A watershed-based classification system for lakes in agriculturally-dominated ecosystems: A case study of Nebraska reservoirs.

摘要

The U.S. Environmental Protection Agency is charged with establishing national standards and criteria for assessing lake water quality. It is increasingly evident that a single set of national water quality standards that do not take into account regional hydrogeologic and ecological differences will not be viable. Lakes clearly have different inherent capacities to meet such standards. The principal objective of this study was to define and test a watershed-based classification procedure for identifying groups of lakes that have similar potential capacity to meet proposed water quality standards. The strategy employs variables such as watershed area, mean watershed slope, soil organic matter, soil pH, and soil erodibility. This study focused on reservoirs in Nebraska, an agriculturally-dominated area of the United States. A preliminary cluster analysis of 78 reservoirs was performed to determine the optimal number of Nebraska reservoir groups. Subsequently, a Classification Trees method was used to describe the structure of reservoir watershed classes and to develop a predictive model that relates watershed conditions to reservoir classes. Results suggest that Nebraska reservoirs can be represented by nine optimal classes, and that soil organic matter content in the watershed is the most important single variable for segregating the reservoirs. The cross-validation prediction error rate of the Classification Tree model was 26.33 percent. The Classification Tree-based watershed classification was then compared with discriminant function analysis (DFA)-based reservoir watershed classification, as well as ecoregions-derived reservoir classes. Overall, both watershed-based classifications were more effective than ecoregions in accounting for variations in lake water quality characteristics. Furthermore, Classification Trees approach was more suited for the ecologically complex datasets than the DFA classification method. Because all geospatial data used in this study are available nationally, the procedure can be adopted throughout the United States. Through model refinement, the Classification Tree interface for watershed-based reservoir classification promises to provide water resources managers an effective decision-support tool in the management of reservoir water quality.