Information Technology Reference
In-Depth Information
which appear within empirical food webs more or less often than expected at ran-
dom. They are unable, however, to address if there is a dynamic justication for
why a particular subgraph appears with greater or less frequency.
It thus follows that a signicant unanswered question in the study of food-webs
is the integration of the bottom-up and top-down approaches to better understand
food-web dynamics and the origins of food-web stability. Is there a direct rela-
tionship between the \local" stability of a subgraph and on whether or not that
subgraph appears more or less frequently in a complete food web? If no direct
relationship exists, it is worth noting that this does not preclude the existence of a
relationship at a \mesoscale" level of combinations of motifs.
The identication of each modules' contribution to community stability will
provide crucial information regarding the forces acting upon the ecosystem. We
will be able to answer an important question regarding the mechanisms responsible
for maintaining stability in the presence of external perturbations. Namely, is a
community stable because it is composed of stable sub-elements, or is a community
stable because of cooperative and synergistic interactions between individual, and
potentially unstable, sub-elements? Do species participate in interactions which
would maximize their own persistence or instead that of the entire community?
14.2.3. Whole food-web dynamics
An ever-present problem in the study of empirical food webs is the relative scarcity
of empirical data. Much of this stems directly from the diculties involved in
collecting data of high quality [Paine (1988)]. These diculties including the long
hours required for direct observation and for data collection to conduct stomach
content analysis or scatology. Additionally, it is still an issue to know when a
scientist may faithfully declare any data as complete; how does one account for the
brief and sole appearance of a migratory bird, for example, within the environment
under investigation? Consider also that these diculties are faced when the food
web will ultimately represent an aggregate over time and space [Lawton (1989)] and
therefore the result is generally absent of dynamic data.
To date, the most frequently utilized approach to quantitatively measure a dy-
namic food-web property is through the interaction strength [Berlow et al. (2004)].
Most commonly this is represented as the fraction of a species' diet or incoming
biomass which comes from particular prey; however, many denitions exist Berlow
et al. (2004). While such interaction strengths are informative and can even help
parametrize dynamics models [Yodzis and Innes (1992); Brose et al. (2006)], they
themselves often represent averages across time and space. This is because of the
fact that ecosystem dynamics necessarily implies dynamics of all aspects of the food
web, from abundances to species' interaction strengths.
For future characterization and understanding of community dynamics, it is
essential that we support initiatives for the collection of dynamic data for entire
ecosystems, potentially even to the scale of the Long Term Ecological Research
