An Experimental Examination of Ecosystem Linearization

Publisher Website
Google Scholar
Abstract
The hypothesis that ecosystems linearize their responses to environmental perturbations was tested using microcosms derived from Narragansett Bay, Rhode Island [USA]. The microcosms consisted of 150 l of water kept at ambient bay temperature and a scaled light intensity, and each contained a benthic sediment core with microcosm water circulating across the sediment surface. Previous experiments established that the phytoplankton and zooplankton were sensitive to turbulence. Consequently, a close approximation to a sinusoidal variation in stirring paddle rotation speed (0-60 RPM) was used as a forcing function on the microcosms. The frequency of this sinusoidal input was 0.1 cycles/day. Phytoplankton, zooplankton and NH3 were measured as response variables. A linear response of these ecosystem components would be indicated if a single periodicity of 0.1 cycles/day were observed in their time-series data. These responses were tested for this periodicity using variance spectral and periodic regression analyses. When tested against the responses of control microcosms receiving constant turbulence inputs, the microcosms receiving the sinusoidal input exhibited statistically significant responses at a number of frequencies, none of which was the 0.1 cycle/day input. Similarly treated microcosms whose structural complexity was reduced by removal of their benthic components showed a less complex frequency response, but still significantly different from the forcing frequency. These results unambiguously demonstrate that these complex microcosms do not respond linearly to inputs. There is no evidence to indicate that the parent ecosystem, Narragansett Bay, responds in a linear manner.

This publication has 2 references indexed in Scilit: