Environmental Engineering Reference
In-Depth Information
mentioned earlier. Indeed, nearly all organisms are
embedded in networks of interactions with other
organisms consisting of literally thousands of species
in many cases; see Olff
et al
. (2009) for an overview of
such 'ecological interaction webs'. Dealing with such
complexity requires much more than merely doing
inventories, or summing up detailed reductionistic
sources of information, however useful that may be.
Over and above such ' alpha - level ' information, restora-
tion ecologists need to be concerned with so-called
emerging properties of the biotic community in the
ecosystem they are concerned with.
One of the fi rst attempts to understand the implica-
tions of ecological interactions within a network of
species was launched in the late 1950s by the discus-
sion on why the world is 'green'. Hairston, Smith and
Slobodkin argued, after weighing against alternative
mechanisms, that the world is green because herbivore
density is generally controlled by predators (the so-
called
HSS hypothesis
; Hairston
et al
. 1960 ). Reduction
of herbivore numbers by predation would release
plants from herbivore control and allow them to reach
a high density, explaining why the (terrestrial) world
generally has a green appearance. In spite of being
fi ercely criticized and discussed (Murdoch 1966), the
notion that predators can indirectly affect the density
of organisms that are the food of the prey they eat (e.g.
two trophic levels lower in the community) has stimu-
lated much research, and led to the development of the
concept of a 'trophic cascade' (Paine 1980), which
implies that consumption not only affects the prey but
also affects still lower trophic levels. Trophic cascades
have subsequently been suggested for various aquatic
ecosystems (e.g. Carpenter
et al
. 1985 ; Power 1992 ),
and terrestrial ecosystem (e.g. Jefferies 1999; Beschta
& Ripple 2009). A recent topic by Terborgh and Ested
(2010) examines trophic cascades in many of the
world's major biomes, including several that are dis-
cussed in Part 3 of this volume. Another interesting
view upon the HSS hypothesis has been brought
forward by Bond and Keeley (2005), who consider fi res
as a sort of 'selective herbivory'. Indeed, as these
authors state, fi re has been burning ecosystems for
hundreds of millions of years, shaping global biome
distributions and signifi cantly altering plant biomass.
The effects of fi re on, for example, plant competition in
consumer-controlled ecosystems are consistent with
the predictions of the HSS hypothesis.
Contrary to what was assumed by scientists previ-
ously, the theorist Robert May pointed out in 1973
(May 1973) that more complex food webs are less stable
(e.g. in terms of their ability to return to equilibrium
after disturbance) than simple ones. His models of food
webs moreover suggest that the
interaction strength
(in
terms of its per capita effect on the prey or the predator)
is of crucial importance to food web stability, and that
weak interactions can, in principle, compensate for
increased food web complexity. May's pivotal work
strongly stimulated research into food webs in a wide
range of ecosystems, and led to important insights into
the processes and properties that stabilize ecological
communities (Pimm 1984). First, food webs were
found to be strongly compartmentalized, for example
interactions within food webs are grouped in smaller
subcommunities, where interactions between species
within a compartment are stronger than between
species of different compartments (Krause
et al
. 2003 ).
This reduces the linking intensity within food webs,
and makes the food web more stable and robust. Other
studies revealed that the relative strength of interac-
tions within food webs follows a pattern: weak bottom-
up effects occur at lower trophic levels, while weak
top-down effects predominate at higher trophic levels
(de Ruiter
et al
. 1995 ; Neutel
et al
. 2007 ). Modelling
exercises of the latter authors showed that the latter
pattern strongly improves the stability of food webs,
when compared to models where strong and weak
interactions are randomly distributed across the web.
Up to the late twentieth century, ecological theory
focused on the distribution of feeding relations within
communities. Obviously, species can interact in a
number of alternative ways, as described in the previ-
ous section. How does this affect our understanding
of the network structure and dynamics of ecological
communities? Studies on mutualistic networks
between plants and animal pollinators and seed dis-
persers revealed that these networks have a different
structure than consumptive networks (e.g. food webs);
they are highly nested, where more generalist plants
and animals interact with each other to form a dense
core of interactions, to which the more specialist
species are attached (Bascompte
et al
. 2003 ). In other
words, mutualistic networks form a nested structure
rather than a compartmented structure that is typical
of food webs (Figure 6.5). How does this affect com-
munity dynamics? Nestedness was found to reduce
interspecifi c competition and enhance the number of
coexisting species. Hence, the structure promotes
species persistence and structural stability, similar to
the importance of compartmentation and interaction
