pfister.plot: Create log-log plots of variance vs. sensitivity and CV vs....

In popbio: Construction and Analysis of Matrix Population Models

Create log-log plots of variance vs. sensitivity and CV vs. elasticity

Description

Create log-log plots of both variance vs. sensitivity and CV vs. elasticity in matrix elements. Plots are based on Figure 2 in Pfister(1998)

Usage

pfister.plot(A)

Arguments

A	A list of two or more projection matrices

Details

Calculates mean, variance and coefficient of variation (CV) of matrix elements from a list of two or more projection matrices. The sensitivity and elasticity matrices are then calculated from the mean matrix using eigen.analysis

Value

Creates two log-log plots similar to Figure 2 in Pfister(1998) and outputs a data.frame with 5 columns listing mean, variance, CV, sensitivity and elasticity for matrix elements with a mean and variance > 0

Author(s)

Chris Stubben

References

Pfister, CA. 1998. Patterns of variance in stage-structured populations: Evolutionary predictions and ecological implications. PNAS 95:213-218.

Examples

## 4 Hudsonia matrices
pfister.plot(hudsonia)
## 3 Mimulus cardinalis matrices at Carlon
mim <- subset(monkeyflower, species == "cardinalis" &
    site == "Carlon" & year != "pooled", select = c(4:19))
## convert data frame to list of matrices using split
mim1 <-split(mim, 2000:2002)
mim2 <-lapply(mim1, matrix, nrow=4, byrow=TRUE)
vr1 <- pfister.plot(mim2)
vr1
## PLOT using labels
plot(vr1$cv, vr1$elas, xlab="CV", ylab="Elasticity", log="xy", type='n')
# Split matrix elements into transitions representing F (fertility),
# S (survival), G (growth), and R (retrogression).
# Fertility on top row, survival on diagonal, growth is above diagonal
# and retrogression below diagonal.
rownames(vr1)
y2 <- expression(S[11],G[21],G[31],G[41],
                 F[12],S[22],G[32],G[42],
                 F[13],R[23],S[33],G[43],
                 F[14],R[34],S[44])
text(vr1$cv, vr1$elas, y2)
### add trend line
 abline(lm(log10(vr1$elas)~log10(vr1$cv)), col="red")
## include Spearman's rank correlation
a <- cor.test(vr1$cv, vr1$elas, method="spearman")
a
text(10, .0015, substitute(rho == x, list(x=round(a$estimate,2))), col="blue")

popbio documentation built on May 29, 2024, 4:35 a.m.