Capturing Feedback in Complex Marine Ecosystems: Two Models.

Teresa R. Johnson

Abstract

Marine systems are complex and highly variable. Feedback is required to learn in and manage these systems. Unfortunately, feedback in complex marine systems is difficult to capture and ambiguous. Feedback is a function of system structure. Conventional fisheries management simplifies this structure by focusing on individual species. It assumes that variability in populations is due solely to changes in the adult population (i.e., that a stock-recruitment relationship exists) and all necessary feedback is available by simply observing the size of the adult population. Unfortunately, this approach does not consider the environment of the species and most marine stocks show poor functional stock-recruitment relationships. Complex systems (hierarchy) theory suggests that this approach may not be the most appropriate way to simplify the system. Hierarchy theory simplifies the system on the basis of nearly decomposable subsystems, whose boundaries are defined by rates of interactions. It implies that feedback can be captured more readily within than between subsystems and that there is more pattern stability at the subsystem level than at the species level in the system. This implies that feedback is best captured from subsystems, not from changes in the abundance of individual species. Fishermen have traditionally dealt with variability by utilizing one of two harvesting strategies. With the Little Box approach, fishermen target a single species across multiple subsystems, averaging the variability. With the Big Box approach, fishermen target multiple species within a subsystem, relying on the relative stability of that subsystem. We hypothesize if less noise exists at the subsystem level than at the species level, then better feedback can be gained with a Big Box approach to management. Two age-structured, multispecies, bioeconomic models were created to explore the two harvesting rights regimes described above in terms of how well they allow decision makers to capture and respond to feedback. The Little Box model allocates rights to single species across subsystems. The Big Box model allocates rights to multiple species within a subsystem. Results of the models illustrate significant advantages for sole ownership compared to open access. The baseline model assumes a sole owner with perfect abilities and high quality and timely information. Results show little difference between Big Box and Little Box management under these "perfect" conditions. The model was then run under various scenarios to reflect imperfect feedback conditions (e.g., measurement errors, delayed response times). Impairment of the sole owner's ability to respond to feedback and degradation of the quality of feedback resulted in advantages for the Big Box sole owner over the Little Box sole owner.