plot.capthist: Plot Detection Histories

In secr: Spatially Explicit Capture-Recapture

Plot Detection Histories

Description

Display a plot of detection (capture) histories or telemetry data over a map of the detectors.

Usage

## S3 method for class 'capthist'
plot(x, rad = 5, hidetraps = FALSE, tracks = FALSE,
    title = TRUE, subtitle = TRUE, add = FALSE, varycol = TRUE, 
    icolours = NULL, randcol = FALSE, lab1cap = FALSE, laboffset = 4, ncap = FALSE, 
    splitocc = NULL, col2 = "green", type = c("petal", "n.per.detector", "n.per.cluster",
        "sightings", "centres", "telemetry", "nontarget"),
    cappar = list(cex = 1.3, pch = 16, col = "blue"), 
    trkpar = list(col = "blue", lwd = 1), 
    labpar = list(cex = 0.7, col = "black"), ...)

plotMCP(x, add = FALSE, col = "black", fill = NA, lab1cap = FALSE, laboffset = 4,
    ncap = FALSE, ...)

Arguments

x	an object of class capthist
rad	radial displacement of dot indicating each capture event from the detector location (used to separate overlapping points)
hidetraps	logical indicating whether trap locations should be displayed
tracks	logical indicating whether consecutive locations of individual animals should be joined by a line
title	logical or character string for title
subtitle	logical or character string for subtitle
add	logical for whether to add to existing plot
varycol	logical for whether to distinguish individuals by colour
icolours	vector of individual colours (when varycol = TRUE), or colour scale (non-petal plots)
randcol	logical to use random colours (varycol = TRUE)
lab1cap	logical for whether to label the first capture of each animal
laboffset	distance by which to offset labels from points
ncap	logical to display the number of detections per trap per occasion
splitocc	optional occasion from which second colour is to be used
col2	second colour (used with splitocc)
type	character string ("petal", "n.per.detector" or "n.per.cluster")
cappar	list of named graphical parameters for detections (passed to par)
trkpar	list of named graphical parameters for tracks (passed to par)
labpar	list of named graphical parameters for labels (passed to par)
...	arguments passed to plot.traps
col	vector of line colour numbers or names (plotMCP only)
fill	vector of fill colour numbers or names (plotMCP only)

Details

By default, a ‘petal’ plot is generated in the style of Density (Efford 2012) using eqscplot from the MASS library.

If type = "n.per.detector" or type = "n.per.cluster" the result is a colour-coded plot of the number of individuals at each unit, pooled over occasions.

If type = "sightings" the sightings of unmarked animals are displayed on a petal-like plot (requires mark-resight data) (see also sightingPlot).

If type = "centres" then a single point is plotted for each animal, jittered on each axis by a random amount (limits +/- rad/2).

If type = "telemetry" and the ‘telemetryxy’ attribute is not NULL then the telemetry locations are plotted.

If type = "nontarget" and the ‘nontarget’ attribute is not NULL then the nontarget captures or interference events are plotted.

If title = FALSE no title is displayed; if title = TRUE, the session identifer is used for the title.

If subtitle = FALSE no subtitle is displayed; if subtitle = TRUE, the subtitle gives the numbers of occasions, detections and individuals.

If x is a multi-session capthist object then a separate plot is produced for each session. Use par(mfrow = c(nr, nc)) to allow a grid of plots to be displayed simultaneously (nr rows x nc columns).

These arguments are used only for petal plots: rad, tracks, varycol, randcol, lab1cap, laboffset, ncap, splitocc, col2, trkpar, and labpar. Call occasionKey to add a key to the petals. From 5.0.1 a warning is issued if rad exceeds 3% of the detector span in either x- or y-dimensions.

If icolours = NULL and varycol = TRUE then a vector of colours is generated automatically as topo.colors((nrow(x)+1) * 1.5). If there are too few values in icolours for the number of individuals then colours will be re-used.

plotMCP plots minimum convex polygons of individual location data over a base plot of detector locations. Usually the data are telemetry locations in the xylist attribute of the capthist object; if this is not present and x is a polygon search capthist then the individual xy data are plotted.

Value

For type = "petal", the number of detections in x. For type = "sightings", the number of sightings of unmarked animals in x. For type = "n.per.detector" or type = "n.per.cluster", a dataframe with data for a legend (see Examples).

plotMCP invisibly returns a list in which each component is a 2-column (x,y) dataframe of boundary coordinates for one individual.

References

Efford, M. G. (2012) DENSITY 5.0: software for spatially explicit capture–recapture. Department of Mathematics and Statistics, University of Otago, Dunedin, New Zealand. https://www.otago.ac.nz/density/.

See Also

capthist, occasionKey, plot.traps, sightingPlot

Examples

demotrap <- make.grid()
tempcapt <- sim.capthist(demotrap, 
    popn = list(D = 5, buffer = 50), 
    detectpar = list(g0 = 0.15, sigma = 30))
plot(tempcapt, border = 10, rad = 3, tracks = TRUE, 
    lab1cap = TRUE, laboffset = 2.5)

## type = n.per.cluster

## generate some captures
testregion <- data.frame(x = c(0,2000,2000,0),
    y = c(0,0,2000,2000))
popn <- sim.popn (D = 10, core = testregion, buffer = 0,
    model2D = "hills", details = list(hills = c(-2,3)))
t1 <- make.grid(nx = 1, ny = 1)
t1.100 <- make.systematic (cluster = t1, spacing = 100,
    region = testregion)
capt <- sim.capthist(t1.100, popn = popn, noccasions = 1)

## now plot captures ...
temp <- plot(capt, title = "Individuals per cluster",
    type = "n.per.cluster", hidetraps = FALSE,
    gridlines = FALSE, cappar = list(cex = 1.5))

if (interactive()) {
    ## add legend; click on map to place top left corner
    legend (locator(1), pch = 21, pt.bg = temp$colour,
        pt.cex = 1.3, legend = temp$legend, cex = 0.8)
}

## Not run: 

## try varying individual colours - requires RColorBrewer
library(RColorBrewer)
plot(infraCH[[2]], icolours = brewer.pal(12, "Set3"), tracks = TRUE,
    bg = "black", cappar = list(cex = 2), border = 10, rad = 2,
    gridlines = FALSE)

## generate telemetry data
te <- make.telemetry()
tr <- make.grid(detector = "proximity")
totalpop <- sim.popn(tr, D = 20, buffer = 100)
tepop <- subset(totalpop, runif(nrow(totalpop)) < 0.05)
teCH <- sim.capthist(te, popn = tepop, renumber=FALSE, detectfn = "HHN",
    detectpar = list(lambda0 = 3, sigma = 25))
plot(teCH, type = 'telemetry', tracks = TRUE)

## simple "centres" example
## polygon data require 'hazard' detection function 14:19
CH <- sim.capthist(make.poly(), nocc = 20, detectfn = 'HHN', 
    detectpar = list(lambda0 = 1, sigma = 10))
plot(CH, cappar = list(col = 'orange'), varycol = FALSE, border = 10)
plot(CH, type = 'centres', add = TRUE, rad = 0)

## End(Not run)

secr documentation built on Nov. 4, 2024, 9:06 a.m.