Information Technology Reference
In-Depth Information
Fig. 14.5. The dynamics of network assembly. The gure illustrates the turnover in number of
nodes and links between two consecutive days. Knowing whether new species tend to become
attached to the already well-connected species is important in bridging structural patterns of
mutualistic networks to their underlying assembly dynamics. (Figure courtesy of J. Olesen.)
ity distribution (as opposed to a power-law) naturally arises. Similarly, Santamar a
and Rodr guez-Girones (2007) have explored a suite of assembly models, concluding
that the best t to the structure of mutualistic networks is performed by a model
that combined species abundance and phenotypic complementarity between mor-
phological traits. In line with this previous result, phenotypic complementarity and
the hierarchical evolutionary relationships between species have also been adduced
to be a good explanation for the values of nestedness observed in nature [Rezende
et al. (2007a)]. This is in agreement with conclusions from observations from the
eld [Stang et al. (2007)]. A combination of species abundance and niche-traits
(e.g., morphological constraints) seems also to account for the levels of nestedness
observed in nature [Krishna et al. (2008)].
We need general models that account not only for a network property such as
nestedness but for a broad combination of network patterns, or, as Allesina et al.
(2008) have recently put it, for the likelihood of reproducing an entire mutualistic
network. We believe it is quite feasible to soon have such a generalized model
of assembly of mutualistic networks which is able to reproduce the full range of
network patterns discovered in the initial studies of mutualistic networks.
14.3.3. Models of mutualistic-network disassembly
A more dicult challenge | and a much more interesting question, as tends to
be the case | would be to develop an experimental setting to examine the disas-
sembly of mutualistic networks as a function of some external driver. This is quite
dicult to realize in the eld, and even in a mesocosm experiment, yet is probably
the most relevant question from the point of view of exploring the consequences of
global change on biodiversity. The most exciting possibility would be an experi-
mental system where one could simulate the eects of species extinction or habitat
