Qualitative modelling and indicators of exploited ecosystems

Abstract

Implementing ecosystem‐based fisheries management requires indicators and models that address the impacts of fishing across entire ecological communities. However, the complexity of many ecosystems presents a challenge to analysis, especially if reliant on quantification because of the onerous task of precisely measuring or estimating numerous parameters. We present qualitative modelling as a complementary approach to quantitative methods. Qualitative modelling clarifies how community structure alone affects dynamics, here of exploited populations. We build an array of models that describe different ecosystems with different harvesting practices, and analyse them to predict responses to various perturbations. This approach demonstrates the utility of qualitative modelling as a means to identify and interpret community‐level indicators for systems that are at or near equilibrium, and for those that are frequently perturbed away from equilibrium. Examining the interaction of ecological and socio‐economic variables associated with commercial fisheries provides an understanding of the main feedbacks that drive and regulate exploited ecosystems. The method is particularly useful for systems where the basic relationships between variables are understood but where precise or detailed data are lacking.

Qualitative mathematical models to support ecosystem‐based management of Australia's Northern Prawn Fishery

“…Fishing fleet searching for dense banana prawn schools (boils) in Gulf of Carpentaria, Australia; fishing nets are held at surface in readiness to ''shoot away'' on banana-prawn boil when located. Photo credit: P. C. Rothlisberg. developed in Dambacher et al (2009). Here we also made a rudimentary attempt to account for spatial dynamics in a fishery by defining the feedbacks associated with fisher behavior and the local allocation of effort.…”

“…Erring on the side of toosimple leads to trivial results, whereas models with extraneous detail can be unwieldy and less informative. Here we sought the relevant subsystem that encapsulated the essential dynamics of the system and was tailored to the specifics of the problem at hand (Dambacher et al 2009). The goal is a parsimonious representation of the feedbacks that encompass the variable(s) of concern and the source of the perturbation.…”

“…In contrast to these quantitative approaches, qualitative mathematical models (Levins 1998) are based on a general understanding of the relationships that connect ecosystem variables (Puccia and Levins 1985, Dambacher et al 2009, Ortiz and Levins 2011, Carey et al 2013). These models contain only the sign (þ, À, 0) of species interactions,…”

“…In considering these economic components of a fishery, the fishing-fleet variable can be expanded into a fishery subsystem (Fig. 5) that includes catch (C ), fishing effort (E), and market price (M ), as distinct model variables (Dambacher et al 2009). In the expanded fleet-stock model ( Fig.…”

Inference reasoning on fishers' knowledge using Bayesian causal maps

“…This model based on timed-automata, was successfully used for scenarios testing and possible futures querying. In fact, aquatic systems have been repeatedly modeled and analysed in a qualitative manner, with various approaches including loop annalysis [Dambacher et al 2003], [Dambacher et al 2009]. …”