Abstract
The idea that much environmental information may be gained by identifying underlying environmen tal system behaviours is examined. Six categories of system behaviour for changes between and within steady-states are identified: disorder maximization; self-organization; linear process-response; stochastic; self- organized criticality; and chaotic. In practical terms, predictability only exists for linear behaviour. Simple frequency distributions of time-series data are used to identify dominant behaviour in the proxy-records from 10 000 years of annual laminations from Holzmaar, Germany. The results suggest that before ~3000 yr cal. BP the catchment-lake system may have reached a state of self-organized criticality where fluctuations in system properties, for instance diatom productivity and sediment flux, were caused by internal reorganization as much as by external forces. Comparison of the records for sediment flux with outputs from the GOLEM landscape model suggests that major shifts in the Holocene record were triggered by human actions rather than by climate. The approach offers a new perspective on the causes of minor and major changes in reconstructed environmental systems and the degree of interactions between subsystems.
Keywords
Get full access to this article
View all access options for this article.
