Ecological heterogeneity in streams: Why variance matters

Authored by MA Palmer, CC Hakenkamp, K NelsonBaker

Date Published: 1997

DOI: 10.2307/1468251

Sponsors: United States National Science Foundation (NSF)

Platforms: No platforms listed

Model Documentation: Other Narrative

Model Code URLs: Model code not found

Abstract

Variability in the spatial and temporal patterns of stream biota has been well documented, yet the causes and consequences of this variability are poorly understood. Most early work focused on how changes in mean environmental parameters over space or time influence the abundance or distribution patterns of biota. We argue for an increased focus on the study of variability in these parameters. To date, only a few lotic studies have used variance as a dependent variable i.e., showing that changes in the variance in some parameter can be related to changes in the mean of that or another parameter. These studies have suggested interesting links between ecological patterns and processes at a variety of scales. The use of variance as an independent variable and its experimental manipulation (i.e., hold the mean of some parameter constant while altering its variance) may also lead to new understandings of how pattern and process are linked. We begin by outlining when and why the study of variance may be important to lotic ecology. Next, we suggest a 3-step process in the use of variance in stream studies: 1) documentation of a pattern with respect to variance; 2) examination of descriptive data to explore potential causes of the pattern; and 3) direct manipulation of variance in experiments. We provide an example of this 3-step process using data from preliminary experiments in it warmwater stream that suggested spatial variability in the abundance of chironomids and copepods could be related to flow variability We close by suggesting that future directions for the use of variance in stream ecology should attempt to couple scale-dependent physical patterns with biological patterns or processes, develop experimental frameworks to determine if species interactions or other biotic processes vary with scale, embrace statistical methods for determining the relative importance of variance-generating processes, develop methods for the quantification of variance in biologically meaningful ways, and focus on the identification and interpretation of domains of homogeneous variance in streams.

Tags

Individual-based model Spatial heterogeneity Dynamics habitat ideal free distribution Variability Flow Disturbance Variable environment Larval settlement