Mask Buffer Diagnostic Plot
Plot effective sampling area (Borchers and Efford 2008) as a function of increasing buffer width.
esa.plot (object, max.buffer = NULL, spacing = NULL, max.mask = NULL,
detectfn, detectpar, noccasions, binomN = NULL, thin = 0.1,
poly = NULL, poly.habitat = TRUE, session = 1, plt = TRUE,
type = c('density', 'esa', 'meanpdot', 'CVpdot'), n = 1, add = FALSE,
overlay = TRUE, conditional = FALSE, ...)object |
|
max.buffer |
maximum width of buffer in metres |
spacing |
distance between mask points |
max.mask |
|
detectfn |
integer code or character string for shape of detection function 0 = halfnormal etc. – see detectfn |
detectpar |
list of values for named parameters of detection function |
noccasions |
number of sampling occasions |
binomN |
integer code for discrete distribution (see
|
thin |
proportion of mask points to retain in plot and output |
poly |
matrix of two columns interpreted as the x and y coordinates of a bounding polygon (optional) |
poly.habitat |
logical as in |
session |
vector of session indices (used if |
plt |
logical to plot results |
type |
character, what to plot |
n |
integer number of distinct individuals detected |
add |
logical to add line to an existing plot |
overlay |
logical; if TRUE then automatically |
conditional |
logical; if TRUE the reported mean and CV are conditional on detection
(see |
... |
graphical arguments passed to plot() and lines() |
Effective sampling area (esa) is defined as the integral of net
capture probability (p.(X)) over a
region. esa.plot shows the effect of increasing region size on
the value of esa for fixed values of the detection parameters. The
max.buffer or max.mask arguments establish the maximum
extent of the region; points (cells) within this mask are sorted by
their distance dk from the nearest detector. esa(buffer) is
defined as the cumulative sum of c * p.(X) for
dk(X) <= buffer, where c
is the area associated with each cell.
The default (type = 'density') is to plot the reciprocal of esa
multiplied by n; this is on a more familiar scale (the density
scale) and hence is easier to interpret.
Because esa.plot uses the criterion 'distance to nearest
detector', max.mask should be constructed to include all
habitable cells within the desired maximum buffer and no others. This
is achieved with type = "trapbuffer" in make.mask. It is
a good idea to set the spacing argument of make.mask
rather than relying on the default based on nx. Spacing may be
small (e.g. sigma/10) and the buffer of max.mask may be quite
large (e.g. 10 sigma), as computation is fast.
Thinning serves to reduce redundancy in the plotted points, and (if
the result is saved and printed) to generate more legible numerical
output. Use thin=1 to include all points.
esa.plot calls the internal function esa.plot.secr when
object is a fitted model. In this case detectfn,
detectpar and noccasions are inferred from
object.
A dataframe with columns
If plt = TRUE the dataframe is returned invisibly.
The response of effective sampling area to buffer width is just one
possible mask diagnostic; it's fast, graphic, and often
sufficient. mask.check performs more intensive checks,
usually for a smaller number of buffer widths.
The old argument 'as.density' was superceded by 'type' in 3.1.7.
Borchers, D. L. and Efford, M. G. (2008) Spatially explicit maximum likelihood methods for capture–recapture studies. Biometrics 64, 377–385.
## with previously fitted model
esa.plot(secrdemo.0)
## from scratch
trps <- make.grid()
msk <- make.mask(trps, buffer = 200, spacing = 5, type = "trapbuffer")
detectpar <- list(g0 = 0.2, sigma = 25)
esa.plot(trps,,, msk, 0, detectpar, nocc = 10, col = "blue")
esa.plot(trps,,, msk, 0, detectpar, nocc = 5, col = "green",
add = TRUE)
esa.plot(trps,,, msk, 0, detectpar, nocc = 5, thin = 0.002, plt = FALSE)Please choose more modern alternatives, such as Google Chrome or Mozilla Firefox.