A diversity based fuzzy systems approach to ecosystem health assessment

Abstract

Ecosystem health assessments generally rely on combining information from variable-oriented surveys and case-studies, which at their extreme represent the ends of an inverse relationship between variables and instances. However, many of the underlying assumptions of both types of analysis, such as homogeneity of systems, simple chains of causality and the representativeness of data, are sufficient to pose fundamental problems in any overall interpretation and analysis. In this paper, an alternative approach is presented which bridges the two extremes and explicitly deals with uncertainty, beliefs and ignorance, complex causality, sufficiency and necessary conditions and diversity amongst classes of attributes and configurations. It adopts a diversity-oriented approach, using fuzzy set theory and logic to construct multiattribute configurations to be able assess different health outcomes. An example of a diversity-based health assessment is presented for the North Sea ecosystem. The results show how sectoral and social benefits over the past 45 years have accrued at the expense of the ecosystem, as evidenced in sets of multiattributes for biodiversity, trophic stability and pollution, despite gains in governance and social cohesion.