FidelityDiv: Live-Dead differences in alpha diversity and evenness

In MJKowalewski/PaleoFidelity: Analyze and Visualize Agreement between Live and Dead Communities

Live-Dead differences in alpha diversity and evenness

Description

FidelityDiv provides estimates of differences in alpha diversity and alpha evenness between pairs of live and dead samples from single sites (or data pooled across multiple sites) using community abundance data. For datasets including multiple site, the function also returns means of live-dead differences across all sites or groups of sites.

Usage

FidelityDiv(
  live,
  dead,
  gp = NULL,
  report = FALSE,
  n.filters = 2,
  t.filters = 1,
  iter = 100,
  CI = 0.5,
  CImean = 0.99,
  outdata = FALSE,
  messages = F
)

Arguments

live	A matrix with counts of live-collected specimens (rows=sites, columns=taxa). Dimensions of 'live' and 'dead' matrices must match exactly.
dead	A matrix with counts of dead-collected specimens (rows=sites, columns=taxa). Dimensions of 'live' and 'dead' matrices must match exactly.
gp	An optional univariate factor defining groups of sites. The length of gp must equal number of rows of 'live' and 'dead' matrices.
report	Logical (default=FALSE) to print report generated by the function FidelitySummary
n.filters	An integer used to filter out small samples (default = 2, all samples with n > 1 kept). Note that for samples with n < 2, Hurlbert's PIE cannot be computed.
t.filters	An integer used to filter out rare taxa (default = 1, taxa >= 1 occurrence kept) Note that removing rare taxa (t.filters > 1) is not advisable when measuring evenness or standardized richness.
iter	An integer defining number of replicate samples (default iter=100)
CI	A numerical value (default = 0.5) defining confidence bars for individual sites. Note: 0.5 - plots bars representing inter-quartile ranges, 0.95 - plots 95 Confidence bars are estimated as percentiles of subsampled estimates of Delta S and Delta PIE.
CImean	A numerical value (default = 0.99) defining confidence bars for means of all sites or groups of sites (if 'gp' factor is provided). Note: 0.5 - bars representing inter-quartile range, 0.99 - plots 99 subsampled estimates of Delta S and Delta PIE based on n (=iter) replicate subsamples.
outdata	Logical (default = FALSE) to determine if data files should be included in the output
messages	Logical (default = FALSE) to enable printing notes generated internally by FidelitySummary function.

Details

FidelityDiv assesses live-dead offsets in evenness/diversity between pairs of sympatric live and dead samples using a bivariate approach described in Olszewski and Kidwell (2007). The estimates of offsets in alpha diversity and evenness are returned as two separate objects.

(1) x - Live-dead offsets in alpha diversity for individual sites. The difference is measured as the difference between natural logarithms of sample-standardized species richness of sympatric dead and live samples:

DELTA S = ln(S[dead]) - ln(S[live])

A negative value indicates that the sample-standardized alpha diversity of a live sample exceeds alpha diversity of the sympatric dead sample and a positive value indicates that a dead sample is more diverse than live sample. Confidence intervals and p-values for the two-tailed null hypothesis: Delta S = 0 are reported. A vegan function "rrarefy" is used to perform sample standardization

(2) y - Live-dead offsets in evenness for individual sites. The difference is measured as the difference between estimates of Hurlbert's PIE for live and dead samples:

DELTA[PIE] = PIE[DEAD] - PIE[LIVE]

A negative value indicates that evenness of live samples exceeds evenness of the sympatric dead sample and a positive value indicates that a dead sample is more even than live sample. Confidence intervals and p values for null hypothesis: Delta PIE = 0 are reported.

Value

A list containing the following components:

live	The post-processed version of 'live' data matrix used in all analyses
dead	The post-processed version of 'dead' data matrix used in all analyses
gp	The post-processed version of 'gp' factor, when provided
x	DELTA S values for each live-dead comparisons (site-level differences in sample standardized species richness)
y	DELTA PIE values for each live-dead comparisons (site-level differences in evenness estimated as Hurlbert's PIE)
xmean	Grand mean of Delta S
ymean	Grand mean of Delta PIE
xgp	Group means of Delta S (when 'gp' factor provided)
ygp	Group means of Delta PIE (when 'gp' factor provided)
p.values	p.values for Null H: Delta.S.p=0 and Delta.PIE.p=0
p.gps	per-group p.values for Null H: Delta.S.p=0 and Delta.PIE.p=0

References

Olszewski, T.D., and Kidwell, S.M., 2007, The preservational fidelity of evenness in molluscan death assemblages. Paleobiology 33: 1:23.

Examples

my.fid <- FidelityDiv(FidData$live[6:9,], FidData$dead[6:9,],
FidData$habitat[6:9], n.filters=20, iter=100, CI=0.95)
my.fid$x # site-level estimates of Delta S with 95% CIs and p values
my.fid$p.gps # p values for means of groups
AlphaPlot(my.fid, col.gp=c('forestgreen', 'coral1'), bgpt='beige')

MJKowalewski/PaleoFidelity documentation built on Aug. 25, 2024, 8:27 p.m.