Abstract
This project was performed to improve understanding of key physical and ecological attributes of natural streams in peninsular Florida in order to derive a practical process-based classification system; and also to provide design aids to assist with stream restoration in rural settings. Fifty-six of the best remaining stream systems in the peninsula were selected for monitoring of more than 120 quantitative variables known to associate with key stream system processes. The sites covered a wide range of physical gradients including soil drainage condition, drainage area, and valley slope, and were observed at watershed, valley, channel and patch scales. Hierarchical cluster analyses were conducted on the full dataset and various subsets to derive the classification groups. Variables were winnowed to a small group explaining the vast majority of stream variability using principal components analysis. The classification system defines streams based on their hydrobiogeomorphology (HBG) and is hierarchical in scale; first categorizing an area within three watershed soil drainage conditions, then based on valley slope, and finally on channel and floodplain surfaces and their dimensions. Regional curve regressions were developed for use in natural channel design. Practical applications of the system for restoration design and construction are presented.
John H. Kiefer, Kristen B. Nowak, AMEC Environment and Infrastructure, Inc.; Joann Mossa, William R. Wise, and Kenneth M. Portier, University of Florida; Thomas L. Crisman, University of South Florida. June 2015.