Multiple Patterns of Food Web Dynamics Revealed by a Minimal Non-deterministic Model

详细信息    查看全文

• 作者：Ulf Lindstrøm ; Benjamin Planque ; Sam Subbey
• 关键词：Barents Sea ecosystem ; trophic interactions ; stochastic ; mass balance ; density dependence ; pattern ; oriented modelling
• 刊名：Ecosystems
• 出版年：2017
• 出版时间：January 2017
• 年：2017
• 卷：20
• 期：1
• 页码：163-182
• 全文大小：
• 刊物类别：Biomedical and Life Sciences
• 刊物主题：Ecology; Plant Sciences; Zoology; Environmental Management; Geoecology/Natural Processes; Hydrology/Water Resources;
• 出版者：Springer US
• ISSN：1435-0629
• 卷排序：20

文摘

Understanding and predicting patterns arising from the dynamics of marine food webs is central to trophic and community ecology and numerical models of food webs constitute a primary tool to simulate these dynamics. Food web simulation models are often highly complex while at the same time often too constrained to reproduce the level of variability observed in real systems. The recently developed non-deterministic network dynamics (NDND) modelling framework has been suggested as a simulation alternative, which can generate multiple patterns of food web variability despite great structural simplicity. Two important aspects of the NDND modelling framework remain unexplored: first the derivation of model input parameters from empirical or theoretical studies and second the evaluation of the model simulations against observations. We provide a methodology for the derivation of model parameters based on empirical observations, the metabolic theory of ecology and life-history theory and apply it to the specific case of the Barents Sea food web. We then evaluate the ability of the NDND simulations to reproduce a wide range of patterns of food web dynamics against observations collected in the Barents Sea during 28 years. Patterns emerging from the simulations include trends and cycles in biomass, trophic levels and transfer efficiency, density-dependent growth, top-down vs bottom-up oscillations, ecosystem level stability and synchrony and trophic functional responses. The ability of the NDND to generate so many patterns observed empirically in the Barents Sea is remarkable given that it is based only on random trophic interactions operating within few constraints set by ecological rules. Our results show that investigations of food web dynamics in marine ecosystems, including the definition of reference states and responses to climate and exploitation pressures, may be achieved with models that are structurally simple and based on few well-established assumptions.