Symmetry, Change, Perturbation, and Observation Mode in Natural Communities

Abstract

Like a kaleidoscope, nature presents continuously fleeting images of natural communities. It has been difficult to discern patterns of community structure in these elusive images. For both theoretical and applied considerations, however, holistic ecosystem properties need to be identified and quantified. An analogy of a kaleidoscope is used here to describe symmetry, change, and perturbation in the plankton community of Delaware Bay. Notions of symmetry are used extensively by physicists but much less so by biologist. For the ecological application give her, symmetry involves the notion of internal order and invariant pattern within the structure of natural communities. The concept of core structure is introduced to illustrate the inherent symmetry of plankton communities arising from consistent linkages and feedback relationships. The core is determined by using the most prevalent variables and links of a series of individual loop diagrams, each describing the community at a given point in time, to form a summary or composite model. Change refers to the seasonal alterations in community structure, whereas perturbation involves the source and entry location of environmental variation to the community network. Observing mode includes both the intrinsic (biological intuition) and extrinsic (loop analysis) models an observer uses to characterize symmetry, change, and perturbation. Loop analysis was used to model the Delaware Bay plankton community at 12 dates in an annual cycle. Loop diagrams consist of qualitative network models based on positive, negative, or zero interactions between variables in a pair for all pairs of variables. The fit of models to the data using the direct change technique resulted in 95% agreement. The core structure contained 18 variables in a three—tiered configuration with distinct subsystems of predators and small—sized algae and their associated herbivores. There was 3 nutrients, 5 algae, and 10 zooplankton variables in the core structure. It is proving to be consistent for several other marine communities. Various network properties related to community structure and stability were computed for the networks. The mean loop length varied from 1.6 to 2.8 variables of the individual networks; a network consisted of 22—31 loops of various lengths. Connectance ranged from 14 to 18%. The individual networks were stable; however, the stability of the core structure was questionable. The uniqueness problem, or verification of the single best—fit model, is presented as a concern central to all types of ecosystem modelling. The notion of three different—sized loop universes (sets of loop models) is introduced to illustrate one approach to this problem.