Landscape position, scaling, and the spatial and temporal variability of ecological parameters: considerations for biological monitoring
Designing and implementing an effective biological monitoring program requires making appropriate choices of what and where to measure. The parameters selected for measurement must obviously be sensitive to the environmental stress or stresses anticipated to occur. However, a change in a parameter in a stressed system must also be distinguishable from the natural variability exhibited by that parameter in an unstressed system. To the extent that sensitive parameters also tend to be more variable, choosing which parameters to monitor will involve a balance between these factors (Frost et al., 1992). Another important consideration in the design and implementation of a biological monitoring program is the ability to extrapolate results to a larger number of locations than can be measured directly. One factor directly influencing the ease to which results can be scaled up to broader regions is the degree to which sytems in different locations behave similarly through time. Thois property, which we term temporal coherence (Magnuson et al. 1990), is a useful concept because the more coherent different locations are, the easier it is to generalize about specific regional responses to environmental stress. Therefore, an ideal parameter is one that is sensitive to the ancitipated stress, has low variability (natural or unpredictable), and is temporally coherent among different locations in the landscape. ... In this paper we assess the sensitivity, variability, and coherence of selected limnological parameters.
CRC Press, Inc.
Boca Raton, Florida