin	Minimum of trap location x-coordinates.
ymax	Maximum of trap location y-coordinates.
ymin	Minimum of trap location y-coordinates.
habitatMask	A binary matrix object indicating which cells are considered as suitable habitat.
log	Logical argument, specifying whether to return the log-probability of the distribution.
n	Integer specifying the number of realisations to generate. Only n = 1 is supported.

Details

The rHabitatMask function returns the value of the habitat mask cell (0 or 1) where the proposed activity center falls. See also M. Meredith: SECR in BUGS/JAGS with patchy habitat.

Value

The log-likelihood value associated with the bivariate activity center location s being in the suitable habitat (i.e. 0 if it falls within the habitat mask and -Inf otherwise).

Author(s)

Daniel Turek

Examples

## define model code
code <- nimbleCode({
    for(i in 1:N) {
        s[i, 1] ~ dunif(0, 100)
        s[i, 2] ~ dunif(0, 100)
        OK[i] ~ dHabitatMask( s = s[i,1:2],
                              xmax = 100,
                              xmin = 0,
                              ymax = 100,
                              ymin = 0,
                              habitatMask = habitatMask[1:100,1:100])
    }
})
 
N <- 20
 
habitatMask <- matrix(rbinom(10000,1,0.75), nrow = 100)
 
constants <- list(N = N, habitatMask = habitatMask)
 
data <- list(OK = rep(1, N))
 
inits <- list(s = array(runif(2*N, 0, 100), c(N,2)))
 
## create NIMBLE model object
Rmodel <- nimbleModel(code, constants, data, inits)
 
## use model object for MCMC, etc.

nimbleSCR documentation built on Dec. 1, 2022, 1:17 a.m.