Stream Ecosystem Theory: A Global Perspective

Consideration of the adequacy of the conceptual basis of stream ecology is facilitated by a comprehensive (global) perspective focusing on the scales of space and time. The full lineal (10^-8 to 10⁷ m) and temporal (10^-7 to 10³ yr) dimensions faced by stream ecologists each extend over 16 orders of magnitude. Examination of both dimensions reveals weaknesses in the process and product of stream ecology through which existing theory is derived. As knowledge about streams developed and the thinking of lotic ecologists matured, conceptualization of stream ecosystem boundaries expanded out from, into, and along the channel resulting in a corresponding change in perception of the factors responsible for the structure and operation of stream ecosystems. Selection of the size of the sampling unit relative to the total sampling area of interest can profoundly influence one's perception of pattern and process in stream ecosystems. Failure to use an appropriate sample scale relative to the size of the intended "universe" may have resulted in some erroneous ideas about ecological relationships in flowing water. Stream ecologists have focused mainly on temporal scales of days, seasons, and years. Other ecologically meaningful scales, such as day/night, diel, and lunar intervals, have been examined much less frequently. Several factors indicate that progressive changes in stream ecosystem dynamics on the order of a hundred years or more may occur. This has important implications in terms of the proper interpretation of results isolated in time ("instantaneous") as well as the development of predictive capabilities for the management of streams. Space and time also function together to shape lotic communities and ecosystems at scales ranging from short-term/local to evolutionary/global. Areas where the simultaneous consideration of space and time in streams is important include longitudinal succession, climate, and disturbance. It is concluded that the idea of longitudinal succession on a geological time scale is wrong but that the concept when considered at the level of the flow-through time of a river is not. Climate operates at several scales of space and time; larger-scale aspects beyond the stream-specific level are just beginning to be addressed but their consideration is necessary if broad-scale patterns in stream ecosystems are to be discerned. Southwood's habitat templet lends itself to the development of hypotheses and the ordering of knowledge along spatiotemporal axes. Application of this model to streams is illustrated for flow-related disturbance. Several misconceptions concerning disturbance in streams seem to have arisen from failure to consider the full range of spatial-temporal dimensions. The explicit recognition and utilization of appropriate spatial and temporal scales is essential to the development of an accurate stream theory having a truly world-wide perspective.