decay.model()
now allows fitting the Gompertz
function (as described in Martín-Devasa et al. 2022, Diversity and
Distributions), in addition to negative exponential and power law
functions. To do this, nonlinear models are now fitted via the
nls.lm()
function in the minpack.lm package
(which uses the Levenberg-Marquardt Nonlinear Least-Squares Algorithm),
instead of glm()
as in the previous versions of
betapart
boot.coefs.decay()
now implements the site-block
resampling method introduced in Martínez-Santalla et al. 2022 (J.
Biogeogr.)
A new function, zdep()
allows assessing the
significance of differences between parameters of two distance-decay
models, as introduced in Martín-Devasa et al. 2022 (Ecol.
Informatics)
plot.decay()
was modified to handle the new
decay.model()
function
decay.model()
to implement the block
permutation described in Martínez-Santalla et al 2022 (J. Biogeogr.) for
assessing the significance of the distance decay modelfunctional.betapart.core()
was fixed to run it
in parallel with multi = TRUEfunctional.betapart.core.pairwise()
to get
vertice coordinates in the output details
functional.betapart.core()
was updated. Options can
be passed to qhull to prevent some crashes and a progress bar can be
displayed. When setting multi=TRUE, the function stop earlier if the
number of communities is too important.
New function to control options passed to qhull for convexhull
estimation: qhull.opt()
New function to compute rapidly pair-wise dissimilarty matrices:
functional.betapart.core.pairwise()
functional.beta.pair()
was updated to integrate
functional.betapart.core.pairwise
functional.betapart.core()
to allow internal
parallel computingbeta.para.control()
functional.betapart.core()
to allow parallel
computingdecay.model()
fits a negative-exponential or mower
law function describing the decay of assemblage similarity with spatial
distance.
plot.decay()
allows plotting the curves fitted with
decay.model()
.
boot.coefs.decay()
bootstraps the parameters of the
functions fitted with decay.model()
.