Authors: Larry Stanislawski*, USGS - CEGIS, Tyler Brockmeyer, USGS - CEGIS, Ethan Shavers, USGS - CEGIS, Barbara P Buttenfield, University of Colorado-Boulder, Barry J Kronenfeld, Eastern Illinois University
Topics: Geographic Information Science and Systems, Quantitative Methods, Geomorphology
Keywords: hydrography, feature extraction, surface-water modeling, geomorphology
Session Type: Paper
Start / End Time: 8:00 AM / 9:40 AM
Room: Washington 1, Marriott, Exhibition Level
Presentation File: No File Uploaded
With continuing threats of severe weather, automated techniques to map surface water features for assessing floods and other environmental hazards or conditions have been a topic of widespread interest. Increased availability of lidar elevation data, such as furnished through the U.S. Geological Survey 3D Elevation Program, along with high-resolution image data, has also spurred interest in this topic. However, with higher resolution data, more complex workflows are needed to handle the additional details. For instance, natural surface water drainage patterns derived from 10-meter resolution elevation data are not altered by roads very much or at all, while those derived from 1-meter resolution elevation data exhibit much more pronounced alterations caused by road barriers. Consequently, more elevation conditioning is required to eliminate road effects for drainage features extracted from high-resolution elevation data. This work will identify and apply automated metrics for assessing geomorphic conditions that can help optimize workflows for hydrographic feature extraction and validation. Automated methods being evaluated for extracting hydrographic features include flow accumulation modelling, and neural network predictions using elevation and other data layers. Lower-order stream features are validated through a 3-dimensional cross-section assessment process. Geomorphic conditions are estimated using magnitude and scale of roughness. The frequency of small to large diameter bends of drainage valleys are estimated with an innovative fractal method. Several datasets distributed over various landscape conditions within the United States are evaluated.