Description Usage Arguments Details
The function fits community weighted mean of species attributes (e.g. traits), calculated for individual samples, onto an ordination, using multiple regression. The relationship is tested by permutation test (defaults to modified permutation test sensu Zeleny & Schaffers 2012).
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 |
ord |
An ordination object or other structure from which the ordination scores can be extracted (including a data frame or matrix of scores). |
cwm |
Community weighted means calculated by function |
wstand |
Logical; should be the cwm calculated on weighted standardized species attributes? Defaults to the same setting as used in |
permutations |
Number of permutations for the permutation test. |
choices |
Axes to be plotted. |
display |
In fitting functions these are ordinary site scores or linear combination scores ("lc") in constrained ordination (cca, rda, capscale). In scores function they are either "vectors" or "factors" (with synonyms "bp" or "cn", resp.). (Description from vegan::envfit help file). |
w |
Weights used in fitting (concerns mainly |
na.rm |
Remove rows in ordination scores or |
test |
Permutation test. |
adjustP |
Logical; should the P-values be adjusted for multiple testing? |
p.adjust.method |
Character string; method of P-value adjustement (check |
p.adjust.n |
Integer; what is the number of comparisons used to adjust the P-values? Default ( |
... |
Parameters passed to |
The function is derived from the function envfit
in library (vegan)
, which fits the environmental vectors or factors onto an ordination and tests their significance using permutation test. The modification is in the permutation schema used to test the fit significance: while the original vegan::envfit
function builds the null distribution of test statistic (r2) by permuting the values of environmental vector (analogy of standard
or rowbased
test), the function weimea::envfit_cwm
additionally offers also modified
or rowbased
test, based on permuting species attributes prior to calculation of cwm
, and max
test combining results of rowbased
and colbased
as P_max = max (P_row, P_col)
. While the null hypothesis of standard
(rowbased
) permutation test that there is "no statistical relationship between mean indicator values and sample ordination scores" is easy to reject in case that CWM's are numerically derived from the same species composition matrix as sample ordination scores (this applies even if some kind of transformation, e.g. log or presence-absence, was applied on the species data before calculating either ordination or CWM), since such relationship is spurious. The null hypothesis of the alternative modified
(rowbased
) permutation test is that there is "no statistical relationship between information in species attributes and sample ordination scores", which is not spurious if species attributes (traits
) are extrinsic to species composition (i.e. not numerically derived from it). The third, max
test, combines these two results together, and is probably not meaningful in the context of relating CWM of species attributes and sample ordination scores, because it includes the standard
(rowbased
) test which has no meaningful null hypothesis (as explained above).
The permutation test requires information about the original vegetation matrix used to calculate the weighted mean of species indicator values, and also original species attributes, which is stored in the object of cwm
class as 'com'
and 'traits'
attributes. If the argument cwm
is not the object of cwm
class, the function will calculate standard vegan::envfit
and return warning message.
The generic functions print
, plot
and scores
for the class envfit can be used to print the results, plot the vectors of CWM onto an ordination diagram and to return the fitting results. Check ?envfit
for details which arguments can be used in these functions.
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.