pdot: Net Detection Probability

In secr: Spatially Explicit Capture-Recapture

Net Detection Probability

Description

Compute spatially explicit net probability of detection for individual(s) at given coordinates (pdot).

Usage

pdot(X, traps, detectfn = 0, detectpar = list(g0 = 0.2,
    sigma = 25, z = 1), noccasions = NULL, binomN = NULL,
    userdist = NULL, ncores = NULL) 

CVpdot(..., conditional = FALSE)    

Arguments

X	vector or 2-column matrix of coordinates
traps	traps object
detectfn	integer code for detection function q.v.
detectpar	a named list giving a value for each parameter of detection function
noccasions	number of sampling intervals (occasions)
binomN	integer code for discrete distribution (see secr.fit)
userdist	user-defined distance function or matrix (see userdist)
ncores	integer number of threads
...	arguments passed to pdot
conditional	logical; if TRUE then computed mean and CV are conditional on detection

Details

If traps has a usage attribute then noccasions is set accordingly; otherwise it must be provided.

The probability computed is p.(\mathbf{X}) = 1 - \prod\limits _{k} \{1 - p_s(\mathbf{X},k)\}^{S} where the product is over the detectors in traps, excluding any not used on a particular occasion. The per-occasion detection function p_s is halfnormal (0) by default, and is assumed not to vary over the S occasions.

For detection functions (10) and (11) the signal threshold ‘cutval’ should be included in detectpar, e.g., detectpar = list(beta0 = 103, beta1 = -0.11, sdS = 2, cutval = 52.5).

The calculation is not valid for single-catch traps because p.(\mathbf{X}) is reduced by competition between animals.

userdist cannot be set if ‘traps’ is any of polygon, polygonX, transect or transectX. if userdist is a function requiring covariates or values of parameters ‘D’ or ‘noneuc’ then X must have a covariates attribute with the required columns.

Setting ncores = NULL uses the existing value from the environment variable RCPP_PARALLEL_NUM_THREADS (see setNumThreads).

CVpdot returns the expected mean and CV of pdot across the points listed in X, assuming uniform population density. X is usually a habitat mask. See Notes for details.

Value

For pdot, a vector of probabilities, one for each row in X.

For CVpdot, a named vector with elements ‘meanpdot’ and ‘CVpdot’.

Note

CVpdot computes the mean \mu and variance V of the location-specific overall detection probability p.(\mathbf{X}) as follows.

\mu = \int p.(\mathbf{X}) f(\mathbf{X}) d\mathbf{X},

V = \int p.(\mathbf{X})^2 f(\mathbf{X}) d\mathbf{X} - \mu^2.

For uniform density and conditional = FALSE, f(\mathbf{X}) is merely a scaling factor independent of \mathbf{X}.

If conditional = TRUE then f(\mathbf{X}) = p.(\mathbf{X}) / \int p.(\mathbf{X}) d\mathbf{X}.

The coefficient of variation is CV = \sqrt{V}/\mu.

See Also

secr, make.mask, Detection functions, pdot.contour, CV

Examples

## Not run: 

  temptrap <- make.grid()
  ## per-session detection probability for an individual centred
  ## at a corner trap. By default, noccasions = 5.
  pdot (c(0,0), temptrap, detectpar = list(g0 = 0.2, sigma = 25),
    noccasions = 5)
    
  msk <- make.mask(temptrap, buffer = 100)
  CVpdot(msk, temptrap, detectpar = list(g0 = 0.2, sigma = 25),
    noccasions = 5)

## End(Not run)

secr documentation built on Oct. 18, 2023, 1:07 a.m.