rtcc3: Clustering signal significance.

In RTCC: Detecting Trait Clustering in Environmental Gradients

Description

For a given trait and environmental variable, this function creates a null model of the clustering/overdispersion pattern in order to test if the observed pattern statistically differs from the expected by random. For this, it sequentially remove random samples from the metacommunity and computes at each step the remaining metacommunity h-index. This index is based on the percentage of samples on a metacoomunity presenting significant trait clustering/overdispersion. After h iterations, computes a 95 obtained h-index for each point of the environmental gradient.

Usage

rtcc3(
  table1,
  table2,
  table3,
  species_abundances,
  trait_col_number,
  min_env_col,
  max_env_col,
  env_var_col,
  h_iteration,
  repetitions,
  model
)

Arguments

table1	A data frame containing organisms names on the first column and its trait values on the consecutive ones. It also has to contain two columns with the maximum and the minimum values of the tested environmental variable where the organisms have been observed.
table2	A presence-absence observations table with the organisms names on the first column and the sample names as consecutive colnames.
table3	A dataframe containing sample names on the first column and environmental parameters on the consecutive ones.
species_abundances	A vector containing the relative abundance of the organisms on the whole data set on the same order as appear on Table 1.
trait_col_number	Table 1 column number of the tested trait.
min_env_col	Table 1 column number indicating the minimum value of the environmental variable were each organism has been observed.
max_env_col	Table 1 column number indicating the maximum value of the environmental variable were each organism has been observed.
env_var_col	Table 2 column number indicating the tested environmental variable.
h_iteration	Number of h-index calculations for computing a confidence interval.
repetitions	Number of simulated synthetic communities distributions.
model	Model selection. All models build synthetic communities based on the organisms richness of the observed communities. - Model 1: organism are selected randomly from the global pool. - Model 2: organism are selected randomly with a probability based on its relative abundance on the global pool. - Model 3: organism are selected randomly, but only those whose environmental range includes the value of the simulated community are elegible. - Model 4: organism are selected randomly, but only those whose environmental range includes the value of the simulated community are elegible and the selection probability is based on its relative abundance on the global pool.

Value

The function returns a dataframe with the maximum value of environmental variable corresponding to the same number of samples on the ordered remova, h-index calculation for each environmental value, and the percentiles 0.025, 0.5 and 0.975 of the obtained distribution for each point (mean value and 95

Examples

data(group_information)
data(table_presence_absence)
data(metadata)
rtcc3(group_information, table_presence_absence, metadata, group_information$sums,
9, 12, 13, 2, 50, 20, model = 1)

RTCC documentation built on July 1, 2020, 7:06 p.m.