Authors: Larry Stanislawski*, USGS - CEGIS, Ethan J Shavers, USGS - CEGIS, Barbara P Buttenfield, University of Colorado-Boulder, Barry J Kronenfeld, Eastern Illinois University
Topics: Water Resources and Hydrology, Geomorphology, Geographic Information Science and Systems
Keywords: geomorphology, National Hydrography Dataset, stream sinuosity, Fourier transform
Session Type: Virtual Paper
Start / End Time: 9:35 AM / 10:50 AM
Room: Virtual 3
Presentation File: No File Uploaded
Geomorphic conditions of a stream environment reveal information about the hydraulic and biogeochemical rates of exchange between streams, groundwater, and the atmosphere. Landscape geomorphic conditions can inform a variety of natural resource evaluations, such as flood inundation, landslide and biodiversity studies. Stream meander patterns are affected by terrain slope, stream flow, sediment loads, and resistivity. Classification of stream meander pattern is a standard variable of stream landform categorization strategies. However, an efficient approach to classify meander patterns of streams within large regions is not available. The National Hydrography Dataset (NHD) provides a comprehensive database of vector representations of the surface water features in the United States. We investigate two approaches for automated classification of stream meander patterns for single-line stream features represented by NHD flowlines. The first approach applies the scale-specific sinuosity (S3) metric, which is derived from the Richardson plot. The Richardson plot is an accepted method for estimating the fractal dimension of a natural line feature that compares feature length for increasing step size (or stride length), displayed on a log-log plot. At each step interval, S3 is computed as -1 times the slope between two adjacent Richardson plot points. It is shown that S3 values are related to the frequency of meanders over a range of meander wavelengths. For comparison, a second method applies Fourier analysis to measure the relative frequency of meander wavelengths in vector stream features. For verification, methods are applied to a series of features simulated from several forms of sine wave functions.