Steele, J.H.1and A.R. Beet1
1Marine Policy Center, Woods Hole Oceanographic
Institution, Woods Hole, MA 02543
2tbd
Linear network analysis of food webs is often carried out by
defining the links in the web and calculating the flux of energy,
nutrients or biomass through the nodes. The problem is that we
generally know more about the nodes than the links. To remedy this,
a form of inverse modeling has been used where the links are the
unknowns except for upper and lower bounds provided by available
data. Because of the large number of unknowns relative to
equations, a "global" constraint or objective function is required.
Previous inverse models have minimized the sum of squares of the
fluxes, but this does not have an obvious ecological basis. As an
alternative we have applied the concept of resilience by calculating
the network that maximizes the rate of return of the system after
perturbation. We show that the structures determined for a pelagic
microbial food web with these alternative objective functions can be
quite distinct. We discuss the ecological implications and propose
possible applications.
A powerpoint version and an html version of this presentation are available on-line.