| sample.points {DSpat} | R Documentation |
Create a dataframe of observations by simulating distance sampling of a point process
with a systematic set of lines over a rectangular grid. The transects,
lines and point process(points.ppp) are input arguments. Detection
of observations is specified with a user-defined detection function which takes
a distance vector and set of parameters det.par as its arguments.
sample.points(transects,lines,points.ppp,detfct=NULL,det.par=NULL,
det.formula=~1,covariates=NULL)
hndetfct(x,scale)
transects |
list of transect polygons |
lines |
dataframe of lines |
points.ppp |
simulated point process |
detfct |
detection function name |
det.par |
parameters for the detection function |
det.formula |
formula of covariates to use for scale of distance if det.formula=~-1, uses a strip transect |
covariates |
a matrix with columns x,y and any number of covariates x and y are the mid points of the grid cells; the order of the rows must match the formulation for function im |
x |
perpendicular distance for detection function |
scale |
scale for detection function |
Definition for half-normal detection function (hndetfct) is exp(-(x^2/(2*exp(scale)^2)))
observation dataframe with fields label,x,y,distance for line label, x,y coordinates of the observation location and its perpendicular distance from the line
Jeff Laake
simCovariates,simPts,create.lines
study.area=owin(xrange=c(0,100),yrange=c(0,100))
hab.range=30
probs=c(1/3,2/3)
covariates = simCovariates(hab.range, probs)
xlines=create.lines(study.area,nlines=10,width=5,angle=45)
ls=lines_to_strips(xlines,study.area)
plot(ls$lines,lty=2)
plot(owin(poly=ls$transects),add=TRUE)
xpp=simPts(covariates=covariates,int.formula=~factor(habitat),int.par=c(0,1,2),EN=1000)
obs=sample.points(transects=ls$transects,lines=xlines,points.ppp=xpp,
hndetfct,c(1),covariates=covariates)
plot(ppp(x=obs$x,y=obs$y,window=study.area),add=TRUE,pch=20)