Given a two-dimensional vegetation data frame and the x- and y-coordinates of the releves (the rows in the data frame) in geographical space, mantel correlation (function mantel
in the vegan package) is evaluated at regular intervals of direction. Direction versus correlation is plotted including 95 percent confidence interval.
1 2 3 4 5 6 7 |
veg |
A data frame of vegetation releves (rows) by species (columns) |
x.axis |
This is the x-coordinate in geographical space |
y.axis |
This is the y-coordinate in geographical space |
step |
The step length in degrees, used to draw the above mentioned function |
x |
An object of class "dircor" |
... |
Further variables used for printing |
The method presently uses correlation as distance, as.dist((1-cor(t(sveg^2.0)))/2), as distance measure for vegetation releves.
An output list of class "dircor" with at least the following intems:
steps |
The explicit steps used in degrees, 0 - 180 degrees |
mean.correlation |
Mantel correlation at each step |
lower.limit |
The lower confidence limits |
upper.limit |
The upper confidence limits |
Computation time is fairly long because function mantel() evaluates the confidence limits for each directional step between 0 and 180 degrees.
Otto Wildi
Legendre, P. & Fortin, M.-J. 1989. Spatial analysis and ecological modeling. Vegetatio 80: 107–138.
Wildi, O. 2013. Data Analysis in Vegetation Ecology. 2nd ed. Wiley-Blackwell, Chichester.
The function mantel() in the vegan package.
1 2 3 4 |
Questions? Problems? Suggestions? Tweet to @rdrrHQ or email at ian@mutexlabs.com.
All documentation is copyright its authors; we didn't write any of that.