RI: Return an expected remigration interval.

In phenology: Tools to Manage a Parametric Function that Describes Phenology and More

Return an expected remigration interval.

Description

Model of remigration interval
Note that r, s and t are conditional probabilities. If c is null, then return probabilities are estimated from r. r can be named vector. For example:
r <- c(r1=0.5, r2=0.60, r3=1) is equivalent to c <- c(c1=0.5, c2=0.3, c3=0.2)
The vector of r described the probability that a female returned after 1, 2, 3 years among those who have not nested before. The vector of r is the same but defining the return probability for an initial female.

Usage

RI(s, t, r = NULL, c = NULL, p)

Arguments

s	Time-conditional probability of survival
t	Time-conditional probability of tag retention
r	Time-conditional probability of return
c	Probability of return
p	Annual probability of observation

Details

RI returns an expected remigration interval

Value

Return a remigration interval.

Author(s)

Marc Girondot

See Also

Other Model of Remigration Interval: Bayesian.remigration(), LnRI_norm(), plot.Remigration()

Examples

## Not run: 
library(phenology)
# Example
s <- c(s1=1, s2=1, s3=1, s4=1, s5=1)
t <- c(t1=0.95, t2=1, t3=1, t4=1, t5=1) 
r <- c(r1=0.1, r2=0.8, r3=0.7, r4=0.7, r5=1)
p <- c(p1=0.6, p2=0.6, p3=0.6, p4=0.6, p5=0.6)

# r is equivalent to 
c <- c(c1=0.1, c2=0.72, c3=0.126, c4=0.0378, c5=0.0162)
# Then the true remigration interval is:
ri_true <- sum(1:5*c[1:5])

s_ri  <- NULL
for (sx in seq(from=0.01, to=1, by=0.01)) {
  s[] <- sx
  ri1 <- RI(s=s, t=t, r=r, p=p)
  s_ri  <- c(s_ri,sum(1:5*ri1)/sum(ri1))
}

par(mar=c(4, 4, 1, 1)+0.4)

plot(x=seq(from=0.01, to=1, by=0.01), y=s_ri, type="l", 
     las=1, bty="n", ylim=c(0, 4), 
     xlab="Annuual survival probabilities", ylab="Naive Remigration Interval", 
    main="")
segments(x0=0.01, x1=1, y0=ri_true, y1=ri_true, lty=2, col="red")
legend("topright", legend="True remigration interval", lty=2, col="red")


## End(Not run)

phenology documentation built on Oct. 16, 2023, 9:06 a.m.