黑料网

Abstract

Ecology has probably borrowed tools and concepts from physics since its origins. Powerful physics approaches have particularly helped incorporate challenges related to ecosystems, including ecosystem functioning and scale issues. I conducted a survey of physical theories and concepts applied to ecosystem ecology to identify fruitful borrowings and past traps. I left aside differential equations and all mathematical tools developed in physics but also used in ecology. Building on information theory, thermodynamics and statistical physics on the one hand, and on dynamical systems, self-organisation, and complexity on the other, my first aim was to identify a trend in this longlasting collaboration between physics and ecology. For example, some physical concepts are now widely recognisedto have failed to help understand and/or to manage an ecosystem as a whole:  information theory, thermodynamics and extremal principles belong to this category. More recent physical theories have emerged in ecology and not yet failed: dynamical systems and statistical physics, complexity and graph theories belong to this category.

The second aim of the survey was to identify some of the reasons for the only partial success of otherwise powerful physical concepts in ecology. The ecosystem is a dual object composed of living (biotic) and inert (abiotic) components in close interaction. Although a basic tenet, an ecosystem cannot simply be understood, in practice, as a purely physical (or purely biological) system. Consequently, a difficult theoretical question needs to be addressed: could a revival of the interface between physics and ecology finally reveal how to understand and manage ecosystems? Or will we need radically new concepts (and more generic tools) to understand the ecological organisation of matter and energy in an ecosystem? Strategies for achieving this goal and for advancing theoretical ecosystem ecology are discussed.

Keywords: Thermodynamics; Statistical physics; Complexity; Dynamical system; Graph; Information; Entropy; Self-organisation