Organize data for the community occupancy model fit by occuComm

Organize detection/non-detection data and covariates for use with the community occupancy model

unmarkedFrameOccuComm(y, siteCovs = NULL, obsCovs = NULL, speciesCovs = NULL)

Arguments

y

The detection-nondetection data for multiple species. This data may be in one of two forms: either a named list of S M x J matrices (M = sites, J = occasions, S = species) or an M x J x S array.

siteCovs

A data.frame of covariates that vary at the site level. This should have M rows and one column per covariate.

obsCovs

Either a named list of data.frames of covariates that vary within sites, or a data.frame with M x J rows in site-major order.

speciesCovs

Covariates that also vary by species. This must be provided as a named list. Each list element can have one of three possible dimensions: a vector of length S (e.g., mean species body mass); a matrix M x S (covariates that vary by site and species); or an array M x J x S (covariates that vary by site, occasion, and species).

Value

an object of class unmarkedFrameOccuComm

See also

occuComm

Examples

# Simulate some multispecies data
nsite <- 300
nocc <- 5
nsp <- 30

set.seed(123)

# Create a site by species covariate
x <- matrix(rnorm(nsite*nsp), nsite, nsp)

mu_0 <- 0
sd_0 <- 0.4
beta0 <- rnorm(nsp, mu_0, sd_0)

mu_x <- 1
sd_x <- 0.3
beta_x <- rnorm(nsp, mu_x, sd_x)

mu_a <- 0
sd_a <- 0.2
alpha0 <- rnorm(nsp, mu_a, sd_a)

ylist <- list()
z <- matrix(NA, nsite, nsp)
for (s in 1:nsp){
  psi <- plogis(beta0[s] + beta_x[s] * x[,s])
  z[,s] <- rbinom(nsite, 1, psi)

  p <- plogis(alpha0[s])

  y <- matrix(NA, nsite, nocc)
  for (m in 1:nsite){
    y[m,] <- rbinom(nocc, 1, p * z[m,s])
  }
  ylist[[s]] <- y
}
names(ylist) <- paste0("sp", sprintf("%02d", 1:nsp))
sc <- data.frame(a=factor(sample(letters[1:5], nsite, replace=TRUE)))

# Species covs need to be a list of named elements
spc <- list(x = x)

umf <- unmarkedFrameOccuComm(ylist, sc, speciesCovs = spc)
summary(umf)
#> unmarkedFrameOccuComm Object
#> 
#> 300 sites
#> 30 species
#> Maximum number of observations per site: 5 
#> Mean number of observations per site: 5 
#> Sites with at least one detection, quantiles by species:
#>     0%    25%    50%    75%   100% 
#> 104.00 130.25 142.50 171.00 208.00 
#> 
#> Site-level covariates:
#>  a     
#>  a:64  
#>  b:59  
#>  c:57  
#>  d:68  
#>  e:52  
#> 
#> Species-level covariates:
#> List of 1
#>  $ x: num [1:300, 1:30] -0.5605 -0.2302 1.5587 0.0705 0.1293 ...