dist.decay: Fit an exponential distance decay model to assemblage...

In chihlinwei/ddecay: Fit an exponential distance decay model to assemblage similarities

Description

Fit a generalized dissimilarity model in ecological similarity of assemblages (including zero similarity points) between two sites as a function of their distance apart along an environmental gradient.

Usage

dist.decay(gradient, counts, coords, nboots = 1000, dis.fun = "vegdist",
  method = "bray", like.pairs = T)

Arguments

gradient	A vector of environmental values of interests
counts	A dataframe of community matrix of species abundance data
coords	A dataframe of 2 column with x and y coordinates of the sampling site. The same sites will be remove from bootsrap procedures if like.pairs = TRUE
nboots	A integer number indicating the numbers of bootsrap resampling
dis.fun	Matching function to compute dissimilarity. The default is vegdist
method	Dissimilarity index, partial match to "manhattan", "euclidean", "canberra", "bray", "kulczynski", "jaccard", "gower", "altGower", "morisita", "horn", "mountford", "raup" , "binomial", "chao", "cao" or "mahalanobis".
like.pairs	Logical, whether to remove the like pairs or not. During bootstrap resampling, because the sites are randomly resampled with replacement and hence there may be multiple copies of some sites. like.pairs = TRUE removes those site pairs having zero separation distance.

Details

This funtion fits a generalized linear models (GLMs) using quasibinomial distribution (with a log-link function) for ecological similarity of assemblages between any two sites as a function of their distance apart along a environmental gradient. A modified bootstrap method was used to estimate the bootraped mean and standard error of model parameters (alpha, beta), s=alpha*exp(-beta*d), where s is assemblage similarity and d is environmental distance, and then use these parameters to calculate the similarity at zero distance (s0) and the halving distance (halfd) for which the similarity between sites decreased by 50% following the method decribed by Millar et al. (2010).

Value

A list of data frames, including:

Results

A data frame of bootstrap estimate of GLM coefficients, (intercept) and x, as well as beta diversity at zero distance (s0) and halving distance (halfd

Summary

A data frame of bootsrap summary statistics

Predictions

A data frame of distance difference, bootstrap predictions and 95% confidence intervals

CtrlList

Other information

Note

This function models the complement of dissimilarity (similarity).

Author(s)

Chih-Lin Wei <[e-mail]>

References

• Millar, R.B., Anderson, M.J., Tolimieri, N., 2011. Much ado about nothings: using zero similarity points in distance-decay curves. Ecology 92, 1717<U+FFFD><U+FFFD>1722. doi:10.1890/11-0029.1

See Also

beta.decay

Examples

# Ostrocode exponential distance decay
data(os)
dd <- dist.decay(gradient=os$dist, counts=os[, -1:-7], coords=os[, c("longitude", "latitude")], nboots=1000, dis.fun = "vegdist", method = "bray", like.pairs=T)
x <- vegdist(os$dist, method = "euclidean")
y <- 1-vegdist(os[, -1:-7], method = "bray")
plot(x, y)
lines(dd$Predictions[, "x"], dd$Predictions[,"mean"], col="red", lwd=2)

chihlinwei/ddecay documentation built on May 13, 2019, 4:48 p.m.