Authors: Leo Harry Meirose, III*, Quantum Spatial, Barnali Dixon, University of South Florida Saint Petersburg, Christopher Brown, Worcester Polytechnic Institute, Alvan Karlin, Dewberry
Topics: Geographic Information Science and Systems, Water Resources and Hydrology, Environmental Science
Keywords: DEM resolution, resampling, stream delineation, fractal analyses, statistics, spatial location analysis
Session Type: Paper
Presentation File: No File Uploaded
Digital elevation model (DEM) resolution is critical to hydrological modeling and stream delineation. Changes in DEM resolution can result in differences in modeled stream delineation, which can result in differences in modeled stream spatial location, geometric complexity and statistical parameters. These differences in modeled stream spatial location, geometric complexity and statistical parameters can manifest differently between flat and mountainous terrains. Therefore, there is a need to analyze how stream delineation changes with DEM resolution, terrain and how stream geometric complexity changes with DEM resolution. Since spatial location of streams matters, it is necessary to evaluate the application of spatial location analysis of streams compared to traditional statistical testing at different resolutions of DEMs. A comprehensive sensitivity analysis using statistical tools for evaluating irregular geometric complexities or topographies was performed on multiple DEM resolutions, characterizing and comparing spatial and statistical differences between delineated modeled and real-world streams. Spatial and statistical results of modeled streams generated from original and resampled 1.5, 10, 30 and 90m DEMs were compared, measuring spatial and statistical consistency with real-world streams. Results showed in both flat and mountainous terrains, modeled streams generated from DEMs having identical cell size yielded statistically comparable length and complexity. However, while statistics indicated similarities, the streams did not share spatial correspondence in length or complexity. Mismatches between spatial and statistical results suggest limited utility in applying statistical complexity indices characterizing spatial changes in delineated streams, as statistical comparability may not guarantee spatial comparability. Spatial comparability can only be guaranteed using spatial analyses.