In tributaries impacted by non-point source pollution, flux rates often are proportional to flow. Since adequate pollutant concentration data are lacking for many river systems, but flow data are often available, measures of flow variability may be the only source of information for planning monitoring programs or evaluating the behavior of tributaries on a basin-wide scale. Seven measures of flow variability were developed, and their characteristics were examined using flow data from Great Lakes tributaries. Classifications of the tributaries based on flow variability were compared with an event-responsiveness classification of earlier investigators. Lack of agreement is conspicuous, and can be attributed in part to an inadequate chemical data base for many rivers when the earlier classification was established. Cluster analysis was used to reveal the natural groupings of 118 U.S. and Canadian Great Lakes tributaries, based on measures of flow variability. The resulting classification of the tributaries is geographically coherent, and demonstrates that large areas of high flow variability are found in the western and central Lake Erie drainage basin, probably as a result of intense agricultural land use in association with heavy soils. Tributaries draining urban areas also tend to have high flow variability. Most of the tributaries which drain the northern Great Lakes watershed have low flow variability. © 1990, International Association for Great Lakes Research. All rights reserved.
